Technical Reference Guide Atoll

Atoll
RF Planning and Optimisation Software

Version 2.7.1
Technical
Reference
Guide
AT271_TRG_E6
Technical Reference Guide
Contact Information
Forsk (Head Office) 7 rue des Briquetiers
31700 Blagnac
France
Forsk (USA Office) 200 South Wacker Drive

Suite 3100
Chicago, IL 60606
USA
Forsk (China Office) Suite 302, 3/F, West Tower,

Jiadu Commercial Building,
No.66 Jianzhong Road,
Tianhe Hi-Tech Industrial Zone,
Guangzhou, 510665,
Peoples Republic of China
www.forsk.com
sales@forsk.com
helpdesk@forsk.com
+33 (0) 562 74 72 10
+33 (0) 562 74 72 25
+33 (0) 562 74 72 11
Web
Sales and pricing information
Technical support
General
Technical support
Fax
sales_us@forsk.com
support_us@forsk.com
+1 312 674 4846
+1 888 GoAtoll (+1 888 462 8655)
+1 312 674 4847
Sales and pricing information

Technical support
General
Technical support
Fax
www.forsk.com.cn
enquiries@forsk.com.cn
+86 20 8553 8938
+86 20 8553 8285
+86 10 6513 4559
Web
Information and enquiries
Telephone
Fax (Guangzhou)
Fax (Beijing)
Atoll 2.7.1 Technical Reference Guide Release AT271_TRG_E6
Copyright 1997 - 2009 by Forsk

The software described in this document is provided under a licence agreement. The software may only be used/copied
under the terms and conditions of the licence agreement. No part of this document may be copied, reproduced or
distributed in any form without prior authorisation from Forsk.
The product or brand names mentioned in this document are trademarks or registered trademarks of their respective
registering parties.
About Technical Reference Guide

This document is targeted at readers with a prior knowledge of Atoll, its operation and basic functioning. It is not the User
Manual for Atoll, and does not teach how to operate and use Atoll. It is a supplementary document containing detailed
descriptions of models, algorithms and concepts adopted in Atoll. Therefore, it concerns only the appropriate personnel.
Atoll Technical Reference Guide is divided into three parts with each part comprising similar topics. The first part contains
descriptions of general terms, entities, ideas and concepts in Atoll that are encountered throughout its use. It is followed
by the second part that consists of descriptions of entities common to all types of networks and the algorithms that are
technology independent and are available in any network type. Lastly, the guide provides detailed descriptions of each
basic type of network that can be modelled and studied in Atoll.
Forsk 2009
AT271_TRG_E6
iii
iv
AT271_TRG_E6
Forsk 2009
Table of Contents
Table of Contents
1
1.1
1.1.1
1.1.1.1
1.1.1.2
1.1.1.3
1.1.1.4
1.1.1.5
1.1.1.6
1.1.2
1.1.2.1
1.1.2.2
1.1.2.3
1.1.3
1.1.3.1
1.1.3.2
1.1.3.3
1.1.3.4
1.1.3.5
1.1.4
1.2
1.2.1
1.2.2
1.3
2
2.1
2.1.1
2.1.1.1
2.1.1.2
2.1.1.2.1
2.1.1.2.2
2.1.1.3
2.1.1.3.1
2.1.1.3.2
2.1.1.3.3
2.1.1.3.4
2.1.1.4
2.1.1.5
2.1.1.6
2.1.1.7
2.1.2
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.2.5
2.2.6
2.2.7
2.2.8
2.2.8.1
2.2.8.2
2.2.8.3
2.2.9
2.2.9.1
3
3.1
Forsk 2009
Coordinate Systems and Units ....................................................... 25

Coordinate Systems............................................................................................................................... 25
Description of Coordinate Systems .................................................................................................. 25
Geographic Coordinate System.................................................................................................. 25
Datum ......................................................................................................................................... 25
Meridian ...................................................................................................................................... 25
Ellipsoid ...................................................................................................................................... 25
Projection.................................................................................................................................... 26
Projection Coordinate System .................................................................................................... 26
Coordinate Systems in Atoll ............................................................................................................. 26
Projection Coordinate System .................................................................................................... 26
Display Coordinate System ........................................................................................................ 26
Internal Coordinate Systems ...................................................................................................... 26
File Formats ..................................................................................................................................... 27
Unit Codes .................................................................................................................................. 27
Datum Codes.............................................................................................................................. 28
Projection Method Codes ........................................................................................................... 29
Ellipsoid Codes ........................................................................................................................... 29
Projection Parameter Indices...................................................................................................... 30
Creating a Coordinate System ......................................................................................................... 30
Units ....................................................................................................................................................... 30
Power Units ...................................................................................................................................... 30
Length Units ..................................................................................................................................... 31
BSIC Format .......................................................................................................................................... 31
Geographic and Radio Data ........................................................... 35

Geographic Data .................................................................................................................................... 35
Data Type......................................................................................................................................... 35
Digital Terrain Model (DTM) ....................................................................................................... 35
Clutter (Land Use) ...................................................................................................................... 36
Clutter Classes...................................................................................................................... 36
Clutter Heights ...................................................................................................................... 36
Traffic Data ................................................................................................................................. 36
Environment Traffic Maps ..................................................................................................... 36
User Profile Traffic Maps....................................................................................................... 36
Live Traffic Maps................................................................................................................... 36
User Density Traffic Maps..................................................................................................... 37
Vector Data................................................................................................................................. 37
Scanned Images......................................................................................................................... 37
Population................................................................................................................................... 37
Other Geographic Data............................................................................................................... 37
Supported Geographic Data Formats .............................................................................................. 37
Radio Data ............................................................................................................................................. 38
Site ................................................................................................................................................... 38
Antenna ............................................................................................................................................ 38
Transmitter ....................................................................................................................................... 38
Repeater........................................................................................................................................... 38
Remote Antenna .............................................................................................................................. 39
Station .............................................................................................................................................. 39
Hexagonal Design ............................................................................................................................ 39
GSM GPRS EGPRS Documents ..................................................................................................... 39
TRX............................................................................................................................................. 39
Subcell ........................................................................................................................................ 39
Cell Type..................................................................................................................................... 39
All CDMA and WiMAX BWA Documents ......................................................................................... 39
Cell.............................................................................................................................................. 39
File Formats .................................................................................... 43

BIL Format ............................................................................................................................................. 43
AT271_TRG_E6
3.1.1
3.1.1.1
3.1.1.2
3.1.1.2.1
3.1.1.2.2
3.1.1.2.3
3.2
3.2.1
3.2.2
3.2.2.1
3.3
3.3.1
3.3.1.1
3.3.1.2
3.3.1.2.1
3.3.2
3.3.3
3.3.3.1
3.4
3.4.1
3.4.2
3.4.2.1
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.12.1
3.12.2
3.13
3.14
3.14.1
3.14.1.1
3.14.1.2
3.14.2
3.14.2.1
3.14.2.2
3.14.3
3.14.3.1
3.14.3.2
3.14.4
3.14.5
3.15
3.15.1
3.15.2
3.16
3.16.1
3.16.2
3.17
3.17.1
3.17.1.1
3.17.1.1.1
3.17.1.1.2
3.17.1.1.3
3.17.1.2
3.17.2
3.18
3.18.1
3.18.1.1
3.18.1.1.1
3.18.1.1.2
3.18.1.1.3
3.18.1.2
3.18.2
3.19
vi
HDR Header File...............................................................................................................................43

Description ..................................................................................................................................43
Samples ......................................................................................................................................44
Digital Terrain Model..............................................................................................................44
Clutter Classes File................................................................................................................44
BIL File...................................................................................................................................44
TIF Format ..............................................................................................................................................45
TFW Header File...............................................................................................................................45
Sample ..............................................................................................................................................46
Clutter Classes File .....................................................................................................................46
BMP Format............................................................................................................................................46
BMP File Description.........................................................................................................................46
BMP File Structure ......................................................................................................................46
BMP Raster Data Encoding ........................................................................................................48
Raster Data Compression Descriptions.................................................................................48
BPW/BMW Header File Description..................................................................................................49
Sample ..............................................................................................................................................49
Generic Raster Header File (.wld) ..........................................................................................................49
WLD File Description ........................................................................................................................49
Sample ..............................................................................................................................................49
DXF Format ............................................................................................................................................50
SHP Format ............................................................................................................................................50
MIF Format .............................................................................................................................................50
TAB Format ............................................................................................................................................50
ECW Format ...........................................................................................................................................51
Erdas Imagine Format ............................................................................................................................51
Planet EV/Vertical Mapper Geographic Data Format .............................................................................51
ArcView Grid Format ..............................................................................................................................52
ArcView Grid File Description ...........................................................................................................52
Sample ..............................................................................................................................................52
Other Supported Geographic Data File Formats ....................................................................................52
Planet Format .........................................................................................................................................52
DTM File............................................................................................................................................53
Description ..................................................................................................................................53
Sample ........................................................................................................................................53
Clutter Class Files .............................................................................................................................53
Description ..................................................................................................................................53
Sample ........................................................................................................................................54
Vector Files .......................................................................................................................................54
Description ..................................................................................................................................54
Sample ........................................................................................................................................55
Image Files........................................................................................................................................55
Text Data Files ..................................................................................................................................55
MNU Format ...........................................................................................................................................56
Description ........................................................................................................................................56
Sample ..............................................................................................................................................56
XML Table Export/Import Format ...........................................................................................................56
Index.xml File ....................................................................................................................................56
XML File ............................................................................................................................................57
Externalised Propagation Results Format ..............................................................................................58
DBF File ............................................................................................................................................58
DBF File Format ..........................................................................................................................58
DBF Structure ........................................................................................................................58
DBF Header (Variable Size - Depends on Field Count) ........................................................58
Each DBF Record (Fixed Length)..........................................................................................60
DBF File Content.........................................................................................................................60
LOS File ............................................................................................................................................61
Externalised Tuning Files .......................................................................................................................61
DBF File ............................................................................................................................................61
DBF File Format ..........................................................................................................................61
DBF Structure ........................................................................................................................62
DBF Header (Variable Size - Depends on Field Count) ........................................................62
Each DBF Record (Fixed Length)..........................................................................................63
DBF File Content.........................................................................................................................63
PTS File ............................................................................................................................................64
Interference Histograms File Formats ....................................................................................................64
AT271_TRG_E6
Forsk 2009
Table of Contents
3.19.1
3.19.1.1
3.19.2
3.19.2.1
3.19.2.1.1
3.19.2.1.2
3.19.2.2
3.19.2.2.1
3.19.2.2.2
3.19.3
3.19.3.1
3.19.4
3.19.4.1
4
4.1
4.2
4.2.1
4.2.1.1
4.2.1.2
4.2.2
4.2.2.1
4.2.2.2
4.2.3
4.3
4.3.1
4.3.2
4.3.2.1
4.3.2.2
4.3.3
4.3.3.1
4.3.3.1.1
4.3.3.1.2
4.3.3.2
4.4
4.4.1
4.4.1.1
4.4.1.2
4.4.1.3
4.4.2
4.4.2.1
4.4.2.2
4.4.2.2.1
4.4.2.2.2
4.4.2.2.3
4.4.2.3
4.4.3
4.4.3.1
4.4.3.2
4.4.3.2.1
4.4.3.2.2
4.4.3.2.3
4.4.3.2.4
4.4.3.2.5
4.4.3.2.6
4.4.3.2.7
4.4.3.3
4.4.3.3.1
4.4.3.3.2
4.4.3.4
4.4.4
4.4.4.1
4.4.4.2
4.4.4.2.1
4.4.4.2.2
4.4.5
4.4.5.1
4.4.5.2
Forsk 2009
One Histogram per Line (.im0) Format............................................................................................. 64

Sample........................................................................................................................................ 65
One Value per Line with Dictionary File (.clc) Format ...................................................................... 65
CLC File...................................................................................................................................... 66
Description ............................................................................................................................ 66
Sample .................................................................................................................................. 66
DCT File...................................................................................................................................... 67
Description ............................................................................................................................ 67
Sample .................................................................................................................................. 67
One Value per Line (Transmitter Name Repeated) (.im1) Format ................................................... 68
Sample........................................................................................................................................ 68
Only Co-Channel and Adjacent Values (.im2) Format ..................................................................... 69
Sample........................................................................................................................................ 69
Calculations .................................................................................... 73
Overview ................................................................................................................................................ 73
Path Loss Matrices................................................................................................................................. 74
Calculation Area Determination........................................................................................................ 75
Computation Zone ...................................................................................................................... 75
Use of Polygonal Zones in Coverage Prediction Reports........................................................... 75
Calculate / Force Calculation Comparison ....................................................................................... 76
Calculate..................................................................................................................................... 76
Force Calculation........................................................................................................................ 76
Matrix Validity ................................................................................................................................... 76
Path Loss Calculations........................................................................................................................... 77
Ground Altitude Determination ......................................................................................................... 77
Clutter Determination ....................................................................................................................... 78
Clutter Class ............................................................................................................................... 78
Clutter Height.............................................................................................................................. 78
Geographic Profile Extraction........................................................................................................... 78
Extraction Methods ..................................................................................................................... 78
Radial Extraction ................................................................................................................... 78
Systematic Extraction ........................................................................................................... 79
Profile Resolution: Multi-Resolution Management...................................................................... 80
Propagation Models ............................................................................................................................... 82
Okumura-Hata and Cost-Hata Propagation Models......................................................................... 83
Hata Path Loss Formula ............................................................................................................. 83
Corrections to the Hata Path Loss Formula................................................................................ 83
Calculations in Atoll .................................................................................................................... 83
ITU 529-3 Propagation Model .......................................................................................................... 84
ITU 529-3 Path Loss Formula..................................................................................................... 84
Corrections to the ITU 529-3 Path Loss Formula ....................................................................... 84
Environment Correction ........................................................................................................ 84
Area Size Correction ............................................................................................................. 84
Distance Correction .............................................................................................................. 85
Standard Propagation Model (SPM) ................................................................................................ 85
SPM Path Loss Formula............................................................................................................. 85
Visibility and Distance Between Transmitter and Receiver................................................... 86
Effective Transmitter Antenna Height ................................................................................... 86
Effective Receiver Antenna Height ....................................................................................... 89
Correction for Hilly Regions in Case of LOS ......................................................................... 89
Diffraction .............................................................................................................................. 90
Losses due to Clutter ............................................................................................................ 90
Recommendations ................................................................................................................ 91
Automatic SPM Calibration......................................................................................................... 91
General Algorithm ................................................................................................................. 92
Sample Values for SPM Path Loss Formula Parameters ..................................................... 92
Unmasked Path Loss Calculation............................................................................................... 93
WLL Propagation Model ................................................................................................................... 94
WLL Path Loss Formula ............................................................................................................. 94
Free Space Loss ................................................................................................................... 94
Diffraction .............................................................................................................................. 94
ITU-R P.526-5 Propagation Model ................................................................................................... 95
ITU 526-5 Path Loss Formula..................................................................................................... 95
AT271_TRG_E6
vii
4.4.5.2.1
4.4.5.2.2
4.4.6
4.4.6.1
4.4.6.2
4.4.6.2.1
4.4.6.2.2
4.4.7
4.4.7.1
4.4.7.2
4.4.7.3
4.4.8
4.4.8.1
4.4.8.1.1
4.4.8.1.2
4.4.8.1.3
4.4.8.1.4
4.4.8.1.5
4.4.8.1.6
4.4.9
4.4.10
4.4.10.1
4.4.10.2
4.4.10.2.1
4.4.10.2.2
4.4.10.2.3
4.4.10.2.4
4.4.10.2.5
4.5
4.6
4.6.1
4.6.2
4.6.3
4.6.4
4.6.4.1
4.7
4.7.1
4.7.1.1
4.7.1.2
4.7.2
4.7.2.1
4.7.2.1.1
4.7.2.1.2
4.7.2.2
4.7.2.2.1
4.7.2.2.2
4.8
4.8.1
4.8.1.1
4.8.1.2
4.8.1.3
4.8.2
5
5.1
5.1.1
5.1.2
5.1.2.1
5.1.2.2
5.1.3
5.1.3.1
5.1.3.1.1
5.1.3.1.2
5.1.3.1.3
5.1.3.1.4
5.1.3.1.5
5.1.3.1.6
viii
Free Space Loss....................................................................................................................95

Diffraction...............................................................................................................................95
ITU-R P.370-7 Propagation Model....................................................................................................95
ITU 370-7 Path Loss Formula .....................................................................................................95
Calculations in Atoll .....................................................................................................................95
Free Space Loss....................................................................................................................95
Corrected Standard Loss.......................................................................................................95
Erceg-Greenstein (SUI) Propagation Model .....................................................................................96
SUI Terrain Types .......................................................................................................................97
Erceg-Greenstein (SUI) Path Loss Formula................................................................................97
ITU-R P.1546-2 Propagation Model..................................................................................................98
Step 1: Determination of Graphs to be Used.........................................................................99
Step 2: Calculation of Maximum Field Strength.....................................................................99
Step 3: Determination of Transmitter Antenna Height ...........................................................99
Step 4: Interpolation/Extrapolation of Field Strength .............................................................99
Step 5: Calculation of Correction Factors ............................................................................101
Step 6: Calculation of Path Loss..........................................................................................102
Sakagami Extended Propagation Model.........................................................................................103
Appendices .....................................................................................................................................104
Free Space Loss .......................................................................................................................104
Diffraction Loss..........................................................................................................................104
Knife-Edge Diffraction..........................................................................................................104
3 Knife-Edge Deygout Method.............................................................................................105
Epstein-Peterson Method ....................................................................................................107
Deygout Method with Correction .........................................................................................107
Millington Method.................................................................................................................107
Path Loss Tuning..................................................................................................................................108
Antenna Attenuation Calculation ..........................................................................................................109
Calculation of Azimuth and Tilt Angles............................................................................................109
Antenna Pattern 3-D Interpolation...................................................................................................110
Additional Electrical Downtilt Modelling...........................................................................................111
Antenna Pattern Smoothing ............................................................................................................111
Smoothing Algorithm .................................................................................................................113
Shadowing Model .................................................................................................................................113
Shadowing Margin Calculation........................................................................................................116
Shadowing Margin Calculation in Predictions ...........................................................................116
Shadowing Margin Calculation in Monte-Carlo Simulations......................................................117
Macro-Diversity Gains Calculation ..................................................................................................118
Uplink Macro-Diversity Gain Evaluation ....................................................................................118
Shadowing Error PDF (n Signals)........................................................................................118
Uplink Macro-Diversity Gain ................................................................................................121
Downlink Macro-Diversity Gain Evaluation ...............................................................................121
Shadowing Error PDF (n Signals)........................................................................................121
Downlink Macro-Diversity Gain............................................................................................123
Appendices ...........................................................................................................................................124
Transmitter Radio Equipment .........................................................................................................124
UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, and
TD-SCDMA Documents124
GSM GPRS EGPRS Documents ..............................................................................................125
WiMAX 802.16d and WiMAX 802.16e Documents ...................................................................126
Secondary Antennas.......................................................................................................................127
GSM GPRS EGPRS Networks......................................................131

General Prediction Studies ...................................................................................................................131
Calculation Criteria..........................................................................................................................131
Point Analysis..................................................................................................................................131
Profile Tab .................................................................................................................................131
Reception Tab ...........................................................................................................................131
Coverage Studies............................................................................................................................132
Service Area Determination ......................................................................................................132
All Servers ...........................................................................................................................132
Best Signal Level and a Margin ...........................................................................................132
Second Best Signal Level and a Margin..............................................................................132
Best Signal Level per HCS Layer and a Margin ..................................................................133
HCS Servers and a Margin..................................................................................................133
Highest Priority HCS Server and a Margin ..........................................................................133
AT271_TRG_E6
Forsk 2009
Table of Contents
5.1.3.1.7
5.1.3.1.8
5.1.3.2
5.1.3.2.1
5.1.3.2.2
5.2
5.2.1
5.2.1.1
5.2.1.1.1
5.2.1.1.2
5.2.1.2
5.2.1.2.1
5.2.1.2.2
5.2.1.3
5.2.1.3.1
5.2.1.3.2
5.2.2
5.2.2.1
5.2.2.1.1
5.2.2.1.2
5.2.2.1.3
5.2.2.2
5.2.2.2.1
5.2.2.2.2
5.2.2.2.3
5.3
5.3.1
5.3.1.1
5.3.1.2
5.3.1.2.1
5.3.1.2.2
5.3.1.2.3
5.3.2
5.3.2.1
5.3.2.1.1
5.3.2.1.2
5.3.2.2
5.3.2.2.1
5.3.2.2.2
5.3.2.2.3
5.3.2.2.4
5.3.2.2.5
5.3.2.2.6
5.4
5.4.1
5.4.1.1
5.4.1.2
5.4.1.3
5.4.2
5.4.2.1
5.4.2.1.1
5.4.2.1.2
5.4.2.1.3
5.4.2.1.4
5.4.2.1.5
5.4.2.1.6
5.4.2.2
5.4.2.2.1
5.4.2.2.2
5.4.2.2.3
5.4.2.2.4
5.4.2.2.5
5.4.2.2.6
5.5
5.5.1
5.5.2
5.6
5.6.1
Forsk 2009
Second Best Signal Level per HCS Layer and a Margin .................................................... 134
Best Idle Mode Reselection Criterion (C2).......................................................................... 134
Coverage Display ..................................................................................................................... 135
Plot Resolution .................................................................................................................... 135
Display Types ..................................................................................................................... 135
Traffic Analysis..................................................................................................................................... 136
Traffic Distribution .......................................................................................................................... 136
Normal Cells (Nonconcentric, No HCS Layer).......................................................................... 136
Circuit Switched Services.................................................................................................... 136
Packet Switched Services................................................................................................... 136
Concentric Cells........................................................................................................................ 136
HCS Layers .............................................................................................................................. 136
Calculation of the Traffic Demand per Subcell ............................................................................... 137
Traffic Maps Based on Environments and User Profiles .......................................................... 137
Normal Cells (Nonconcentric, No HCS Layer) .................................................................... 137
Concentric Cells .................................................................................................................. 137
HCS Layers......................................................................................................................... 138
Traffic Maps Based on Transmitters and Services ................................................................... 142
Normal Cells (Nonconcentric, No HCS Layer) .................................................................... 142
Concentric Cells .................................................................................................................. 142
HCS Layers......................................................................................................................... 143
Network Dimensioning ......................................................................................................................... 146
Dimensioning Models and Quality Graphs ..................................................................................... 146
Circuit Switched Traffic ............................................................................................................. 146
Packet Switched Traffic ............................................................................................................ 146
Throughput.......................................................................................................................... 146
Delay ................................................................................................................................... 149
Blocking Probability............................................................................................................. 149
Network Dimensioning Process ..................................................................................................... 150
Network Dimensioning Engine.................................................................................................. 150
Inputs .................................................................................................................................. 150
Outputs ............................................................................................................................... 151
Network Dimensioning Steps.................................................................................................... 151
Step 1: Timeslots Required for CS Traffic........................................................................... 151
Step 2: TRXs Required for CS Traffic and Dedicated PS Timeslots................................... 151
Step 3: Effective CS Blocking, Effective CS Traffic Overflow and Served CS Traffic ......... 152
Step 4: TRXs to Add for PS Traffic ..................................................................................... 152
Step 5: Served PS Traffic ................................................................................................... 154
Step 6: Total Traffic Load.................................................................................................... 154
Key Performance Indicators Calculation .............................................................................................. 154
Circuit Switched Traffic................................................................................................................... 155
Erlang B .................................................................................................................................... 155
Erlang C.................................................................................................................................... 155
Served Circuit Switched Traffic................................................................................................. 155
Packet Switched Traffic .................................................................................................................. 155
Case 1: Total Traffic Demand > Dedicated + Shared Timeslots .............................................. 155
Traffic Load ......................................................................................................................... 155
Packet Switched Traffic Overflow ....................................................................................... 156
Throughput Reduction Factor ............................................................................................. 156
Delay ................................................................................................................................... 156
Served Packet Switched Traffic .......................................................................................... 156
Case 2: Total Traffic Demand < Dedicated + Shared Timeslots .............................................. 156
Traffic Load ......................................................................................................................... 156
Packet Switched Traffic Overflow ....................................................................................... 156
Throughput Reduction Factor ............................................................................................. 156
Delay ................................................................................................................................... 156
Served Packet Switched Traffic .......................................................................................... 157
Neighbour Allocation ............................................................................................................................ 157
Global Allocation for All Transmitters ............................................................................................. 157
Allocation for a Group of Transmitters or One Transmitter ............................................................ 160
Interference Prediction Studies ............................................................................................................ 160
Coverage Studies ........................................................................................................................... 160
AT271_TRG_E6
ix
5.6.1.1
5.6.1.1.1
5.6.1.1.2
5.6.1.1.3
5.6.1.1.4
5.6.1.2
5.6.1.2.1
5.6.1.2.2
5.6.1.2.3
5.6.1.2.4
5.6.1.3
5.6.1.3.1
5.6.1.3.2
5.6.1.4
5.6.1.4.1
5.6.1.4.2
5.6.1.4.3
5.6.1.4.4
5.6.2
5.7
5.7.1
5.7.1.1
5.7.1.2
5.7.1.3
5.7.1.4
5.7.1.5
5.7.2
5.7.2.1
5.7.2.2
5.7.2.3
5.7.2.4
5.7.3
5.7.3.1
5.7.3.1.1
5.7.3.1.2
5.7.3.2
5.7.3.2.1
5.7.3.2.2
5.7.3.3
5.7.3.3.1
5.7.3.3.2
5.7.3.4
5.7.3.4.1
5.7.3.4.2
5.7.3.5
5.7.3.5.1
5.7.3.5.2
5.7.3.6
5.7.3.6.1
5.7.3.6.2
5.7.4
5.7.4.1
5.7.4.1.1
5.7.4.1.2
5.7.4.2
5.7.4.2.1
5.7.4.2.2
5.7.4.2.3
5.7.4.2.4
5.7.4.2.5
5.7.4.2.6
5.7.4.2.7
5.8
5.8.1
5.8.2
5.8.2.1
5.8.2.2
5.8.2.3
Service Area Determination ......................................................................................................161

All Servers ...........................................................................................................................161
Best Signal Level per HCS Layer and a Margin ..................................................................161
Best Signal Level of the Highest Priority HCS Layer and a Margin .....................................161
Second Best Signal Level per HCS Layer and a Margin .....................................................161
Carrier to Interference Ratio Calculation ...................................................................................161
Carrier Power Level .............................................................................................................162
Interference Calculation.......................................................................................................163
Collision Probability for Non Hopping Mode ........................................................................164
Collision Probability for BBH and SFH Modes.....................................................................164
Coverage Area Determination...................................................................................................165
Interference Condition Satisfied by At Least One TRX........................................................165
Interference Condition Satisfied by The Worst TRX ............................................................165
Coverage Area Display .............................................................................................................165
C/I Level...............................................................................................................................165
Max C/I Level.......................................................................................................................165
Min C/I Level........................................................................................................................165
Transmitter...........................................................................................................................165
Point Analysis..................................................................................................................................166
GPRS EGPRS Coverage Studies ........................................................................................................166
Coverage Area Determination.........................................................................................................166
All Servers .................................................................................................................................166
Best Signal Level per HCS Layer and a Margin ........................................................................166
Second Best Signal Level per HCS Layer and a Margin...........................................................166
HCS Servers and a Margin .......................................................................................................167
Highest Priority HCS Server and a Margin ................................................................................167
Calculation Options .........................................................................................................................167
Calculations Based on C ...........................................................................................................167
Calculations Based on C/I Without Considering Thermal Noise ...............................................167
Calculations Based on C/I Considering Thermal Noise ............................................................168
Ideal Link Adaptation (ILA) ........................................................................................................168
Coverage Study Scenarios .............................................................................................................168
GPRS/EGPRS Studies Based on C Without ILA ......................................................................168
Coding Scheme Based on C Without ILA............................................................................168
Throughput Based on C Without ILA ...................................................................................168
GPRS/EGPRS Studies Based on C With ILA ...........................................................................169
Coding Scheme Based on C With ILA.................................................................................169
Throughput Based on C With ILA ........................................................................................169
GPRS/EGPRS Studies Based on C/I Without ILA and Thermal Noise .....................................169
Coding Scheme Based on C/I Without ILA and Thermal Noise ..........................................169
Throughput Based on Worst Case Between C and C/I Without ILA....................................170
GPRS/EGPRS Studies Based on C/I With ILA and Without Thermal Noise.............................170
Coding Scheme Based on C/I With ILA and Without Thermal Noise ..................................170
Throughput Based on Worst Case Between C and C/I With ILA.........................................170
GPRS/EGPRS Studies Based on C/I Without ILA and With Thermal Noise.............................170
Coding Scheme Based on C/I Without ILA and With Thermal Noise ..................................170
Throughput Based on Interpolation Between C/N and C/(I+N) Without ILA ........................171
GPRS/EGPRS Studies Based on C/I With ILA and Thermal Noise..........................................171
Coding Scheme Based on C/I With ILA and Thermal Noise ...............................................171
Throughput Based on Interpolation Between C/N and C/(I+N) With ILA .............................172
Coverage Display............................................................................................................................172
GPRS/EGPRS Coding Schemes Study Display Types ............................................................172
Coding Schemes .................................................................................................................172
Best Coding Schemes .........................................................................................................173
RLC/MAC and Application Throughput/Timeslot Studies Display Types ..................................173
Relation Between RLC/MAC and Application Throughputs.................................................173
Throughput/Timeslot............................................................................................................173
Best Throughput/Timeslot....................................................................................................173
Average Throughput/Timeslot .............................................................................................173
Block Error Rate Computation .............................................................................................173
BLER Percentage ................................................................................................................173
Maximum BLER Percentage ...............................................................................................174
Circuit Quality Indicators Studies..........................................................................................................174
Circuit Quality Indicators .................................................................................................................174
Coverage Area Determination.........................................................................................................175
All Servers .................................................................................................................................176
Best Signal Level per HCS Layer and a Margin ........................................................................176
Second Best Signal Level per HCS Layer and a Margin...........................................................176
AT271_TRG_E6
Forsk 2009
Table of Contents
5.8.2.4
5.8.2.5
5.8.3
5.8.3.1
5.8.3.2
5.8.3.3
5.8.4
5.8.4.1
5.8.4.2
5.8.4.3
5.8.4.4
5.8.5
5.8.5.1
5.8.5.1.1
5.8.5.1.2
6
6.1
6.1.1
6.1.2
6.1.2.1
6.1.2.2
6.1.3
6.1.3.1
6.1.3.1.1
6.1.3.1.2
6.1.3.1.3
6.1.3.2
6.1.3.2.1
6.1.3.2.2
6.2
6.2.1
6.2.2
6.2.3
6.2.4
6.3
6.4
6.4.1
6.4.1.1
6.4.1.1.1
6.4.1.1.2
6.4.1.2
6.4.1.2.1
6.4.1.2.2
6.4.2
6.4.2.1
6.4.2.2
6.4.2.3
6.4.2.3.1
6.4.2.3.2
6.4.2.3.3
6.4.2.3.4
6.4.2.3.5
6.4.2.4
6.4.2.4.1
6.4.2.4.2
6.4.2.4.3
6.4.2.5
6.4.3
6.4.3.1
6.4.3.2
6.4.3.2.1
6.4.3.2.2
6.4.3.2.3
6.4.3.2.4
6.4.4
6.4.4.1
6.4.4.2
Forsk 2009
HCS Servers and a Margin....................................................................................................... 176

Highest Priority HCS Server and a Margin ............................................................................... 176
Calculation Options ........................................................................................................................ 176
Calculations Based on C/N....................................................................................................... 177
Calculations Based on C/(I+N) ................................................................................................. 177
Ideal Link Adaptation (ILA) ....................................................................................................... 178
Calculation Scenarios..................................................................................................................... 178
CQI Study Based on C/N Without ILA ...................................................................................... 178
CQI Study Based on C/N With ILA ........................................................................................... 178
CQI Study Based on C/(I+N) Without ILA................................................................................. 178
CQI Study Based on C/(I+N) With ILA...................................................................................... 179
Coverage Display ........................................................................................................................... 179
Circuit Quality Indicators Study Display Types ......................................................................... 179
FER/BER/MOS ................................................................................................................... 179
Max FER/Max BER/Max MOS ............................................................................................ 179
UMTS HSPA Networks ................................................................. 183

General Prediction Studies .................................................................................................................. 183
Calculation Criteria ......................................................................................................................... 183
Point Analysis ................................................................................................................................. 183
Profile Tab ................................................................................................................................ 183
Reception Tab .......................................................................................................................... 183
Service Area Determination...................................................................................................... 184
All Servers........................................................................................................................... 184
Best Signal Level and a Margin .......................................................................................... 184
Second Best Signal Level and a Margin ............................................................................. 184
Plot Resolution .................................................................................................................... 185
Display Types ..................................................................................................................... 185
Definitions and Formulas ..................................................................................................................... 186
Inputs.............................................................................................................................................. 186
Ec/Io Calculation ............................................................................................................................ 190
DL Eb/Nt Calculation ...................................................................................................................... 191
UL Eb/Nt Calculation ...................................................................................................................... 192
Active Set Management ....................................................................................................................... 192
Simulations........................................................................................................................................... 193
Generating a Realistic User Distribution ........................................................................................ 193
Simulations Based on Raster Traffic and Vector Traffic Maps ................................................. 193
Circuit Switched Service (i) ................................................................................................. 194
Packet Switched Service (j) ................................................................................................ 194
Simulations Based on Traffic Map per Service and per Transmitter ........................................ 197
Power Control Simulation............................................................................................................... 199
Algorithm Initialization............................................................................................................... 200
R99 Part of the Algorithm ......................................................................................................... 200
HSDPA Part of the Algorithm.................................................................................................... 207
HSDPA Power Allocation .................................................................................................... 207
Number of HS-SCCH Channels and Maximum Number of HSDPA Users......................... 207
HSDPA Bearer Allocation Process ..................................................................................... 208
Fast Link Adaptation Modelling ........................................................................................... 209
Scheduling Algorithms ........................................................................................................ 215
HSUPA Part of the Algorithm.................................................................................................... 216
HSUPA Users ..................................................................................................................... 217
Admission Control ............................................................................................................... 217
HSUPA Bearer Allocation Process ..................................................................................... 218
Convergence Criteria................................................................................................................ 220
Results ........................................................................................................................................... 220
R99 Related Results................................................................................................................. 220
HSPA Related Results.............................................................................................................. 222
Statistics Tab ...................................................................................................................... 222
Mobiles Tab ........................................................................................................................ 222
Cells Tab ............................................................................................................................. 224
Sites Tab ............................................................................................................................. 225
Appendices..................................................................................................................................... 226
Admission Control in the R99 Part............................................................................................ 226
Resources Management........................................................................................................... 226
AT271_TRG_E6
xi
6.4.4.2.1
6.4.4.2.2
6.4.4.3
6.4.4.3.1
6.4.4.3.2
6.4.4.4
6.4.4.5
6.5
6.5.1
6.5.1.1
6.5.1.1.1
6.5.1.1.2
6.5.1.1.3
6.5.2
6.5.2.1
6.5.2.1.1
6.5.2.1.2
6.5.2.2
6.5.2.2.1
6.5.2.2.2
6.5.2.3
6.5.2.3.1
6.5.2.3.2
6.5.2.4
6.5.2.4.1
6.5.2.4.2
6.5.2.4.3
6.5.2.5
6.5.2.5.1
6.5.2.5.2
6.5.2.6
6.5.2.6.1
6.5.2.6.2
6.5.2.6.3
6.6
6.6.1
6.6.2
6.7
6.7.1
6.7.1.1
6.7.1.2
6.7.1.2.1
6.7.1.2.2
6.7.1.3
6.7.1.3.1
6.7.1.3.2
6.7.1.3.3
6.7.2
6.7.2.1
6.7.2.1.1
6.7.2.1.2
6.7.2.1.3
6.7.2.1.4
6.7.2.2
6.8
6.8.1
6.8.2
6.8.2.1
6.8.2.2
6.8.2.3
6.8.2.3.1
6.8.2.3.2
7
7.1
7.1.1
7.1.2
7.1.2.1
xii
OVSF Codes Management..................................................................................................226

Channel Elements Management .........................................................................................227
Downlink Load Factor Calculation .............................................................................................228
Downlink Load Factor per Cell.............................................................................................228
Downlink Load Factor per Mobile ........................................................................................229
Uplink Load Factor Due to One User ........................................................................................229
Inter-carrier Power Sharing Modelling .......................................................................................231
UMTS HSPA Prediction Studies...........................................................................................................232
Point Analysis..................................................................................................................................232
AS Analysis Tab ........................................................................................................................232
Bar Graph and Pilot Sub-Menu............................................................................................232
Downlink Sub-Menu.............................................................................................................234
Uplink Sub-Menu .................................................................................................................238
Coverage Studies............................................................................................................................241
Pilot Reception Analysis ............................................................................................................242
Prediction Study Inputs........................................................................................................242
Study Display Options .........................................................................................................242
Downlink Service Area Analysis ................................................................................................243
Uplink Service Area Analysis ....................................................................................................245
Downlink Total Noise Analysis ..................................................................................................246
Study Inputs.........................................................................................................................247
Analysis on All Carriers........................................................................................................247
Analysis on a Specific Carrier..............................................................................................247
HSDPA Prediction Study ...........................................................................................................248
HSUPA Prediction Study ...........................................................................................................252
Calculation Options..............................................................................................................252
Display Options....................................................................................................................252
Automatic Neighbour Allocation............................................................................................................254
Global Allocation for All Transmitters ..............................................................................................254
Allocation for a Group of Transmitters or One Transmitter .............................................................258
Primary Scrambling Code Allocation ....................................................................................................258
Automatic Allocation Description.....................................................................................................259
Options and Constraints ............................................................................................................259
Allocation Process.....................................................................................................................260
Single Carrier Network.........................................................................................................260
Multi-Carrier Network...........................................................................................................261
Priority Determination ................................................................................................................261
Cell Priority ..........................................................................................................................261
Transmitter Priority ..............................................................................................................264
Site Priority ..........................................................................................................................264
Allocation Examples........................................................................................................................264
Allocation Strategies and Use a Maximum of Codes ................................................................264
Strategy: Clustered ..............................................................................................................265
Strategy: Distributed ............................................................................................................265
Strategy: One Cluster per Site ............................................................................................266
Strategy: Distributed per Site ..............................................................................................266
Allocate Carriers Identically .......................................................................................................266
Automatic GSM-UMTS Neighbour Allocation .......................................................................................267
Overview .........................................................................................................................................267
Algorithm Based on Distance ....................................................................................................268
Algorithm Based on Coverage Overlapping ..............................................................................268
Appendices................................................................................................................................270
Delete Existing Neighbours Option......................................................................................270
Calculation of Inter-Transmitter Distance ............................................................................270
IS-95 cdmaOne and CDMA2000 Networks ...................................273

General Prediction Studies ...................................................................................................................273
Calculation Criteria..........................................................................................................................273
Point Analysis..................................................................................................................................273
Profile Tab .................................................................................................................................273
AT271_TRG_E6
Forsk 2009
Table of Contents
7.1.2.2
7.1.3
7.1.3.1
7.1.3.1.1
7.1.3.1.2
7.1.3.1.3
7.1.3.2
7.1.3.2.1
7.1.3.2.2
7.2
7.2.1
7.2.1.1
7.2.1.2
7.2.1.3
7.2.1.4
7.2.1.5
7.2.2
7.2.2.1
7.2.2.2
7.2.2.3
7.2.2.4
7.3
7.4
7.4.1
7.4.1.1
7.4.1.1.1
7.4.1.1.2
7.4.1.2
7.4.1.3
7.4.2
7.4.2.1
7.4.2.1.1
7.4.2.1.2
7.4.2.1.3
7.4.2.2
7.4.2.2.1
7.4.2.2.2
7.4.2.2.3
7.4.2.3
7.4.2.3.1
7.4.2.3.2
7.4.2.3.3
7.4.3
7.4.3.1
7.4.3.2
7.4.3.2.1
7.4.3.2.2
7.4.3.3
7.4.3.3.1
7.4.3.3.2
7.5
7.5.1
7.5.1.1
7.5.1.2
7.5.1.2.1
7.5.1.2.2
7.5.1.2.3
7.5.1.3
7.5.1.3.1
7.5.1.3.2
7.5.1.3.3
7.5.2
7.5.2.1
7.5.2.1.1
7.5.2.1.2
7.5.2.2
7.5.2.2.1
7.5.2.2.2
Forsk 2009
Reception Tab .......................................................................................................................... 274

Service Area Determination...................................................................................................... 274
All Servers........................................................................................................................... 274
Best Signal Level and a Margin .......................................................................................... 274
Second Best Signal Level and a Margin ............................................................................. 274
Plot Resolution .................................................................................................................... 275
Display Types ..................................................................................................................... 275
Definitions and Formulas ..................................................................................................................... 276
Parameters Used for IS-95 cdmaOne and CDMA2000 1xRTT Modelling ..................................... 276
Inputs ........................................................................................................................................ 276
Ec/I0 Calculation....................................................................................................................... 280
DL Eb/Nt Calculation ................................................................................................................ 280
UL Eb/Nt Calculation ................................................................................................................ 281
Simulation Results .................................................................................................................... 282
Parameters Used for CDMA2000 1xEV-DO Modelling .................................................................. 283
Inputs ........................................................................................................................................ 283
Ec/I0 and Ec/Nt Calculations .................................................................................................... 287
UL Eb/Nt Calculation ................................................................................................................ 287
Simulation Results .................................................................................................................... 288
Active Set Management ....................................................................................................................... 289
Simulations........................................................................................................................................... 290
Generating a Realistic User Distribution ........................................................................................ 290
Number of Users, User Activity Status and User Data Rate..................................................... 290
Simulations Based on Raster Traffic and Vector Traffic Maps............................................ 290
Simulations Based on Traffic Maps per Service and per Transmitter ................................. 293
Transition Flags for 1xEV-DO Rev.0 User Data Rates............................................................. 297
User Geographical Position ...................................................................................................... 297
Network Regulation Mechanism..................................................................................................... 297
IS-95 cdmaOne Power Control Simulation Algorithm............................................................... 297
Algorithm Initialization ......................................................................................................... 298
Presentation of the Algorithm.............................................................................................. 298
Convergence Criterion ........................................................................................................ 302
CDMA2000 1xRTT Power Control Simulation Algorithm.......................................................... 302
CDMA2000 1xEV-DO Power/Data Rate Control Simulation Algorithm .................................... 310
Appendices..................................................................................................................................... 316
Admission Control..................................................................................................................... 316
Resources Management........................................................................................................... 317
Walsh Code Management .................................................................................................. 317
Channel Element Management .......................................................................................... 317
Downlink Load Factor Calculation ............................................................................................ 318
Downlink Load Factor per Cell ............................................................................................ 318
Downlink Load Factor per Mobile........................................................................................ 319
IS-95 cdmaOne and CDMA2000 Prediction Studies ........................................................................... 319
Point Analysis: The AS Analysis Tab ............................................................................................. 319
Bar Graph and Pilot Sub-Menu................................................................................................. 320
Downlink Sub-Menu.................................................................................................................. 321
IS-95 cdmaOne ................................................................................................................... 321
CDMA2000 1xRTT.............................................................................................................. 324
CDMA2000 1xEV-DO ......................................................................................................... 327
Uplink Sub-Menu ...................................................................................................................... 328
IS-95 cdmaOne ................................................................................................................... 328
CDMA2000 1xRTT.............................................................................................................. 331
CDMA2000 1xEV-DO ......................................................................................................... 335
Pilot Reception Analysis ........................................................................................................... 338
1st Case: Analysis Based on all Carriers ............................................................................ 338
2nd Case: Analysis Based on a Specific Carrier................................................................. 338
Downlink Service Area Analysis ............................................................................................... 339
IS-95 cdmaOne and CDMA2000 1xRTT............................................................................. 339
CDMA2000 1xEV-DO ......................................................................................................... 340
AT271_TRG_E6
xiii
7.5.2.3
7.5.2.3.1
7.5.2.3.2
7.5.2.4
7.5.2.4.1
7.5.2.4.2
7.6
7.6.1
7.6.2
7.7
7.7.1
7.7.1.1
7.7.1.2
7.7.1.2.1
7.7.1.2.2
7.7.1.2.3
7.7.1.3
7.7.1.3.1
7.7.1.3.2
7.7.1.3.3
7.7.2
7.7.2.1
7.7.2.2
7.7.2.3
7.8
7.8.1
7.8.2
7.8.2.1
7.8.2.2
7.8.2.3
7.9
7.9.1
7.9.2
7.9.2.1
8
8.1
8.1.1
8.1.2
8.1.3
8.1.4
8.1.5
8.1.6
8.1.7
8.1.8
8.2
8.2.1
8.2.1.1
8.2.1.2
8.2.2
8.2.2.1
8.2.2.2
8.2.2.2.1
8.2.2.2.2
8.2.2.3
8.2.2.4
8.2.2.5
8.2.2.5.1
8.2.2.5.2
8.2.2.6
8.2.2.6.1
8.2.2.6.2
8.2.2.7
8.2.2.7.1
8.2.2.7.2
8.2.2.8
8.3
8.3.1
xiv
Uplink Service Area Analysis ....................................................................................................342

IS-95 cdmaOne and CDMA2000 1xRTT .............................................................................342
CDMA2000 1xEV-DO ..........................................................................................................343
Downlink Total Noise Analysis ..................................................................................................345
Analysis on all Carriers ........................................................................................................346
Analysis on a Specific Carrier..............................................................................................346
Automatic Neighbour Allocation............................................................................................................346
Global Allocation for all Transmitters ..............................................................................................347
Allocation for a Group of Transmitters or One Transmitter .............................................................350
PN Offset Allocation..............................................................................................................................350
Options and Constraints ............................................................................................................351
Allocation Process.....................................................................................................................352
Single Carrier Network.........................................................................................................352
Multi-Carrier Network...........................................................................................................353
Difference between Adjacent and Distributed PN-Clusters .................................................353
Priority Determination ................................................................................................................353
Cell Priority ..........................................................................................................................353
Transmitter Priority ..............................................................................................................355
Site Priority ..........................................................................................................................355
Allocation Examples........................................................................................................................356
Strategy: PN Offset per Cell ......................................................................................................356
Strategy: Adjacent PN-Clusters Per Site...................................................................................357
Strategy: Distributed PN-Clusters Per Site................................................................................357
Automatic GSM-CDMA Neighbour Allocation.......................................................................................357
Overview .........................................................................................................................................357
Delete Existing Neighbours Option ...........................................................................................360
Automatic cdmaOne-CDMA2000 Neighbour Allocation .......................................................................360
Overview .........................................................................................................................................360
Delete Existing Neighbours Option ...........................................................................................362
TD-SCDMA Networks....................................................................365
Definitions and Formulas ......................................................................................................................365
Inputs ..............................................................................................................................................365
P-CCPCH Eb/Nt and C/I Calculation ..............................................................................................369
DwPCH C/I Calculation ...................................................................................................................369
DL TCH Eb/Nt and C/I Calculation..................................................................................................370
UL TCH Eb/Nt and C/I Calculation..................................................................................................370
Interference Calculation ..................................................................................................................371
HSDPA Dynamic Power Calculations .............................................................................................371
Smart Antenna Modelling................................................................................................................371
Signal Level Based Calculations ..........................................................................................................372
Point Analysis..................................................................................................................................372
Profile Tab .................................................................................................................................373
Reception Tab ...........................................................................................................................373
RSCP Based Coverage Predictions................................................................................................373
Calculation Criteria ....................................................................................................................373
P-CCPCH RSCP Coverage Prediction .....................................................................................374
Coverage Condition .............................................................................................................374
Coverage Display ................................................................................................................374
Best Server P-CCPCH Coverage Prediction.............................................................................374
P-CCPCH Pollution Coverage Prediction..................................................................................375
DwPCH RSCP Coverage Prediction .........................................................................................375
UpPCH RSCP Coverage Prediction..........................................................................................376
Baton Handover Coverage Prediction .......................................................................................376
Scrambling Code Interference Analysis ....................................................................................377
Monte Carlo Simulations.......................................................................................................................377
Generating a Realistic User Distribution .........................................................................................377
AT271_TRG_E6
Forsk 2009
Table of Contents
8.3.1.1
8.3.1.1.1
8.3.1.1.2
8.3.1.2
8.3.1.2.1
8.3.1.2.2
8.3.2
8.3.2.1
8.3.2.2
8.3.2.2.1
8.3.2.2.2
8.3.2.2.3
8.3.2.2.4
8.3.2.2.5
8.3.2.2.6
8.3.2.2.7
8.3.2.3
8.3.2.3.1
8.3.2.3.2
8.3.2.3.3
8.3.2.3.4
8.3.2.3.5
8.3.2.4
8.4
8.4.1
8.4.2
8.4.3
8.4.4
8.4.5
8.4.6
8.4.7
8.4.8
8.4.9
8.4.10
8.4.11
8.5
8.5.1
8.5.1.1
8.5.1.2
8.5.1.3
8.5.1.4
8.5.1.4.1
8.5.1.4.2
8.5.1.5
8.5.2
8.5.3
8.6
8.6.1
8.7
8.7.1
8.7.2
8.8
8.8.1
8.8.1.1
8.8.1.2
8.8.1.3
8.8.1.3.1
8.8.1.3.2
8.8.1.4
8.8.1.4.1
8.8.1.4.2
8.8.1.4.3
8.8.2
8.8.2.1
8.8.2.1.1
8.8.2.1.2
8.8.2.1.3
8.8.2.1.4
Forsk 2009
Simulations Based on Raster and Vector Traffic Maps ............................................................ 378

Simulations Based on Traffic Map Based on Transmitters and Services ................................. 381
Power Control Simulation............................................................................................................... 383
Algorithm Initialisation............................................................................................................... 383
R99 Part of the Algorithm ......................................................................................................... 384
Determination of Mis Best Server (SBS(Mi))...................................................................... 384
Dynamic Channel Allocation ............................................................................................... 384
Uplink Power Control .......................................................................................................... 386
Downlink Power Control...................................................................................................... 387
Uplink Signals Update......................................................................................................... 389
Downlink Signals Update .................................................................................................... 390
Control of Radio Resource Limits (Downlink Traffic Power and Uplink Load) .................... 390
HSDPA Part of the Algorithm.................................................................................................... 390
HSDPA Power Allocation .................................................................................................... 390
Connection Status and Number of HSDPA Users .............................................................. 392
HSDPA Admission Control.................................................................................................. 392
HSDPA Dynamic Channel Allocation.................................................................................. 393
Ressource Unit Saturation .................................................................................................. 393
Convergence Criteria................................................................................................................ 393
TD-SCDMA Prediction Studies ............................................................................................................ 394
P-CCPCH Reception Analysis (Eb/Nt) or (C/I) ............................................................................... 394
DwPCH Reception Analysis (C/I) ................................................................................................... 395
Downlink TCH RSCP Coverage ..................................................................................................... 397
Uplink TCH RSCP Coverage ......................................................................................................... 398
Downlink Total Noise...................................................................................................................... 398
Downlink Service Area (Eb/Nt) or (C/I)........................................................................................... 399
Uplink Service Area (Eb/Nt) or (C/I) ............................................................................................... 401
Effective Service Area (Eb/Nt) or (C/I) ........................................................................................... 402
Cell to Cell Interference .................................................................................................................. 403
UpPCH Interference ....................................................................................................................... 404
HSDPA Coverage .......................................................................................................................... 404
Smart Antenna Modelling..................................................................................................................... 405
Modelling in Simulations................................................................................................................. 405
Grid of Beams Modelling .......................................................................................................... 405
Adaptive Beam Modelling ......................................................................................................... 406
Statistical Modelling .................................................................................................................. 407
Optimum Beamformer Model.................................................................................................... 408
Downlink ............................................................................................................................. 408
Uplink .................................................................................................................................. 409
3rd Party Smart Antenna Modelling.......................................................................................... 411
Construction of the Geographic Distributions ................................................................................. 411
Modelling in Coverage Predictions ................................................................................................. 412
N-Frequency Mode and Carrier-Type Allocation.................................................................................. 413
Automatic Carrier-Type Allocation.................................................................................................. 413
Neighbour Allocation ............................................................................................................................ 413
Global Allocation for All Transmitters ............................................................................................. 414
Allocation for a Group of Transmitters or One Transmitter ............................................................ 417
Scrambling Code Allocation ................................................................................................................. 417
Automatic Allocation Description .................................................................................................... 418
Allocation Constraints and Options........................................................................................... 418
Allocation Strategies ................................................................................................................. 418
Allocation Process .................................................................................................................... 419
Single Carrier Network ........................................................................................................ 419
Multi-Carrier Network .......................................................................................................... 420
Priority Determination ............................................................................................................... 420
Cell Priority.......................................................................................................................... 420
Transmitter Priority.............................................................................................................. 423
Site Priority.......................................................................................................................... 423
IScrambling Code Allocation Example ........................................................................................... 423
Single Carrier Network.............................................................................................................. 423
Strategy: Clustered ............................................................................................................. 424
Strategy: Distributed per Cell .............................................................................................. 424
Strategy: One SYNC_DL Code per Site ............................................................................. 425
Strategy: Distributed per Site .............................................................................................. 425
AT271_TRG_E6
xv
8.8.2.2
8.9
8.9.1
8.9.1.1
8.9.1.2
8.9.1.3
8.9.1.3.1
8.9.1.3.2
9
9.1
9.1.1
9.1.2
9.1.3
9.1.3.1
9.1.3.2
9.1.3.3
9.1.3.4
9.1.3.5
9.1.4
9.1.4.1
9.1.4.2
9.1.4.3
9.1.4.4
9.1.4.5
9.1.4.6
9.1.4.7
9.1.4.8
9.1.4.9
9.1.4.10
9.1.5
9.1.5.1
9.1.5.2
9.1.5.3
9.1.6
9.1.6.1
9.1.7
9.2
9.2.1
9.2.1.1
9.2.1.2
9.2.2
9.2.2.1
9.2.2.1.1
9.2.2.1.2
9.2.2.1.3
9.2.2.2
9.2.2.2.1
9.2.2.2.2
9.2.3
9.2.3.1
9.2.3.2
9.2.3.3
9.2.3.3.1
9.2.3.3.2
9.2.3.3.3
9.2.4
9.2.5
9.2.5.1
9.2.5.1.1
9.2.5.1.2
9.2.5.2
9.2.6
9.2.6.1
9.2.6.2
9.2.6.3
9.2.6.3.1
9.2.6.3.2
xvi
Multi Carrier Network.................................................................................................................425

Automatic GSM/TD-SCDMA Neighbour Allocation ..............................................................................426
Appendices................................................................................................................................428
Delete Existing Neighbours Option......................................................................................428
Calculation of Inter-Transmitter Distance ............................................................................429
WiMAX BWA Networks..................................................................433

Definitions and Formulas ......................................................................................................................433
Input ................................................................................................................................................433
Co- and Adjacent Channel Overlaps Calculation............................................................................436
Preamble Signal Quality Calculations .............................................................................................437
Preamble Signal Level Calculation............................................................................................437
Preamble Noise Calculation ......................................................................................................438
Preamble Interference Calculation ............................................................................................438
Preamble C/N Calculation .........................................................................................................438
Preamble C/(I+N) Calculation....................................................................................................438
Traffic and Pilot Signal Quality Calculations ...................................................................................438
Traffic and Pilot Signal Level Calculation (DL) ..........................................................................438
Noise Calculation (DL) ..............................................................................................................439
Traffic and Pilot Interference Calculation (DL) ..........................................................................439
Traffic and Pilot C/N Calculation (DL) .......................................................................................440
Traffic and Pilot C/(I+N) Calculation (DL) ..................................................................................440
Traffic Signal Level Calculation (UL) .........................................................................................441
Noise Calculation (UL) ..............................................................................................................441
Traffic Interference Calculation (UL) .........................................................................................441
Traffic C/N Calculation (UL) ......................................................................................................441
Traffic C/(I+N) Calculation (UL) .................................................................................................442
Channel Throughput Calculation.....................................................................................................442
Calculation of Cell Capacity (Total Resources) .........................................................................442
Bearer Determination ................................................................................................................443
Channel Throughput Calculation...............................................................................................443
Scheduling and Radio Resource Management...............................................................................443
User Throughput Calculation.....................................................................................................445
Optimum Beamformer Model ..........................................................................................................445
Calculation Processes ..........................................................................................................................446
Point Analysis..................................................................................................................................446
Profile Tab .................................................................................................................................447
Reception Tab ...........................................................................................................................447
Preamble Signal Level Coverage Predictions.................................................................................447
All Servers ...........................................................................................................................447
Best Signal Level and a Margin ...........................................................................................447
Second Best Signal Level and a Margin..............................................................................448
Coverage Display ......................................................................................................................448
Coverage Resolution ...........................................................................................................448
Display Types ......................................................................................................................448
Signal Quality Analysis Coverage Predictions ................................................................................449
Coverage Parameter Calculation ..............................................................................................450
Signal Quality Analysis (DL) Display Types.........................................................................450
Signal Quality Analysis (UL) Display Types.........................................................................451
Calculations on Subscriber Lists .....................................................................................................451
Monte Carlo Simulations .................................................................................................................452
Generating a Realistic User Distribution ...................................................................................452
Simulations Based on Raster Traffic Maps, Vector Traffic Maps, and Subscriber Lists......452
Simulations Based on Traffic Map Based on Transmitters and Services ............................454
Simulation Process....................................................................................................................455
C/(I+N)-Based Coverage Predictions..............................................................................................459
Coverage Parameter Calculation ..............................................................................................460
Coverage by C/(I+N) Level (DL) Display Types ..................................................................460
AT271_TRG_E6
Forsk 2009
Table of Contents
9.2.6.3.3
9.2.6.3.4
9.2.6.3.5
9.2.6.3.6
9.2.6.3.7
9.3
9.3.1
9.3.1.1
9.3.1.2
9.3.1.3
9.3.1.4
9.3.1.5
9.3.2
9.3.2.1
9.3.2.2
9.3.2.3
9.3.2.4
9.3.2.5
9.3.3
9.3.4
9.3.5
9.3.6
9.3.6.1
9.3.6.2
9.3.6.3
9.3.6.3.1
9.3.6.3.2
9.3.6.4
9.3.6.5
9.3.6.6
9.3.6.7
9.3.6.8
9.3.6.8.1
9.3.6.8.2
9.3.6.9
9.3.6.10
9.3.7
9.3.7.1
9.3.7.1.1
9.3.7.1.2
9.3.7.1.3
9.3.7.1.4
9.3.7.2
9.3.7.3
9.3.8
9.3.8.1
9.3.8.2
9.3.9
9.3.9.1
9.3.9.2
9.3.9.3
9.4
9.4.1
9.4.2
9.4.2.1
9.4.2.2
9.4.2.3
9.5
10
10.1
10.1.1
10.1.1.1
10.1.1.1.1
10.1.1.1.2
10.1.1.1.3
10.1.1.2
10.1.1.2.1
Forsk 2009
Coverage by Best Bearer (DL) Display Types .................................................................... 461

Coverage by Channel Throughput (DL) Display Types ...................................................... 461
Coverage by C/(I+N) Level (UL) Display Types.................................................................. 461
Coverage by Best Bearer (UL) Display Types .................................................................... 462
Coverage by Channel Throughput (UL) Display Types ...................................................... 462
Calculation Algorithms ......................................................................................................................... 462
Co- and Adjacent Channel Overlaps Calculation ........................................................................... 462
Conversion From Channel Numbers to Start and End Frequencies ........................................ 463
Co-Channel Overlap Calculation .............................................................................................. 464
Adjacent Channel Overlap Calculation ..................................................................................... 464
FDD TDD Overlap Ratio Calculation ..................................................................................... 465
Total Overlap Ratio Calculation ................................................................................................ 465
Preamble Signal Quality Calculations ............................................................................................ 466
Preamble Signal Level Calculation ........................................................................................... 466
Preamble Noise Calculation ..................................................................................................... 467
Preamble Interference Calculation ........................................................................................... 469
Preamble C/N Calculation ........................................................................................................ 470
Preamble C/(I+N) Calculation ................................................................................................... 470
Best Server Determination ............................................................................................................. 471
Service Area Calculation ................................................................................................................ 471
Permutation Zone Selection (WiMAX 802.16e).............................................................................. 471
Traffic and Pilot Signal Quality Calculations................................................................................... 472
Traffic and Pilot Signal Level Calculation (DL) ......................................................................... 473
Noise Calculation (DL).............................................................................................................. 474
Traffic and Pilot Interference Calculation (DL).......................................................................... 475
Traffic and Pilot Interference Signal Levels Calculation (DL).............................................. 475
Effective Traffic and Pilot Interference Calculation (DL) ..................................................... 478
Traffic and Pilot C/N Calculation (DL)....................................................................................... 482
Traffic C/(I+N) Calculation (DL) ................................................................................................ 483
Traffic Signal Level Calculation (UL) ........................................................................................ 484
Noise Calculation (UL).............................................................................................................. 485
Traffic Interference Calculation (UL)......................................................................................... 486
Traffic Interference Signal Levels Calculation (UL)............................................................. 486
Noise Rise Calculation (UL) ................................................................................................ 487
Traffic C/N Calculation (UL)...................................................................................................... 487
Traffic C/(I+N) Calculation (UL) ................................................................................................ 490
Channel Throughput Calculation.................................................................................................... 493
Calculation of Cell Capacity (Total Resources) ........................................................................ 493
Calculation of Sampling Frequency .................................................................................... 493
Calculation of Symbol Duration........................................................................................... 493
Calculation of Cell Capacity - TDD Networks...................................................................... 494
Calculation of Cell Capacity - FDD Networks...................................................................... 495
Bearer Determination................................................................................................................ 496
Channel Throughput Calculation .............................................................................................. 496
Scheduling and Radio Resource Management .............................................................................. 498
Scheduling and Radio Resource Allocation.............................................................................. 499
User Throughput Calculation .................................................................................................... 504
Optimum Beamformer Model ......................................................................................................... 505
Introduction ............................................................................................................................... 505
Downlink ................................................................................................................................... 506
Uplink........................................................................................................................................ 507
Automatic Allocation Algorithms........................................................................................................... 509
Automatic Neighbour Allocation ..................................................................................................... 509
Automatic Preamble Index Allocation............................................................................................. 511
Constraint and Relationship Weights........................................................................................ 512
Calculation of Cost Between TBA and Related Cells ............................................................... 512
Automatic Allocation Algorithm ................................................................................................. 514
Glossary of WiMAX Terms................................................................................................................... 514
Repeaters and Remote Antennas................................................. 519

Modelling Repeaters ............................................................................................................................ 519
CDMA Documents.......................................................................................................................... 519
Over the Air............................................................................................................................... 519
Signal Level Received From Repeaters.............................................................................. 519
Gain Automatic Calculation................................................................................................. 520
Donor Side Parameter Automatic Calculation..................................................................... 521
Microwave Link ......................................................................................................................... 522
Signal Level Received From Repeaters.............................................................................. 522
AT271_TRG_E6
xvii
10.1.1.2.2
10.1.1.3
10.1.1.3.1
10.1.1.3.2
10.1.1.4
10.1.1.4.1
10.1.1.4.2
10.1.1.4.3
10.1.2
10.1.2.1
10.1.2.1.1
10.1.2.1.2
10.1.2.1.3
10.1.2.2
10.1.2.2.1
10.1.2.2.2
10.1.2.3
10.1.2.3.1
10.1.2.3.2
10.1.2.4
10.1.2.4.1
10.2
10.2.1
10.2.1.1
10.2.1.2
10.2.2
10.2.2.1
10.2.2.2
xviii
Gain Automatic Calculation .................................................................................................523

Fibre Link...................................................................................................................................524
Signal Level Received From Repeaters ..............................................................................524
Gain Automatic Calculation .................................................................................................524
Appendices................................................................................................................................525
Automatic Controls ..............................................................................................................525
Carrier Power and Interference Calculation.........................................................................526
Consideration of Repeater Noise Figure .............................................................................528
GSM Documents.............................................................................................................................528
Over the Air ...............................................................................................................................528
EIRP Automatic Calculation.................................................................................................529
Donor Side Parameter Automatic Calculation .....................................................................530
Microwave Link..........................................................................................................................531
Fibre Link...................................................................................................................................532
Appendices................................................................................................................................533
Automatic Controls ..............................................................................................................533
Modelling Remote Antennas.................................................................................................................534
CDMA Documents ..........................................................................................................................534
Signal Level Received From Repeaters ....................................................................................534
Gain Automatic Calculation .......................................................................................................534
GSM Documents.............................................................................................................................535
Signal Level Received From Repeaters ....................................................................................535
EIRP Automatic Calculation ......................................................................................................536
AT271_TRG_E6
Forsk 2009
List of Figures
List of Figures
Figure 2.1:
Figure 2.2:
Figure 2.3:
Figure 4.1:
Figure 4.2:
Figure 4.3:
Figure 4.4:
Figure 4.5:
Figure 4.6:
Figure 4.7:
Figure 4.8:
Figure 4.9:
Figure 4.10:
Figure 4.11:
Figure 4.12:
Figure 4.13:
Figure 4.14:
Figure 4.15:
Figure 4.16:
Figure 4.17:
Figure 4.18:
Figure 4.19:
Figure 4.20:
Figure 4.21:
Figure 4.22:
Figure 4.23:
Figure 4.24:
Figure 4.25:
Figure 4.26:
Figure 4.27:
Figure 4.28:
Figure 5.1:
Figure 5.2:
Figure 5.3:
Figure 5.4:
Figure 5.5:
Figure 5.6:
Figure 5.7:
Figure 5.8:
Figure 5.9:
Figure 5.10:
Figure 5.11:
Figure 5.12:
Figure 5.13:
Figure 6.1:
Figure 6.2:
Figure 6.3:
Figure 6.4:
Figure 6.5:
Figure 6.6:
Figure 6.7:
Figure 6.8:
Figure 6.9:
Figure 6.10:
Figure 6.11:
Forsk 2009
Digital Terrain Model.................................................................................................................................. 35

Schematic view of a DTM file .................................................................................................................... 35
Clutter Classes .......................................................................................................................................... 36
Example 1: Single Calculation Area .......................................................................................................... 75
Example 2: Multiple Calculation Areas ...................................................................................................... 75
Ground Altitude Determination - 1 ............................................................................................................. 77
Clutter Height............................................................................................................................................. 78
Radial calculation method.......................................................................................................................... 79
Site-bin centre profile................................................................................................................................. 79
Radial calculation method.......................................................................................................................... 80
Enhanced Slope at Receiver ..................................................................................................................... 87
Losses due to Clutter................................................................................................................................. 90
Tx-Rx profile .............................................................................................................................................. 91
Knife-Edge Diffraction.............................................................................................................................. 105
Deygout Construction 1 Obstacle ......................................................................................................... 106
Deygout Construction 3 Obstacles ....................................................................................................... 106
Epstein-Peterson Construction ................................................................................................................ 107
Millington Construction ............................................................................................................................ 108
Azimuth and Tilt Computation.................................................................................................................. 109
Vertical Pattern Transformation due to Electrical Downtilt....................................................................... 111
Vertical Antenna Pattern.......................................................................................................................... 112
Peaks and Nulls in the Antenna Pattern .................................................................................................. 112
Log-normal Probability Density Function ................................................................................................. 113
Normalised Margin .................................................................................................................................. 117
Margin - Probability (Case of 2 Signals) .................................................................................................. 123
Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 1dB) ............................................... 123
Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 2dB) ............................................... 123
Reference Point - Location of the Transmission/Reception parameters ................................................. 124
Representation of a Concentric Cell TXi.................................................................................................. 138
Representation of Micro and Macro Layers............................................................................................. 139
Concentric Cells....................................................................................................................................... 140
Concentric Cells....................................................................................................................................... 144
Reduction of Throughput per Timeslot .................................................................................................... 147
Reduction Factor for Different Packet Switched Traffic Loads (Lp, X-axis)............................................. 148
Blocking Probability for Different Packet Switched Traffic Loads (Lp, X-axis)......................................... 150
Network Dimensioning Process............................................................................................................... 150
Minimum Throughput Reduction Factor .................................................................................................. 154
Overlapping Zones .................................................................................................................................. 158
FER vs. C/I Graphs.................................................................................................................................. 175
BER vs. C/I Graphs ................................................................................................................................. 175
MOS vs. C/I Graphs................................................................................................................................. 175
Description of a Packet Session .............................................................................................................. 195
UMTS HSPA Power Control Algorithm.................................................................................................... 200
Connection status of HSDPA users......................................................................................................... 208
Bearer Allocation Process for Scheduled Users...................................................................................... 208
HSDPA Radio Bearers Table .................................................................................................................. 212
HSDPA UE Categories Table .................................................................................................................. 212
HSUPA UE Categories Table .................................................................................................................. 217
HSUPA Radio Bearers Table .................................................................................................................. 218
HSUPA Bearer SelectionTable................................................................................................................ 218
OVSF Code Tree Indices (Not OVSF Code Numbers)............................................................................ 227
Overlapping Zone for Intra-carrier Neighbours ........................................................................................ 255
AT271_TRG_E6
xix
Figure 6.12:
Figure 6.13:
Figure 6.14:
Figure 6.15:
Figure 6.16:
Figure 7.1:
Figure 7.2:
Figure 7.3:
Figure 7.4:
Figure 7.5:
Figure 7.6:
Figure 7.7:
Figure 7.8:
Figure 8.1:
Figure 8.2:
Figure 8.3:
Figure 8.4:
Figure 8.5:
Figure 8.6:
Figure 8.7:
Figure 8.8:
Figure 8.9:
Figure 8.10:
Figure 8.11:
Figure 8.12:
Figure 8.13:
Figure 8.14:
Figure 8.15:
Figure 8.16:
Figure 8.17:
Figure 9.1:
Figure 9.2:
Figure 9.3:
Figure 9.4:
Figure 9.5:
Figure 9.6:
Figure 9.7:
Figure 9.8:
Figure 9.9:
Figure 9.10:
Figure 9.11:
Figure 9.12:
Figure 9.13:
Figure 9.14:
Figure 9.15:
Figure 9.16:
Figure 9.17:
Figure 9.18:
Figure 10.1:
Figure 10.2:
Figure 10.3:
Figure 10.4:
Figure 10.5:
Figure 10.6:
Figure 10.7:
Figure 10.8:
Figure 10.9:
Figure 10.10:
Figure 10.11:
Figure 10.12:
Figure 10.13:
Figure 10.14:
xx
Overlapping Zone for Inter-carrier Neighbours - 1st Case .......................................................................

Overlapping Zone for Inter-carrier Neighbours - 2nd Case ......................................................................
Neighbourhood Constraints......................................................................................................................
Primary Scrambling Codes Allocation ......................................................................................................
Inter-Transmitter Distance Computation ..................................................................................................
IS-95 cdmaOne Power Control Algorithm ................................................................................................
CDMA2000 1xRTT Power Control Algorithm ...........................................................................................
CDMA2000 1xEVDO Power Control Algorithm ........................................................................................
Walsh Code Tree Indices (Not Walsh Code Numbers) ............................................................................
Overlapping Zones - 1st Case..................................................................................................................
Overlapping Zones - 2nd Case ................................................................................................................
PN Offset Allocation .................................................................................................................................
Description of a Packet Session...............................................................................................................
TD-SCDMA Power Control Algorithm ......................................................................................................
Grid Of Beams Modelling .........................................................................................................................
GOB Modelling - Determination of the Best Beam...................................................................................
Adaptive Beam Modelling - Determination of the Best Beam ..................................................................
Linear Adaptive Array System ..................................................................................................................
Downlink Beamforming ............................................................................................................................
Uplink Adaptive Algorithm ........................................................................................................................
Construction of the Geographic Distribution of Downlink Traffic Power ...................................................
Geographic Distribution of Downlink Traffic Power ..................................................................................
Geographic Distribution of downlink traffic power and uplink load ...........................................................
Intra-carrier Neighbours ...........................................................................................................................
Overlapping Coverages............................................................................................................................
Scrambling Code Allocation Example ......................................................................................................
Scrambling Code Allocation to All Carriers ..............................................................................................
Inter-Transmitter Distance Computation ..................................................................................................
WiMAX Simulation Algorithm ...................................................................................................................
Victim and Interfering Mobiles ..................................................................................................................
Simulation Convergence Stability Factor .................................................................................................
Co-Channel and Adjacent Channel Overlaps ..........................................................................................
Downlink C/(I+N) calculation in Simulations.............................................................................................
Downlink C/(I+N) calculation in Coverage Predictions .............................................................................
Segmentation ...........................................................................................................................................
Segmentation Interference Scenarios ......................................................................................................
Uplink Subchannelization in WiMAX 802.16e Networks (Example) .........................................................
Uplink Subchannelization in WiMAX 802.16e Networks (Example) .........................................................
Linear Adaptive Array System ..................................................................................................................
Downlink Beamforming ............................................................................................................................
Uplink Adaptive Algorithm ........................................................................................................................
Determination of Adjacent Cells ...............................................................................................................
Overlapping Zones ...................................................................................................................................
Weighted Distance Between Cells ...........................................................................................................
Importance Based on Distance Relation ..................................................................................................
WiMAX Frame ..........................................................................................................................................
CDMA Documents - Over the Air Repeater .............................................................................................
Over the Air Repeater - Downlink Total Gain ...........................................................................................
Over the Air Repeater - Uplink Total Gain................................................................................................
Angle from North (Azimuth) ......................................................................................................................
Positive/Negative Mechanical Downtilt.....................................................................................................
Tilt Angle Computation .............................................................................................................................
CDMA Documents - Microwave Link Repeater ........................................................................................
Microwave Link Repeater - Downlink Total Gain .....................................................................................
Microwave Link Repeater - Uplink Total Gain ..........................................................................................
CDMA Documents - Fibre Link Repeater .................................................................................................
Fibre Link Repeater - Downlink Total Gain ..............................................................................................
Fibre Link Repeater - Uplink Total Gain ...................................................................................................
GSM Documents - Over the Air Repeater ................................................................................................
Over the Air Repeater - EIRP...................................................................................................................
AT271_TRG_E6
256
256
262
264
270
298
303
310
317
348
348
354
356
379
383
406
406
407
408
408
410
412
412
413
415
415
421
423
426
429
456
457
457
463
477
478
479
481
489
492
505
506
507
509
510
513
514
515
519
520
521
521
522
522
522
523
523
524
525
525
529
529
Forsk 2009
List of Figures
Figure 10.15:
Figure 10.16:
Figure 10.17:
Figure 10.18:
Figure 10.19:
Figure 10.20:
Figure 10.21:
Figure 10.22:
Figure 10.23:
Figure 10.24:
Figure 10.25:
Figure 10.26:
Forsk 2009
Angle from North (Azimuth) .....................................................................................................................

Positive/Negative Mechanical Downtilt ....................................................................................................
Tilt Angle Computation ............................................................................................................................
GSM Documents - Microwave Link Repeater..........................................................................................
Microwave Link Repeater - EIRP.............................................................................................................
GSM Documents - Fibre Link Repeater...................................................................................................
Fibre Link Repeater - EIRP......................................................................................................................
CDMA Documents - Remote Antenna Signal Level ................................................................................
Remote Antennas - Downlink Total Gain.................................................................................................
Remote Antennas - Uplink Total Gain .....................................................................................................
GSM Documents - Remote Antenna Signal Level...................................................................................
Remote Antennas - EIRP ........................................................................................................................
AT271_TRG_E6
530
530
530
531
532
532
533
534
534
535
535
536
xxi
xxii
AT271_TRG_E6
Forsk 2009
Chapter 1
Coordinate Systems and Units
This chapter presents the different coordinate systems available in Atoll by default. It describes the projection,
display, and internal coordinate systems, and describes the format of the coordinate systems files. This chapter
also provides details of the different power and length units available in Atoll.
Atoll
Atoll
Microwave
RF PlanningLink
andPlanning
Optimisation
Software
Microwave
Software
24
AT271_TRG_E6
Forsk 2009
Chapter 1: Coordinate Systems and Units
Coordinate Systems and Units
1.1
Coordinate Systems
A map or a geo-spatial database is a flat representation of data collected from a curved surface. A projection is a means
for producing all or part of a spheroid on a flat sheet. This projection cannot be done without distortion. Therefore, the
cartographer must choose the characteristic (distance, direction, scale, area, or shape) that he wants to be shown accurately at the expense of the other characteristics, or compromise on several characteristics [1-3]. The projected zones are
referenced using cartographic coordinates (meter, yard, etc.). Two projection methods are widely used:
The Lambert Conformal-Conic Method: A portion of the earth is mathematically projected on a cone conceptually secant at one or two standard parallels. This projection method is useful for representing countries or regions
that have a predominant east-west expanse.
The Universal Transverse Mercator (UTM) Method: A portion of the earth is mathematically projected on a cylinder tangent to a meridian (which is transverse or crosswise to the equator). This projection method is useful for
mapping large areas that are oriented north-south.
The geographic system is not a projection. It is only a representation of a location on the surface of the earth in geographic
coordinates (degree-minute-second, grade) giving the latitude and longitude in relation to the meridian origin (e.g., Paris
for NTF system and Greenwich for ED50 system). The locations in the geographic system can be converted into other
projections.
References:
[1] Snyder, John. P., Map Projections Used by the US Geological Survey, 2nd Edition, United States Government
Printing Office, Washington, D.C., 313 pages, 1982.
[2] http://www.colorado.edu/geography/gcraft/notes/gps/gps_f.html
[3] http://www.posc.org/Epicentre.2_2/DataModel/ExamplesofUsage/eu_cs34.html
[4] http://www.ign.fr/telechargement/Pi/SERVICES/transfo.pdf (Document in French)
1.1.1
Description of Coordinate Systems

A Geographic coordinate system is a latitude and longitude coordinate system. The latitude and longitude are related to
an ellipsoid, a geodetic datum, and a prime meridian. The geodetic datum provides the position and orientation of the ellipsoid relative to the earth.
Cartographic coordinate systems are obtained by transforming each (latitude, longitude) value into an (easting, northing)
value. A projection coordinate system is obtained by transforming each (latitude, longitude) value into an (easting, northing) value. Projection coordinate systems are geographic coordinate systems that provide longitude and latitude, and the
transformation method characterised by a set of parameters. Different methods may require different sets of parameters.
For example, the parameters required for Transverse Mercator coordinate systems are:
The longitude of the natural origin (central meridian)

The latitude of the natural origin
The False Easting value
The False Northing value
A scaling factor at the natural origin (central meridian)
Basic definitions are presented below.
1.1.1.1
Geographic Coordinate System

The geographic coordinate system is a datum and a meridian. Atoll enables you to choose the most suitable geographic
coordinate system for your geographic data.
1.1.1.2
Datum
The datum consists of the ellipsoid and its position relative to the WGS84 ellipsoid. In addition to the ellipsoid, translation,
rotation, and distortion parameters define the datum.
1.1.1.3
Meridian
The standard meridian is Greenwich, but some geographic coordinate systems are based on other meridians. These
meridians are defined by the longitude with respect to Greenwich.
1.1.1.4
Ellipsoid
The ellipsoid is the pattern used to model the earth. It is defined by its geometric parameters.
Forsk 2009
AT271_TRG_E6
25
1.1.1.5
Projection
The projection is the transformation applied to project the ellipsoid of the earth on to a plane. There are different projection
methods that use specific sets of parameters.
1.1.1.6
Projection Coordinate System

The projection coordinate system is the result of the application of a projection to a geographic coordinate system. It associates a geographic coordinate system and a projection. Atoll enables you to choose the projection coordinate system
matching your geographic data.
1.1.2
Coordinate Systems in Atoll

Depending on the working environment, there can be either two or four coordinate systems used in Atoll. If you are working
with stand-alone documents, i.e., documents not connected to databases, there are two coordinate systems used in Atoll:
Projection coordinate system

Display coordinate system
If you are working in a multi-user environment, Atoll uses four coordinate systems:
1.1.2.1
Projection coordinate system for the Atoll document

Display coordinate system for the Atoll document
Internal projection coordinate system for the database
Internal display coordinate system for the database
Projection Coordinate System

The projection coordinate system is the coordinate system of the available raster geographic data files. You should set the
projection coordinate system of your Atoll document so that it corresponds to the coordinate system of the available raster
geographic data. You can set the projection coordinate system of your document in the Options dialog.
All the raster geographic data files that you want to import and use in an Atoll document must have the same coordinate
system. You cannot work with raster geographic data files with different coordinate systems in the same document.
Note:
If you import vector geographic data (e.g., traffic, measurements, etc.) with different
coordinate systems, it is possible to convert the coordinate systems of these data into the
projection coordinate system of your Atoll document.
The projection coordinate system is used to keep the coordinates of sites (radio network data) consistent with the
geographic data.
When you import a raster geographic data file, Atoll reads the geo-referencing information from the file (or from its header
file, depending on the geographic data file format), i.e., its Northwest pixel, to determine the coordinates of each pixel. Atoll
does not use any coordinate system during the import process. However, the geo-referencing information of geographic
data files are considered to be provided in the projection coordinate system of the document.
1.1.2.2
Display Coordinate System

The display coordinate system is the coordinate system used for the display, e.g., in dialogs, in the Map window rulers, in
the status bar, etc. The coordinates of each pixel of geographic data are converted to the display coordinate system from
the projection coordinate system for display. The display coordinate system is also used for sites (radio network data). You
can set the display coordinate system of your document in the Options dialog.
If you import sites data, the coordinate system of the sites must correspond to the display coordinate system of your Atoll
document.
If you change the display coordinate system in a document which is not connected to a database, the coordinates of all
the sites are converted to the new display system.
Note:
1.1.2.3
If the coordinate systems of all your geographic data files and sites (radio network data) are
the same, you do not have to define the projection and display coordinate systems
separately. By default, the two coordinate systems are the same.
Internal Coordinate Systems

The internal coordinate systems are the projection and the display coordinate systems stored in a database. The projection
and display coordinate systems set by the administrator in the central Atoll project are stored in the database when the
database is created, and cannot be modified by users. Only the administrator can modify the internal coordinate systems
manually by editing the entries in the CoordSys and the Units tables. All Atoll documents opened from a database will have
the internal coordinate systems of the database as their default projection and display coordinate systems.
When exporting an Atoll project to a database, the currently chosen display coordinate system becomes the internal
display coordinate system for the database, and the currently chosen projection coordinate system becomes the internal
projection coordinate system for the database.
26
AT271_TRG_E6
Forsk 2009

Although Atoll stores both the coordinate systems in the database, i.e., the projection and the display coordinate systems,
the only relevant coordinate system for the database is the internal display coordinate system because this coordinate
system is the one used for the coordinates of sites (radio network data).
Users working on documents connected to a database can modify the coordinate systems in their documents locally, and
save these changes in their documents, but they cannot modify the coordinate systems stored in the database.
If you change the display coordinate system in a document which is not connected to a database, the coordinates of all
the sites are converted to the new display system.
If you change the display coordinate system in a document which is connected to a database, the coordinates of all the
sites are converted to the new coordinate system in the Atoll document locally but not in the database because the internal
coordinate systems cannot be changed.
Atoll uses the internal coordinates systems in order to keep the site coordinates consistent in the database which is usually
accessed by a large number of users in a multi-user environment.
1.1.3
File Formats
The Coordsystems folder located in the Atoll installation directory contains all the coordinate systems, both geographic
and cartographic, offered in the tool. Coordinate systems are grouped by regions. A catalogue per region and a "Favourites" catalogue are available in Atoll. The Favourites catalogue is initially empty and can be filled by the user by adding
coordinate systems to it. Each catalogue is described by an ASCII text file with .cs extension. In a .cs file, each coordinate
system is described in one line. The line syntax for describing a coordinate system is:
Code = "Name of the system"; Unit Code; Datum Code; Projection Method Code,
Projection Parameters; "Comments"
Examples:
4230 = "ED50"; 101; 230; 1; "Europe - west"

32045 = "NAD27 / Vermont"; 2; 267; 6, -72.5, 42.5, 500000, 0, 0.9999643; "United
States - Vermont"
You should keep the following points in mind when editing or creating .cs files:
The identification code enables Atoll to differentiate coordinates systems. In case you create a new coordinate
system, its code must be an integer value higher than 32767.
When describing a new datum, you must enter the ellipsoid code and parameters instead of the datum code in
brackets. There can be 3 to 7 parameters defined in the following order: Dx, Dy, Dz, Rx, Ry, Rz, S. The syntax of
the line in the .cs file will be:
Code = "Name of the system"; Unit Code; {Ellipsoid Code, Dx, Dy, Dz, Rx, Ry,
Rz, S}; Projection Method Code, Projection Parameters; "Comments"
There can be up to seven projection parameters. These parameters must be ordered according to the parameter
index (see "Projection Parameter Indices" on page 30). Parameter with index 0 is the first one. Projection parameters are delimited by commas.
For UTM projections, you must provide positive UTM zone numbers for north UTM zones and negative numbers
for south UTM zones.
You can add all other information as comments (such as usage or region).
Codes of units, data, projection methods, and ellipsoids, and projection parameter indices are listed in the tables below.
1.1.3.1
Unit Codes
Code
Forsk 2009
Cartographic Units
Code
Geographic Units
Metre
100
Radian
Kilometre
101
Degree
Foot
102
Grad
Link
103
ArcMinute
Chain
104
ArcSecond
Yard
Nautical mile
Mile
-1
Unspecified
-1
Unspecified
AT271_TRG_E6
27
1.1.3.2
28
Datum Codes
Code
Datum
Code
Datum
121
Greek Geodetic Reference System 1987
260
Manoca
125
Samboja
261
Merchich
126
Lithuania 1994
262
Massawa
130
Moznet (ITRF94)
263
Minna
131
Indian 1960
265
Monte Mario
201
Adindan
266
M'poraloko
202
Australian Geodetic Datum 1966
267
North American Datum 1927
203
Australian Geodetic Datum 1984
268
NAD Michigan
204
Ain el Abd 1970
269
North American Datum 1983
205
Afgooye
270
Nahrwan 1967
206
Agadez
271
Naparima 1972
207
Lisbon
272
New Zealand Geodetic Datum 1949
208
Aratu
273
NGO 1948
209
Arc 1950
274
Datum 73
210
Arc 1960
275
Nouvelle Triangulation Franaise
211
Batavia
276
NSWC 9Z-2
212
Barbados
277
OSGB 1936
213
Beduaram
278
OSGB 1970 (SN)
214
Beijing 1954
279
OS (SN) 1980
215
Reseau National Belge 1950
280
Padang 1884
216
Bermuda 1957
281
Palestine 1923
217
Bern 1898
282
Pointe Noire
218
Bogota
283
Geocentric Datum of Australia 1994
219
Bukit Rimpah
284
Pulkovo 1942
221
Campo Inchauspe
285
Qatar
222
Cape
286
Qatar 1948
223
Carthage
287
Qornoq
224
Chua
288
Loma Quintana
225
Corrego Alegre
289
Amersfoort
226
Cote d'Ivoire
290
RT38
227
Deir ez Zor
291
South American Datum 1969
228
Douala
292
Sapper Hill 1943
229
Egypt 1907
293
Schwarzeck
230
European Datum 1950
294
Segora
231
European Datum 1987
295
Serindung
232
Fahud
296
Sudan
233
Gandajika 1970
297
Tananarive 1925
234
Garoua
298
Timbalai 1948
235
Guyane Francaise
299
TM65
236
Hu Tzu Shan
300
TM75
237
Hungarian Datum 1972
301
Tokyo
238
Indonesian Datum 1974
302
Trinidad 1903
239
Indian 1954
303
Trucial Coast 1948
240
Indian 1975
304
Voirol 1875
241
Jamaica 1875
305
Voirol Unifie 1960
242
Jamaica 1969
306
Bern 1938
243
Kalianpur
307
Nord Sahara 1959
244
Kandawala
308
Stockholm 1938
245
Kertau
309
Yacare
247
La Canoa
310
Yoff
248
Provisional South American Datum 1956
311
Zanderij
AT271_TRG_E6
Forsk 2009

249
1.1.3.3
1.1.3.4
Forsk 2009
Lake
312
Militar-Geographische Institut
250
Leigon
313
Reseau National Belge 1972
251
Liberia 1964
314
Deutsche Hauptdreiecksnetz
252
Lome
315
Conakry 1905
253
Luzon 1911
322
WGS 72
254
Hito XVIII 1963
326
WGS 84
255
Herat North
901
Ancienne Triangulation Franaise
256
Mahe 1971
902
Nord de Guerre
903
NAD 1927 Guatemala/Honduras/Salvador

(Panama Zone)
257
Makassar
258
European Reference System 1989
Projection Method Codes

Code
Projection Method
Code
Projection Method
Undefined
Oblique Stereographic
No projection > Longitude / Latitude
New Zealand Map Grid
Lambert Conformal Conical 1SP
10
Hotine Oblique Mercator
Lambert Conformal Conical 2SP
11
Laborde Oblique Mercator
Mercator
12
Swiss Oblique Cylindrical
Cassini-Soldner
13
Oblique Mercator
Transverse Mercator
14
UTM Projection
Transverse Mercator South Oriented
Ellipsoid Codes
Code
Name
Major Axis
Minor Axis
Airy 1830
6377563.396
6356256.90890985
Airy Modified 1849
6377340.189
6356034.44761111
Australian National Spheroid
6378160
6356774.71919531
Bessel 1841
6377397.155
6356078.96261866
Bessel Modified
6377492.018
6356173.50851316
Bessel Namibia
6377483.865
6356165.38276679
Clarke 1858
6378293.63924683
6356617.98173817
Clarke 1866
6378206.4
6356583.8
Clarke 1866 Michigan
6378693.7040359
6357069.45104614
10
Clarke 1880 (Benoit)
6378300.79
6356566.43
11
Clarke 1880 (IGN)
6378249.2
6356515
12
Clarke 1880 (RGS)
6378249.145
6356514.86954978
13
Clarke 1880 (Arc)
6378249.145
6356514.96656909
14
Clarke 1880 (SGA 1922)
6378249.2
6356514.99694178
15
Everest 1830 (1937 Adjustment)
6377276.345
6356075.41314024
16
Everest 1830 (1967 Definition)
6377298.556
6356097.5503009
17
Everest 1830 (1975 Definition)
6377301.243
6356100.231
18
Everest 1830 Modified
6377304.063
6356103.03899315
19
GRS 1980
6378137
6356752.31398972
20
Helmert 1906
6378200
6356818.16962789
21
Indonesian National Spheroid
6378160
6356774.50408554
22
International 1924
6378388
6356911.94612795
23
International 1967
6378160
6356774.71919530
24
Krassowsky 1940
6378245
6356863.01877305
25
NWL 9D
6378145
6356759.76948868
26
NWL 10D
6378135
6356750.52001609
27
Plessis 1817
6376523
6355862.93325557
28
Struve 1860
6378297
6356655.84708038
AT271_TRG_E6
29

29
1.1.3.5
1.1.4
War Office
6378300.583
6356752.27021959
30
WGS 84
6378137
6356752.31398972
31
GEM 10C
6378137
6356752.31398972
32
OSU86F
6378136.2
6356751.51667196
33
OSU91A
6378136.3
6356751.61633668
34
Clarke 1880
6378249.13884613
6356514.96026256
35
Sphere
6371000
6371000
Projection Parameter Indices

Index
Projection Parameter
Index
Projection Parameter
UTM zone number
Scale factor at origin
Longitude of origin
Latitude of 1st parallel
Latitude of origin
Azimuth of central line
False Easting
Latitude of 2nd parallel
False Northing
Angle from rectified to skewed grid
Creating a Coordinate System

Atoll provides a large catalogue of default coordinate systems. Nevertheless, it is possible to add the description of
geographic and cartographic coordinate systems. New coordinate systems can be created from scratch or initialised on
the basis of an existing one.
To create a new coordinate system from scratch:
1. Select Tools > Options. The Options dialog opens.
2. Select the Coordinates tab.
3. Click the browse button (...) on the right of the Projection field.
4. Click the New button. The Coordinate System dialog opens.
5. In the Coordinate System dialog,
a. Select the coordinate systems catalogue to which you want to add the new coordinate system.
b. In the General properties section: Enter a name for the new coordinate system, select a unit. You can also
enter any comments about its usage. Atoll assigns the code automatically.
c. In the Category section: Select the type of coordinate system. Enter the longitude and latitude for a geographic coordinate system, or the type of projection and its set of associated parameters for a cartographic coordinate system (false easting and northing, and the first and second parallels).
d. In the Geo section: Specify the meridian and choose a datum for the coordinate system. The associated ellipsoid is automatically selected. You can also describe a geodetic datum by selecting "..." in the Datum list.
In this case, you must provide parameters (Dx, Dy, Dz, Rx, Ry, Rz, and S) needed for the transformation of
the datum into WGS84, and an ellipsoid.
6. Click OK. The new coordinate system is added to the selected coordinate system catalogue.
To create a new coordinate system based on an existing system, select a coordinate system in the Coordinate Systems
dialog before clicking New in step 4. The new coordinate system is initialised with the values of the selected coordinate
system.
1.2
Units
1.2.1
Power Units
Depending on the working environment, there can be either one or two types of units for transmission and reception
powers. If you are working with stand-alone documents, i.e., documents not connected to databases, there is only one unit
used in Atoll:
Display power units
If you are working in a multi-user environment, Atoll uses two type of units:
Display power units for the Atoll document

Internal power units for the database
The display units are used for the display in dialogs and tables, e.g., reception thresholds (coverage prediction properties,
etc.), and received signal levels (measurements, point analysis, coverage predictions etc.). You can set the display units
for your document in the Options dialog.
30
AT271_TRG_E6
Forsk 2009

The internal units are the power units stored in a database. The power units set by the administrator in the central Atoll
project are stored in the database when the database is created, and cannot be modified by users. Only the administrator
can modify the internal units manually by editing the entries in the Units tables. All Atoll documents opened from a database will have the internal units of the database as their default power units.
Users working on documents connected to a database can modify the units in their documents locally, and save these
changes in their documents, but they cannot modify the units stored in the database.
1.2.2
Length Units
There are two types of units for distances, heights, and offsets:
Display length units

Internal length units
The display length units are used to display distances, heights, and offsets in dialogs, tables, and the status bar. You can
set the display units for your document in the Options dialog.
The internal unit for lengths is metre for all Atoll documents whether they are connected to databases or not. The internal
unit is not stored in the databases. The internal unit cannot be changed.
1.3
BSIC Format
Depending on the working environment, there can be either one or two types of BSIC formats. If you are working with
stand-alone documents, i.e., documents not connected to databases, there is only one BSIC format:
Display BSIC format
If you are working in a multi-user environment, Atoll uses two type of formats:
Display BSIC format for the Atoll document

Internal BSIC format for the database
The display format is used for the display in dialogs and tables. You can set the display format for your document from the
Transmitters folders context menu.
The internal format is the BSIC format stored in a database. The BSIC format set by the administrator in the central Atoll
project is stored in the database when the database is created, and cannot be modified by users. Only the administrator
can modify the internal format manually by editing the corresponding entry in the Units tables. All Atoll documents opened
from a database will have the internal format of the database as their default BSIC format.
Users working on documents connected to a database can modify the format in their documents locally, and save this
change in their documents, but they cannot modify the format stored in the database.
Forsk 2009
AT271_TRG_E6
31
32
AT271_TRG_E6
Forsk 2009
Chapter 2
Geographic and Radio Data
This chapter defines the different types of data with which you can work in Atoll. These data can be
geographic data, such as maps, and radio network data, such as sites, antennas, other equipment and
parameters.
Atoll
Atoll
Microwave
RF PlanningLink
andPlanning
Optimisation
Software
Microwave
Software
34
AT271_TRG_E6
Forsk 2009
Chapter 2: Geographic and Radio Data
Geographic and Radio Data
2.1
Geographic Data
2.1.1
Data Type
Atoll manages several geographic data types; DTM (Digital Terrain Model), clutter (Land-Use), scanned images, vector
data, traffic data, population, and any other generic data.
2.1.1.1
Digital Terrain Model (DTM)

The DTM (Digital Terrain Model or height) files describe the ground elevation above the sea level. DTM files supported by
Atoll are 16 bits/pixel relief maps in .tif, .bil, Planet and Erdas Imagine formats and 8 bits/pixel relief maps in .tif, .bil, Erdas
Imagine and .bmp formats. DTM maps are taken into account in path loss calculations by Atoll propagation models.
DTM file provides altitude value (z stated in metre) on evenly spaced points. Abscissa and ordinate axes are respectively
oriented in right and downwards directions. Space between points is defined by pixel size (P stated in metre). Pixel size
must be the same in both directions. First point given in the file corresponds to the centre of the upper-left pixel of the map.
This point refers to the northwest point geo-referenced by Atoll. Four points (hence, four altitude values) are necessary to
describe a bin; these points are bin vertices.
Figure 2.1: Digital Terrain Model

Therefore, a n*n bin DTM file requires (n)2 points (altitude values).
Figure 2.2: Schematic view of a DTM file

Notes:
Forsk 2009
Altitude values differ within a bin. Method used to calculate altitudes is described in the
Path loss calculations: Altitude determination part. Concerning DTM map display, Atoll
takes altitude of the southwest point of each bin to determine its colour.
In most documents, Digital Elevation Model (DEM) and Digital Terrain Model (DTM) are
differentiated and do not have the same meaning. By definition, DEM refers to altitude
above sea level including, both, ground and clutter while DTM just corresponds to the
ground height above sea level. In Atoll, the DEM term may be used instead of DTM term.
AT271_TRG_E6
35
2.1.1.2
Clutter (Land Use)

You may import two types of clutter files in ATL documents. These files indicate either the clutter class or the clutter height
on each bin of the map.
2.1.1.2.1
Clutter Classes
Atoll supports 8 bits/pixel (255 classes) raster maps in .tif, .bil, .bmp, Erdas Imagine formats or 16 bits/pixel raster maps
in Planet format. This kind of clutter file describes the land cover (dense urban, buildings, residential, forest, open,
villages, ). A grid map represents ground and each bin of the map is characterised by a code corresponding to a main
type of cover (a clutter class). Atoll automatically lists all the clutter classes of the map. It is possible to specify an average
clutter height for each clutter class manually during the map description step. Clutter maps are taken into account in path
loss calculations by Atoll propagation models.
Clutter file provides a clutter code per bin. Bin size is defined by pixel size (P stated in metre). Pixel size must be the same
in both directions. Abscissa and ordinate axes are respectively oriented in right and downwards directions. First point given
in the file corresponds to the centre of the upper-left pixel of the image. This point refers to the northwest point geo-referenced by Atoll.
Figure 2.3: Clutter Classes

Therefore, a n*n bin Clutter file requires (n)2 code values.
Note:
2.1.1.2.2
The clutter code is the same inside a bin.
Clutter Heights
Files supported by Atoll for clutter heights are 8 or 16 bits/pixel raster maps in .tif, .bil and Erdas Imagine formats. The file
provides clutter height value on evenly spaced points. Abscissa and ordinate axes are respectively oriented in right and
downwards directions. Space between points is defined by pixel size (P in metre). Pixel size must be the same in both
directions. First point given in the file corresponds to the centre of the upper-left pixel of the map. This point refers to the
northwest point geo-referenced by Atoll.
These maps are taken into account in path loss calculations by Atoll propagation models.
Note:
2.1.1.3
Atoll considers the clutter height of the nearest point in calculations (see Path loss
calculations: Clutter determination part). For map display, Atoll takes clutter height of the
southwest point of each bin to determine its colour.
Traffic Data
Atoll offers different kinds of traffic data:
2.1.1.3.1
Environment Traffic Maps

Atoll supports 8 bits/pixel (256 class) traffic raster maps in .tif, .bil, .bmp, Erdas Imagine formats. These maps provide
macroscopic traffic estimation. Each pixel is assigned an environment class, which is a list of user profiles with a defined
mobility type and a density.
2.1.1.3.2
User Profile Traffic Maps

Atoll supports vector traffic maps with .dxf, Planet, .shp, .mif and .agd formats. These maps are detailed traffic estimations (lines, polygons or points carrying a specific traffic). Each polygon, line or point is assigned a specific user profile with
associated mobility type and density. They can be built from population density vector maps.
2.1.1.3.3
Live Traffic Maps

Atoll supports maps with .agd format. This kind of map is based on the network feedback. It provides actual information
on connections (and not just subscriber estimation) from the network. It is built from a coverage by transmitter prediction
36
AT271_TRG_E6
Forsk 2009

study that defines sector boundaries for the traffic distribution in each sector. In UMTS, CDMA2000 and IS95-CDMA, either
data rates or the number of users per service are indicated for each transmitter service area. In GSM/TDMA, Atoll expects
a number of Erlangs in case of voice service and data rate values for packet-switched services for each transmitter service
area.
2.1.1.3.4
User Density Traffic Maps

This kind of map is only available in GSM/TDMA documents. Atoll supports 16 and 32 bits/pixel traffic raster maps in .tif,
.bil, .bmp, Planet and Erdas Imagine formats. This map is also based on the network feedback as it deals with network
users information as well. Each pixel is assigned a number of users with a given service, terminal and mobility type.
In GSM documents, traffic maps are taken into account for traffic analysis and network dimensioning. In UMTS,
CDMA2000 and IS95-CDMA documents, they are used by the Monte-Carlo simulator to model user distributions and evaluate related network parameters (cell power, mobile terminal power, ).
2.1.1.4
Vector Data
These data represent either polygons (regions, etc.), lines (roads, coastlines, etc.) or points (towns, etc.). Atoll supports
vector data files in .dxf, Planet, .shp, .mif and .agd formats. These maps are only used for display and provide information about the geographic environment.
2.1.1.5
Scanned Images
These geographic data include the road maps and the satellite images. They are only used for display and provide information about the geographic environment. Atoll supports scanned image files in .tif (1, 4, 8, 24-bits/pixel), .bil (1, 4, 8, 24bits/pixel), Planet (1, 4, 8, 24-bits/pixel), .bmp (1-24-bits/pixel), Erdas Imagine (1, 4, 8, 24-bits/pixel) and .ecw (24-bits/
pixel) formats.
2.1.1.6
Population
Atoll deals with vector population files (polygons, lines or points) in .mif, .shp and .agd formats or 8, 16, 32 bits/pixel raster
population files in .tif, .bil, .bmp and Erdas Imagine formats. Population map describes the population distribution. They
are considered in clutter statistics and in coverage prediction reports.
2.1.1.7
Other Geographic Data

It is possible to import generic geographic data types, other than those listed above, (Customer density, revenue density,
) in Atoll. These data can be either vector files in .mif, .shp and .agd formats or 8, 16, 32 bits/pixel raster files in .tif, .bil,
.bmp and Erdas Imagine formats. These maps are taken into account in clutter statistics and in coverage prediction
reports.
The ArcView Grid format (.txt) is an ASCII format dedicated to define raster maps. It may be used to export any raster map
such as DTM, images, Clutter Classes and/or Heights, Population, Generic data maps and even coverage predictions.
The contents of an ArcView Grid file are in ASCII and consist of a header, describing the content, followed by the content
in the form of cell values.
Notes:
2.1.2
The minimum resolution supported by Atoll is 1m for any raster maps, excepted for
scanned images, for which it is unlimited.
DTM and clutter map resolution must be an integer.
All the raster maps you want to import in an ATL document must be represented in the
same projection system.
Supported Geographic Data Formats

Atoll offers Import/Export filters for the most commonly used geographic data formats. The different filters are:
Forsk 2009
File format
Import/
Export
Can contain
Georeferenced
.bil
Both
DTM, Clutter classes and heights, Traffic,

Image, Population, Other data
Yes via .hdr files
.tif
Both

Yes via associated .tfw files if

they exist
Planet
Both
DTM, Clutter classes, Image, Vector data
Yes via index files
.bmp
Both
DTM, Clutter heights, Clutter classes, Traffic,

Yes via .bpw (or .bmw) files
.dxf
Import Only
Vector data, Vector traffic
Yes
.shp
Both
Vector data, Vector traffic, Population, Other

data
Yes
.mif/.mid
Both

data
Yes
AT271_TRG_E6
37

Erdas Imagine
Import Only

Yes
ArcView Grid
Export Only

Yes automatically embedded in

the data file
.agd
Both

data

the data file
Vertical Mapper
(.grd, .grc)
Both


the data file
.ecw
Import Only
Images
Yes via ers file (not mandatory)
Note:
The .wld files may be used as georeferencement file for any type of binary raster file.
Tiled .tif format is not supported.
Thus, to sum up, you can import:
DTM files in .tif (16-bits, 8-bits), .bil (16-bits, 8-bits), Planet (16-bits), Erdas Imagine (16-bits, 8-bits), Vertical
Mapper (.grd, .grc) and .bmp (8-bits) formats.
Clutter heights files in .tif (16-bits, 8-bits), .bil (16-bits, 8-bits), Planet (16-bits), Erdas Imagine (16-bits, 8-bits),
Vertical Mapper (.grd, .grc) and .bmp (8-bits) formats.
Clutter classes and traffic raster files in .tif (8-bits), .bil (8-bits), .bmp (8-bit), Erdas Imagine (8-bits) and Vertical
Mapper (.grd, .grc) and Planet format (16-bits) are also supported.
Vector data files in .dxf, Planet, .shp, .mif and .agd formats.
Vector traffic files in .dxf, Planet, .shp, .mif and .agd formats.
Scanned image files in .tif (1, 4, 8, 24-bits), .bil (1, 4, 8, 24-bits), Planet (1, 4, 8, 24-bits), .bmp (1-24-bits), Erdas
Imagine (1, 4, 8, 24-bits), Vertical Mapper (.grd, .grc) and .ecw (Enhanced Compressed Wavelet) (24 bits) formats.
Population files in .mif, .shp, .agd, .tif (8, 16, 32-bits), .bil (8, 16, 32-bits), .bmp (8, 32-bits), Vertical Mapper (.grd,
.grc) and Erdas Imagine (8, 16, 32-bits) formats.
Other generic data types in .mif, .shp, .agd, .tif (8, 16, 32-bits), .bil (8, 16, 32-bits), .bmp (8, 32-bits), Vertical
Mapper (.grd, .grc) and Erdas Imagine (8, 16, 32-bits) formats.
Note:
2.2
It is possible to import Packbit, FAX-CCITT3 and LZW compressed .tif files. However, in
case of DTM and clutter, we recommend not to use compressed files in order to avoid poor
performances. If uncompressed files are too big, it is better to split them.
Radio Data
Atoll manages several radio data types; sites, transmitters, antennas, stations and hexagonal designs. Data definition in
Atoll is detailed hereafter.
2.2.1
Site
A site is a geographical point where one or several transmitters (multi-sectored site or station) equipped with antennas are
located.
2.2.2
Antenna
An antenna is a device used for transmitting or receiving electromagnetic waves.
2.2.3
Transmitter
A transmitter is a group of radio devices located at a site. Transmitters are equipped with antenna(s) and other equipment
such as feeder, tower mounted amplifiers (TMA) and BTS.
2.2.4
Repeater
A repeater is a device that receives, amplifies and transmits the radiated or conducted RF carrier both in downlink and
uplink. It comprises a donor side and a server side. The donor side receives the signal from a donor transmitter. This signal
may be carried by different types of links such as radio link, microwave link, or optic fibre. The server side transmits the
repeated signal.
38
AT271_TRG_E6
Forsk 2009
2.2.5
Remote Antenna
The use of remote antennas allows antenna positioning at locations that would normally require prohibitively long runs of
feeder cable. A remote antenna is connected to the base station via an optic fibre. The main difference from a repeater is
that a remote antenna generates its own cell whereas a repeater extends the coverage of an existing cell.
2.2.6
Station
A station can represent one transmitter on a site or a group of transmitters on a same site sharing the same properties.
You can define station templates and build your network from stations instead of single transmitters.
2.2.7
Hexagonal Design
A hexagonal design is a group of stations created from the same station template.
2.2.8
GSM GPRS EGPRS Documents
2.2.8.1
TRX
A base station (transmitter) consists of several transceivers or TRXs. One TRX supports as many timeslots as the multiplexing factor defined in properties of your frequency band (8 timeslots in GSM networks). Three types of TRXs are
modelled in Atoll:
2.2.8.2
The BCCH TRX type: carries the BCCH,

The TCH TRX type: which is the default traffic carrier,
The TCH_INNER TRX type: this TRX type is an inner traffic carrier.
Subcell
A subcell corresponds to a group of TRXs having the same radio characteristics, the same quality (C/I) requirements, and
common settings. A subcell is characterised by the transmitter-TRX type pair. Each transmitter may have one or more
subcells. The most common configurations are the {BCCH, TCH} configuration or the {BCCH, TCH, TCH_INNER} one.
2.2.8.3
Cell Type
A cell type describes the subcells (types of TRXs) that a cell can use and their parameters, which can be different. In the
current Atoll version, the cell type definition must include a TRX type as the BCCH carrier (BCCH TRX type) and another
TRX type as the default traffic carrier (TCH TRX type). Only one TRX type carrying the broadcast and only one TRX type
carrying the default TCH are supported.
2.2.9
All CDMA and WiMAX BWA Documents
2.2.9.1
Cell
Cell comprises the carrier characteristics of a transmitter. Cell is characterised by the transmitter-carrier pair. The transmitter-carrier pair must be unique.
Forsk 2009
AT271_TRG_E6
39
40
AT271_TRG_E6
Forsk 2009
Chapter 3
File Formats
Atoll supports a set of file formats for each type of data, may it be geographic data or calculation results.
This chapter contains details of these file formats, their usage, availability, and limitations.
Atoll
Atoll
Microwave
RF PlanningLink
andPlanning
Optimisation
Software
Microwave
Software
42
AT271_TRG_E6
Forsk 2009
Chapter 3: File Formats
File Formats
3.1
BIL Format
Band Interleaved by Line is a method of organizing image data for multi-band images. It is a schema for storing the actual
pixel values of an image in a file. The pixel data is typically preceded by a file header that contains auxiliary data about the
image, such as the number of rows and columns in the image, a colour map, etc. .bil data stores pixel information band
by band for each line, or row, of the image. Although .bil is a data organization schema, it is treated as an image format.
An image description (number of rows and columns, number of bands, number of bits per pixel, byte order, etc.) has to be
provided to be able to display the .bil file. This information is included in the header .hdr file associated with the .bil file. A
.hdr file has the same name as the .bil file it refers to, and should be located in the same directory as the source file. The
.hdr structure is simple; it is an ASCII text file containing eleven lines. You can open a .hdr file using any ASCII text editor.
Atoll supports the following objects in .bil format:
Digital Terrain Model (8 or 16 bits)

Clutter heights (8 or 16 bits)
Clutter classes and Environment traffic maps (8 bits)
Traffic density maps (16 or 32 bits)
Raster images (1, 4, 8, 24 bits)
Population maps (8, 16, 32 bits)
Other generic geographic data (8, 16, 32 bits)
Path loss or received signal level value matrices (16 bits)
3.1.1
HDR Header File
3.1.1.1
Description
The header file is a text file that describes how data are organised in the .bil file. The header file is made of rows, each
row having the following format:
keyword
value
where keyword corresponds to an attribute type, and value defines the attribute value.
Keywords required by Atoll are described below. Other keywords are ignored.
nrows
Number of rows in the image.
ncols
Number of columns in the image.
nbands
Number of spectral bands in the image, (1 for DTM data and

8 bit pictures).
nbits
Number of bits per pixel per band; 8 or 16 for DTMs or

Clutter heights (altitude in metres), 8 for clutter classes file (clutter code), 16 for path loss matrices (path
loss in dB, field value in dBm, dBV and DBV/m).
byteorder
Byte order in which image pixel values are stored. Accepted values are M (Motorola byte order) or I (Intel byte order).
layout
Must be bil.
skipbytes
Byte to be skipped in the image file in order to reach the

beginning of the image data. Default value is 0.
ulxmap
x coordinate of the centre of the upper-left pixel.
ulymap
y coordinate of the centre of the upper-left pixel.
xdim
x size in metre of a pixel.
ydim
y size in metre of a pixel.
Four additional keywords may be optionally managed.
datatype
Type of data read (in addition to the length)
It can be:
Forsk 2009
AT271_TRG_E6
43

I1
Integer
1 bit
I2
Integer
2 bits
I4
Integer
4 bits
I8
Integer
8 bits
I16
Integer
16 bits
I32
Integer
32 bits
R32
Real
32 bits
R64
Real
64 bits
RGB24
Integer
3 colour components on 24 bits
By default, integer data types are chosen with respect to the pixel length (nbits).
valueoffset
Real value to be added to the read value (Vread)
valuescale
Scaling factor to be applied to the read value
So, we have V = V read valuescale + valueoffset
nodatavalue
3.1.1.2
Value corresponding to NO DATA
Samples
Here, the data is 20m.
3.1.1.2.1
3.1.1.2.2
3.1.1.2.3
Digital Terrain Model

nrows
1500
ncols
1500
nbands
nbits
8 or 16
byteorder
layout
bil
skipbytes
ulxmap
975000
ulymap
1891000
xdim
20.00
ydim
20.00
Clutter Classes File

nrows
1500
ncols
1500
nbands
nbits
byteorder
layout
bil
skipbytes
ulxmap
975000
ulymap
1891000
xdim
20.00
ydim
20.00
BIL File
.bil files are usually binary files without header. Data are stored starting from the Northwest corner of the area. The skipbytes value defined in the header file allows to skip records if the data do not start at the beginning of the file.
44
AT271_TRG_E6
Forsk 2009
3.2
TIF Format
Tagged Image File Format graphics filter supports all image types (monochrome, greyscale, palette colour, and RGB full
colour images) and Packbit, LZW or fax group 3-4 compressions. .tif files are not systematically geo-referenced. You have
to enter spatial references of the image manually during the import procedure (x and y-axis map coordinates of the centre
of the upper-left pixel, pixel size); an associated file with .tfw extension will be simultaneously created with the same name
and in the same directory as the .tif file it refers to. Atoll will then use the .tfw file during the import procedure for an automatic geo-referencing.
Note:
Atoll also supports .tif files using the Packbit, FAX-CCITT3 and LZW compression modes.
You can modify the colour palette convention used by Atoll when exporting .tif files. This can be helpful when working on
.tif files exported by Atoll in other tools. In the default palette, the first colour indexes represent the useful information and
the remaining colour indexes represent the background. It is possible to export .tif files with a palette which defines the
background colour at the colour index 0, and then the colour indexes necessary to represent useful information. Add the
following lines in the Atoll.ini file to set up the new palette convention:
[TiffExport]
PaletteConvention=Gis
Please refer to the Administrator Manual for more details about the Atoll.ini file.
Notes:
Using compressed geo data formats (compressed .tif, Erdas Imagine, or .ecw) can cause
performance loss due to real-time decompression. However, you can recover this loss in
performance by:
- Either, hiding the status bar, which provides geographic data information in real time, by
unchecking the Status Bar item in the View menu.
- Or, not displaying some of the information, such as altitude, clutter class and clutter
height, in the status bar. This can be done through the Atoll.ini file, by adding the following
lines:
[StatusBar]
DisplayZ=0
DisplayClutterClass=0
DisplayClutterHeight=0
You can also save the produced map in an uncompressed format.
Atoll supports the following objects in .tif format:
Digital Terrain Model (8 or 16 bits)

Clutter heights (8 or 16 bits)
Clutter classes and Environment traffic maps (8 bits)
Traffic density maps (16 or 32 bits)
Population maps (8, 16, 32 bits)
Other generic geographic data (8, 16, 32 bits)
.tfw file contains the spatial reference data of an associated .tif file. The .tfw file structure is simple; it is an ASCII text file
that contains six lines. You can open a .tfw file using any ASCII text editor.
3.2.1
TFW Header File

The .tfw files contain spatial reference data for the associated .tif file. The header file is a text file that describes how data
are organised in the .tif file. You can open a .tfw file using any ASCII text editor. The header file consists of six lines, with
each line having the following description:
Line
Forsk 2009
Description
x dimension of a pixel in map units
amount of translation
amount of rotation
negative of the y dimension of a pixel in map units
x-axis map coordinate of the centre of the upper-left pixel
y-axis map coordinate of the centre of the upper-left pixel
AT271_TRG_E6
45

Note:
Atoll does not use the lines 2 and 3 when importing a .tif format geographic file.
3.2.2
Sample
3.2.2.1

100.00
0.00
0.00
-100.00
60000.00
2679900.00
3.3
BMP Format
This is the MS-Windows standard format. It holds black & white, 16-, 256- and True-colour images. The palletized 16colour and 256-colour images may be compressed via run length encoding (though compressed .bmp files are quite rare).
The image data itself can either contain pointers to entries in a colour table or literal RGB values. .bmp files are not systematically geo-referenced. You have to enter spatial references of the image manually during the import procedure (x and yaxis map coordinates of the centre of the upper-left pixel, pixel size). When exporting (saving) a .bmp file, an associated
file with .bpw extension is created with the same name and in the same directory as the .bmp file it refers to. Atoll stores
the georeferencing information in this file for future imports of the .bmp so that the .bpw file can be used during the import
procedure for automatic geo-referencing. Atoll also supports .bmw extension for the .bmp related world files.
Atoll supports the following objects in .tif format:
3.3.1
Digital Terrain Model (8 bits)

Clutter Heights (8 bits)
Clutter classes and traffic density maps (8 bits)
Population maps (8, 32 bits)
Other generic geographic data (8, 32 bits)
BMP File Description

A .bmp file contains of the following data structures:
3.3.1.1
BITMAPFILEHEADER
bmfh
BITMAPINFOHEADER
RGBQUAD
BYTE
bmih
aColors[]
aBitmapBits[]
Contains some information about the bitmap file (about the file, not
about the bitmap itself).
Contains information about the bitmap (such as size, colours, etc.).
Contains a colour table.
Image data (whose format is specified by the bmih structure).
BMP File Structure

The following tables give exact information about the data structures. The Start-value is the position of the byte in the file
at which the explained data element of the structure starts, the Size-value contains the number of bytes used by this data
element, the Name column contains both generic name and the name assigned to this data element by the Microsoft API
documentation, and the Description column gives a short explanation of the purpose of this data element.
Start
Size
1
3
Name
Description
Generic
MS API
Signature
bfType
Must always be set to 'BM' to declare that this is a .bmp-file.
FileSize
bfSize
Specifies the size of the file in bytes.
Reserved1
bfReserved1
Unused. Must be set to zero.
Reserved2
bfReserved2
Unused. Must be set to zero.
11
DataOffset
bfOffBits
Specifies the offset from the beginning of the file to the bitmap (raster)
data.
46
BITMAPFILEHEADER (Header - 14 bytes):
BITMAPINFOHEADER (InfoHeader - 40 bytes):
AT271_TRG_E6
Forsk 2009
Start
Size
15
Name
Description
Generic
MS API
Size
biSize
Specifies the size of the BITMAPINFOHEADER structure, in

bytes (= 40 bytes).
19
Width
biWidth
Specifies the width of the image, in pixels.
23
Height
biHeight
Specifies the height of the image, in pixels.
biPlanes
Specifies the number of planes of the target device, must be

set to zero or 1.
biBitCount
Specifies the number of bits per pixel.

1 = monochrome pallete. # of colours = 1
4 = 4-bit palletized. # of colours = 16
24 = 24-bit palletized. # of colours = 16M
27
29
Planes
BitCount
31
Compression
biCompression
Specifies the type of compression, usually set to zero.

0 = BI_RGB no compression
1 = BI_RLE8 8-bit RLE encoding
2 = BI_RLE4 4-bit RLE encoding
35
ImageSize
biSizeImage
Specifies the size of the image data, in bytes. If there is no

compression, it is valid to set this element to zero.
39
XpixelsPerM
biXPelsPerMeter
Specifies the the horizontal pixels per meter.
43
YpixelsPerM
biYPelsPerMeter
Specifies the the vertical pixels per meter.
47
ColoursUsed
biClrUsed
Specifies the number of colours actually used in the bitmap. If

set to zero the number of colours is calculated using the
biBitCount element.
51
ColoursImportant
biClrImportant
Specifies the number of colour that are 'important' for the

bitmap. If set to zero, all colours are considered important.
Note:
biBitCount actually specifies the colour resolution of the bitmap. It also decides if there is a
colour table in the file and how it looks like.
- In 1-bit mode the colour table has to contain 2 entries (usually white and black). If a bit in
the image data is clear, it points to the first palette entry. If the bit is set, it points to the
second.
- In 4-bit mode the colour table must contain 16 colours. Every byte in the image data
represents two pixels. The byte is split into the higher 4 bits and the lower 4 bits and each
value of them points to a palette entry.
- In 8-bit mode every byte represents a pixel. The value points to an entry in the colour
table which contains 256 entries.
- In 24-bit mode three bytes represent one pixel. The first byte represents the red part, the
second the green and the third the blue part. There is no need for a palette because every
pixel contains a literal RGB-value, so the palette is omitted.
RGBQUAD array (ColorTable):
Start
Size
Name
Description
Generic
MS API
Blue
rgbBlue
Specifies the blue part of the colour.
Green
rgbGreen
Specifies the green part of the colour.
Red
rgbRed
Specifies the red part of the colour.
Reserved
rgbReserved
Must always be set to zero.
Note:
In a colour table (RGBQUAD), the specification for a colour starts with the blue byte, while
in a palette a colour always starts with the red byte.
Pixel data:
The interpretation of the pixel data depends on the BITMAPINFOHEADER structure. It is important to know that the rows
of a .bmp are stored upside down meaning that the uppermost row which appears on the screen is actually the lowermost
row stored in the bitmap. Another important thing is that the number of bytes in one row must always be adjusted by
appending zero bytes to fit into the border of a multiple of four (16-bit or 32-bit rows).
Forsk 2009
AT271_TRG_E6
47
3.3.1.2
BMP Raster Data Encoding

Depending on the image BitCount and on the Compression flag there are 6 different encoding schemes. In all of them,
Pixels are stored bottom-up, left-to-right.

Pixel lines are padded with zeros to end on a 32-bit boundary.
For uncompressed formats every line will have the same number of bytes.
Colour indices are zero based, meaning a pixel colour of 0 represents the first colour table entry, a pixel colour of
255 (if there are that many) represents the 256th entry. For images with more than 256 colours there is no colour
table.
Encoding type
BitCoun Compressio
t
n
1-bit
B&W images
4-bit
16 colour images
8-bit
256 colour images
Every byte holds 1 pixel. There are 256 colour table entries.
Padding each line with zeros up to a 32-bit boundary will result in up
to 3 bytes of zeros = 3 'wasted pixels'.
16-bit
High colour images
16
Every 2 bytes hold 1 pixel. There are no colour table entries.

Padding each line with zeros up to a 16-bit boundary will result in up
to 2 zero bytes.
Every 4 bytes hold 1 pixel. The first holds its red, the second its
green, and the third its blue intensity. The fourth byte is reserved
and should be zero. There are no colour table entries. No zero
padding necessary.
Pixel data is stored in 2-byte chunks. The first byte specifies the
number of consecutive pixels with the same pair of colour. The
second byte defines two colour indices. The resulting pixel pattern
will have interleaved high-order 4-bits and low order 4 bits
(ABABA...). If the first byte is zero, the second defines an escape
code. The End-of-Bitmap is zero padded to end on a 32-bit
boundary. Due to the 16bit-ness of this structure this will always be
either two zero bytes or none.
The pixel data is stored in 2-byte chunks. The first byte specifies the
number of consecutive pixels with the same colour. The second byte
defines their colour indices. If the first byte is zero, the second
defines an escape code. The End-of-Bitmap is zero padded to end
on a 32-bit boundary. Due to the 16bit-ness of this structure this will
always be either two zero bytes or none.
24
4-bit
16 colour images
8-bit
256 colour images
Raster Data Compression Descriptions
4-bit / 16 colour images
n (Byte 1)
48
Every byte holds 8 pixels, its highest order bit representing the
leftmost pixel of these 8. There are 2 colour table entries. Some
readers assume that 0 is black and 1 is white. If you are storing
black and white pictures you should stick to this, with any other 2
colours this is not an issue. Remember padding with zeros up to a
32-bit boundary.
Every byte holds 2 pixels, its high order 4 bits representing the left of
those. There are 16 colour table entries. These colours do not have
to be the 16 MS-Windows standard colours. Padding each line with
zeros up to a 32-bit boundary will result in up to 28 zeros = 7 'wasted
pixels'.
24-bit
True colour images
3.3.1.2.1
Remarks
c (Byte 2)
Description
>0
any
n pixels to be drawn. The 1st, 3rd, 5th, ... pixels' colour is in c's high-order 4 bits, the
even pixels' colour is in c's low-order 4 bits. If both colour indices are the same, it
results in just n pixels of colour c.
End-of-line
End-of-Bitmap
Delta. The following 2 bytes define an unsigned offset in x and y direction (y being up).
The skipped pixels should get a colour zero.
>=3
The following c bytes will be read as single pixel colours just as in uncompressed files.
Up to 12 bits of zeros follow, to put the file/memory pointer on a 16-bit boundary again.
8-bit / 256 colour images
n (Byte 1)
c (Byte 2)
Description
>0
any
n pixels of colour number c
AT271_TRG_E6
Forsk 2009
3.3.2
End-of-line
End-of-Bitmap
Delta. The following 2 bytes define an unsigned offset in x and y direction (y being up).
The skipped pixels should get a colour zero.
>=3
The following c bytes will be read as single pixel colours just as in uncompressed files.
A zero follows, if c is odd, putting the file/memory pointer on a 16-bit boundary again.
BPW/BMW Header File Description

The header file is a text file that describes how data are organised in the .bmp file. The header file is made of rows, each
row having the following description:
Line
Description
amount of rotation
Atoll supports .bpw and .bmw header file extensions for Import, but exports headers with .bpw file extensions.
3.3.3
Sample
3.3.3.1

100.00
0.00
0.00
-100.00
60000.00
2679900.00
3.4
Generic Raster Header File (.wld)

.wld is a new Atoll specific header format that can be used for any raster data file for georeferencing. At the time of import
of any raster data file, Atoll can use the corresponding .wld file to read the georeferencing information related to the raster
data file. The .wld file contains the spatial reference data of any associated raster data file. The .wld file structure is simple;
it is an ASCII text file containing six lines. You can open a .wld file using any ASCII text editor.
3.4.1
WLD File Description

The .wld file is a text file that describes how data are organised in the associated raster data file. The header file is made
of rows, each row having the following description:
Line
Description
amount of rotation
3.4.2
Sample
3.4.2.1

100.00
Forsk 2009
AT271_TRG_E6
49
0.00
0.00
-100.00
60000.00
2679900.00
3.5
DXF Format
Atoll is capable of importing and working with AutoCAD drawings in the Drawing Interchange Format (DXF). .dxf files
can have ASCII or binary formats. But only the ASCII .dxf files can be used in Atoll.
.dxf files are composed of pairs of codes and associated values. The codes, known as group codes, indicate the type of
value that follows. .dxf files are organized into sections of records containing the group codes and their values. Each group
code and value is a separate line.
Each section starts with a group code 0 followed by the string, SECTION. This is followed by a group code 2 and a string
indicating the name of the section (for example, HEADER). Each section ends with a 0 followed by the string ENDSEC.
3.6
SHP Format
ESRI (Environmental Systems Research Institute, Inc.) ArcView GIS Shapefiles have a simple, non-topological format
for storing geometric locations and attribute information of geographic features. A shapefile is one of the spatial data
formats that you can work with in ArcExplorer. .shp data files usually have associated .shx and .dbf files.
Among these three files:
The .shp file stores the feature geometry

The .shx file stores the index of the feature geometry.
The .dbf (dBASE) file stores the attribute information of features. When a shapefile is added as a theme to a view,
this file is displayed as a feature table.
You can define mappings between the coordinate system used for the ESRI vector files, defined in the corresponding .prj
files, and Atoll. In this way, when you import a vector file, Atoll can detect the correct coordinate system automatically. For
more information about defining the mapping between coordinate systems, please refer to the Administrator Manual.
3.7
MIF Format
MapInfo Interchange Format (.mif) allows various types of data to be attached to a variety of graphical items. These ASCII
files are editable, easy to generate, and work on all platforms supported by MapInfo. Vector objects with a .mif extension
may be imported in Atoll.
Two files, a .mif and a .mid, contain MapInfo data. Graphics reside in the .mif file while the text contents are stored in the
.mid file. The text data is delimited with one row per record, and Carriage Return, Carriage Return plus Line Feed, or Line
Feed between lines. The .mif file has two sections, the file header and the data section. The .mid file is optional. When
there is no .mid file, all fields are blank.
You can find more information at http://www.mapinfo.com.
You can define mappings between the coordinate system used for the MapInfo vector files, defined in the corresponding
.mif files, and Atoll. In this way, when you import a vector file, Atoll can detect the correct coordinate system automatically.
For more information about defining the mapping between coordinate systems, please refer to the Administrator Manual.
3.8
TAB Format
TAB files (MapInfo Tables) are the native format of MapInfo. They actually consist of a number of files with extensions
such as .TAB, .DAT and .MAP. All of these files need to be present and kept together for the table to work. These are
defined as follows:
.TAB: table structure in ASCII format

.DAT: table data storage in binary format
.MAP: storage of map objects in binary format
.ID: index to the MapInfo graphical objects (.MAP) file
.IND: index to the MapInfo tabular (DAT) file
You can find more information at http://www.mapinfo.com.

You can define mappings between the coordinate system used for the MapInfo vector files, defined in the corresponding
.mif files, and Atoll. In this way, when you import a vector file, Atoll can detect the correct coordinate system automatically.
For more information about defining the mapping between coordinate systems, please refer to the Administrator Manual.
TAB files are also supported as georeference information files for raster files (.bmp and .tif).
50
AT271_TRG_E6
Forsk 2009
3.9
ECW Format
The Enhanced Compressed Wavelet file format is supported in Atoll. .ecw files are geo-referenced image files, which can
be imported in Atoll. This is an Open Standard wavelet compression technology, developed by Earth Resource Mapping,
which can compress images with up to a 100-to-1 compression ratio. Each compressed image file contains a header carrying the following information about the image:
3.10
The image size expressed as the number of cells across and down
The number of bands (RGB images have three bands)
The image compression rate
The cell measurement units (meters, degrees or feet)
The size of each cell in measurement units
Coordinate space information (Projection, Datum etc.)
Erdas Imagine Format

Atoll supports Erdas Imagine data files in order to import DTM (8 or 16 bit/pixel), clutter (8 bit/pixel), traffic (8 bit/pixel), and
image (1-24 bit/pixel) files with the .img format. These files use the Erdas Imagine Hierarchical File Format (HFA) structure.
For any type of file, if there are pyramids (storage of different resolution layers), they are used to enhance performance
when decreasing the resolution of the display. Some aspects of working with Erdas Imagine format in Atoll are:
Atoll supports uncompressed as well as compressed (or partially compressed) DTM .img files.
You can create a .mnu file to improve the clutter class map loading.
The colour-to-code association (raster maps) may be automatically imported from the .img file.
These files are automatically geo-referenced, i.e., they do not require any additional file for geo-reference.
For image files, the number of supported bands is either 1 (colour palette is defined separately) or 3 (no colour palette but
direct RGB information for each pixel). In case of 3 bands, only 8 bit per pixel format is supported. Therefore, 8-bit images,
containing RGB information (three bands are provided: the first band is for Blue, the second one is for Green and the third
for Red), can be considered as 24 bit per pixel files. 32 bit per pixel files are not supported.
Notes:
Using compressed geo data formats (compressed .tif, Erdas Imagine, or .ecw) can cause
performance loss due to real-time decompression. However, you can recover this loss in
performance by:
- Either, hiding the status bar, which provides geographic data information in real time, by
unchecking the Status Bar item in the View menu.
- Or, not displaying some of the information, such as altitude, clutter class and clutter
height, in the status bar. This can be done through the Atoll.ini file, by adding the following
lines:
[StatusBar]
DisplayZ=0
DisplayClutterClass=0
DisplayClutterHeight=0
3.11
You can also save the produced map in an uncompressed format.
Planet EV/Vertical Mapper Geographic Data Format

Vertical Mapper offers two types of grids:
Numerical continuous grids, which contain numerical information (such as DTM), and are stored in files with the
.grd extension.
Classified grids, which contain alphanumeric (characters) information, and are stored in files with the .grc extension.
Atoll is capable of supporting the Vertical Mapper Classified Grid (GRC) and Vertical Mapper Continuous Grid (GRD) file
formats in order to import and export:
GRD: DTM, image, population, traffic density, and other data types.
GRC: DTM, clutter classes, clutter heights, environment traffic, image, population, and other data types.
It is also possible to export coverage prediction studies in GRD and GRC formats.
This is the geographic data format used by Planet EV. So, it is possible to directly import geographic data from Planet EV
to Atoll using this format.
Forsk 2009
AT271_TRG_E6
51
3.12
ArcView Grid Format

The ArcView Grid format (.txt) is an ASCII format dedicated to defining raster maps. It may be used to export any raster
map such as DTM, images, clutter classes and/or heights, population, other data maps, and even coverage predictions.
The contents of an ArcView Grid file are in ASCII and consist of a header, describing the content, followed by the content
in the form of cell values.
3.12.1
ArcView Grid File Description

The format of this file is as follows:
ncols XXX
Number of columns of the grid (XXX columns).
nrows XXX
Number of rows of the grid (XXX rows).
xllcenter XXX OR
xllcorner XXX
Significant value relative to the bin centre or corner.
yllcenter OR
yllcorner XXX
Significant value relative to the bin centre or corner.
cellsize XXX
Grid resolution.
nodata_value XXX
Optional value corresponding to no data (no information).
//Row 1
Top of the raster. Description of the first row. Syntax:
ncols number of values separated by spaces.
:
:
//Row N
3.12.2
Bottom of the raster.
Sample
ncols 303
nrows 321
xllcorner 585300.000000
yllcorner 5615700.000000
cellsize 100.000000
nodata_value 0
...
3.13
Other Supported Geographic Data File Formats

Other than the .bil, .tif, Planet, .dxf, .shp, .mif, .img, and .ecw formats, Atoll supports 3 other formats.
The .ist and .dis formats are ASCII files used for Digital Terrain Model only. .ist images come from Istar, whereas .dis
images come from IGN (Institut Gographique National). The .ist format works in exactly the same way as the .bil format,
except for DTM images. For DTM images, the .ist format uses a decimetric coding for altitudes, whereas .bil images use
only a metric coding.
3.14
Planet Format
The Planet geographic data are described by a set of files grouped in a Planet directory. The directory structure depends
on the geographic data type.
Atoll supports the following objects in Planet format:
52
Digital Terrain Model (8 and 16 bits)

Clutter class maps (16 bits)
Raster images (1, 4, 8 and 24 bits)
Vector data
Text data
AT271_TRG_E6
Forsk 2009
3.14.1
DTM File
3.14.1.1
Description
The DTM directory consists of three files; the height file and two other files detailed below:
The index file structure is simple; it is an ASCII text file that holds position information about the file. It contains
five columns. You can open an index file using any ASCII text editor. The format of the index file is as follows:
Field
Acceptable values
Description
File name
Text
Name of file referenced by the index file
East min
Float
x-axis map coordinate of the centre of the upper-left pixel in meters
East max
Float
x-axis map coordinate of the centre of the upper-right pixel in meters
North min
Float
y-axis map coordinate of the centre of the lower-left pixel in meters
North max
Float
y-axis map coordinate of the centre of the upper-left pixel in meters
Square size
Float
Dimension of a pixel in meters
The projection file provides information about the projection system used. This file is optional. It is an ASCII text
file with four lines maximum.
Line
Description
Spheroid
Zone
Projection
Central meridian
Latitude and longitude of projection central meridian and equivalent x and y coordinates in meters
(optional)
Note:
3.14.1.2
In the associated binary file, the value -9999 corresponds to No data which is supported
by Atoll.
Sample
Index file associated with height file (DTM data):
sydney1
303900 343900 6227900 6267900 50
Projection file associated with height file (DTM data):
Australian-1965
56
UTM
0 153 500000 10000000
3.14.2
Clutter Class Files
3.14.2.1
Description
The Clutter directory consists of three files; the clutter file and two other files detailed below:
Field
Type
Description
Clutter-code
Integer (>1)
Identification code for clutter class
Feature-name
Text (up to 32 characters in length)
Name associated with the clutter-code. (It may contain

spaces)
Forsk 2009
The menu file, an ASCII text file, defines the feature codes for each type of clutter. It consists of as many lines
(with the following format) as there are clutter codes in the clutter data files. This file is optional.
The index file gives clutter spatial references. The structure of clutter index file is the same as the structure of DTM
index file.
AT271_TRG_E6
53

Note:
3.14.2.2
In the associated binary file, the value -9999 corresponds to No data which is supported
by Atoll.
Sample
Menu file associated with the clutter file:
open
sea
inlandwater
residential
meanurban
denseurban
buildings
village
industrial
10
openinurban
11
forest
12
parks
13
denseurbanhigh
14
blockbuildings
15
denseblockbuild
16
rural
17
mixedsuburban
3.14.3
Vector Files
3.14.3.1
Description
Vector data comprises terrain features such as coastlines, roads, etc. Each of these features is stored in a separate vector
file. Four types of files are used, the vector file, where x and y coordinates of vector paths are stored, and three other files
detailed below:
The menu file, an ASCII text file, lists the vector types stored in the database. The menu file is composed of one
or more records with the following structure:
Field
Type
Description
Vector type code
Integer > 0
Identification code for the vector type
Vector type name
Name of the vector type
The fields are separated by space character.
The index file, an ASCII text file, lists the vector files and associates each vector file with one vector type, and
optionally with one attribute file. The index file consists of one or more records with the following structure:
Field
Type
Vector file name
Name of the vector file
Name of attribute file associated with the vector file

(optional)
Dimensions
Real
vector file eastmin: minimum x-axis coordinate of all

vector path points in the vector file
vector file eastmax: maximum x-axis coordinate of all
vector file northmin: minimum y-axis coordinate of all
vector file northmax: maximum y-axis coordinate of all
vector path points
Vector type name
Name of the vector type with which the vector file is

associated. This one must match exactly a vector type
name field in the menu file.
Attribute file name
54
AT271_TRG_E6
Description
Forsk 2009

The fields are separated by spaces.
3.14.3.2
The attribute file stores the height and description properties of vector paths. This file is optional.
Sample
Index file associated with the vector files
3.14.4
sydney1.airport
313440 333021 6239426 6244784 airport
sydney1.riverlake
303900 342704 6227900 6267900 riverlake
sydney1.coastline
322837 343900 6227900 6267900 coastline
sydney1.railways
303900 336113 6227900 6267900 railways
sydney1.highways
303900 325155 6240936 6267900 highways
sydney1.majstreets
303900 342770 6227900 6267900 majstreets
sydney1.majorroads
303900 342615 6227900 6267900 majorroads
Image Files
The image directory consists of two files, the image file with .tif extension and an index file with the same structure as the
DTM index file structure.
3.14.5
Text Data Files

The text data directory consists of:
The text data files are ASCII text files with the following format:
Airport
637111.188 3094774.00
Airport
628642.688 3081806.25
Each file contains a line of text followed by easting and northing of that text, etc.
The index file, an ASCII text file, stores the position of each text file. It consists of one or more records with the
following structure:
Field
Type
Description
File name
File name of the text data file
East Min
Real
Minimum x-axis coordinate of all points listed in the text

data file
East Max
Real
Maximum x-axis coordinate of all points listed in the text

data file
North Min
Real
Minimum y-axis coordinate of all points listed in the text

data file
North Max
Real
Maximum y-axis coordinate of all points listed in the text

data file
Text feature
This field is omitted in case no menu file is available.
The fields are separated by spaces.
railwayp.txt -260079 693937 2709348 3528665 Railway_Station

airport.txt -307727 771663 2547275 3554675 Airport
ferryport.txt 303922 493521 2667405 3241297 Ferryport
Forsk 2009
The menu file, an ASCII text file, contains the text features. This file is optional.
Airport
Ferryport
Railway_Station
AT271_TRG_E6
55
3.15
MNU Format
3.15.1
Description
A .mnu file is useful when importing clutter classes or raster traffic files in .tif, .bil and .img formats. It gives the correspondence between the clutter (or traffic) code and the class name. It is a text file with the same name as the clutter (or traffic)
file with .mnu extension. It must be stored at the same location as the clutter (or traffic) file. It has the same structure as
the menu file used in the Planet format.
Field
Type
Description
Class code
Integer > 0
Identification code for the clutter (or traffic) class
Class name
Name of the clutter (or traffic) class. It may contain spaces.
Separator used can either be a space character or a tab.
3.15.2
Sample
A .mnu file associated to a clutter classes file:
3.16
none
open
sea
inland_water
residential
meanurban
XML Table Export/Import Format

All the data tables in an Atoll document can be exported to XML files.
Atoll creates the following files when exporting data tables to XML files:
One index.xml file which contains the mapping between the data tables in Atoll and the corresponding XML file
created by the export.
One XML file per data table which contains the data table format (schema) and the data.
The XML import does not modify the active document table and field definitions. Therefore, the Networks and CustomFields tables, although exported, are not imported.
The following sections describe the structures of these two types of XML files created at export.
3.16.1
Index.xml File
The index.xml file stores the system (GSM, UMTS, etc.) and the technology (TDMA, CDMA, etc.) of the document, and
the version of Atoll used for exporting the data tables to XML files. It also contains the mapping between the data tables
in the Atoll document and the XML file corresponding to each data table.
The root tag <Atoll_XML_Config...> of the index.xml file contains the following attributes:
Attribute
Description
Atoll_File_System
Corresponds to the SYSTEM_ field of the Networks table of the exported document
Atoll_File_Technology
Corresponds to the TECHNOLOGY field of the Networks table of the exported

document
Atoll_File_Version
Corresponds to the Atoll version
The index file also contains a list of mapping between the tables exported from Atoll and the XML files corresponding to
each table. This list is sorted in the order the Atoll tables are to be imported.
The list is composed of <XML_Table.../> tags with the following attributes:
Attribute
Description
XML_File
Corresponds to the exported XML file name (e.g., "Sites.xml")
Atoll_Table
Corresponds to the exported Atoll table name (e.g., "Sites")
A sample extract of the index.xml is given below:
56
AT271_TRG_E6
Forsk 2009
<Atoll_XML_Config Atoll_File_System="UMTS" Atoll_File_Technology="CDMA"

Atoll_File_Version="2.7.0 build 2334">
<XML_Table XML_File="CustomFields.xml" Atoll_Table="CustomFields" />
<XML_Table XML_File="CoordSys.xml" Atoll_Table="CoordSys" />
...
</Atoll_XML_Config>
Note that no closing tag </XML_Table> is required.
3.16.2
XML File
Atoll creates an XML file per exported data table. This XML file has two sections, one for storing the description of the table
structure, and the second for the data itself. The XML file uses the standard XML rowset schema (schema included in the
XML file between <s:Schema id=RowsetSchema> and </s:Schema> tags).
Rowset Schema
The XML root tag for XML files using the rowset schema is the following:
<xml xmlns:s='uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882'
xmlns:dt='uuid:C2F41010-65B3-11d1-A29F-00AA00C14882'
xmlns:rs='urn:schemas-microsoft-com:rowset'
xmlns:z='#RowsetSchema'>
The schema definition follows the root tag and is enclosed between the following tags:
<s:Schema id=RowsetSchema>
<!-Schema is defined here, using <s:ElementType> and <s:AttributeType> tags ->
</s:Schema>
In the rowset schema, after the schema description, the data are enclosed between <rs:data> and </rs:data>.
Between these tags, each record is handled by a <z:row /> tag having its attributes set to the record field values since
in the rowset schema, values are handled by attributes. Note that no closing tag </z:row> is required.
A sample extract of a Sites.xml file containing the Sites table with only one site is given below:
<xml xmlns:s='uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882'
xmlns:dt='uuid:C2F41010-65B3-11d1-A29F-00AA00C14882'
xmlns:rs='urn:schemas-microsoft-com:rowset'
xmlns:z='#RowsetSchema'>
<s:Schema id='RowsetSchema'>
<s:ElementType name='row' content='eltOnly' rs:updatable='true'>
<s:AttributeType name='NAME' rs:number='1' rs:maydefer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecolumn='NAME' rs:keycolumn='true'>
<s:datatype dt:type='string' dt:maxLength='50'/>
</s:AttributeType>
<s:AttributeType name='LONGITUDE' rs:number='2' rs:maydefer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecolumn='LONGITUDE'>
<s:datatype dt:type='float' dt:maxLength='8' rs:precision='15' rs:fixedlength='true'/>
</s:AttributeType>
<s:AttributeType name='LATITUDE' rs:number='3' rs:maydefer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecolumn='LATITUDE'>
<s:datatype dt:type='float' dt:maxLength='8' rs:precision='15' rs:fixedlength='true'/>
</s:AttributeType>
<s:AttributeType name='ALTITUDE' rs:number='4' rs:nullable='true' rs:maydefer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecolumn='ALTITUDE'>
Forsk 2009
AT271_TRG_E6
57
<s:datatype
edlength='true'/>
dt:type='r4'
dt:maxLength='4'
rs:precision='7'
rs:fix-
</s:AttributeType>
<s:AttributeType name='COMMENT_' rs:number='5' rs:nullable='true' rs:maydefer='true'
rs:writeunknown='true'
rs:basetable='Sites'
rs:basecolumn='COMMENT_'>
<s:datatype dt:type='string' dt:maxLength='255'/>
</s:AttributeType>
<s:extends type='rs:rowbase'/>
</s:ElementType>
</s:Schema>
<rs:data>
<rs:insert>
<z:row NAME='Site0' LONGITUDE='8301' LATITUDE='-9756'/>
</rs:insert>
</rs:data>
</xml>
3.17
Externalised Propagation Results Format

Propagation results, i.e. the path loss matrices, may be stored in an external folder. This folder consists of a dBASE III
based file named pathloss.dbf that contains calculation parameters of all the transmitters considered and one file (or two
when calculating main and extended path loss matrices) per transmitter taken into account. This is a binary file with .los
extension and contains the path loss values for a transmitter.
Note:
3.17.1
Each transmitter path loss matrix is calculated on the area where calculation radius
intersects the computation zone (see: Computation zone).
DBF File
dBASE III file (pathloss.dbf) has a standard .dbf format described below. Its content can be checked by opening it in MSAccess. The format is detailed hereafter.
3.17.1.1
DBF File Format

For general information, the format of .dbf files in any Xbase language is described.
Following notations are used in tables:
FS = FlagShip
D3 = dBaseIII+
Fb = FoxBase
D4 = dBaseIV
Fp = FoxPro
D5 = dBaseV
CL = Clipper
3.17.1.1.1
3.17.1.1.2
58
DBF Structure
Byte
Description
0...n
.dbf header (see next part for size, byte 8)
n+1
1st record of fixed length (see next parts)

2nd record (see next part for size, byte10)
last record
last
optional: 0x1a (eof byte)
If .dbf is not empty
DBF Header (Variable Size - Depends on Field Count)

Byte
Size
Contents
Description
Applies for (supported by)
00
0x03
plain .dbf
FS, D3, D4, D5, Fb, Fp, CL
0x04
plain .dbf
D4, D5 (FS)
0x05
plain .dbf
D5, Fp (FS)
AT271_TRG_E6
Forsk 2009

with .dbv memo var size
FS
0xB3
with .dbv and .dbt memo
FS
0x83
with .dbt memo
0x8B
with .dbt memo in D4 format
D4, D5
0x8E
with SQL table
D4, D5
0xF5
with .fmp memo
Fp
YYMMDD
Last update digits
All
ulong
Number of records in file
All
ushort
Header size in bytes
All
10
ushort
Record size in bytes
All
12
0,0
Reserved
All
14
0x01
Begin transaction
D4, D5
0x00
End Transaction
D4, D5
0x00
ignored
FS, D3, Fb, Fp, CL
0x01
Encrypted
D4, D5
01
04
08
15
0x00
normal visible
All
16
12
0 (1)
multi-user environment use
D4,D5
28
0x01
production index exists
Fp, D4, D5
0x00
index upon demand
All
language driver ID
D4, D5
0x01
codepage437 DOS USA
Fp
0x02
codepage850 DOS Multi ling
Fp
0x03
codepage1251 Windows ANSI
Fp
0xC8
codepage1250 Windows EE
Fp
0x00
ignored
FS, D3, Fb, Fp, CL
29
30
32
n*32
+1
0,0
0x0D
reserved
All
Field Descriptor, (see next paragraph)
all
Header Record Terminator
all
Field descriptor array in the .dbf header (32 bytes for each field)
Byte
Size
Contents
Description
11
ASCI
field name, 0x00 termin
all
11
ASCI
field type (see next paragraph)
all
12
n,n,n,n
Fld address in memory
D3
n,n,0,0
offset from record begin
Fp
0,0,0,0
ignored
FS, D4, D5, Fb, CL
16
byte
Field length, bin (see next paragraph)
all \ FS,CL: for C field type
17
byte
decimal count, bin
all / both used for fld lng
18
0,0
reserved
all
20
21
23
2
1
byte
Work area ID
D4, D5
0x00
unused
FS, D3, Fb, Fp, CL
n,n
multi-user dBase
D3, D4, D5
0,0
ignored
FS, Fb, Fp, CL
0x01
Set Fields
D3, D4, D5
0x00
ignored
FS, Fb, Fp, CL
24
0...0
reserved
all
31
0x01
Field is in .mdx index
D4, D5
0x00
ignored
FS, D3, Fb, Fp, CL
Forsk 2009
0x43
Field type and size in the .dbf header, field descriptor (1 byte)
Size
Type
Description/Storage
C 1...n
Char
ASCII (OEM code page chars)

rest= space, not \0 term.
all
AT271_TRG_E6
59

n = 1...64kb (using deci count)
FS
Fp, CL
n = 1...254
all
Date
8 ASCII digits (0...9) in the YYYYMMDD format
all
Numeric
ASCII digits (-.0123456789)

variable pos. of float.point
n = 1...20
FS, D4, D5, Fp
N 1...n Numeric

fix posit/no float.point
all
D8
F 1...n
n = 1...20
FS, Fp, CL
n = 1...18
D3, D4, D5, Fb
ASCII chars (YyNnTtFf space)
FS, D3, Fb, Fp, CL
ASCII chars (YyNnTtFf?)
D4, D5 (FS)
Memo
10 digits repres. the start block posit. in .dbt file, or 10 spaces if

no entry in memo
all
V 10
Variable
Variable, bin/asc data in .dbv

4bytes bin= start pos in memo
4bytes bin= block size
1byte = subtype
1byte = reserved (0x1a)
10 spaces if no entry in .dbv
FS
P 10
Picture
binary data in .ftp

structure like M
Fp
B 10
Binary
binary data in .dbt

structure like M
D5
G 10
General
OLE objects
structure like M
D5, Fp
22
short int
binary int max +/- 32767
FS
44
long int
FS
88
double
binary signed double IEEE
FS
L1
Logical
M 10
3.17.1.1.3
3.17.1.2
Each DBF Record (Fixed Length)

Byte
Size
Description
deleted flag "*" or not deleted " "
all
1n
x-times contents of fields, fixed length, unterminated.

For n, see (2) byte 1011
All
DBF File Content

The .dbf file provides information that is needed to check validity of each path loss matrix.
Field
Type
Description
TX_NAME
Text
Name of the transmitter
FILE_NAME
Text
Name (and optionally, path) of .los file
MODEL_NAME
Text
Name of propagation model used to calculate path loss
MODEL_SIG
Text
Signature (identity number) of model used in calculations. You may check it in the
propagation model properties (General tab).
The Model_SIG is used for the purpose of validity. A unique Model_SIG is
assigned to each propagation model. When model parameters are modified, the
associated model ID changes. This enables Atoll to detect path loss matrix
invalidity. In the same way, two identical propagation models in different projects
do not have the same model IDa.
60
ULXMAP
Float
X-coordinate of the top-left corner of the path loss matrix upper-left pixel
ULYMAP
Float
Y-coordinate of the top-left corner of the path loss matrix upper-left pixel
RESOLUTION
Float
Resolution of path loss matrix in metre
NROWS
Float
Number of rows in path loss matrix
NCOLS
Float
Number of columns in path loss matrix
FREQUENCY
Float
Frequency band
TILT
Float
Transmitter antenna mechanical tilt
AZIMUTH
Float
Transmitter antenna azimuth
AT271_TRG_E6
Forsk 2009

TX_HEIGHT
Float
Transmitter height in metre
TX_POSX
Float
X-coordinate of the transmitter
TX_POSY
Float
Y-coordinate of the transmitter
ALTITUDE
Float
Ground height above sea level at the transmitter in metre
RX_HEIGHT
Float
Receiver height in metre
ANTENNA_SI
Float
Logical number referring to antenna pattern. Antennas with the same pattern will
have the same number.
MAX_LOS
Float
Maximum path loss stated in 1/16 dB. This information is used, when no
calculation radius is set, to check the matrix validity.
CAREA_XMIN
Float
Lowest x-coordinate of centre pixel located on the calculation radiusb
CAREA_XMAX
Float
Highest x-coordinate of centre pixel located on the calculation radius
CAREA_YMIN
Float
Lowest y-coordinate of centre pixel located on the calculation radius
CAREA_YMAX
Float
Highest y-coordinate of centre pixel located on the calculation radius
WAREA_XMIN
Float
Lowest x-coordinate of centre pixel located in the computation zonec
WAREA_XMAX
Float
Highest x-coordinate of centre pixel located in the computation zone
WAREA_YMIN
Float
Lowest y-coordinate of centre pixel located in the computation zone
WAREA_YMAX
Float
Highest y-coordinate of centre pixel located in the computation zone
LOCKED
Boolean
Locking status
0: path loss matrix is not locked
1: path loss matrix is locked.
Boolean
Atoll indicates if losses due to the antenna pattern are taken into account in the
path loss matrix.
0: antenna losses not taken into account
1: antenna losses included
INC_ANT
a.
b.
c.
3.17.2
In order to benefit from the calculation sharing feature, users must retrieve the propagation models from the same
central database. This can be done using the Open from database command for a new document or the Refresh
command for an existing one. Otherwise, Atoll generates different model_ID (even if same parameters are applied
on the same kind of model) and calculation sharing become unavailable due to inconsistency.
These coordinates enable Atoll to determine the area of calculation for each transmitter.
These coordinates enable Atoll to determine the rectangle including the computation zone.
LOS File
The data file is a 16 bits binary row file organized in a standard row-column structure. It contains an integer path loss value,
with a 1/16 dB unit. Data are stored starting from the southwest to the northeast corner of the area.
3.18
Externalised Tuning Files

Atoll can tune path loss matrices obtained from propagation results by the use of real measurements (CW Measurements
or Test Mobile Data). For each measured transmitter, Atoll tries to merge measurements and predictions on the same
points and to smooth the surrounding points of the path loss matrices for homogeneity reasons. A transmitter path loss
matrix can be tuned several times by the use of several measurement paths. All these tuning paths are stored in a catalogue. This catalogue is stored under a .tuning folder containing a .dbf file and one .pts file per corrected transmitter. Since
a tuning file can contain several measurement paths, all these measurements are added to the tuning file.
For more information on the path loss tuning algorithm, See "Path Loss Tuning" on page 108.
3.18.1
DBF File
dBASE III file (pathloss.dbf) has a standard .dbf format described below. Its content can be checked by opening it in MSAccess. The format is detailed hereafter.
3.18.1.1
DBF File Format

For general information, the format of .dbf files in any Xbase language is described.
Following notations are used in tables:
FS = FlagShip
D3 = dBaseIII+
Fb = FoxBase
D4 = dBaseIV
Fp = FoxPro
D5 = dBaseV
CL = Clipper
Forsk 2009
AT271_TRG_E6
61
3.18.1.1.1
3.18.1.1.2
DBF Structure
Byte
Description
0...n
.dbf header (see next part for size, byte 8)
n+1
1st record of fixed length (see next parts)

2nd record (see next part for size, byte10)
last record
last
optional: 0x1a (eof byte)
DBF Header (Variable Size - Depends on Field Count)

Byte
Size
Contents
Description
00
0x03
plain .dbf
0x04
plain .dbf
D4, D5 (FS)
0x05
plain .dbf
D5, Fp (FS)
0x43
with .dbv memo var size
FS
0xB3
with .dbv and .dbt memo
FS
0x83
with .dbt memo
0x8B
with .dbt memo in D4 format
D4, D5
0x8E
with SQL table
D4, D5
0xF5
with .fmp memo
Fp
01
YYMMDD
Last update digits
All
04
ulong
Number of records in file
All
08
ushort
Header size in bytes
All
10
ushort
Record size in bytes
All
12
0,0
Reserved
All
14
0x01
Begin transaction
D4, D5
0x00
End Transaction
D4, D5
0x00
ignored
FS, D3, Fb, Fp, CL
0x01
Encrypted
D4, D5
0x00
normal visible
All
15
16
12
0 (1)
multi-user environment use
D4,D5
28
0x01
production index exists
Fp, D4, D5
0x00
index upon demand
All
29
language driver ID
D4, D5
0x01
codepage437 DOS USA
Fp
0x02
codepage850 DOS Multi ling
Fp
0x03
codepage1251 Windows ANSI
Fp
0xC8
codepage1250 Windows EE
Fp
0x00
ignored
FS, D3, Fb, Fp, CL
0,0
reserved
All
Field Descriptor, (see next paragraph)
all
Header Record Terminator
all
30
32
n*32
+1
62
If .dbf is not empty
0x0D
Field descriptor array in the .dbf header (32 bytes for each field)
Byte
Size
Contents
Description
11
ASCI
field name, 0x00 termin
all
11
ASCI
field type (see next paragraph)
all
12
n,n,n,n
Fld address in memory
D3
n,n,0,0
offset from record begin
Fp
0,0,0,0
ignored
FS, D4, D5, Fb, CL
16
byte
Field length, bin (see next paragraph)
all \ FS,CL: for C field type
17
byte
decimal count, bin
all / both used for fld lng
18
0,0
reserved
all
20
byte
Work area ID
D4, D5
AT271_TRG_E6
Forsk 2009
21
23
FS, D3, Fb, Fp, CL
n,n
multi-user dBase
D3, D4, D5
0,0
ignored
FS, Fb, Fp, CL
0x01
Set Fields
D3, D4, D5
0x00
ignored
FS, Fb, Fp, CL
0...0
reserved
all
31
0x01
Field is in .mdx index
D4, D5
0x00
ignored
FS, D3, Fb, Fp, CL
Field type and size in the .dbf header, field descriptor (1 byte)
Size
Type
Description/Storage
C 1...n
Char
ASCII (OEM code page chars)

rest= space, not \0 term.
all
FS
Fp, CL
n = 1...254
all
D8
Date
8 ASCII digits (0...9) in the YYYYMMDD format
all
F 1...n
Numeric

variable pos. of float.point
n = 1...20
FS, D4, D5, Fp
N 1...n Numeric

fix posit/no float.point
all
n = 1...20
FS, Fp, CL
n = 1...18
D3, D4, D5, Fb
Logical
ASCII chars (YyNnTtFf space)
FS, D3, Fb, Fp, CL
ASCII chars (YyNnTtFf?)
D4, D5 (FS)
Memo
10 digits repres. the start block posit. in .dbt file, or 10 spaces if

no entry in memo
all
V 10
Variable
Variable, bin/asc data in .dbv

4bytes bin= start pos in memo
4bytes bin= block size
1byte = subtype
1byte = reserved (0x1a)
10 spaces if no entry in .dbv
FS
P 10
Picture
binary data in .ftp

structure like M
Fp
B 10
Binary
binary data in .dbt

structure like M
D5
G 10
General
OLE objects
structure like M
D5, Fp
22
short int
FS
44
long int
FS
88
double
binary signed double IEEE
FS
L1
M 10
Each DBF Record (Fixed Length)

Byte
Size
0
1n
3.18.1.2
unused
24
3.18.1.1.3
0x00
Description
deleted flag "*" or not deleted " "
all
x-times contents of fields, fixed length, unterminated.

For n, see (2) byte 1011
All
DBF File Content

The .dbf file provides information about the measured transmitters participating in the tuning.
Forsk 2009
Field
Type
Description
TX_NAME
Text
FILE_NAME
Text
Name (and optionally, path) of .pts file
AREA_XMIN
Float
Not used
AREA_XMAX
Float
Not used
AT271_TRG_E6
63
3.18.2
AREA_YMIN
Float
Not used
AREA_YMAX
Float
Not used
PTS File
The tuning file contains a header and the list of points.
The contents of the header is:
4 bytes : version
4 bytes : flag (can be used to manage flags like active flag)
50 bytes : GUID
4 bytes : Number of points
255 bytes : original measurements name (with prefix Num : for test mobile data and CW: for CW measurements)
256 bytes : comment
4 bytes : X_RADIUS
4 bytes : Y_RADIUS
4 bytes : Gain : measurement gain - losses
4 bytes : Global error
4 bytes : Rx height
4 bytes : Frequency
8 bytes : Tx Position
The list of points contains following 4-uplet for all points
3.19
4 bytes : X
4 bytes : Y
4 bytes : Measurement value
4 bytes : Incidence angle.
Interference Histograms File Formats

Interference histograms required by automatic frequency planning tools can be imported and exported.
Notes:
3.19.1
No validity check is carried out when importing an interference histogram file.
Atoll only imports interference histograms related to loaded transmitters.
The lines starting with the symbol "#" are considered as comments.
The interferer TRX type is not specified. In fact, the subcells of the interferer transmitter
differ by their power offsets. If the power offset of a subcell is X with respect to the BCCH,
then its interference C/I histogram will be shifted by X with respect to the BCCH
interference histogram. It contains no further information; therefore, the interferer TRX type
is always BCCH.
For each interfered subcell-interferer subcell pair, Atoll saves probabilities for several C/I
values (between 6 to 24 values). Five of these values are fixed; probabilities are calculated
for C/I values equal to 9, 1, 8, 14, and 22 dB. Then, between each fixed C/I value, there
can be up to three additional values (this number depends on the probability variation
between the fixed values). The C/I values have 0.5 dB accuracy and probability values are
calculated and stored with an accuracy of 0.002 for probabilities between 1 and 0.05, and
with an accuracy of 0.0001 for probabilities lower than 0.05.
If no power offset is defined on the Interfered TRX type, it is possible to use the "All" value.
The values of probability should be absolute (between 0 and 1), and not in precentage
(between 0 and 100%).
One Histogram per Line (.im0) Format

This file contains one histogram per line for each interfered/interfering subcell pair. The histogram is a list of C/I values
with associated probabilities.
The .im0 file consists of two parts:
The first part is a header used for format identification. It must start with and contain the following lines:
# Calculation Results Data File.

# Version 1.1, Tab separated format. Commented lines start with #.
The second part details interference histogram of each interfered subcell-interferer subcell pair.
The lines after the header are considered as comments if they start with the symbol "#". If not, they must have the following
format:
64
AT271_TRG_E6
Forsk 2009
<Column1><tab><Column2><tab><Column3><tab><Column4><newline>
The 4 tab-separated columns are defined in the table below:
3.19.1.1
Column name
Description
Column1
Interfered transmitter
Name of the interfered transmitter.
Column2
Interfering transmitter
Name of the interferer transmitter.
Column3
Interfered TRX type
Interfered subcell. In order to save storage, all subcells with no power

offset are not duplicated (e.g. BCCH, TCH).
Column4
C/I Probability
C/I value and the probability associated to this value separated by a space
character. This entry cannot be null.
Sample
# Remark:
C/I results do not incorporate power offset values.
# Fields are:
#-----------------------------------------------------------------------#Transmitter
Interferer
TRX type
{C/I Probability} values
#-----------------------------------------------------------------------#
# Warning, The parameter settings of this header can be wrong if
# the "export" is performed following an "import". They
# are correct when the "export" follows a "calculate".
#
# Service Zone Type is "Best signal level of the highest priority HCS layer".
# Margin is 5.
# Cell edge coverage probability 75%.
# Traffic spreading was Uniform
##---------------------------------------------------------------------#
#
Site0_2
Site0_1
BCCH,TCH -10 1 -9 0.996 -6 0.976 -4 0.964 -1 0.936

0 0.932 1 0.924 4 0.896 7 0.864 8 0.848
9 0.832 10 0.824 11 0.804 14 0.712 17 0.66
Site0_2
Site0_3
BCCH,TCH -10 1 -9 0.996 -6 0.976 -4 0.972 -1 0.948

0 0.94 1 0.928 4 0.896 7 0.856 8 0.84
11 0.772 13 0.688 14 0.636 15 0.608 18 0.556
Site0_3
Site0_1
BCCH,TCH -10 1 -9 0.996 -6 0.98 -3 0.948 0 0.932

1 0.924 4 0.892 7 0.852 8 0.832 9 0.816
10 0.784 11 0.764 14 0.644 15 0.616 18 0.564
Site0_3
Site0_2
BCCH,TCH -9 1 -6 0.972 -3 0.964 -2 0.96 0 0.94

1 0.932 4 0.904 7 0.876 8 0.86 9 0.844
11 0.804 13 0.744 14 0.716 15 0.692 18 0.644
3.19.2
One Value per Line with Dictionary File (.clc) Format

Atoll creates two ASCII text files in a specified directory: xxx.dct and xxx.clc (xxx is the user-specified name).
Note:
Forsk 2009
When importing interference histograms with standard format, you must specify the .clc file
to be imported. Atoll looks for the associated .dct file in the same directory and uses it to
decode transmitter identifiers. If this file is unavailable, Atoll assumes that the transmitter
identifiers are the transmitter names. In this case, the columns 1 and 2 of the .clc file must
contain the names of the interfered and interferer transmitters instead of their identification
numbers.
AT271_TRG_E6
65
3.19.2.1
CLC File
3.19.2.1.1
Description
The .clc file consists of two parts:

format:
<Column1><tab><Column2><tab><Column3><tab><Column4><tab><Column5><newline>
Column name
Description
Column1
Identification number of the interfered transmitter. If the column is empty,

its value is identical to the one of the line above.
Column2
Identification number of the interferer transmitter. If the column is null, its

value is identical to the one of the line above.
Column3
Interfered TRX type
Interfered subcell. If the column is null, its value is identical to the one of
the line above. In order to save storage, all subcells with no power offset
are not duplicated (e.g. BCCH, TCH).
Column4
C/I threshold
C/I value. This column cannot be null.
Column5
Probability C/I > Threshold
Probability to have C/I the value specified in column 4 (C/I threshold). This
field must not be empty.
Note:
3.19.2.1.2
The columns 1, 2, and 3 must be defined only in the first line of each histogram.
Sample
# Version 1.1,
# Remark:
Tab separated format. Commented lines start with #.
# Fields are:
##------------#------------#------------#-----------#------------------#
#| Interfered | Interfering| Interfered | C/I
#| Transmitter| Transmitter| Trx type
| Probability
| Threshold | C/I >= Threshold |
##------------#------------#------------#-----------#------------------#
#
#
# Margin is 5.
##---------------------------------------------------------------------#
1
TCH_INNER
8
9
66
1
0.944
10
0.904
11
0.892
14
0.844
AT271_TRG_E6
Forsk 2009
BCCH,TCH
15
0.832
16
0.812
17
0.752
22
0.316
25
0.292
0.944
10
.904
13
0.872
14
0.84
17
0.772
Note:
A new interference matrix histograms format has been introduced in Atoll 2.3.1 to improve
the import and export features and the overall performance. In this format, if the TCH and
BCCH histograms are the same, they are no longer duplicated. Atoll keeps a single record
of these histograms indicating that they belong to TCH and BCCH both. For example,
- Old format histograms between victim 1 and interferer 2:
1 2 TCH
-9.5
1 2 BCCH
-9.5
- 9
- 9
1
1
- 6
- 6
1
1
- New format histograms between victim 1 and interferer 2:
1 2 TCH,BCCH
3.19.2.2
DCT File
3.19.2.2.1
Description
-9.5
- 9
- 6
The .dct file is divided into two parts:
# Calculation Results Dictionary File.

The second part provides information about transmitters taken into account in AFP.
format:
<Column1><tab><Column2><newline>
Column name
Type
Description
Column1
Transmitter name
Text
Column2
Transmitter Identifier
Integer
Identification number of the transmitter
Column3
BCCH during calculation
Integer
BCCH used in calculations
Column4
BSIC during calculation
Integer
BSIC used in calculations
Column5
% of vic coverage
Float
Percentage of overlap of the victim service area
Column6
% of int coverage
Float
Percentage of overlap of the interferer service area
The last four columns describe the interference matrix scope. One transmitter per line is described separated with a tab
character.
3.19.2.2.2
Sample
# Calculation Results Dictionary File.
# Version 2.1,
Tab separated format. Commented lines start with #.
# Fields are:
##-----------#-----------#-----------#-----------#---------#---------#
#|Transmitter|Transmitter|BCCH during|BSIC during|% of vic'|% of int'|
#|Name
|Identifier |calculation|calculation|coverage |coverage |
##-----------#-----------#-----------#-----------#---------#---------#
Forsk 2009
AT271_TRG_E6
67
#
#
# Service Zone Type is "Best signal level per HCS layer".
# Margin is 5.
# Cell edge coverage probability is 75%.
# Traffic spreading was Uniform (percentage of interfered area)
##---------------------------#
3.19.3
Site0_0
-1
-1
100
100
Site0_1
-1
-1
100
100
Site0_2
-1
-1
100
100
Site1_0
-1
-1
100
100
Site1_1
-1
-1
100
100
Site1_2
-1
-1
100
100
Site2_0
-1
-1
100
100
Site2_1
-1
-1
100
100
One Value per Line (Transmitter Name Repeated) (.im1)

Format
This file contains one C/I threshold and probability pair value per line for each interfered/interfering subcell pair. The histogram is a list of C/I values with associated probabilities.
The .im1 file consists of two parts:

format:
<Column1><tab><Column2><tab><Column3><tab><Column4><tab><Column5><newline>
3.19.3.1
Column name
Description
Column1
Column2
Column3
Interfered TRX type
Interfered subcell. In order to save storage, all subcells with no power

offset are not duplicated (e.g. BCCH, TCH).
Column4
C/I threshold
C/I value. This column cannot be null.
Column5
Probability C/I > Threshold
Probability to have C/I the value specified in column 4 (C/I threshold). This
field must not be empty.
Sample
# Remark:
# Fields are:
#-----------------------------------------------------------------------#Transmitter
68
Interferer
TRX type
AT271_TRG_E6
C/I
Probability
Forsk 2009
#-----------------------------------------------------------------------#
#
# Margin is 5.
##---------------------------------------------------------------------#
Site0_2
Site0_1
BCCH,TCH
-10
Site0_2
Site0_1
BCCH,TCH
-9
0.996
Site0_2
Site0_1
BCCH,TCH
-6
0.976
Site0_2
Site0_1
BCCH,TCH
-4
0.964
Site0_2
Site0_1
BCCH,TCH
-1
0.936
Site0_2
Site0_1
BCCH,TCH
0.932
Site0_2
Site0_1
BCCH,TCH
0.924
Site0_2
Site0_1
BCCH,TCH
0.896
Site0_2
Site0_1
BCCH,TCH
0.864
Site0_2
Site0_1
BCCH,TCH
0.848
Site0_2
Site0_1
BCCH,TCH
0.832
Site0_2
Site0_1
BCCH,TCH
10
0.824
...
3.19.4
Only Co-Channel and Adjacent Values (.im2) Format

In this case, there is only one .im2 file containing co-channel and adjacent channel interference probabilities specified for
each interfered transmitter interferer transmitter pair. There is only one set of values for all the subcells of the interfered
transmitter.
Each line must have the following format:
<Column1><SEP><Column2><SEP><Column3><SEP><Column4><newline>
Where the separator (<SEP>) can either be a tab or a semicolon.
The four columns are defined in the table below:
Column name
Description
Column2
Column3
Co-channel interference
probability
Column4
Adjacent channel
interference probability
Column1
Probability of having C I
Probability of having C I
Max
BCCH ,TCH
Max
BCCH ,TCH
C I req
C I req F
C I req corresponds to the required C/I threshold. This parameter is defined for each subcell.
F is the adjacent channel protection level.
3.19.4.1
Sample
# Remark:
# Fields are:
#------------------------------------------------------------------------
Forsk 2009
AT271_TRG_E6
69
#Transmitter
Interferer
Co-channel
Adjacent channel
#-----------------------------------------------------------------------#
#
# Margin is 5.
##---------------------------------------------------------------------#
Site0_2
Site0_1
0.226667
0.024
Site0_2
Site0_3
0.27
0.024
Site0_3
Site0_1
0.276
0.02
Site0_3
Site0_2
0.226
0.028
The columns in the sample above are separated with a tab. These columns can also be separated with a semilcolon:
Site0_2;Site0_1;0.226667;0.024
Site0_2;Site0_3;0.27;0.024
Site0_3;Site0_1;0.276;0.02
Site0_3;Site0_2;0.226;0.028
70
AT271_TRG_E6
Forsk 2009
Chapter 4
Calculations
This chapter describes in detail the calculation of path losses, the propagation models implemented in Atoll
by default, the calculation of antenna attenuation according to antenna patterns, and other calculation
algorithms in Atoll.
Atoll
Atoll
Microwave
RF PlanningLink
andPlanning
Optimisation
Software
Microwave
Software
72
AT271_TRG_E6
Forsk 2009
Chapter 4: Calculations
Calculations
4.1
Overview
Three kinds of predictions are available in Atoll:
Point analysis enables you to visualise transmitter-receiver profile and to get predictions for a user-defined
receiver in real time anywhere on a geographic map (Point analysis window: Profile tab).
Coverage studies consider each bin of calculation areas as a potential receiver you can define. Therefore, covered
bins correspond to areas where a criterion on the predicted received signal is fulfilled.
Point analysis based on path loss matrices enables you to get parameters derived from predicted values in coverage studies (field received, path loss, C/I, UMTS parameters) for a receiver anywhere inside a calculation area
(Point analysis window: Reception, Interference, AS analysis tabs).
An overview of different analysis methods is presented in the table below:
Coverage studies
Point analysis
Point analysis based on path loss

matrices
Any study
Profile
Reception, Results,
Interference, AS analysis
Receiver
position
At the centre of each

calculation bin within
calculation areas
Anywhere. Even beyond

computation zone
Anywhere inside the calculation areas
Calculation
Path loss matrix

calculation
Real time
No calculation: result coming from path

loss matrices
Profile
extractiona
Radial except when

using SPM
Systematic
Method used for coverage studies: radial

except when using SPM
Result
One value inside a

calculation bin
Different values inside a

calculation bin
One value inside a calculation bin
a.
When using SPM, you can choose either radial or systematic calculation option.
Notes:
In coverage studies, Atoll calculates path loss for every bin within calculation areas.
However, only results on calculation bins inside the computation zone are displayed.
Profile point analysis is calculated in real time. Therefore, prediction is always consistent
with the network. On the other hand, if you modify any parameter (radio or geo), which may
make matrices invalid, consider updating the matrices before using point analysis based on
path loss matrices.
Due to different calculation methods, you can get different results at a same point when
performing a point analysis in profile or reception mode.
In any case, prediction is performed in three steps:

1st step: First of all, Atoll calculates the path loss ( L path ), using the selected propagation model.
L path = L model + L ant
Tx
+ L ant
Rx
L model is the loss on the transmitter-receiver path calculated through the propagation model. L model value depends on
the selected propagation model.
L ant
Tx
L ant
Rx
is the transmitter antenna attenuation (from antenna patterns).

is the receiver antenna attenuation ( L ant
Rx
= 0 ) (from antenna patterns).
Notes:
In any project, Atoll considers that the receiver antenna is in the transmitter antenna axis.
Therefore, the receiver antenna attenuation is supposed to be zero.
Transmitter antenna attenuation may not be considered in this step. It depends on

propagation model provider, who may choose to include this parameter in L path
calculation. However, all the propagation models available in Atoll calculate L path by
considering transmitter antenna attenuation.
2nd step: When the option Shadowing taken into account is selected, Atoll evaluates a shadowing margin,
M Shadowing model , from the user-defined model standard deviation at the receiver and the cell edge coverage probability.
Forsk 2009
AT271_TRG_E6
73

Note:
For a cell edge coverage probability of 50%, the shadowing margin is always zero. In this
case, Atoll still works as above.
3rd step: Then, Atoll determines the prediction criterion and displays coverage.
For a signal level study,
The signal level at the receiver ( P Rec ) is calculated. We have (in dBm):
P Rec = EIRP L path M Shadowing model L Indoor + G ant
Where EIRP = P Tx + G ant
Tx
Rx
L Rx
L Tx
EIRP is the effective isotropic radiated power of the transmitter.

P Tx is the transmitter power.
G ant
Tx
is the transmitter antenna gain.
L Tx are transmitter losses.

M Shadowing model is the shadowing margin.
L Indoor are the indoor losses.These losses are defined for each clutter class.They are taken into account when the option
Indoor coverage is selected,
L Rx are receiver losses.
G ant
Rx
is the receiver antenna gain.

Notes:
In UMTS, CDMA2000 and IS95-CDMA documents, P Tx = P Pilot and L Tx = L total DL .
In UMTS, CDMA2000 and IS95-CDMA documents, Atoll considers that G ant
Rx
and L Rx
equal zero when calculating the received signal level (in point analysis, Profile and
Reception tabs, and in common coverage studies such as Coverage per transmitter,
Coverage by field level, Overlapping).
In GSM_EGPRS documents, L Tx = L total DL .
In GSM_EGPRS documents, receiver is equipped with an antenna with zero gain.
The prediction is performed for a user-defined cell edge coverage probability (x%). This means that the measured criterion
exceeds the predicted criterion for x% of time. The prediction is reliable during x% of time.
Note:
4.2
In case of interference studies, only signal from interfered transmitter (C) is downgraded by
the shadowing margin. We consider that interference value (I) is not altered by the
shadowing margin.
Path Loss Matrices

Atoll is able to calculate two path loss matrices per transmitter, a first matrix over a smaller radius computed with a high
resolution and a propagation model (main matrix), and a second matrix over a larger radius computed with a low resolution
and another propagation model (extended matrix).
To be considered for calculations, a transmitter must fulfil the following conditions:
It must be active,
It must satisfy filter criteria defined in the Transmitters folder, and
It must have a calculation area.
In the rest of the document, a transmitter fulfilling the conditions detailed above will be called TBC transmitter.
The path loss matrix size of a TBC transmitter depends on its calculation area. Atoll determines a path loss value ( L path )
on each calculation bin (calculation bin is defined by the resolution) of the calculation area of the TBC transmitter. You may
have one or two path loss matrices per TBC transmitter.
74
AT271_TRG_E6
Forsk 2009
4.2.1
Calculation Area Determination
4.2.1.1
Computation Zone
Transmitter calculation area is made of a rectangle or a square depending on transmitter calculation radius and the computation zone.
Calculation radius enables Atoll to define a square around the transmitter. One side of the square equals twice the entered
calculation radius.
Since the computation zone can be made of one or several polygons, transmitter calculation area corresponds to the intersection area between its calculation square and the rectangle containing the computation zone area(s).
Figure 4.1: Example 1: Single Calculation Area
Figure 4.2: Example 2: Multiple Calculation Areas

Computation zone(s)
Rectangle containing the computation zone(s)
Calculation area defined (square)
Transmitter
Calculation area: real area for which Atoll calculates path losses
4.2.1.2
Use of Polygonal Zones in Coverage Prediction Reports

Prediction statistics are evaluated over the focus zone, if existing, then over the computation zone, if existing, or over the
whole covered area. The area of the focus and computation zones are calculated by decomposition in triangles.
The area of each prediction is calculated by counting its pixels inside the focus (resp. computation) zone. This number of
pixels multiplied by the area of one of its pixels gives the total area.
This area depends on the study resolution. At the border of the focus (resp. computation) zone, pixels are considered
either IN or OUT of the zone. A pixel is IN if its centre is inside the focus zone.
If a prediction covers the entire focus (resp. computation) zone, its area should be equal to the focus (resp. computation)
zone area, but as these 2 different methods differ, the results may be slightly different. If it happens that the value of the
prediction area is higher than the focus zone area, then the calculated percentage value is higher than 100%. In that case,
Atoll automatically replaces it by 100%.
Forsk 2009
AT271_TRG_E6
75
4.2.2
Calculate / Force Calculation Comparison
4.2.2.1
Calculate
The Calculate feature (F7) enables you:
1. To calculate prediction studies
The first time you click Calculate (no path loss matrices exist), Atoll computes path loss matrices for each TBC
transmitter. Then, it calculates created and unlocked coverage prediction studies inside the computation zone.
2. To check result validity and update calculations
If calculations have been performed once and you have changed some parameters such as radio data or calculation area, Atoll automatically detects path loss matrices to be recalculated. These are either one or several path
loss matrices that become invalid due to certain modifications. Then Atoll calculates the prediction study, or just
the prediction study if matrices were all still valid.
4.2.2.2
Force Calculation
With the Force calculation feature (Ctrl+F7), Atoll deletes all the path loss matrices even if they are valid, recalculates them
and then updates the results of prediction studies.
Note:
4.2.3
Geographic data (DTM, clutter) modification makes path loss matrices invalid. However,
Atoll does not detect this invalidity just by using Calculate. Therefore, to update
calculations, you must click the Force calculation command.
Matrix Validity
Atoll manages path loss matrix validity transmitter by transmitter, even in case of transmitters with two path loss matrices
(main and extended matrices). Therefore, even if only one path loss matrix of the transmitter is invalid, Atoll will recalculate
both of them. All the geographic data modifications and some radio data changes can make matrices invalid. This table
lists these modifications and also changes that have an impact only on prediction studies.
Modification
Matrix validity
Impact on
Calculate
Force
calculation
Frequency
Invalid
Path loss matrices
Sufficient
Not necessary
Antenna* coordinates (site coordinate:

X and Y, Dx and Dy)
Invalid
Path loss matrices
Sufficient
Not necessary
Antennaa height
Invalid
Path loss matrices
Sufficient
Not necessary
Invalid
Path loss matrices
Sufficient
Not necessary
Invalid
Path loss matrices
Sufficient
Not necessary
Antenna pattern
Downtilt
Invalid
Path loss matrices
Sufficient
Not necessary
% Power (when there is other

antennas)
Invalid
Path loss matrices
Sufficient
Not necessary
Site position/altitude
Invalid
Path loss matrices
Sufficient
Not necessary
Grid resolution (main or/and

extended)
Invalid
Path loss matrices
Sufficient
Not necessary
Propagation model (main or/and

extended)
Invalid
Path loss matrices
Sufficient
Not necessary
Propagation model parameters
Invalid
Path loss matrices
Sufficient
Not necessary
Calculation areas
1. Calculation areas gets smaller
Valid
Prediction study
Sufficient
Not necessary
Calculation areas
2. Calculation areas gets larger
Invalid
Path loss matrices
Sufficient
Not necessary
Receiver height
Invalid
Path loss matrices
Sufficient
Not necessary
Receiver losses
Valid
Prediction study
Sufficient
Not necessary
Receiver gain
Valid
Prediction study
Sufficient
Not necessary
Prediction study
Sufficient
Not necessary
Azimuth
Receiver antenna
Rx
= 0
Geographic layer order
Invalid
Path loss matrices Insufficientb
Necessary
Geographic file resolution
Invalid
Necessary
Invalid
Insufficientb
Necessary
New DTM map
76
Valid because L ant
AT271_TRG_E6
Path loss matrices
Forsk 2009
New clutter class edition
Invalid
Coverage study resolution
Valid
Prediction study
Sufficient
Not necessary
Cell edge coverage probability
Valid
Prediction study
Sufficient
Not necessary
Coverage study conditions
Valid
Prediction study
Sufficient
Not necessary
Coverage study display options
Valid
Prediction study
Sufficient
Not necessary
Necessary
a.Modification of any parameter related to main or other antennas makes matrix invalid.
b.Except if this action has an impact on the site positions/altitudes.
Tip 1
Calculate or Force Calculation?
If you modify radio data or calculation areas, use the Calculate button. On the other hand, if you change geographic
data, it is necessary to use Force calculation.
Tip 2
Calculation area management
When performing prediction studies, it is recommended to follow this methodology to minimise recalculations:
1st step: Calculate without computation zone.
2nd step: Draw a computation zone and calculate.
3rd step: Decrease the calculation radius and calculate.
4.3
Path Loss Calculations
4.3.1
Ground Altitude Determination

Atoll determines reception and transmission site altitude from Digital Terrain Model map. The method used to evaluate site
altitude is based on a bilinear interpolation. It is described below.
Let us suppose a site S located inside a bin. Atoll knows the altitudes of four bin vertices, S1, S1, S2 and S2, from the
DTM file (Centre of each DTM pixel).
Figure 4.3: Ground Altitude Determination - 1

1st step: Atoll draws a vertical line through S. This line respectively intersects (S1,S1) and (S2, S2) lines at S1 and S2.

2nd step: Atoll determines the S1 and S2 altitudes using a linear interpolation method.
Forsk 2009
AT271_TRG_E6
77

3rd step: Atoll performs a second linear interpolation to evaluate the S altitude.
4.3.2
Clutter Determination
Some propagation models need clutter class and clutter height as information at receiver or along a transmitter-receiver
profile.
4.3.2.1
Clutter Class
Atoll uses clutter classes file to determine the clutter class.
4.3.2.2
Clutter Height
To evaluate the clutter height, Atoll uses clutter heights file if available in the .atl document; clutter height of a site is the
height of the nearest point in the file.
Example: Let us suppose a site S. In the clutter heights file, Atoll reads clutter heights of four points around the site, S1,
S1, S2 and S2. Here, the nearest point to S is S2; therefore Atoll takes the S2 clutter height as clutter height of S.
Figure 4.7: Clutter Height

If you do not have any clutter height file, Atoll takes clutter height information in clutter classes file. In this case, clutter
height is an average height related to a clutter class.
4.3.3
Geographic Profile Extraction

Geographic profile extraction is needed in order to calculate diffraction losses. Profiles can be based on DTM only or on
DTM and clutter both. In fact, it depends on the selected propagation model.
4.3.3.1
Extraction Methods
4.3.3.1.1
Radial Extraction
Atoll draws radials from the site (where transmitter is located) to each calculation bin located along the transmitter calculation area border. In other words, Atoll determines a geographic profile between site and each bin centre.
78
AT271_TRG_E6
Forsk 2009
Figure 4.8: Radial calculation method

Transmitter
Radial: Atoll will extract a geographic profile for each radial
Centre of a bin located on the calculation border
Receiver: it may be anywhere in point analysis or at the centre of each calculation bin in coverage studies
Figure 4.9: Site-bin centre profile

The receiver may be located either anywhere within a calculation bin (Point prediction) or at the centre of a calculation bin
(Coverage study). Therefore, according to the receiver position, Atoll chooses the nearest profile and uses it (receiver is
considered as located on the profile) to perform prediction study at the receiver.
4.3.3.1.2
Systematic Extraction
In this case, Atoll systematically extracts a geographic profile between the site (where transmitter resides) and the receiver.
Forsk 2009
AT271_TRG_E6
79
Figure 4.10: Radial calculation method

Transmitter
Geographic profiles
Receiver: it may be anywhere in point analysis or at the centre of each calculation bin in coverage studies
4.3.3.2
Profile Resolution: Multi-Resolution Management

Geographic profile resolution depends on resolution of geographic data used by the propagation model (DTM and/or clutter).
1. 1st case: If the chosen propagation model considers both DTM and clutter heights along the profile, the profile
resolution will be the highest of the two.
Example 1: Standard Propagation Model is used to perform predictions. A DTM map with a 40 m resolution and
a clutter heights map with a 20 m resolution are available.
Both DTM and clutter maps are considered when using the Standard propagation model. Therefore, here, the
profile resolution will be 20 m. It means that Atoll will extract geographic information, ground altitude and clutter
height, every 20 m. To get ground altitude every 20m, Atoll uses the bilinear interpolation method described in
"Ground Altitude Determination" on page 77. Clutter heights are read from the clutter heights map. Atoll takes the
clutter height of the nearest point every 20m (see Path loss calculations: Clutter determination).
Example 2: Standard Propagation Model is used to perform predictions. A DTM map with a 40 m resolution and
a clutter classes map with a 20 m resolution are available. No clutter height file has been imported in .atl document.
Both DTM and clutter maps are considered when using the Standard propagation model. Therefore, here, the
profile resolution will be 20 m. It means that Atoll will extract geographic information, ground altitude and clutter
height, every 20 m. To get ground altitude every 20 m, Atoll uses the bilinear interpolation method described in
"Ground Altitude Determination" on page 77. Atoll uses the clutter classes map to determine clutter height. Every
20 m, it determines clutter class and takes associated average height.
2. 2nd case: If the chosen propagation model takes into account only DTM map along the profile, profile resolution
will be the highest resolution among the DTM files.
Example: Cost-Hata is used to perform predictions. Both DTM maps with 40 m and 25 m resolutions and a clutter
map with a 20 m resolution are available.
80
AT271_TRG_E6
Forsk 2009
Explorer window
DTM
DTM 1 (25m)
DTM 2 (40m)
Clutter
Clutter (20m)
Work space
Only DTM maps are considered along the whole profile when using Cost-Hata model. Therefore, here, the profile
resolution will be 25 m. It means that Atoll will extract geographic information, only the ground altitude, every 25 m.
DTM 1 is on the top of DTM 2. Thus, Atoll will consider ground elevation read from DTM 1 in the definition area of
DTM 1 and DTM 2 elsewhere. To get ground altitude every 25 m, Atoll uses the bilinear interpolation method
described in "Ground Altitude Determination" on page 77.
Notes:
Forsk 2009
The selected profile resolution does not depend on the geographic layer order. In the last
example, whatever the DTM file order you choose, profile resolution will always be 25m.
On the other hand, the geographic layer order will influence the usage of data to establish
the profile.
The calculation bin of path loss matrices defined by the grid resolution is independent of
geographic file resolution.
AT271_TRG_E6
81
82
AT271_TRG_E6
Macro cell
Rooftop
Fixed
Cell size
Receiver
location
Receiver
Broadcast
Profile
extraction
mode
Use
Diffraction
calculation
method
Profile
based on
Free space loss +

Corrections
Free space loss

Corrected
standard
loss
Physical
phenomena
Fixed
WLL
Broadcast
Land and maritime
Mobile
Street
Macro cell
Radial
DTM
WLL
WiMAX
Fixed
Street
Rooftop
Radial
DTM
Clutter
Deygout
(3 obstacles)
Mobile
GSM900
CDMA2000
Mobile and Fixed

GSM900
GSM1800
UMTS
CDMA2000
WiMAX
Street
WiMAX in Urban
and Suburban
Fixed
Street
Macro cell
Mini cell
Macro cell
Mini cell
Macro cell
Mini cell
Street
Rooftop
Radial
DTM
Radial
DTM
Deygout
(1 obstacle)
L(d, f, HTx, HRx)

(per environment)
Diffraction loss
1900-6000 MHz
GSM900
GSM1800
UMTS
CDMA2000
Mobile
Street
Macro cell
Mini cell
Radial
DTM
Deygout
(1 obstacle)
L(d, f, HRx)
(per environment)
Diffraction loss
150-2000 MHz
Erceg-Greenstein
COST-Hata
(SUI)
Okumura-Hata
Radial
Systematic
DTM
Clutter
Deygout
(1 obstacle)
Deygout (3 obstacles)
Epstein-Peterson (3 obstacles)
Deygout corrected (3 obstacles)
Millington (1 obstacle)
300-1500 MHz
ITU 529-3
Deygout
(3 obstacles)
Deygout corrected
(3 obstacles)
150-3500 MHz
Standard Propagation Model
L(d, f, HRx)
Free space loss
L(d, HTxeff, HRxeff, Diff loss, clutter) (per environment)
Diffraction loss
Diffraction loss
30-10000 MHz
WLL
Free space loss

Diffraction loss
30-10000 MHz
ITU 526-5
Mobile
Rooftop
Macro cell
30-3000 MHz
100-400 MHz
Frequency
band
ITU 1546
ITU 370-7
(Vienna 93)
4.4
Propagation
model
Propagation Models
Propagation models available in Atoll are listed in the table below along with their main characteristics.
Forsk 2009
Notes:
In formulas described above, L model is stated in dB.
Under Physical phenomena, L(...) expressions refer to formulas customisable in Atoll.
SUI stands for Stanford University Interim models.
4.4.1
Okumura-Hata and Cost-Hata Propagation Models
4.4.1.1
Hata Path Loss Formula

Hata formula empirically describes the path loss as a function of frequency, receiver-transmitter distance and antenna
heights for an urban environment. This formula is valid for flat, urban environments and 1.5 metre mobile antenna height.
Path loss (Lu) is calculated (in dB) as follows:
Lu = A 1 + A 2 log f + A 3 log h Tx + B 1 + B 2 log h Tx log d
f is the frequency (MHz).
hTx is the transmitter antenna height above ground (m) (Hb notation is also used in Atoll).
d is the distance between the transmitter and the receiver (km).
The parameters A1, A2, A3, B1 and B2 can be user-defined. Default values are proposed in the table below:
Parameters
Okumura-Hata
f 1500 MHz
Cost-Hata
f > 1500 MHz
A1
69.55
46.30
A2
26.16
33.90
A3
-13.82
-13.82
B1
44.90
44.90
B2
-6.55
-6.55
Default Hata parameters
4.4.1.2
Corrections to the Hata Path Loss Formula

As described above, the Hata formula is valid for urban environment and a receiver antenna height of 1.5m. For other environments and mobile antenna heights, corrective formulas must be applied.
L model1 = Lu a h Rx for large city and urban environments
f 2
L model1 = Lu a h Rx 2 log ------ 5.4 for suburban area
28
L model1 = Lu a h Rx 4.78 log f + 18.33 log f 40.94 for rural area

a(hRx) is a correction for a receiver antenna height different from 1.5m.
Environment
a(Hr)
Rural/Small city
1.1 log f 0.7 h Rx 1.56 log f 0.8
Large city
3.2 log 11.75h Rx 4.97
Note:
4.4.1.3
When receiver antenna height equals 1.5m, a(hRx) is close to 0 dB regardless of

frequency.
Calculations in Atoll
Hata models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the receiver.
1st step: For each calculation bin, Atoll determines the clutter bin on which the receiver is located. This clutter bin corresponds to a clutter class. Then, it uses the Hata formula assigned to this clutter class to evaluate L model1 .
2nd step: This step depends on whether the Add diffraction loss option is checked.
If the Add diffraction loss option is unchecked, Atoll stops calculations.

L model = L model1
Forsk 2009
AT271_TRG_E6
83
If the Add diffraction loss option is selected, Atoll proceeds as follows:

a. It extracts a geographic profile between the transmitter and the receiver based on the radial calculation mode.
b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates losses due to diffraction L model2 .
L model = L model1 + L model2
Note:
Like for any Hata-based model, L model is, by default, limited to the computed free space
loss value. It is also possible to avoid this option (option in the related scrolling menu of
Configuration tab).
4.4.2
ITU 529-3 Propagation Model
4.4.2.1
ITU 529-3 Path Loss Formula

The ITU 529.3 model is a Hata-based model. For this reason, its formula empirically describes the path loss as a function
of frequency, receiver-transmitter distance and antenna heights for a urban environment. This formula is valid for flat,
urban environments and 1.5 metre mobile antenna height.
The standard ITU 529-3 formula, for a receiver located on a urban environment, is given by:
E = 69.82 6.16 log f + 13.82 log h Tx 44.9 6.55 log h Tx log d
where:
E is the field strength for 1 kW ERP
f is the frequency (MHz).
h Tx is the transmitter antenna height above ground (m) (Hb notation is also used in Atoll)
h Rx is the receiver antenna height above ground (m)
d is the distance between the transmitter and the receiver (km)
b is the distance correction
The domain of validity of such is formula is:
Frequency range: 300-1500 MHz

Base Station height: 30-200 m
Mobile height: 1-10 m
Distance range: 1-100 km
Since Atoll needs the path loss (Lu) formula, a conversion has to be made. One can find the following conversion formula:
Lu = 139.37 + 20 log f E
which gives the following path loss formula for the ITU 529-3 model:
Lu = 69.55 + 26.16 log f 13.82 log h Tx + 44.9 6.55 log h Tx log d
4.4.2.2
Corrections to the ITU 529-3 Path Loss Formula
4.4.2.2.1
Environment Correction
As described above, the Hata formula is valid for urban environment. For other environments and mobile antenna heights,
corrective formulas must be applied.
L model1 = Lu a h Rx for large city and urban environments
2
f
L model1 = Lu a h Rx 2 log ------ 5.4 for suburban area
28
L model1 = Lu a h Rx 4.78 log f + 18.33 log f 40.94 for rural area
4.4.2.2.2
Area Size Correction

In the formulas above, a h Rx is the environment correction and is defined according to the area size
84
Environment
a(Hr)
Rural/Small city
1.1 log f 0.7 h Rx 1.56 log f 0.8
Large city
3.2 log 11.75h Rx 4.97
AT271_TRG_E6
Forsk 2009
4.4.2.2.3
Distance Correction
The distance correction refers to the term b above.
Distance
d<20 km
1
4
3
d 0.8
b = 1 + 0.14 + 1.87 10 f + 1.07 10 h' Tx log ------
20
d>20 km
4.4.2.3
h Tx
where h' Tx = -------------------------------------------6 2
1 + 7 10 h Tx
Hata-based models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the
receiver.
1st step: For each calculation bin, Atoll determines the clutter bin on which the receiver is located. This clutter bin corresponds to a clutter class. Then, it uses the ITU 529-3 formula assigned to this clutter class to evaluate L model1 .
2nd step: This step depends on whether the Add diffraction loss option is checked.
If the Add diffraction loss option is unchecked, Atoll stops calculations.

L model = L model1

a. It extracts a geographic profile between the transmitter and the receiver based on the radial calculation mode.
b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates losses due to diffraction L model2 .
L model = L model1 + L model2
Note:
Configuration tab)
4.4.3
Standard Propagation Model (SPM)
4.4.3.1
SPM Path Loss Formula

SPM is based on the following formula:
L model = K 1 + K 2 log d + K 3 log H Txeff + K 4 DiffractionLoss + K 5 log d log H Txeff +
K 6 H Rxeff + K 7 log H Rxeff + K clutter f clutter
with,
K1: constant offset (dB).
K2: multiplying factor for log(d).
d: distance between the receiver and the transmitter (m).
K3: multiplying factor for log(HTxeff).
HTxeff: effective height of the transmitter antenna (m).
K4: multiplying factor for diffraction calculation. K4 has to be a positive number.
Diffraction loss: loss due to diffraction over an obstructed path (dB).
K5: multiplying factor for log d log H Txeff
K6: multiplying factor for H Rxeff .
K7: multiplying factor for log H Rxeff .
H Rxeff : effective mobile antenna height (m).
Kclutter: multiplying factor for f(clutter).
f(clutter): average of weighted losses due to clutter.
Forsk 2009
AT271_TRG_E6
85
4.4.3.2
4.4.3.2.1
Visibility and Distance Between Transmitter and Receiver

For each calculation bin, Atoll determines:
The distance between the transmitter and the receiver.
If the distance Tx-Rx is less than the maximum user-defined distance (break distance), the receiver is considered to be
near the transmitter. Atoll will use the set of values marked Near transmitter.
If the distance Tx-Rx is greater than the maximum distance, receiver is considered far from transmitter. Atoll will use the
set of values Far from transmitter.
Whether the receiver is in the transmitter line of sight or not.
If the receiver is in the transmitter line of sight, Atoll will take into account the set of values (K1,K2)LOS.
If the receiver is not in the transmitter line of sight, Atoll will use the set of values (K1,K2)NLOS.
4.4.3.2.2
Effective Transmitter Antenna Height

Effective transmitter antenna height (HTxeff) may be calculated with six different methods.
Height Above Ground

The transmitter antenna height is above the ground (HTx in m).
HTxeff = HTx
Height Above Average Profile

The transmitter antenna height is determined relative to an average ground height calculated along the profile between a
transmitter and a receiver. The profile length depends on distance min and distance max values and is limited by the transmitter and receiver locations. Distance min and Distance max are minimum and maximum distances from the transmitter
respectively.
H Txeff = H Tx + H 0Tx H 0
where,
H 0Tx is the ground height (ground elevation) above sea level at transmitter (m).
H 0 is the average ground height above sea level along the profile (m).
Note:
If the profile is not located between the transmitter and the receiver, HTxeff equals HTx only.
Slope at Receiver Between 0 and Minimum Distance

The transmitter antenna height is calculated using the ground slope at receiver.
H Txeff = H Tx + H 0Tx H 0Rx + K d
where,
H 0Rx is the ground height (ground elevation) above sea level at receiver (m).
K is the ground slope calculated over a user-defined distance (Distance min). In this case, Distance min is a distance from
receiver.
Notes:
If H Txeff 20m then, Atoll uses 20m in calculations.
If H Txeff 200m then, Atoll takes 200m.
Spot Ht
If H 0Tx H 0Rx then, H Txeff = H Tx + H 0Tx H 0Rx
If H 0Tx H 0Rx then, H Txeff = H Tx
86
AT271_TRG_E6
Forsk 2009
Absolute Spot Ht
H Txeff = H Tx + H 0Tx H 0Rx
Note:
Distance min and distance max are set to 3000 and 15000 m according to ITU
recommendations (low frequency broadcast f < 500 Mhz) and to 0 and 15000 m according
Okumura recommendations (high frequency mobile telephony).
These values are only used in the two last methods and have different meanings according to the method.
Enhanced Slope at Receiver

Atoll offers a new method called Enhanced slope at receiver to evaluate the effective transmitter antenna height.
Figure 4.11: Enhanced Slope at Receiver

Let x-axis and y-axis respectively represent positions and heights. We assume that x-axis is oriented from the transmitter
(origin) towards the receiver.
This calculation is achieved in several steps:
1st step: Atoll determines line of sight between transmitter and receiver.
The LOS line equation is:
H 0Tx + H Tx H 0Rx + H Rx
Los i = H 0Tx + H Tx ------------------------------------------------------------------------------- Res i
d
where,
H Rx is the receiver antenna height above the ground (m).
i is the point index.
Res is the profile resolution (distance between two points).
2nd step: Atoll extracts the transmitter-receiver terrain profile.
3rd step: Hills and mountains are already taken into account in diffraction calculations. Therefore, in order for them not to
unfavourably influence the regression line calculation, Atoll filters the terrain profile.
Atoll calculates two filtered terrain profiles; one established from the transmitter and another from the receiver. It determines filtered height of every profile point. Profile points are evenly spaced on the basis of profile resolution. To determine
filtered terrain height at a point, Atoll evaluates ground slope between two points and compares it with a threshold set to
0.05; where three cases are possible.
Some notations defined hereafter are used in next part.
H filt is the filtered height.
H orig is the original height. Original terrain height is determined from extracted ground profile.
- Filter starting from transmitter
Let us assume that H filt Tx Tx = H orig Tx
For each point, we have three different cases:
H orig i H orig i 1
1st case: If H orig i H orig i 1 and ------------------------------------------------------ 0.05 ,
Res
Then, H filt Tx i = H filt Tx i 1 + H orig i H orig i 1
Forsk 2009
AT271_TRG_E6
87

H orig i H orig i 1
2nd case: If H orig i H orig i 1 and ------------------------------------------------------ 0.05
Res
Then, H filt Tx i = H filt Tx i 1
3rd case: If H orig i H orig i 1
Then, H filt Tx i = H filt Tx i 1
If H filt i H orig i additionally
Then, H filt Tx i = H orig i
- Filter starting from receiver
Let us assume that H filt Rx = H orig Rx
For each point, we have three different cases:
H orig i H orig i + 1
1st case: If H orig i H orig i + 1 and ------------------------------------------------------- 0.05 ,
Res
Then, H filt Rx i = H filt Rx i + 1 + H orig i H orig i + 1
H orig i H orig i + 1
2nd case: If H orig i H orig i + 1 and ------------------------------------------------------- 0.05
Res
Then, H filt Rx i = H filt Rx i + 1
3rd case: If H orig i H orig i + 1
Then, H filt Rx i = H filt Rx i + 1
If H filt i H orig i additionally
Then, H filt Rx i = H orig i
Then, for every point of profile, Atoll compares the two filtered heights and chooses the higher one.
H filt i = max H filt Tx i H filt Rx i
4th step: Atoll determines the influence area, R. It corresponds to the distance from receiver at which the original terrain
profile plus 30 metres intersects the LOS line for the first time (when beginning from transmitter).
The influence area must satisfy additional conditions:
R 3000m ,
R 0.01 d ,
R must contain at least three bins.
Notes:
When several influence areas are possible, Atoll chooses the highest one.
If d < 3000m, R = d.
5th step: Atoll performs a linear regression on the filtered profile within R in order to determine a regression line.
The regression line equation is:
y = ax + b
d i dm Hfilt i Hm
i
a = ----------------------------------------------------------------------- and b = H m ad m
2
d i dm
where,
1
H m = --n
Hfilt i
i
i is the point index. Only points within R are taken into account.
R
d m = d ---2
d(i) is the distance between i and the transmitter (m).
Then, Atoll extends the regression line to the transmitter location. Therefore, its equation is:
88
AT271_TRG_E6
Forsk 2009
regr i = a i Res + b
6th step: Then, Atoll calculates effective transmitter antenna height, H Txeff (m).
H 0Tx + H Tx b
H Txeff = ------------------------------------2
1+a
If HTxeff is less than 20m, Atoll recalculates it with a new influence area, which begins at transmitter.
Notes:
In case H Txeff 1000m , 1000m will be used in calculations.
If H Txeff is still less than 20m, an additional correction is taken into account (7th step).
7th step: If H Txeff is still less than 20m (even negative), Atoll evaluates path loss using H Txeff = 20m and applies a
correction factor.
Therefore, if H Txeff 20m ,
L model = L model H Txeff = 20m d f + K lowant
20 1 H Txeff 20
d
where, K lowant = --------- 0.3 H Txeff 20 -----------------------------------------------------------------------------5
d - 6.93 + -----------d -
9.63 + -----------10
1000
1000
4.4.3.2.3
Effective Receiver Antenna Height

H Rxeff = H Rx + H 0Rx H 0Tx
where,
H Rx is the receiver antenna height above the ground (m).
H 0Rx is the ground height (ground elevation) above sea level at the receiver (m).
H 0Tx is the ground height (ground elevation) above sea level at the transmitter (m).
Note:
The calculation of effective antenna heights ( H Rxeff and H Txeff ) is based on extracted
DTM profiles. They are not properly performed if you have not imported heights (DTM file)
beforehand.
4.4.3.2.4
Correction for Hilly Regions in Case of LOS

An optional corrective term enables Atoll to correct path loss for hilly regions when the transmitter and the receiver are in
Line-of-sight.
Therefore, if the receiver is in the transmitter line of sight and the Hilly terrain correction option is active, we have:
L model = K 1 LOS + K 2 LOS log d + K 3 log H Txeff + K 5 log H Txeff log d + K 6 H Rx + K clutter f clutter + K hill LOS
When the transmitter and the receiver are not in line of sight, the path loss formula is:
L model =K 1 NLOS + K 2 NLOS log d + K 3 log H Txeff + K 4 Diffraction + K 5 log H Txeff log d + K 6 H Rx + K clutter f clutter
K hill LOS is determined in three steps. Influence area, R, and regression line are supposed available.
1st step: For every profile point within influence area, Atoll calculates height deviation between the original terrain profile
and regression line. Then, it sorts points according to the deviation and draws two lines (parallel to the regression line),
one which is exceeded by 10% of the profile points and the other one by 90%.
2nd step: Atoll evaluates the terrain roughness, h; it is the distance between the two lines.
3rd step: Atoll calculates K hill LOS .
We have K hill LOS = K h + K hf
If 0 h 20m , K h = 0
2
Else K h = 7.73 log h 15.29 log h + 6.746

If 0 h 10m , K hf = 2 0.1924 H 0Rx + H Rx regr i Rx
H 0Rx + H Rx regr i Rx
2
Else K hf = 2 1.616 log h + 14.75 log h 11.21 -----------------------------------------------------------h
Forsk 2009
AT271_TRG_E6
89

iRx is the point index at receiver.
4.4.3.2.5
Diffraction
Four methods are available to calculate diffraction loss over the transmitter-receiver profile. They are detailed in the
Appendices.
Along the transmitter-receiver profile, you may consider:
4.4.3.2.6
Either ground altitude and clutter height (Consider heights in diffraction option),
In this case, Atoll uses clutter height information from clutter heights file if available in the .atl document. Otherwise,
it considers average clutter height specified for each clutter class in the clutter classes file description.
Or only ground altitude.
Losses due to Clutter

n
Atoll calculates f(clutter) over a maximum distance from receiver: f clutter =
Li wi
i=1
where,
L: loss due to clutter defined in the Clutter tab by the user (in dB).
w: weight determined through the weighting function.
n: number of points taken into account over the profile. Points are evenly spaced depending on the profile resolution.
Four weighting functions are available:
1
Uniform weighting function: w i = --n
di
Triangular weighting function: w i = ------------n
dj
j=1
d i = D d' i , where di is the distance between the receiver and the ith point and D is the maximum distance
defined.
d
log ----i + 1
D
Logarithmic weighting function: w i = ------------------------------------n

d
log ----j + 1
D
j=1
di
---D
e 1
Exponential weighting function: w i = -------------------------n
dj
---D
j=1
The chart below shows the weight variation with the distance for each weighting function.
Figure 4.12: Losses due to Clutter
90
AT271_TRG_E6
Forsk 2009
4.4.3.2.7
Recommendations
Beware that the clutter influence may be taken into account in two terms, Diffraction loss and f(clutter) at the same time.
To avoid this, we advise:
1. Not to consider clutter heights to evaluate diffraction loss over the transmitter-receiver profile if you specify losses
per clutter class.
This approach is recommended if the clutter height information is statistical (clutter roughly defined, no altitude).
Or
2. Not to define any loss per clutter class if you take clutter heights into account in the diffraction loss.
In this case, f(clutter)=0. Losses due to clutter are only taken into account in the computed Diffraction loss term.
This approach is recommended if the clutter height information is either semi-deterministic (clutter roughly
defined, altitude defined with an average height per clutter class) or deterministic (clutter sharply defined, altitude
defined with an average height per clutter class or - even better - via a clutter height file).
In case of semi-deterministic clutter information, specify receiver clearance (m) per clutter class. Both ground altitude and
clutter height are considered along the whole transmitter-receiver profile except over a specific distance around the
receiver (clearance), where Atoll proceeds as if there was only the DTM map. The clearance information is used to model
streets.
Figure 4.13: Tx-Rx profile

In the above figure, the ground altitude and clutter height (in this case, average height specified for each clutter class in
the clutter classes map description) are taken into account along the profile.
Clearance definition is not necessary in case of deterministic clutter height information. Clutter height information is accurate enough to be used directly without additional information such as clearance. Two cases can be considered:
1. If the receiver is in the street (clutter height lower than receiver height), Atoll calculates the path loss by considering
potentially some diffraction loss at reception.
2. If the receiver is supposed to be inside a building (clutter height higher than receiver height), Atoll does not consider any difraction (and clearance) from the building but takes into account the clutter class indoor loss as an additional penetration loss.
Notes:
To consider indoor losses in building only when using a deterministic clutter map (clutter
height map), the 'Indoor Coverage' box must not be checked in predictions unless this loss
will be counted twice inside buildings (on the entire reception clutter class and not only
inside the building).
Configuration tab)
4.4.3.3
Even with no clearance, the clutter height (extracted either from clutter class or clutter
height folders) is never considered at the last profile point.
Automatic SPM Calibration

The goal of this tool is to calibrate parameters and methods of the SPM formula in a simple and reproducible way. Calibration is based on imported CW measurement data. It is the process of limiting the difference between predicted and
measured values. For a complete description of the calibration procedure (including the very important prerequisite filtering
work on the CW measurement points), please refer to the User Manual and the SPM Calibration Guide.
The following SPM formula parameters can be estimated:
Forsk 2009
K1, K2, K3, K4, K5, K6 and K7
AT271_TRG_E6
91
Losses per clutter class (Kclutter must be user-defined)

Effective antenna height method
Diffraction method
Automatic model calibration provides a mathematical solution. The relevance of this mathematical solution with a physical
and realistic solution must be determined before committing these results.
You must keep in mind that the model calibration and its result (standard deviation and root mean square) strongly depend
on the CW measurement samples you use. A calibrated model must restore the behaviour of CW measurements depending on their configuration on a large scale, and not just totally coincide with a few number of CW measurements. The calibrated model has to give correct results for every new CW measurement point in the same geographical zone, without
having been calibrated on these new CW measurements.
4.4.3.3.1
General Algorithm
Propagation model calibration is a special case of the more general Least-Square problems, i.e. given a real m x n matrix
A, and a real m-vector b, find a real n-vector x0 that minimises the Euclidean length of Ax - b.
Here,
m is the number of measurement points,
n is the number of parameters to calibrate,
A is the values of parameter associated variables (log(d), log(heff), etc.) at each measurement point, and
b is the vector of measurement values.
The vector x0 is the set of parameters found at the end of the calibration.
The theoretical mathematical solution of this problem was found by Gauss (around 1830). Further enhancements to the
original method were proposed in the 60's in order to solve the numerical instability problem.
In 1974, Lawson & Hanson [2] proposed a theoretical solution of the least-square problem with general linear inequality
constraints on the vector x0. Atoll implementation is based on this method, which is explained in detail in [1].
References:
[1] Bjrck A. Numerical Methods for Least Square Problems, SIAM, 1996.
[2] Lawson C.L., Hanson R.J. Solving Least Squares Problems, SIAM, 1974.
4.4.3.3.2
Sample Values for SPM Path Loss Formula Parameters

The following tables list some sample orders of magnitudes for the different parameters composing the Standard Propagation Model formula.
Minimum
Typical
Maximum
K1
Variable
Variable
Variable
K2
20
44.9
70
K3
-20
5.83
20
K4
0.5
0.8
K5
-10
-6.55
K6
-1
K7
-10
K1 depends on the frequency and the technology. Here are some sample values:
Project type
Frequency (MHz)
K1
GSM 900
935
12.5
GSM 1800
1805
22
GSM 1900
1930
23
UMTS
2110
23.8
1xRTT
1900
23
2300
24.7
2500
25.4
2700
26.1
3300
27.8
3500
28.3
WiMAX
92
AT271_TRG_E6
Forsk 2009
The above K1 values for WiMAX are extrapolated estimates for different frequency ranges. It is highly recommended to
calibrate the SPM using measurement data collected on the field for WiMAX networks before using the SPM for predictions.
All K paramaters can be defined by the automatic calibration wizard. Since Kclutter is a constant, its value is strongly
dependant on the values given to the losses per clutter classes. From experienced users, the typical losses (in dB) per
clutter class are:
Dense urban
From 4 to 5
Woodland
From 2 to 3
Urban
Suburban
From -5 to -3
Industrial
From -5 to -3
Open in urban
From -6 to -4
Open
From -12 to -10
Water
From -14 to -12
These values have to be entered only when considering statistical clutter class maps only.
If you want to calibrate the losses per clutter class (Kclutter != 0) and press the Identify (Clutter row selected), a warning
message appears asking you to force the Max distance to 0 (in the Clutter tab), if it is not already set to 0. In fact, Atoll
uses the following process on these constants:
1st step: Atoll makes groups of measurement points according to the clutter class on which they are located.
2nd step: Atoll calculates the mean error for each of this group with Max distance = 0 (i.e. the clutter loss is applied
only on the reception bin).
3rd step: For each group, the mean error is then automatically shifted to 0 by manipulating the corresponding loss.
For example, if the mean error on a certain clutter class is 5 dB and if the initial loss for this clutter class is 2 dB,
then the calibrated loss becomes 7 dB.
Therefore, the global mean error on all the measurement points is null. In short, it is not possible to calibrate the clutter
losses if Max distance != 0. This parameter must be forced to 0.
Note:
4.4.3.4
The Standard Propagation Model is deduced from the Hata formulae, valid in the case of
an urban environment. The above values are consistent since they are normalized with
respect to the urban clutter class (0 dB for urban clutter class). Positive values correspond
to denser clutter classes and negative values to less dense clutter classes.
Unmasked Path Loss Calculation

You can use the SPM to calculate unmasked path losses. Unmasked path losses are calculated by not taking into account
the transmitter antenna patterns, i.e., the attenuation due to the transmitter antenna pattern is not included. Such path
losses are useful when using path loss matrices calculated by Atoll with automatic optimisation tools.
The instance of the SPM available by default, under the Propagation Models folder in the Modules tab, has the following
characteristics:
Signature:
Type:
{D5701837-B081-11D4-931D-00C04FA05664}
Atoll.StdPropagModel.1
You can access these parameters in the Propagation Models table by double-clicking the Propagation Models folder in
the Modules tab.
To make the SPM calculate path losses excluding the antenna pattern attenuation, you have to change the type of the
SPM to:
Type:
Atoll.StdPropagModelUnmasked.1
However, changing the type only does not invalidate the already calculated path loss matrices, because the signature of
the propagation model is still the same. If you want Atoll to recognize that the SPM has changed, and to invalidate the path
loss matrices calculated with this model, you have to change the signature of the model as well. The default signature for
the SPM that calculates unmasked path loss matrices is:
Signature:
{EEE060E5-255C-4C1F-B36C-A80D3D972583}
The above signature is a default signature. Atoll automatically creates different signatures for different instances of the
same propagation model. Therefore, it is possible to create different instances of the SPM, with different parameter
settings, and create unmasked versions of these instances.
You can change the signature and type of the original instance of the SPM, but it is recommended to make a copy, i.e., a
new instance, of the SPM in order not to lose the original SPM parameters. So, you will be able to keep different versions
of the SPM, those that calculate path losses with antenna pattern attenuation, and others that calculate path losses without
it.
Forsk 2009
AT271_TRG_E6
93

The usual process flow of an ACP working on an Atoll document through the API would be to:
1. Backup the storage directory of path loss matrices.
2. Set a different storage directory for calculating and storing unmasked path loss matrices.
3. Select the SPM used, backup its signature, and change its signature and type as shown above.
4. Perform optimisation using the path loss matrices calculated by the unmasked version of the SPM.
5. Restore the type and the signature of the SPM.
6. Reset the path loss storage directory to the original one.
Notes:
It is not possible to calibrate the unmasked version of the SPM using measurement data.
You can also use Atoll.ini options, AngleCalculation = 2000 and AngleCalculation = 3000,
for calculating unmasked path losses and angles of incidence, respectively. These options
are only available for the propagation models available with Atoll by default. Please refer to
the Administrator Manual for details.
Using the SPM, you can also calculate the angles of incidence by creating a new instance
of the SPM with the following characteristics:
Type: Atoll.StdPropagModelIncidence.1
Signature: {659F0B9E-2810-4e59-9F0D-DA9E78E1E64B}
Important:
The "masked" version of the algorithm has not been changed. It still takes into account
Atoll.ini options. However, the "unmasked" version does not take Atoll.ini options into
account.
Its highly recommended to use one method (Atoll.ini options) or the other one (new
identifier & signature) but not to combine both.
4.4.4
WLL Propagation Model
4.4.4.1
WLL Path Loss Formula

L model = FreeSpaceLoss + DiffractionLoss
4.4.4.2
4.4.4.2.1
Free Space Loss

Please refer to the Appendices for further details about free space loss calculation.
4.4.4.2.2
Diffraction
Atoll calculates diffraction loss along the transmitter-receiver profile built from DTM and clutter maps. Therefore, losses
due to clutter are taken into account in diffraction losses. Atoll takes clutter height information from the clutter heights file
if available in the .atl document. Otherwise, it considers average clutter height specified for each clutter class in the clutter
classes file description.
The Deygout construction (considering 3 obstacles) is used. This method is detailed in the Appendices.
Receiver Clearance
Define receiver clearance (m) per clutter class when clutter height information is either statistical or semi-deterministic. Both ground altitude and clutter height are considered along the whole profile except over a specific distance
around the receiver (clearance), where Atoll proceeds as if there was only the DTM map (see SPM part). Atoll uses the
clearance information to model streets.
If the clutter is deterministic, do not define any receiver clearance (m) per clutter class. In this case, clutter height
information is accurate enough to be used directly without additional information such as clearance (Atoll can locate
streets).
Receiver Height
Entering receiver height per clutter class enables Atoll to consider the fact that receivers are fixed and located on the roofs.
Visibility
If the option Line of sight only is not selected, Atoll computes Lmodel on each calculation bin using the formula defined
above. When selecting the option Line of sight only, Atoll checks for each calculation bin if the Diffraction loss (as defined
in the Diffraction loss: Deygout part) calculated along profile equals 0.
94
AT271_TRG_E6
Forsk 2009
In this case, receiver is considered in line of sight and Atoll computes Lmodel on each calculation bin using the
formula defined above.
Otherwise, Atoll considers that Lmodel tends to infinity.
4.4.5
ITU-R P.526-5 Propagation Model
4.4.5.1

L model = FreeSpaceLoss + DiffractionLoss
4.4.5.2
4.4.5.2.1
Free Space Loss

4.4.5.2.2
Diffraction
Atoll calculates diffraction loss along the transmitter-receiver profile is built from the DTM map. The Deygout construction
(considering 3 obstacles), with or without correction, is used. These methods are detailed in the Appendices.
4.4.6
4.4.6.1

If d<1 km, L model = FreeSpaceLoss
If d>1000 km, L model = 1000
If 1<d<1000 km, L model = max FreeSpaceLoss CorrectedS tan dardLoss
d is the distance between the transmitter and the receiver (km).
4.4.6.2
4.4.6.2.1
Free Space Loss

4.4.6.2.2
Corrected Standard Loss

This formula is given for a 60 dBm (1kW) transmitter power.
CorrectedS tan dardLoss = 60 C n A H
Rxeff
A cl 108.75 + 31.54 20 log f
where,
Cn is the field strength received in dBV/m,
AH
Rxeff
is a correction factor for effective receiver antenna height (dB),
Acl is the correction for terrain clearance angle (dB),

f is the frequency in MHz.
Cn Calculation
The Cn value is determined from charts Cn=f(d, HTxeff).
In the following part, let us assume that Cn=En(d,HTxeff) (where En(d,HTxeff) is the field received in dBV/m) is read from
charts for a distance, d (in km), and an effective transmitter antenna height, HTxeff (in m).
First of all, Atoll evaluates the effective transmitter antenna height, H Txeff , as follows:
If 0 d 3km , H Txeff = H 0Tx + H Tx H 0Rx
If 3 d 15km , H Txeff = H 0Tx + H Tx H 0 3 ;d
If 15 d , H Txeff = H 0Tx + H Tx H 0 3 ;15
where,
H Tx is the transmitter antenna height above the ground (m).
Forsk 2009
AT271_TRG_E6
95

H 0Tx is the ground height (ground elevation) above sea level at the transmitter (m).
H 0 3 ;d is the average ground height (m) above sea level for the profile between a point 3 km from transmitter and the
receiver (located at d km from transmitter).
H 0 3 ;15 is the average ground height (m) above sea level for the profile between a point 3 km and another 15 km from
transmitter.
Then, depending on d and HTxeff, Atoll determines Cn using bilinear interpolation as follows.
If 37.5 HTxeff 1200, Cn= En(d,HTxeff)
Otherwise, Atoll considers d horizon = 4.1 H Txeff (d is stated in km)
Therefore,
If HTxeff < 37.5
If d d horizon , we have C n = E n d + 25 d horizon 37.5
Else Cn=En(d, 37.5) En(dhorizon, 37.5) + En(25, 37.5)
If HTxeff > 1200
If d d horizon , we have C n = E n d + 142 d horizon 1200
Else Cn=En(d, 1200) En(dhorizon, 1200) + En(142, 1200)
AHRxeff Calculation
AH
Rxeff
H Rx
c
= --- 20 log ----------
10
6
where,
HRx is the user-defined receiver height,
c is the height gain factor.
Note:
c values are provided in the recommendation 370-7; for example, c=4 in a rural case.
Acl Calculation
2
If f 300 MHz, A cl = 8.1 6.9 + 20 log 0.1 + 1 + 0.1

2
Otherwise, A cl = 14.9 6.9 + 20 log 0.1 + 1 + 0.1

f
With = 4000 ---------300
where,
is the clearance angle (in radians) determined according to the recommendation 370-7 (figure 19),
f is the frequency stated in MHz.
4.4.7
Erceg-Greenstein (SUI) Propagation Model

Erceg-Greenstein propagation model is a statistical path loss model derived from experimental data collected at 1.9 GHz
in 95 macrocells. The model is for suburban areas, and it distinguishes between different terrain categories called the Stanford University Interim Terrain Models. This propagation model is well suited for distances and base station antenna
heights that are not well-covered by other models. The path loss model applies to base antenna heights from 10 to 80 m,
base-to-terminal distances from 0.1 to 8 km, and three distinct terrain categories.
The basic path loss equation of the Erceg-Greenstein propagation model is:
d
PL = A + 10 a H BS Log 10 ------
d 0
4d 0
Where A = 20 Log 10 ------------- . This is a fixed quantity which depends upon the frequency of operation. d is the distance

between the base station antenna and the receiver terminal and d0 is a fixed reference distance (100 m). a(HBS) is the
correction factor for base station antenna heights, HBS:
c
a HBS = a b H BS + ---------H BS
96
AT271_TRG_E6
Forsk 2009
Where 10 m HBS 80 m , and a, b, and c are correction coefficients which depend on the SUI terrain type.
The Erceg-Greenstein propagation model is further developed through the correction factors introduced by the Stanford
University Interim model. The standards proposed by the IEEE working group 802.16 include channel models developed
by Stanford University. The basic path loss equation with correction factors is presented below:
d
PL = A + 10 a H BS Log 10 ------ + a f a H R
d 0
f
Where a(f) is the correction factor for the operating frequency, a f = 6 Log 10 ------------- , with f being the operating
2000
HR
frequency in MHz. a(HR) is the correction factor for the receiver antenna height, a H R = X Log 10 ------- , where d
2
depends on the terrain type.
Note:
a(HR) = 0 for HR = 2 m.
References:
[1] V. Erceg et. al, An empirically based path loss model for wireless channels in suburban environments, IEEE J.
Select Areas Commun., vol. 17, no. 7, July 1999, pp. 1205-1211.
[2] Abhayawardhana, V.S.; Wassell, I.J.; Crosby, D.; Sellars, M.P.; Brown, M.G.; "Comparison of empirical propagation
path loss models for fixed wireless access systems," Vehicular Technology Conference, 2005. IEEE 61st Volume 1, 30
May-1 June 2005 Page(s):73 - 77 Vol. 1
4.4.7.1
SUI Terrain Types

The SUI models are divided into three types of terrains1, namely A, B and C.
Type A is associated with maximum path loss and is appropriate for hilly terrain with moderate to heavy tree
densities.
Type B is characterised with either mostly flat terrains with moderate to heavy tree densities or hilly terrains
with light tree densities.
Type C is associated with minimum path loss and applies to flat terrain with light tree densities.
The constants used for a, b, and c are given in the table below.
Model Parameter
Terrain A
Terrain B
Terrain C
4.6
4.0
3.6
0.0075
0.0065
0.005
c (m)
12.6
17.1
20
10.8
10.8
20
4.4.7.2
(m-1)
Erceg-Greenstein (SUI) Path Loss Formula

The Erceg-Greenstein (SUI) propagation model formula can be simplified from the following equation:
4d 0
d
PL = 20 Log 10 ------------- + 10 a H BS Log 10 ------ + a f a H R
d 0

(1)
to the equation below:

PL = 7.366 + 26 Log 10 f + 10 a H BS 1 + Log 10 d a H R
(2)
Where,
f is the operating frequency in MHz

d is the distance from the transmitter to the received in m in equation (1) and in km in equation (2)
HBS is the transmitter height in m
HR is the receiver height in m
The above equation is divided into two parts in Atoll:

PL = Lu a H R
Where,
Lu = 7.366 + 26 Log 10 f + 10 a H BS 1 + Log 10 d
1.
The word terrain is used in the original definition of the model rather than environment. Hence it is used
interchangeably with environment in this subsection.
Forsk 2009
AT271_TRG_E6
97

The above path loss formulas are valid for d > d0, i.e. d > 100 m. For d < 100 m, the path loss has been restricted to the
free space path loss with correction factors for operating frequency and receiver height:
4d
4d
PL = 20 Log 10 ------------------ + a f a H R instead of PL = 20 Log 10 ------------------

Where a(f) and a(Hr) have the same definition as given above. Simplifying the above equation, we get,
PL = 12.634 + 26 Log 10 f + 20 Log 10 d a H R , or Lu = 12.634 + 26 Log 10 f + 20 Log 10 d
The above equation is not user-modifiable in Atoll except for the coefficient of Log 10 f , i.e. 26. Atoll uses the same coefficient as the one you enter for Log 10 f in Atoll for the case d > d0.
Note:
You can get the same equation, i.e., Lu = 12.634 + 26 Log 10 f + 20 Log 10 d , by
setting a(hBS) = 2.
4.4.7.3
The Erceg-Greenstein (SUI) propagation model takes DTM into account between the transmitter and the receiver, and it
can also take clutter into account at the receiver location.
1st step: For each pixel in the calculation radius, Atoll determines the clutter bin on which the receiver is located. This clutter
bin corresponds to a clutter class. Atoll uses the Erceg-Greenstein (SUI) path loss formula assigned to this clutter class to
evaluate path loss.
2nd step: This step depends on whether the Add diffraction loss option is selected or not.
If the Add diffraction loss option is not selected, 1st step gives the final path loss result.
a. It extracts a geographic profile between the transmitter and the receiver based on the radial calculation method.
b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates losses due to diffraction L Diffraction . For more information on the Deygout method, see "3 Knife-Edge Deygout
Method" on page 105.
The final path loss is the sum of the path loss determined in 1st step and L Diffraction .
Shadow fading is computed in Atoll independent of the propagation model. For more information on the shadow fading
calculation, see "Shadowing Model" on page 113.
4.4.8

This propagation model is based on the P.1546-2 recommendations of the ITU-R. These recommendations extend the
P.370-7 recommendations, and are suited for operating frequencies from 30 to 3000 MHz. The path loss is calculated by
this propagation model with the help of graphs available in the recommendations. The graphs provided in the recommendations represent field (or signal) strength, given in db V m , as a function of distance for:
Nominal frequencies, f n : 100, 600, and 1000 MHz

The graphs provided for 100 MHz are applicable to frequencies from 30 to 300 MHz, those for 600 MHz are applicable to frequencies from 300 to 1000 MHz, and the graphs for 1000 MHz are applicable to frequencies from 1000
to 3000 MHz. The method for interpolation is described in the recommendations (Annex 5, 6).
Transmitter antenna heights, h 1 : 10, 20, 37.5, 75, 150, 300, 600, and 1200 m
For any values of h 1 from 10 to 3000 m, an interpolation or extrapolation from the appropriate two curves is used,
as described in the recommendations (Annex 5, 4.1). For h 1 below 10 m, the extrapolation to be applied is given
in Annex 5, 4.2. It is possible for the value of h 1 to be negative, in which case the method is given in Annex 5,
4.3.
Time variability, t : 1, 10, and 50 %

The propagation curves represent the field strength values exceeded for 1, 10 and 50 % of time.
Receiver antenna height, h 2 : 10 m

For land paths, the graphs represent field strength values for a receiver antenna height above ground, equal to
the representative height of the clutter around the receiver. The minimum value of the representative height of clutter is 10 m. For sea paths, the graphs represent field strength values for a receiver antenna height of 10 m.
For other values of receiver antenna height, a correction is applied according to the environment of the receiver.
The method for calculating this correction is given in Annex 5, 9.
98
AT271_TRG_E6
Forsk 2009
These recommendations are not valid for transmitter-receiver distances less than 1 km or greater than 1000 km. Therefore
in Atoll, the path loss between a transmitter and a receiver over less than 1 km is the same as the path loss over 1 km.
Similarly, the path loss between a transmitter and a receiver over more than 1000 km is the same as the path loss over
1000 km.
Moreover, these recommendations are not valid for transmitter antenna heights less than the average clutter height
surrounding the transmitter.
Notes:
4.4.8.1
The cold sea graphs are used for calculations over warm and cold sea both.
The mixture of land and sea paths is not supported by Atoll.
The input to the propagation model are the transmission frequency, transmitter and receiver heights, the distance between
the transmitter and the receiver, the precentage of time the field strength values are exceeded, the type of environment
(i.e., land or sea), and the clutter at the receiver location.
In the following calculations, f is the transmission frequency, d is the transmitter-receiver distance, and t is the percentage of time for which the path loss has to be calculated.
The following calculations are performed in Atoll to calculate the path loss using this propagation model.
4.4.8.1.1
Step 1: Determination of Graphs to be Used

First of all, the upper and lower nominal frequencies are determined for any given transmission frequency. The upper and
lower nominal frequencies are the nominal frequencies (100, 600, and 2000 MHz) between which the transmission
frequency is located, i.e., f n1 f f n2 .
Once f n1 and f n1 are known, along with the information about the percentage of time t and the type of path (land or sea),
the sets of graphs which will be used for the calculation are also known.
4.4.8.1.2
Step 2: Calculation of Maximum Field Strength

A field strength must not exceed a maximum value, E Max , which is given by:
E Max = E FS = 106.9 20 Log d for land paths, and
E Max = E FS + E SE = 106.9 20 Log d + 2.38 1 exp d 8.94 Log 50 t for sea paths.
Where E FS is the free space field strength for 1 kW ERP, E SE is an enhancement for sea graphs.
4.4.8.1.3
Step 3: Determination of Transmitter Antenna Height

The transmitter antenna height to be used in the calculation depends on the type and length of the path.
Land paths
h 1 = h eff
Sea paths
h 1 = Max 1 h a
Here, all antenna heights (i.e., h 1 , h eff , and h a ) are in expressed in m. h a is the antenna height above ground and h eff
is the effective height of the transmitter antenna, which is its height over the average level of the ground between distances
of 0.2 d and d km from the transmitter in the direction of the receiver.
4.4.8.1.4
Step 4: Interpolation/Extrapolation of Field Strength

The interpolations are performed in series in the same order as described below. The first interpolation/extrapolation is
performed over the field strength values, E , from the graphs for transmitter antenna height to determine E h1 . The second
interpolation/extrapolation is performed over the interpolated/extrapolated values of E h1 to determine E d . And, the thrid
and final interpolation/extrapolation is performed over the interpolated/extrapolated values of E d to determine E f .
Step 4.1: Interpolation/Extrapolation of Field Strength for Transmitter Antenna Height

If the value of h 1 coincides with one of the eight heights for which the field strength graphs are provided, namely 10, 20,
37.5, 75, 150, 300, 600, and 1200 m, the required field strength is obtained directly from the corresponding graph. Otherwise:
If 10 m h 1 3000 m
The field strength is interpolated or extrapolated from field strengths obtained from two curves using the following
equation:
Forsk 2009
AT271_TRG_E6
99

Log h 1 h Low
E h1 = E Low + E Up E Low -----------------------------------------Log h Up h Low
Where h Low = 600 m if h 1 1200 m , otherwise h Low is the nearest nominal effective height below h 1 ,
h Up = 1200 m if h 1 1200 m , otherwise h Up is the nearest nominal effective height above h 1 , E Low is the field
strength value for h Low at the required distance, and E Up is the field strength value for h Up at the required
distance.
If 0 m h 1 10 m
-
For land path if the transmitter-receiver distance is less than the smooth-Earth horizon distance
d H h 1 = 4.1 h 1 , i.e., if d 4.1 h 1 ,
E h1 = E 10 d H 10 + E 10 d E 10 d H h 1 , or
E h1 = E 10 12.9 km + E 10 d E 10 d H h 1 because d H 10 = 12.9 km
For land path if the transmitter-receiver distance is greater than or equal to the smooth-Earth horizon distance
d H h 1 = 4.1 h 1 , i.e., if d 4.1 h 1 ,
E h1 = E 10 d H 10 + d d H h 1 , or E h1 = E 10 12.9 km + d d H h 1 because d H 10 = 12.9 km
Where E x y is the field strength value read for the transmitter-receiver distance of y from the graph available
for the transmitter antenna height of x.
If in the above equation, d H 10 + d d H h 1 1000 km even though d 1000 km , the field strength is determined from linear extrapolation for Log (distance) of the graph given by:
Log d D Low
E h1 = E Low + E Up E Low -------------------------------------------Log D Up D Low
Where D Low is penultimate tabulation distance (km), D Up is the final tabulation distance (km), E Low is the
field strength value for D Low , and E Up is the field strength value for D Up .
For sea path, h 1 should not be less than 1 m. This calculation requires the distance at which the path has 0.6
of the first Fresnel zone just unobstructed by the sea surface. This distance is given by:
D h1 = D 0.6 f h 1 h 2 = 10 m (km)
Df Dh
Where D 0.6 = Max 0.001 ------------------- (km) with D f = 0.0000389 f h 1 h 2 (frequency-dependent term),
D f + D h
and D h = 4.1 h 1 + h 2 (asymptotic term defined by the horizon distance).
If d D h1 the 0.6 Fresnel clearance distance for the sea path where the transmitter antenna height is 20 m is
also calculated as:
D 20 = D 0.6 f h 1 = 20 m h 2 = 10 m (km)
Once D h1 and D 20 are known, the field strength for the required distance is given by:
E h1
E Max
Log d D h1
= E D + E D E D -------------------------------------h1
20
h1
Log D 20 D h1
E' 1 F S + E'' F S
for d D h1
for D h1 d D 20
for d D 20
Where E Max is the maximum field strength at the required distance as calculated in "Step 2: Calculation of
Maximum Field Strength" on page 99, E D
ED
20
h1
is E Max for d = D h1 ,
Log h1 10
Log h1 10
= E 10 D 20 + E 20 D 20 E 10 D 20 ---------------------------------- , E' = E 10 d + E 20 d E 10 d ---------------------------------- ,
Log 20 10
Log 20 10
and E'' is the field strength calculated as described for land paths. E 10 y and E 20 y are field strengths
interpolated for distance y and h 1 = 10 m and 20 m , respectively, and F S = d D 20 d .
If h 1 0 m
A correction is applied to the field strength, E h1 , calculated in the above description in order to take into account
the diffraction and tropospheric scattering. This correction is the maximum of the diffraction correction,, and tropospheric scattering correction, .
100
AT271_TRG_E6
Forsk 2009
C h1 = Max C h1d C h1t
Where
C h1d = 6.03 J
with
J = 6.9 + 20 Log 0.1 + 1 + 0.1
and
= K eff2 ,
h1
eff2 = arc tan ------------- , and K is 1.35 for 100 MHz, 3.31 for 600 MHz, 6.00 for 2000 MHz.
9000
e
180 d
C h1t = 30 Log ------------------------ with e = ---------------------- , a = 6370 km (radius of the Earth), and k = 4 3 is the effec e + eff2
ak
tive Earth radius factor for mean refractivity conditions.
Step 4.2: Interpolation/Extrapolation of Field Strength for Transmitter-Receiver Distance

In the field strength graphs in the recommendations, the field strength is plotted against distance from 1 km to 1000 km.
The distance values for which field strengths are tabulated are 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200,
225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, 500, 525, 550, 575, 600, 625, 650, 675, 700, 725, 750, 775, 800,
825, 850, 875, 900, 925, 950, 975, 1000. If the transmitter-receiver distance is a value from this list, then interpolation of
field strength is not required and the field strength can be directly read from the graphs.
If the transmitter-receiver distance does not coincide with the list of distances for which the field strengths are accurately
available from the graphs, the field strength are linearly interpolated or extrapolated for the logarithm of the distance using
the following equation:
Log d d Low
E d = E Low + E Up E Low -----------------------------------------Log d Up d Low
Where d Low is the lower value of the nearest tabulated distance to d , d Up is the higher value of the nearest tabulated
distance to d , E Low is the field strength value for d Low , and E Up is the field strength value for d Up .
Step 4.3: Interpolation/Extrapolation of Field Strength for Transmission Frequency

The field strength at the transmission frequency is interpolated from the graphs available for the upper and lower nominal
frequencies as follows:
Log f f Low
E f = E Low + E Up E Low --------------------------------------Log f Up f Low
Where f Low is the lower nominal frequency (100 MHz if f < 600 MHz, 600 MHz otherwise), f Up is the higher nominal
frequency (600 MHz if f < 600 MHz, 2000 MHz otherwise), E Low is the field strength value for f Low , and E Up is the field
strength value for f Up .
In the case of transmission frequencies below 100 MHz or above 2000 MHz, the field strength values are extrapolated
from the two nearer nominal frequency values. The above equation is used for all land paths and sea paths.
4.4.8.1.5
Step 5: Calculation of Correction Factors

Step 5.1: Correction for Receiver Antenna Height
The receiver antenna height correction depends on the type of path and clutter in which the receiver is located. The field
strength values given by the graphs for land paths are for a reference receiver antenna at a height, R (m), representative
of the height of the clutter surrounding the receiver, subject to a minimum height value of 10 m. Examples of reference
heights are 20 m for an urban area, 30 m for a dense urban area, and 10 m for a suburban area. For sea paths the notional
value of R is 10 m.
For land paths, the elevation angle of the arriving ray is taken into account by calculating a modified representative clutter
1000 d R 15 h 1
height R' , given by R' = Max 1 --------------------------------------------------------------- .
1000 d 15
Note that for h 1 6.5 d + R , R' R .
The different correction factors are calculated as follows:
For land path in urban and suburban zones

6.03 J
for h 2 R'
C Receiver =
h
3.2 + 6.2 Log f Log -----2- for h 2 R'
R'
R' h 2
2
With J = 6.9 + 20 Log 0.1 + 1 + 0.1 and = 0.0108 f R' h 2 arc tan ----------------- .
27
Forsk 2009
AT271_TRG_E6
101

10
If R' 10 m , C Receiver is reduced by 3.2 + 6.2 Log f Log ------ .
R'
For land path other zones

h2
C Receiver = 3.2 + 6.2 Log f Log ------
10
For sea path

d 10 and d h2 are determined as distances at which at which the path has 0.6 of the first Fresnel zone just unobstructed by the sea surface with h 2 = 10 m and variable h 2 , respectively. These distances are given by
Df Dh
d 10 = D 0.6 f h 1 h 2 = 10 m and d h2 = D 0.6 f h 1 h 2 (km), respectively. Here D 0.6 = Max 0.001 -------------------
D f + D h
as explained earlier.
-
h2
If h 2 10 m , C Receiver = 3.2 + 6.2 Log f Log ------
10
h2
If h 2 10 m and d d 10 , C Receiver = 3.2 + 6.2 Log f Log ------
10
If h 2 10 m and d d 10 and d d h2 , C Receiver = 0
h2
Log d d h2
If h 2 10 m and d d 10 and d d h2 , C Receiver = 3.2 + 6.2 Log f Log ------ -------------------------------------
10 Log d 10 d h2
Step 5.2: Correction for Short Urban/Suburban Paths

This correction is only applied when the path loss is to be calculated over land paths, over a transmitter-receiver distance
less than 15 km, in urban and suburban zones. This correction takes into account the presence of buildings in these zones.
The buildings are assumed to be of uniform height.
The correction represents a reduction in the field strength due to building clutter. It is added to the field strength and is
given by:
C Building = 3.3 Log f 1 0.85 Log d 1 0.46 Log 1 + h a R
Where h a is the antenna height above the ground, and R is the clutter height of the clutter class where the receiver is
located. This correction is only applied when d 15 km and h 1 R 150 m .
Step 5.3: Correction for Receiver Clearance Angle

This correction is only applied when the path loss is to be calculated over land paths, and over a transmitter-receiver
distance less than 16 km. This correction gives more precise field strength prediction over small reception areas. The
correction is added to the field strength and is given by:
C Clearance = J ' J
2
Where J = 6.9 + 20 Log 0.1 + 1 + 0.1 , ' = 0.036 f , and = 0.065 Clearance f
Clearance is the clearance angle in degrees determined from:
: The elevation angle of the line from the receiver which just clears all terrain obstructions in the direction of the
transmitter over a distance of up to 16 km but not going beyond the transmitter.
h 1S h 2S
Ref : The reference angle, Ref = arc tan ------------------------ .
1000 d
Where h 1S and h 2S are the heights of the transmitter and the receiver above sea level, respectively.
4.4.8.1.6
Step 6: Calculation of Path Loss

First, the final field strength is calculated from the interpolated/extrapolated field strength, E f , by applying the corrections
calculated earlier. The calculated field strength is given by:
E Calc = E f + C Receiver + C Building + C Clearance
The resulting field strength is given by E = Min E Calc E Max , from which the path loss (basic transmission loss, L B ) is
calculated as follows:
L B = 139 E + 20 Log f
102
AT271_TRG_E6
Forsk 2009
4.4.9
Sakagami Extended Propagation Model

The Sakagami extended propagation model is based on the simplification of the extended Sakagami-Kuboi propagation
model. The Sakagami extended propagation model is valid for frequencies above 3 GHz. Therefore, it is only available in
WiMAX 802.16d and WiMAX 802.16e documents by default.
The Sakagami-Kuboi propagation model requires detailed information about the environment, such as widths of the streets
where the receiver is located, the angles formed by the street axes and the directions of the incident waves, heights of the
buildings close to the receiver, etc. The path loss formula for the Sakagami-Kuboi propagation model is [1]:
H 2
L Model = 100 7.1 Log W + 0.023 + 1.4 Log h s + 6.1 Log H 1 24.37 3.7 -------- Log h b +
h b0
43.2 3.1 Log h b Log d + 20 Log f + e
13 Log f 3.23
Where,
W is the width (in meters) of the streets where the receiver is located
is the angle (in degrees) formed by the street axes and the direction of the incident wave
hs is the height (in meters) of the buildings close to the receiver
H1 is the average height (in meters) of the buildings close to the receiver
hb is the height (in meters) of the transmitter antenna with respect to the observer
hb0 is the height (in meters) of the transmitter antenna with respect to the ground level
H is the average height (in meters) of the buildings close to the base station
d is the separation (in kilometres) between the transmitter and the receiver
f is the frequency (in MHz)
The Sakagami-Kuboi propagation model is valid for:

5m
<W<
50 m
< <
90
5m
< hs <
80 m
5m
< H1 <
50 m
20 m
< hb <
100 m
0.5 km
<d<
10 km
450 MHz
<f<
2200 MHz
h b0 H
Studies [2] have shown that the Sakagami-Kuboi propagation model can be extended to frequencies higher than 3 GHz,
which also allows a simplification in terms of the input required by the model.
The path loss formula for the extended Sakagami-Kuboi propagation model is:
L Model = 54 + 40 Log d 30 Log h b + 21 Log f + a
Where a is a corrective factor with three components:
H0
hm
W
a = a H 0 + a W + a h m = 11 Log ------- 7.1 Log ------ 5 Log --------
20
1.5
20
W is the width (in meters) of the streets where the receiver is located
H0 (= hs = H1) is the height (in meters) of the buildings close to the receiver
hb (= hb0) is the height (in meters) of the transmitter antenna with respect to the ground
hm is the height (in meters) of the receiver antenna
H is the average height (in meters) of the buildings close to the base station
d is the separation (in metres) between the transmitter and the receiver
f is the frequency (in GHz)
The extended Sakagami-Kuboi propagation model is valid for:

5m
<W<
50 m
10 m
< H0 <
30 m
10 m
< hb <
100 m
0.1 km
<d<
3 km
0.8 GHz
<f<
8 GHz
1.5 m
< hm <
5m
Studies also show that above 3 GHz, the path loss predicted by the extended model is almost independant of the input
parameters such as street widths and angles. Therefore, the extended Sakagami-Kuboi propagation model can be simplified to the extended Sakagami propagation model:
Forsk 2009
AT271_TRG_E6
103

L Model = 54 + 40 Log d 30 Log h b + 21 Log f 5 Log h m
The extended Sakagami propagation model is valid for:
10 m
< hb <
100 m
0.1 km
<d<
3 km
3 GHz
<f<
8 GHz
1.5 m
< hm <
5m
The path loss calculation formula of the Sakagami extended propagation model resembles the formula of the Standard
Propagation Model. In Atoll, this model is in fact a copy of the Standard Propagation Model with the following values
assigned to the K coefficients:
K1
65.4 (calculated for 3.5 GHz)
K2
40
K3
-30
K4
K5
K6
K7
-5
For more information on the Standard Propagation Model, see "Standard Propagation Model (SPM)" on page 85.
References:
[1] Manuel F. Catedra, Jesus Perez-Arriaga, "Cell Planning for Wireless Communications," Artech House Publishers,
1999.
[2] Koshiro Kitao, Shinichi Ichitsubo, "Path Loss Prediction Formula for Urban and Suburban Areas for 4G Systems,"
IEEE, 2006.
4.4.10
Appendices
4.4.10.1
Free Space Loss

The calculation of free space loss is based on ITU 525 recommendations.
FreeSpaceLoss = 32.4 + 20 log f + 20 log d
where,
f is the frequency in MHz,
d is the Tx-Rx distance in km,
Free space loss is stated in dB.
4.4.10.2
Diffraction Loss
The calculation of diffraction is based on ITU 526-5 recommendations. General method for one or more obstacles (knifeedge diffraction) is used to evaluate diffraction losses (Diffraction loss in dB). Four construction modes are implemented
in Atoll. All of them are based on this same physical principle presented hereafter, but differ in the way they consider one
or several obstacles. Calculations take the earth curvature into account through the effective Earth radius concept (K
factor=1.333).
4.4.10.2.1
Knife-Edge Diffraction
The procedure checks whether a knife-edge obstructs the first Fresnel zone constructed between the transmitter and the
receiver. The diffraction loss, J(), depends on the obstruction parameter (), which corresponds to the ratio of the obstruction height (h) and the radius of the Fresnel zone (R).
104
AT271_TRG_E6
Forsk 2009
Figure 4.14: Knife-Edge Diffraction
R =
c0 n d1 d2
--------------------------------f d1 + d2
where,
n is the Fresnel zone index,
c0 is the speed of light (2.99792 x108 ms-1),
f is the frequency in Hz
d1 is the distance from the transmitter to obstacle in m,
d2 is the distance from obstacle to receiver in m.
We have: = h
--r
where,
R
r = ------2
h is the obstruction height (height from the obstacle top to the Tx-Rx axis).
Hence,
2
For 1 knife-edge method, if 0.7 , J = 6.9 + 20 log 0.1 + 1 + 0.1

Else, J = 0
Note:
4.4.10.2.2
In case of multiple-knife edge method, the minimum required to estimate diffraction loss
is -0.78.
3 Knife-Edge Deygout Method

The Deygout construction, limited to a maximum of three edges, is applied to the entire profile from transmitter to receiver.
This method is used to evaluate path loss incurred by multiple knife-edges. Deygout method is based on a hierarchical
knife-edge sorting used to distinguish the main edges, which induce the largest losses, and secondary edges, which have
a lesser effect. The edge hierarchy depends on the obstruction parameter () value.
Forsk 2009
AT271_TRG_E6
105
1 Obstacle
Figure 4.15: Deygout Construction 1 Obstacle

A straight line between transmitter and receiver is drawn and the height of the obstacle above the Tx-Rx axis, hi, is calculated. The obstruction position, di, is also recorded. i are evaluated from these data. The point with the highest value is
termed the principal edge, p, and the corresponding loss is J(p).
Therefore, we have
DiffractionLoss = J P
3 Obstacles
Then, the main edge (point p) is considered as a secondary transmitter or receiver. Therefore, the profile is divided in two
parts: one half profile, between the transmitter and the knife-edge section, another half, constituted by the knife-edgereceiver section.
Figure 4.16: Deygout Construction 3 Obstacles

The same procedure is repeated on each half profile to determine the edge with the higher . The two obstacles found,
(points t and r), are called secondary edges. Losses induced by the secondary edges, J(t) and J(r), are then calculated.
Once the edge hierarchy is determined, the total loss is evaluated by adding all the intermediary losses obtained.
Therefore, if P 0
we have DiffractionLoss = J P + J t + J r
Otherwise, If P 0.7 , DiffractionLoss = J P
Note:
In case of ITU 526-5 and WLL propagation models, Diffraction loss term is determined as
follows:
- If P 0.78 , we have
DiffractionLoss = J P + J t + J r t
J P
Where, t = min -------------- 1
6
- Otherwise DiffractionLoss = 0
106
AT271_TRG_E6
Forsk 2009
4.4.10.2.3
Epstein-Peterson Method
The Epstein-Peterson construction is limited to a maximum of three edges. First, Deygout construction is applied to determine the three main edges over the whole profile as described above. Then, the main edge height, hp, is recalculated
according to the Epstein-Peterson construction. hp is the height above a straight line connecting t and r points. The main
edge position dp is recorded and p and J(p) are evaluated from these data.
Figure 4.17: Epstein-Peterson Construction

Therefore, we have
DiffractionLoss = J P + J t + J r
4.4.10.2.4
Deygout Method with Correction

The Deygout method with correction (ITU 526-5) is based on the Deygout construction (3 obstacles) plus an empirical
correction, C.
Therefore, If P 0 ,
we have DiffractionLoss = J P + J t + J r + C
Otherwise DiffractionLoss = J P + C
Note:
In case of ITU 526-5 propagation model, Diffraction loss term is determined as follows:
- If P 0.78 , we have DiffractionLoss = J P + t J t + J r + C
J P
Where, t = min -------------- 1
6
C = 8.0 + 0.04d (d: distance stated in km between the transmitter and the receiver).
- Otherwise DiffractionLoss = 0
4.4.10.2.5
Millington Method
The Millington construction, limited to a single edge, is applied over the entire profile. Two horizon lines are drawn at the
transmitter and at the receiver. A straight line between the transmitter and the receiver is defined and the height of the
intersection point between the two horizon lines above the Tx-Rx axis, hh, is calculated. The position dh is recorded and
then, from these values, h and J(h) are evaluated using the same previous formulas.
Therefore, we have
DiffractionLoss = J h
Forsk 2009
AT271_TRG_E6
107
Figure 4.18: Millington Construction
4.5
Path Loss Tuning

Atoll can tune path loss matrices obtained from propagation results by the use of real measurements (CW Measurements
or Test Mobile Data). For each measured transmitter, Atoll tries to merge measurements and predictions on the same
points and to smooth the surrounding points of the path loss matrices for homogeneity reasons. A transmitter path loss
matrix can be tuned several times by the use of several measurement paths. All these tuning paths are stored in a catalogue. This catalogue is stored under a .tuning folder containing a .dbf file and one .pts file per corrected transmitter. Since
a tuning file can contain several measurement paths, all these measurements are added to the tuning file.
For more information on the tuning files, See "Externalised Tuning Files" on page 61.
The same algorithm is used for CW Measurement and Test Mobile Data. It is also the same for main and extended matrices.
Path Losses tuning will be done using two steps.
1.
Global matrix correction
A mean error is calculated between each measured value and the corresponding bin in the pathloss matrix. Mean error is
calculated for each pathloss matrix (main and extended) of each transmitter. This mean error is then applied to all the
matrix bins. This correction is done to smooth the local corrections (step 2) of measured values and not the corrected bins.
2.
Correction for each measured value
For each measured value, an ellipse is used to define the pathloss area which has to be tuned. The main axis of the ellipse
is oriented to the transmitter.The ellipse is user-defined by two parameters :
The radius of the axis parallel to the Profile (A)

The radius of the axis perpendicular to the Profile (B)
Lets take M a measurement value and P i the path loss value at point i, before any correction.
The squared elliptic distance between i and M is given by:
2
Xi XM
Yi YM
D i = -------------------------+ -------------------------2
2
A
B
where:
X i and X M are the X-coordinates of i and M respectively
Y i and Y M are the Y-coordinates of i and M respectively
The mean error for the first correction is given by:
1
E = ---
n
ei
i
where:
e i is the error between measurement and prediction at point i
Then, the path loss value is corrected using E:
Pi
new
= Pi
old
+E
Finally, a second correction ( R i ) is applied where:

Ri = 1 Di M g Pi
108
new
AT271_TRG_E6
so R i = 1 D i M g P i
old
+ E
Forsk 2009
where g is (measurement gain - losses).
So, the final corrected path loss is:
Pi
tuned
= Pi
new
+ R i so P i
tuned
= Pi
old
+ E + Ri
When several ellipses overlap a pathloss bin, the final corrected path loss is given by:
1 d j P j
tuned
j
Pi
= ---------------------------------------------------tuned
d j
n
Where:
n is the number of overlapping ellipses
4.6
Antenna Attenuation Calculation

The modelling method used to evaluate transmitter antenna attenuation, L antTx , is described below. Atoll calculates the
accurate azimuth and tilt angles and then, performs a 3-D interpolation of horizontal and vertical patterns to determine the
attenuation of antenna.
Furthermore, you will find explanations about the remote electrical downtilt modelling.
4.6.1
Calculation of Azimuth and Tilt Angles

From the direction of the transmitter antenna and the receiver position relative to the transmitter, Atoll determines the
receiver position relative to the direction of the transmitter antenna (i.e. the direction of the transmitter-receiver path in the
transmitter antenna coordinate system).
aTx and eTx are respectively the transmitter (Tx) antenna azimuth and tilt in the coordinate system S 0 x y z .
aRx and eRx are respectively the azimuth and tilt of the receiver (Rx) in the coordinate system S 0 x y z .
d is the distance between the transmitter (Tx) and the receiver (Rx).
Figure 4.19: Azimuth and Tilt Computation

In the coordinate system S 0 x y z , the receiver coordinates are:
x Rx
cos e Rx sin a Rx d
y Rx =
cos e Rx cos a Rx d
z Rx
(1)
sin e Rx d
Let az and el respectively be the azimuth and tilt of the receiver in the transmitter antenna coordinate system
S Tx x'' y'' z'' . These angles describe the direction of the transmitter-receiver path in the transmitter antenna coordinate
system. Therefore, the receiver coordinates in S Tx x'' y'' z'' are:
Forsk 2009
AT271_TRG_E6
109

x'' Rx
y'' Rx =
z'' Rx
cos el sin az d
cos el cos az d
sin el d
(2)
According to the figure above, we have the following relations:

x'
y' =
z'
cos a Tx sin a Tx 0
x
sin a Tx cos a Tx 0 y
z
0
0
1
(3)
and
1
0
0
x''
x'
y'' = 0 cos e Tx sin e Tx y'
z''
0 sin e Tx cos e Tx
z'
(4)
Therefore, the relation between the system S 0 x y z and the transmitter antenna system S Tx x'' y'' z'' is:
cos a Tx sin a Tx 0
1
0
0
x''
x
=
0
e
cos
sin
y''
sin a Tx cos a Tx 0 y
Tx
Tx
z''
0 sin e Tx cos e Tx
z
0
0
1
(5)
We get,
x''
y'' =
z''
cos a Tx
sin a Tx
0
x
cos e Tx sin a Tx cos e Tx cos a Tx sin e Tx y
z
sin e Tx sin a Tx sin e Tx cos a Tx cos e Tx
(6)
Then, substituting the receiver coordinates in the system S0 from Eq. (1) and the receiver coordinates in the system STx
from Eq. (2) in Eq. (6) leads to a system where two solutions are possible:
1st solution: If a Rx = a Tx , then az = 0 and el = e Rx e Tx
2nd solution: If a Rx a Tx , then
1
az = atan -----------------------------------------------------------------------------------------------cos e Tx
sin e Tx tan e Rx
-------------------------------------- + -------------------------------------------------tan a Rx a Tx
sin a Rx a Tx
and
cos e Tx tan e Rx
sin e Tx
el = atan sin az ------------------------------------- + ---------------------------------------------------
sin a Rx a Tx
tan a Rx a Tx
If sin az sin a Rx a Tx 0 , then az = az + 180
4.6.2
Antenna Pattern 3-D Interpolation

The direction of the transmitter-receiver path in the transmitter antenna coordinate system is given by angle values, az and
el. Atoll considers these values in order to determine transmitter antenna attenuations in the horizontal and vertical
patterns. It reads the attenuation H(az) in the horizontal pattern for the calculated azimuth angle az and the attenuation
V(el) in the vertical pattern for the calculated tilt angle el. Then, it calculates the antenna total attenuation, L antTx az el .
180 az
az
L antTx az el = H az ------------------------- H 0 V el + ---------- H 180 V 180 el
180
180
Notes:
We assume that the horizontal and vertical patterns are two cross-sections of the 3-D
pattern. In other words, the description of the antenna pattern must satisfy the following:
H(0)=V(0) and H()=V()
In case of an electrical tilt, , the horizontal pattern is a conical section with a degrees
elevation off the horizontal plane. Here, horizontal and vertical patterns must satisfy the
following:
H(0)=V() and H()=V(-)
If the constraints listed above are satisfied, this implies that:
110
AT271_TRG_E6
Forsk 2009
1. Interpolated horizontal and vertical patterns respectively fit in with the entered horizontal
and vertical patterns, even in case of electrical tilt,
2. The contribution of both the vertical pattern back and front parts are taken into account.
Otherwise, only the second point is guaranteed.
4.6.3
Atoll uses this modelling method from the Atoll 2.1 version (inclusive) and above. In Atolls
versions prior to the 2.1, another modelling method was available to evaluate angles and
losses due to antenna pattern. The user has the option to choose between these two
methods through Atoll.ini file (see Atoll administration files). For further information about
the old modelling method, please refer to the Technical Reference Guide 2.2.
The above interpolation is performed in dBs.
Angle values in formulas are stated in degrees.
The above interpolation is not used in case the transmitter antenna is described by a 3-D
antenna pattern.
Additional Electrical Downtilt Modelling

The additional electrical downtilt, AEDT, also referred to as remote electrical downtilt or REDT, introduces a conical transformation of the 3-D antenna pattern in the vertical axis. In order to take it into account, the vertical pattern is transformed
as follows:
V x = V x AEDT when x [ 90,90]
V x = V x + AEDT when x [90,270]
Where, the angle values are in degrees.
The vertical pattern transformation is represented below. The left picture shows the initial vertical pattern when there is no
electrical downtilt and the right one shows the vertical pattern transformation due to an electrical downtilt of 10.
Then, Atoll proceeds as explained in the previous section. It determines the antenna attenuation in the transformed vertical
pattern for the calculated tilt angle (V(el)) and applies the 3-D interpolation formula in order to calculate the antenna total
attenuation, L antTx az el .
Figure 4.20: Vertical Pattern Transformation due to Electrical Downtilt
4.6.4
Antenna Pattern Smoothing

Empirical propagation models, like the Standard Propagation Model (SPM), require antenna pattern smoothing in the vertical plane to simulate the effects of reflections and diffractions. Signal level predictions can be improved by smoothing the
high-attenuation points of the vertical pattern. You can smooth vertical as well as horizontal antenna patterns in Atoll.
The antenna pattern smoothing algorithm in Atoll first determines the peaks and nulls in the pattern within the smoothing
angle (ASmoothing) defined by the user. Peaks (P) are the lowest attenuation angles and nulls (N) are the highest attenuation angles in the pattern. Then, it determines the nulls to be smoothed (NSmoothing) and their corresponding angles according to the defined Peak-to-Null Deviation (DPeak-to-Null). DPeak-to-Null is the minimum difference of attenuation in dBs
between two peaks and a null between them. Finally, Atoll smooths the pattern between 0 and the smoothing angle
(ASmoothing) by applying the smoothing to a certain smoothing factor (FSmoothing) defined by the user.
Lets take an example of an antenna pattern to be smoothed, as shown in Figure 4.21: on page 112. Let DPeak-to-Null be
10 dB, ASmoothing = 90 degrees, and FSmoothing = 0.5.
Forsk 2009
AT271_TRG_E6
111
Figure 4.21: Vertical Antenna Pattern

Atoll first determines the peaks and nulls in the part of the pattern to be smoothed by verifying the slopes of the pattern
curve at each angle.
Figure 4.22: Peaks and Nulls in the Antenna Pattern

Peaks (P) and Nulls (N):
Angle ()
Attenuation (dB)
0.1
15
33.5
21
13.2
30
37.6
38
16.9
49
32.2
67
15.6
Then, Atoll verifies whether the difference of attenuation at a given angle is DPeak-to-Null less than the before and after it.
This comparison determines the nulls to be smoothed (NSmoothing).
Nulls to be smoothed (NSmoothing):
Angle ()
Attenuation (dB)
15
33.5
30
37.6
49
32.2
Once the nulls are known, Atoll applies the smoothing algorithm to all the attenuation values at all the angles between the
first peak, the null, and the last peak.
112
AT271_TRG_E6
Forsk 2009
4.6.4.1
Smoothing Algorithm
For all nulls n N Smoothing surrounded by two peaks P1 and P2 at angles 1 and 2 ,
A 2 A 1
- i 1
A i Smoothed = A i F Smoothing A i A + ----------------------1
2 1
Where,
i is the angle in degrees from 1 to 2 incremented by 1 degree,
AAngle is the attenuation at any given angle which can be i, 1 or 2 , and
FSmoothing is the smoothing factor defined by the user.
4.7
Shadowing Model
Propagation models predict the mean path loss as a function of transmission and reception parameters such as frequency,
antenna heights, and distance, etc. Therefore, the predicted path loss between a transmitter and a receiver is constant, in
a given environment and for a given distance. However, in reality different types of clutter may exist in the transmitterreceiver path. Therefore, the path losses for the same distance could be different along paths that pass throught different
types of environments. The location of the receiver in different types of clutter causes variations with respect to the mean
path loss values given by the path loss models. Some paths undergo more loss while others are less obstructed and may
have higher received signal strength. The variation of path loss with respect to the mean path loss values predicted by the
propagation models, depending on the type of environment is called shadow fading (shadowing) or slow fading. "Slow"
fading implies that the variations in the path loss due to shadow fading occur comparatively slower than the fast fading
effect (Rayleigh fading), which is due to the mobile receiving multipath copies of a signal.
Different types of clutter (buildings, hills, etc.) make large shadows that cause variations in the path loss over long
distances. As a mobile passes under a shadow, the path loss to the mobile keeps varying from point to point. Shadow
fading varies as the mobile moves, while fast fading can vary even if the mobile remains at the same location or moves
over very small distances. It is crucial to account for the shadow fading in order to predict the reliability of coverage
provided by any mobile cellular system.
The shadowing effect is modelled by a log-normal (Gaussian) distribution, as shown in Figure 4.23: on page 113, whose
standard deviation depends on the type of clutter.
Figure 4.23: Log-normal Probability Density Function

Different clutter types have different shadowing effects. Therefore, each clutter type in Atoll can have a different standard
deviation representing its shadowing characteristics. For different standard deviations, the shape of the Gaussian distribution curve remains similar, as shown in Figure 4.23: on page 113.
The accuracy of this model depends upon:
Forsk 2009
The suitability of the range of standard deviation used for each clutter class,
The definition (bin size) of the digital map,
How up-to-date the digital map is,
The number of clutter classes,
The accuracy of assignment of clutter classes.
AT271_TRG_E6
113

Shadowing is applied to the predicted path loss differently depending on the technology, and whether it is applied to predictions or simulations. The following sections explain how shadowing margins are calculated and applied to different technology documents.
Shadowing margins are calculated for a given cell edge coverage probability. The cell edge coverage probability is the
probability of coverage at a pixel located at the cell edge, and corresponds to the reliability of coverage that you are planning to achieve at the cell edge. For example, a cell edge coverage probability of 75 % means that the users located at
the cell edge will receive adequate signal level during 75 % of the time. Therefore, a coverage prediction with a cell edge
coverage probability of x % means that the signal level predicted on each pixel is reliable x % of the time, and the overall
predicted coverage area is reliable at least x % of the time.
References:
[1] Saunders S. Antennas and propagation for Wireless Communication Systems pp. 180-198
[2] Holma H., Toskala A. WCDMA for UMTS
[3] Jhong S., Leonard M. CDMA systems engineering handbook pp. 309-315, 1051-1053
[4] Remy J.G., Cueugnet J., Siben C. Systmes de radiocommunications avec les mobiles pp. 309-310
[5] Laiho J., Wacker A., Novosad T. Radio network planning and optimisation for UMTS pp. 80-81

The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 116, and
applied to signal level or C/I as explained below.
Signal Level-Based Predictions

Signal level-based predictions include coverage predictions (Coverage by Transmitter, Coverage by Signal Level,
and Overlapping Zones) and calculations in point analysis tabs (Profile and Reception) that require calculation of
the received signal level only, and do not depend on interference.
In these calculations (signal level calculations), a shadowing margin ( M Shadowing model ) is added to the path loss
( L path ) calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage probability,
and depends on the model standard deviation ( model in dB) associated to the clutter class where the receiver is
located.
Interference-Based Predictions
Interference-based predictions include coverage predictions (Coverage by C/I Level, Interfered Zones, GPRS/
EGPRS Coding Schemes, RLC/MAC Throughout/Timeslot, Application Throughput/Timeslot, Circuit Quality Indicators) and calculations in point analysis windows Interference tab that require calculation of the received signal
level and interference received from other base stations.
In these calculations, ( C I calculations), the shadowing margin ( M Shadowing C I ) is added to the ratio of the
carrier power (C) and the interfering signal levels (I) received from the interfering base stations. This shadowing
margin is calculated for a given cell edge coverage probability and depends on the C/I standard deviation ( C I
in dB) associated to the clutter class where the receiver is located.
UMTS HSPA, IS-95 cdmaOne, and CDMA2000 1xRTT 1xEV-DO Documents

The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 116 and
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 117, and applied to signal level, Ec/I0, or Eb/Nt as
explained below.

located.
Interference+noise-Based Predictions
Interference+noise-based predictions include coverage predictions (Pilot Reception Analysis, Downlink Total
Noise, Service Area Analyses, Handoff Status, etc.) and point analysis (AS Analysis tab) that require calculation
of the received signal level and intra-cellular interference and noise received from other base stations.
In these calculations, the shadowing margins ( M Shadowing Ec Io
M Shadowing Eb Nt
114
UL
and
M Shadowing Eb Nt
DL
) , or
) are added to Ec/I0 or Eb/Nt. This shadowing margin is calculated for a given cell edge
AT271_TRG_E6
Forsk 2009
coverage probability and depends on the Ec/I0 or Eb/Nt standard deviations ( Ec Io , Eb Nt
DL
, or Eb Nt
UL
in dB) associated to the clutter class where the receiver is located.
Macro-Diversity Gains
UL
DL
Atoll calculates the uplink and downlink macro-diversity gains ( G macro diversity and G macro diversity ) depending
on the receiver handover status. These gains are respectively taken into account to evaluate the uplink Eb/Nt in
case of soft handover and the downlink Ec/Io from best server. For detailed description of the calculation of macrodiversity gains, please refer to "Macro-Diversity Gains Calculation" on page 118.
Monte-Carlo Simulations
Random values for shadowing margins are calculated for each transmitter-receiver link and added to the predicted
path loss. A shadowing margin for each transmitter-receiver link in each simulation is obtained by taking a random
value from the probability density distribution for the appropriate clutter class. The probability distribution is a lognormal distribution as explained above.
TD-SCDMA Documents
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 117, and applied to signal level or interference+noise
predictions as explained below.

Signal level-based predictions include coverage predictions (Best Server and RSCP P-CCPCH Coverages, PCCPCG Pollution, Baton Handover Coverage, DwPCH and UpPCH Coverages, Cell to Cell Interference, and
Scrambling Code Interference) and calculations in point analysis tabs (Profile and Reception) that require calculation of the received signal level only, and do not depend on interference.
located.
Interference+noise-based predictions include coverage predictions (P-CCPCH Eb/Nt and C/I Coverages, Service
Area Analsyses for downlink and uplink Eb/Nt and C/I, etc.) that require calculation of the received signal level and
interference received from other base stations.
In these calculations, the shadowing margins ( M Shadowing Eb Nt
M Shadowing Eb Nt
UL
P CCPCH
, M Shadowing Eb Nt
DL
, or
) are added to Eb/Nt. This shadowing margin is calculated for a given cell edge coverage
probability and depends on the Eb/Nt standard deviations ( Eb Nt
P CCPCH
, Eb Nt
DL
, or Eb Nt
UL
, in dB)
associated to the clutter class where the receiver is located.
WiMAX 802.16d and WiMAX 802.16e Documents

"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 117 , and applied to signal level or C/(I+N) as
explained below.

located.
Interference-based predictions include coverage predictions (Coverage by C/(I+N) Level, Coverage by Best
Bearer, Coverage by Channel Throughput) that require calculation of the received signal level and interference
received from other base stations.
Forsk 2009
AT271_TRG_E6
115

In these calculations, (C/(I+N) calculations), the shadowing margin ( M Shadowing C I ) is added to the ratio of the
carrier power (C) and the interfering signal levels (I) and noise received from the interfering base stations. This
shadowing margin is calculated for a given cell edge coverage probability and depends on the C/I standard deviation ( C I in dB) associated to the clutter class where the receiver is located.
4.7.1
Shadowing Margin Calculation

The following sections describe the calculation method used for determining different shadowin margins.
The following shadowing margins are calculated using the method described below:
Network Type
GSM GPRS EGPRS
UMTS HSPA
IS-95 cdmaOne
CDMA2000 1xRTT 1xEV-DO
Standard
Deviation
MShadowing
Applied to
model
M Shadowing model
C I
M Shadowing C I
C/I
model
M Shadowing model
Ec Io
M Shadowing Ec Io
Ec/I0
Eb Nt
DL
M Shadowing Eb Nt
DL
Eb/Nt (DL)
Eb Nt
UL
M Shadowing Eb Nt
UL
Eb/Nt (UL)
model
M Shadowing model
Ec Io
M Shadowing Ec Io
Ec/I0
Eb Nt
DL
M Shadowing Eb Nt
DL
Eb/Nt (DL)
Eb Nt
UL
M Shadowing Eb Nt
UL
Eb/Nt (UL)
model
Eb Nt
TD-SCDMA
WiMAX 802.16d
WiMAX 802.16e
4.7.1.1
P CCPCH
M Shadowing model
M Shadowing Eb Nt
P CCPCH
C
Eb/Nt P-CCPCH
Eb Nt
DL
M Shadowing Eb Nt
DL
Eb/Nt (DL)
Eb Nt
UL
M Shadowing Eb Nt
UL
Eb/Nt (UL)
model
M Shadowing model
C I
M Shadowing C I
C/(I+N)
Shadowing Margin Calculation in Predictions

Shadowing margins, MShadowing, are calculated from standard deviation values defined for the clutter class where the pixel
(probe mobile) is located, and required cell edge coverage probability, and added to the path loss, Lpath.
Shadowing Error PDF (1 Signal)

The measured path loss in dB can be expressed as a Gaussian random variable:
L = L path + dB G 0 1
where,
Lpath is the predicted path loss,

dB is the user-defined standard deviation of the error,
G(0,1) is a zero-mean unit-variance Gaussian random variable.
Therefore, the probability density function (pdf) for the random (shadowing) part of path loss is:
2
1
p L x = --------------------- e
dB 2
116
x ------------2
2 dB
AT271_TRG_E6
Forsk 2009
The probability that the shadowing error exceeds z dB is
2
x ------------2
2 dB
PL x z =
p L x dx =
z
1
--------------------- e
dB 2
dx
Normalising x by dividing it bydB:

2
1
P L x z = -----------
2
z-------- dB
x
-----2
z
dx = Q ---------
dB
where Q is the complementary cumulative function.

To ensure a given cell edge coverage probability, R L , for the predicted value, a shadowing margin, M Shadowing , is added
to the link budget.
Confidence in the prediction can be expressed as:
C d = P' Tx L P rec L P' Tx P rec G 0 1 dB M Shadowing
where,
P rec is the signal level predicted at the receiver. P rec = P' Tx L path M Shadowing
P' Tx = EIRP + G antRx L Rx
EIRP is the effective isotropic radiated power of the transmitter.

L Rx are receiver losses.
G antRx is the receiver antenna gain.
The shadowing margin is calculated such that:

M Shadowing
P C d P rec = R L M Shadowing = 1 P L x M Shadowing 0 = 1 Q ------------------------------
dB
A lookup table is used for mapping the values of Q vs. a set of cell edge coverage probabilities.
M Shadowing
Figure 4.24: Normalised Margin M arg in = ---------------------------- dB
In interference-based predictions, where signal to noise ratio is calculated, the shadowing margin is only applied to the
signal from the interfered transmitter (C). We consider that the interference value is not altered by the shadowing margin.
Random variations also exist in the interfering signals, but taking only the average interference gives accurate results. [3]
explains how a certain level of interference is maintained by congestion control in CDMA-based networks.
4.7.1.2
Shadowing Margin Calculation in Monte-Carlo Simulations

Shadowing margins, MShadowing, are calculated from standard deviation values defined for the clutter class where the pixel
(probe mobile) is located, and required cell edge coverage probability, and added to the path loss, Lpath.
Random values are generated during Monte-Carlo simulation. Each user is assigned a service, a mobility type, an activity
status, a geographic position and a random shadowing value.
For each link, path loss (L) can be broken down to L = L path + .
Here, is a zero mean gaussian random variable G 0 dB representing variation due to shadowing. It can be
expressed as the sum of two uncorrelated zero mean gaussian random variables, L and P . L models the error related
to the receivers location (surrounding environment), and remains the same for all links between the receiver and the base
stations from which it is receiving signals. P models the error related to the path between the transmitter and the receiver.
Therefore, in case of two links, we have:
Forsk 2009
AT271_TRG_E6
117

1
1 = L + P for link 1
2
2 = L + P for link 2
i
Standard deviations of L L and P P can be calculated from i , the model standard deviation model , and the
correlation coefficient between 1 and 2 .
Assuming all P have the same standard deviations, we have:
2
model = L + P
2
L
= ---------------2
model
Therefore,
2
P = model 1
L = model
is set to 0.5 in Atoll, which gives:
model
model
L = ---------------- and P = ---------------2
2
Receiver
Therefore, to model shadowing error common to all the signals received at a receiver ( E Shadowing model ), values are
randomly generated for each receiver. These values have a zero-mean gaussian distribution with a standard deviation of
model
---------------- , where model is the model standard deviation associated with the receivers clutter class.
2
Next, Atoll generates another random value for each transmitter-receiver pair. This values represents the shadowing error
Path
not related to the location of the receiver ( E Shadowing model ). These values also have a zero-mean gaussian distribution
model
with a standard deviation ----------------- .
2
So, we have:
Receiver
Path
E Shadowing model = E Shadowing model + E Shadowing model

Random shadowing error has its mean value at zero. Hence, this shadowing modelling method has no impact on the simulated network load. On the other hand, as shadowing errors on the transmitter-receiver links are uncorrelated, the method
influences the calculated macro-diversity gain in case the mobile is in soft handover.
4.7.2
Macro-Diversity Gains Calculation

The following sections explain how uplink and downlink macro-diversity gains are calculated in UMTS HSPA, IS-95
cdmaOne, and CDMA2000 1xRTT 1xEV-DO documents for predictions and AS Analysis tab of the point analysis tool.
4.7.2.1
Uplink Macro-Diversity Gain Evaluation

In UMTS, CDMA2000 and IS95-CDMA, mobiles may be in soft handoff (mobile connected to cells located on different
sites). In this case, we can consider the shadowing error pdf described below.
4.7.2.1.1
Shadowing Error PDF (n Signals)

For each link, path loss (L) can be broken down as:
L = L path +
is a zero mean gaussian random variable G 0 dB representing variation due to shadowing. It can be expressed as
the sum of two uncorrelated zero mean gaussian random variables, L and P . L models error related to the receiver
local environment; it is the same whichever the link. P models error related to the path between transmitter and receiver.
1
1 = L + P for the link 1

2
118
AT271_TRG_E6
Forsk 2009
Knowing i , the uplink Eb/Nt standard deviation Eb Nt and the correlation coefficient between 1 and 2 , we
UL
can calculate standard deviations of L L and P P (assuming all P have the same standard deviations).
We have:
2
Eb Nt
= L + P
UL
L
= ------------------------2
Eb Nt
UL
Therefore,
2
P = Eb Nt
2
UL
L = Eb Nt
UL
2 Signals Without Recombination

In technologies supporting soft handoff (UMTS, CDMA2000, IS95-CDMA), cell is interference limited. As for one link, to
ensure a required cell edge coverage probability R L for the prediction, we add to each link budget a shadowing margin,
2signals
M Shadowing Eb Nt
UL
Prediction reliability in order to have Eb/Nt higher or equal to Eb/Nt from the best server can be expressed as:
Cd
1
1
--------1- = P' Tx1 L 1 N 1 CI pred 1 P' Tx1 L path N 1 CI pred
1
N1
or
Cd
1
1
--------2- = P' Tx2 L 2 N 2 CI pred 2 P' Tx2 L path N 2 CI pred
2
N2
where
i
CI pred is the quality level (signal to noise ratio) predicted at the receiver for link i.
Ni is the noise level for link i.
We note:
2signals
M Shadowing Eb Nt
UL
= P' Txi L path N i CI pred

i
and
2
1 = CI pred CI pred
2
1 is the minimum needed margin on each link.

Therefore, the probability of having a quality at least equal to the best predicted one is:
noMRC
RL
noMRC
RL
Cd
Cd
2signals
1
1
M Shadowing Eb Nt = 1 P L1 L2 --------1- CI pred --------2- CI pred
UL
N
N
1
2
2signals
M Shadowing Eb Nt = 1 P
UL
1 2
2signals
2signals
1 M Shadowing Eb Nt 2 M Shadowing Eb Nt
UL
UL
We can express it using L , P and P

P
1 2
2signals
UL
= P L P
L
1 2
2signals
1
P P
2signals
P M Shadowing Eb Nt
2signals
1
Forsk 2009
2signals
2signals
UL
2signals
UL
1 L = L
L P M Shadowing Eb Nt
= P L P P M Shadowing Eb Nt
L
UL
UL
UL
1 L
UL
1 L = L
2signals
L P P M Shadowing Eb Nt
P
AT271_TRG_E6
UL
1 L
119

noMRC
RL
2signals
M Shadowing Eb Nt
UL
1
2signals
2
2signals
2
= 1 P L P P M Shadowing Eb Nt L P P M Shadowing Eb Nt 1 L d L
L
P
UL
P
UL
2signals
P P M Shadowing Eb Nt
P
1
= ----------------- 2
P
M 2signals
Shadowing Eb Nt
UL
L
2
e
UL
x
---------2
2 P
2signals
M Shadowing Eb Nt UL L
dx = Q ---------------------------------------------------------------------
P
Then, we have:
noMRC
RL
2signals
M Shadowing Eb Nt
UL
2signals
M Shadowing Eb Nt UL
2signals
2
L
M Shadowing Eb Nt UL 1 L
- d L
= 1 P L Q --------------------------------------------------------------------- Q --------------------------------------------------------------------------------
P
P
If we introduce user defined standard deviation Eb Nt and correlation coefficient , and consider that P is a
UL
Gaussian pdf:
noMRC
RL
2signals
M Shadowing Eb Nt
UL
1
= 1 ----------- e
xL
--------2
M 2signals
Shadowing Eb Nt UL
M 2signals
x L Eb Nt
Shadowing Eb Nt UL x L Eb Nt UL 1
UL
- Q ------------------------------------------------------------------------------------------------------------------- dx L
Q ------------------------------------------------------------------------------------------------------
1
1
Eb Nt
Eb Nt
UL
UL
n Signals Without Recombination

We can generalize the previous expression to n signals (n is the number of available signals - Atoll may consider up to 3
signals):
noMRC
RL
nsignals
M Shadowing Eb Nt
UL
1
= 1 ----------- e
xL
--------2
M nsignals
Shadowing Eb Nt UL
M nsignals
x L Eb Nt
Shadowing Eb Nt UL x L Eb Nt UL 1
UL
- Q ------------------------------------------------------------------------------------------------------------------- dx L
Q ------------------------------------------------------------------------------------------------------
1
1
Eb Nt
Eb Nt
UL
UL
The case where softer handoff occurs (two signals from co-site cells) is equivalent to the one signal case. The Softer/soft
case is equivalent to the two signals case. For the path associated with the softer recombination, we will use combined
SNR to calculate the availability of the link.
Correlation Coefficient Determination

There is currently no agreed model for predicting correlation coefficient between 1 and 2 . Two key variables influence correlation:
The angle between the two signals. If this angle is small, correlation is high.
The relative values of the two signal lengths. If angle is 0 and lengths are the same, correlation is zero. Correlation
is different from zero when path lengths differ.
A simple model has been found [1]:

T D1
= ------ -------- when T
D2
T is a function of the mean size of obstacles near the receiver and is also linked to the receiver environment.
In a normal handover status, assuming a hexagonal design for sites, is close to (+/- /3) and D1/D2 is close to 1.
In [1,5], = 0.5 when = 0.3 and T = ------ .

10
In Atoll, is set to 0.5.
120
AT271_TRG_E6
Forsk 2009
4.7.2.1.2
Uplink Macro-Diversity Gain

UL
Atoll determines the uplink macro-diversity gain ( G macro diversity ) from the shadowing margins calculated in case of one
signal and n signals.
Therefore, we have:
UL
nsignals
G macro diversity = M Shadowing Eb Nt
UL
M Shadowing Eb Nt
UL
Where n is the number of cell-mobile signals.
4.7.2.2
Downlink Macro-Diversity Gain Evaluation

In UMTS, CDMA2000 and IS95-CDMA, in case of soft handoff, mobiles are able to switch from one cell to another if the
best pilot drastically fades. To model this function, we have to consider the probability of fading over the shadowing margin,
both for the best signal and for all the other available signals, in the shadowing margin calculation.
Let us consider the shadowing error pdf described below.
4.7.2.2.1
Shadowing Error PDF (n Signals)

For each link, path loss (L) can be broken down as:
L = L path +
is a zero mean gaussian random variable G 0 dB representing variation due to shadowing. It can be expressed as
the sum of two uncorrelated zero mean gaussian random variables, L and P . L models the error related to the receiver
local environment, which is the same for all links. P models the error related to the path between the transmitter and the
receiver.
1

2

Knowing i , the Ec/Io standard deviation Ec I o and the correlation coefficient between 1 and 2 , we can calculate
standard deviations of L L and P P (assuming all P have the same standard deviations).
We have:
2
Ec I o = L + P
2
L
= --------------2
Ec I o
Therefore,
2
P = Ec I o 1
L = Ec I o
2 Available Signals
In technologies supporting soft handoff (UMTS, CDMA2000 and IS95-CDMA), cells are interference limited. As for one
link, to ensure a required cell edge coverage probability R L for the prediction, we add a shadowing margin,
2signals
M Shadowing Ec Io , to each link budget.

Ec
Ec
Prediction reliability to have ------- -------
for the best server can be expressed as:
Io Io pred
Ec
Ec 1
Ec 1
---------1- = P pilot L 1 Io -------
1 P pilot L m Io -------
Io pred
Io pred
1
1
1
Io
Or
Ec
Ec 1
Ec 1
2 P pilot L m Io -------
---------2- = P pilot L 2 Io -------
Io pred
Io pred
2
2
2
Io
We note:
1
Ec
2signals
M Shadowing Ec Io = P pilot L m Io -------
Io pred
i
i
Forsk 2009
AT271_TRG_E6
121

Ec 1
Ec 2
2
1 = -------
-------
Io pred Io pred
2
1 is the minimum needed margin on each link.

Therefore, probability of having a quality at least equal to the best predicted one is:
noMRC
RL
noMRC
RL
Ec 1
Ec 2
Ec 1
Ec 1
2signals
M Shadowing Ec Io = 1 P L1 L2 ---------- -------
---------- -------
Io Io pred Io Io pred
2signals
2signals
2signals
M Shadowing Ec Io = 1 P 1 2 1 M Shadowing Ec Io 2 M Shadowing Ec Io 1

1
We can express it by using L , P and P

2signals
2signals
P 1 2 1 M Shadowing Ec Io 2 M Shadowing Ec Io 1 L = L
= P L P
L
P P
2signals
2signals
P M Shadowing Ec Io L P M Shadowing Ec Io 1 L
2signals
2signals
P 1 2 1 M Shadowing Ec Io 2 M Shadowing Ec Io 1 L = L
1
2signals
2signals
= P L P P M Shadowing Ec Io L P P M Shadowing Ec Io 1 L
L
noMRC
RL
2signals
M Shadowing Ec Io
P L P P MShadowing Ec Io L P P MShadowing Ec Io 1 L dL
1
= 1
2signals
2signals
i
P P
P
2signals
M Shadowing Ec Io
1
L = ----------------- P 2
SHO L
x
---------2
2 P
2signals
M Shadowing Ec Io L
dx = Q -----------------------------------------------------------
P
Then, we have:
noMRC
RL
2signals
M Shadowing Ec Io = 1
2signals
2signals
M Shadowing Ec Io L
M Shadowing Ec Io 1 L
P L Q ----------------------------------------------------------- Q ------------------------------------------------------------------------ d L
L
P
P
If we introduce a user defined Ec/Io standard deviation and a correlation coefficient and consider that P is a
L
Gaussian pdf:
noMRC
RL
2signals
M Shadowing Ec Io
2
1
= 1 ----------2
xL
--------2
2signals
2signals
M Shadowing Ec Io x L Ec I o
M Shadowing Ec Io 1 x L Ec I o
Q ------------------------------------------------------------------------------------ Q ------------------------------------------------------------------------------------------------ dx L
Ec I o 1
Ec I o 1
n Available Signals
We can generalize the previous expression for n signals (n is the number of available signals - Atoll may consider up to 3
signals):
noMRC
RL
nsignals
M Shadowing Ec Io
2
1
= 1 ----------2
122
xL
--------2
nsignals
M Shadowing Ec Io x L Ec I o
Q ------------------------------------------------------------------------------------ x
Ec I o 1
AT271_TRG_E6
i=2
nsignals
M Shadowing Ec Io 1 x L Ec I o
Q ------------------------------------------------------------------------------------------------ dx L
Ec I o 1
Forsk 2009
1 =1 dB
2
1 =5 dB
2
1 =10 dB
Figure 4.25: Margin - Probability (Case of 2 Signals)
2 signals
3
1 =5 dB
3
1 =10 dB
Figure 4.26: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 1dB)
2 signals
3
1 =5 dB
3
1 =10 dB
Figure 4.27: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 2dB)
Correlation Coefficient Determination

For further information about determination of the correlation coefficient, please see "Correlation Coefficient Determination" on page 123.
4.7.2.2.2
Downlink Macro-Diversity Gain

DL
Atoll determines the downlink macro-diversity gain ( G macro diversity ) from the shadowing margins calculated in case of
one signal and n signals.
Therefore, we have:
DL
nsignals
G macro diversity = M Shadowing Ec Io M Shadowing Ec Io

Where n is the number of available signals.
Note:
Forsk 2009
Atoll uses the DL macro-diversity gain to calculate Ec/Io. You can force Atoll not to take it
into account through the Atoll.ini file (see Atoll administration files). You must create this file
and place it in the Atoll installation directory.
AT271_TRG_E6
123
4.8
Appendices
4.8.1
Transmitter Radio Equipment

Radio equipment such as TMA, feeder and BTS, are taken into account to evaluate:
Total UL and DL losses of transmitter ( L total UL L total DL ) and transmitter noise figure NF Tx in UMTS HSPA,
CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, TD-SCDMA, WiMAX 802.16d, and WiMAX 802.16e documents,
Transmitter total losses L Total in GSM GPRS EGPRS documents.
In Atoll, the transmitter-equipment pair is modelled a single entity. The entry to the BTS is considered the reference point
which is the location of the transmission/reception parameters.
Figure 4.28: Reference Point - Location of the Transmission/Reception parameters
4.8.1.1
UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, and

TD-SCDMA Documents
As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we
have:
NF TX = NF BTS
where NF BTS is the BTS noise figure.
Atoll calculates total UL losses as follows:
UL
UL
UL
UL
L Total UL = L Misc + L Feeder + L BTS Conf + NR Repeaters G Ant div G TMA (in dB)
where,
UL
L Misc are the miscellaneous reception losses (Transmitter property),

UL
UL
UL
UL
UL
L Feeder are the feeder reception losses ( L Feeder = L Feeder I Feeder + L Connector , where L Feeder , I Feeder and
UL
L Connector are respectively the feeder loss per metre (Feeder property), the reception feeder length in metre (Transmitter
property) and the connector reception losses,
UL
L BTS Conf are the losses due to BTS configuration (BTS property),
UL
G Ant div is the antenna diversity gain (Transmitter property),

NR Repeaters is the noise rise at transmitter due to repeaters. This parameter is taken into account only if the transmitter
has active repeater(s),
G TMA is the gain due to TMA.
The noise rise at transmitter due to repeaters is calculated as follows:
For each active repeater ( k ), Atoll calculates a noise injection margin ( NIM Rp ). This is the difference between the donor
k
transmitter noise figure ( NF TX ) and the repeater noise figure received at the donor.
Rp k
NIM Rp = NF TX NF Rp + G amp L
r
k
TX Rp k
(in dB)
where,
NF Rp is the repeater noise figure,
k
Rp k
G amp is the repeater amplification gain (repeater property),
124
AT271_TRG_E6
Forsk 2009
L
TX R p k
are the losses between the donor transmitter and the repeater (repeater property).
For each active repeater ( k ), Atoll converts the noise injection margin ( NIM Rp ) to Watt. Then, it uses the values to calcuk
late the noise rise at the donor transmitter due to active repeaters ( NR Repeaters ).
NR Repeaters = 10 Log 1 +
-
-----------------NIM Rp
(in dB)
The gain due to TMA is calculated as follows:

WithoutTMA
WithTMA
G TMA = NF Composite NF Composite (in dB)

where,
WithTMA
WithoutTMA
NF Composite and NF Composite
are the composite noise figures with and without TMA respectively.
Friis' equation is used to calculate the composite noise figure when there is a TMA.
NF Feeder 1
NF BTS 1
WithTMA
NF Composite = NF TMA + -------------------------------------+ --------------------------------------- (not in dB)
UL
UL
UL
G TMA
G TMA G Feeder
And,
WithoutTMA
NF Composite
= NF BTS + NF Feeder (in dB)
where,
NF Feeder is the feeder noise figure,
NF TMA is the TMA noise figure,
NF BTS is the BTS noise figure,
UL
G TMA is the TMA reception gain,

UL
UL
UL
G Feeder is the feeder UL gain; G Feeder = L Feeder .

UL
UL
UL
UL
UL
UL
L Feeder is the feeder reception loss ( L Feeder = L Feeder I Feeder + L Connector , where L Feeder , I Feeder and L Connector
are respectively the feeder loss per metre, the reception feeder length in metre and the connector reception loss),
Notes:
According to the book Radio network planning and optimisation for UMTS by Laiho J.,
Wacker A., Novosad T., the noise figure corresponds to the loss in case of passive
components. Therefore, feeder noise figure is equal to the cable uplink losses.
UL
NF Feeder = L Feeder (in dB)
Loss and gain inputs specified in .atl documents must be positive values.
Atoll calculates total DL losses as follows.

DL
DL
DL
DL
L Total DL = L TMA + L Feeder + L Misc + L BTS Conf (in dB)

where,
DL
L TMA is the TMA transmission loss,

DL
DL
DL
DL
DL
L Feeder is the feeder transmission loss ( L Feeder = L Feeder I Feeder + L Connector , where L Feeder , I Feeder and
DL
L Connector are respectively the feeder loss per metre, the transmission feeder length in metre and the connector transmission losses),
DL
L Misc are the miscellaneous transmission losses,

DL
L BTS Conf are the losses due to BTS configuration (BTS property).
4.8.1.2

Atoll calculates DL total losses as follows:
DL
DL
DL
DL
L Total DL = L TMA + L Feeder + L Misc + L BTS Conf (in dB)

where,
Forsk 2009
AT271_TRG_E6
125

DL

DL
DL
DL
DL
DL
DL
L Connector are respectively the feeder loss per metre, the transmission feeder length in metre and the connector transmission loss),
DL

DL
4.8.1.3
WiMAX 802.16d and WiMAX 802.16e Documents

As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we
have:
NF TX = NF BTS
where NF BTS is the BTS noise figure.
Atoll calculates total UL losses as follows:
UL
UL
UL
UL
L Total UL = L Misc + L Feeder + L BTS Conf G Ant div G TMA (in dB)
where,
UL
L Misc are the miscellaneous reception losses (Transmitter property),

UL
UL
UL
UL
UL
L Feeder are the feeder reception losses ( L Feeder = L Feeder I Feeder + L Connector , where L Feeder , I Feeder and
UL
L Connector are respectively the feeder loss per metre (Feeder property), the reception feeder length in metre (Transmitter
property) and the connector reception losses,
UL
L BTS Conf are the losses due to BTS configuration (BTS property),
UL
G Ant div is the antenna diversity gain (Transmitter property),

G TMA is the gain due to TMA, which is calculated as follows:
WithoutTMA
WithTMA
G TMA = NF Composite NF Composite (in dB)

where,
WithTMA
WithoutTMA
NF Composite and NF Composite
are the composite noise figures with and without TMA respectively.
Friis' equation is used to calculate the composite noise figure when there is a TMA.
NF Feeder 1
NF BTS 1
WithTMA
+ --------------------------------------- (not in dB)
NF Composite = NFTMA + -------------------------------------UL
UL
UL
G TMA
G TMA G Feeder
WithoutTMA
And NF Composite
= NF BTS + NF Feeder (in dB)
where,
NF Feeder is the feeder noise figure,
NF TMA is the TMA noise figure,
NF BTS is the BTS noise figure,
UL
G TMA is the TMA reception gain,

UL
UL
UL
G Feeder is the feeder UL gain; G Feeder = L Feeder .

UL
UL
UL
UL
UL
UL
L Feeder is the feeder reception loss ( L Feeder = L Feeder I Feeder + L Connector , where L Feeder , I Feeder and L Connector
are respectively the feeder loss per metre, the reception feeder length in metre and the connector reception loss),
Notes:
According to the book Radio network planning and optimisation for UMTS by Laiho J.,
Wacker A., Novosad T., the noise figure corresponds to the loss in case of passive
components. Therefore, feeder noise figure is equal to the cable uplink losses.
UL
NF Feeder = L Feeder (in dB)
126
AT271_TRG_E6
Forsk 2009
Loss and gain inputs specified in .atl documents must be positive values.
Atoll calculates total DL losses as follows.

DL
DL
DL
DL
L total DL = L TMA + L Feeder + L Misc + L BTS Conf (in dB)

where,
DL

DL
DL
DL
DL
DL
DL
L Connector are respectively the feeder loss per metre, the transmission feeder length in metre and the connector transmission losses),
DL

DL
4.8.2
Secondary Antennas
When secondary antennas are installed on a transmitter, the signal level received from it is calculated as follows:
G ant mTx
P 1
G ant i
X i ----------------------Tx
Tx
P Tx X i --------------------L Tx
L Tx
i
----------------------------------------------
----------------------------------------------------------------------- +
L ant m az m el m
L ant i az i el i
Tx
Tx
i
= ------------------------------------------------------------------------------------------------------------------------------------------ (not in dB2)

L model
P rec
Where,
PTx is the transmitter power (Ppilot in UMTS, CDMA2000 and IS95-CDMA documents),
i is the secondary antenna index,
xi is the percentage of power dedicated to the secondary antenna, i,
G ant m
Tx
is the gain of the main antenna installed on the transmitter,
LTx are transmitter losses (LTx=Ltotal-DL),

G ant i
Tx
is the gain of the secondary antenna, i, installed on the transmitter,
Lmodel is the path loss calculated by the propagation model,

L ant m az m el m is the attenuation due to main antenna pattern,
Tx
L ant i az i el i is the attenuation due to pattern of the secondary antenna, i.

Tx
The definition of angles, az and el, depends on the used calculation method.
2.
Forsk 2009
Method 1 (must be indicated in an Atoll.ini file):

- azm is the difference between the receiver antenna azimuth and azimuth of the transmitter main antenna,
- elm is the difference between the receiver antenna tilt and tilt of the transmitter main antenna,
- azi is the difference between the receiver antenna azimuth and azimuth of the transmitter secondary antenna,
i,
- eli is the difference between the receiver antenna tilt and tilt of the transmitter secondary antenna, i,
Method 2 (default):
- azm is the receiver azimuth in the coordinate system of the transmitter main antenna,
- elm is the receiver tilt in the coordinate system of the transmitter main antenna,
- azi is the receiver azimuth in the coordinate system of the transmitter secondary antenna, i,
- eli is the receiver tilt in the coordinate system of the transmitter secondary antenna, i,
Formula cannot be directly calculated from components stated in dB and must be converted in linear values.
AT271_TRG_E6
127
128
AT271_TRG_E6
Forsk 2009
Chapter 5
GSM GPRS EGPRS Networks
This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations and
prediction studies available in GSM GPRS EGPRS projects.
Atoll
130
AT271_TRG_E6
Forsk 2009
Chapter 5: GSM GPRS EGPRS Networks
GSM GPRS EGPRS Networks
5.1
General Prediction Studies
5.1.1
Calculation Criteria
Three criteria can be studied in point analysis (Profile tab) and in general coverage studies. Study criteria are detailed in
the table below.
Study criteria
Formulas
Signal level received from a transmitter on a TRX type
Txi
Signal level ( P rec )
Txi
Txi
P rec tt = EIRP tt P tt L path M Shadowing model L Indoor + G ant
Rx
L Rx
Txi
Txi
Path loss ( L path )

Txi
Txi
Tx
Txi
L total = L path + M Shadowing model + L Indoor + L Tx + L Rx G ant
Total losses ( L total )
Tx
+ G ant
Rx
where,
EIRP is the effective isotropic radiated power of the transmitter,
L model is the loss on the transmitter-receiver path (path loss) calculated by the propagation model,
L ant
Tx
is the transmitter antenna attenuation (from antenna patterns),
M Shadowing model is the shadowing margin. This parameter is taken into account when the option Shadowing taken into
account is selected,
L Indoor are the indoor losses. These losses are defined for each clutter class. They are taken into account when the option
L Rx are the receiver losses,
G ant
Rx
is the receiver antenna gain,
P is the power offset defined for the selected TRX type in the transmitter property dialog,
tt is the TRX type (in the GSM GPRS EGPRS.mdb document template, there are three possible TRX types, BCCH, TCH
and inner TCH).
5.1.2
Point Analysis
5.1.2.1
Profile Tab
Txi
Atoll displays the signal level received from the selected transmitter on a TRX type ( P rec tt ).
Notes:
If power offsets of subcells are identical, field level received from a selected transmitter will
be the same for all the studied TRX types.
For a selected transmitter, it is also possible to study the path loss, L path , or the total
Txi
Txi
losses, L total . Path loss and total losses are the same on any TRX type.
5.1.2.2
Reception Tab
Analysis provided in the Reception tab is based on path loss matrices. So, you can study reception from TBC transmitters
for which path loss matrices have been computed on their calculation areas.
Txi
For each transmitter, Atoll displays the signal level received on a TRX type, ( P rec tt ).
Reception bars are displayed in a decreasing signal level order. The maximum number of reception bars depends on the
signal level received from the best server. Only reception bars of transmitters whose signal level is within a 30 dB margin
from the best server can be displayed.
Forsk 2009
AT271_TRG_E6
131

Notes:
If power offsets of subcells are identical, field level received from a given transmitter will be
the same whichever the studied TRX type.
It is also possible to study the path loss, L path , or the total losses, L total of each
You can use a value other than 30 dB for the margin from the best server signal level, for
example a smaller value for improving the calculation speed. For more information on
defining a different value for this margin, see the Administrator Manual.
Txi
Txi
transmitter. Path loss and total losses are the same on any TRX type.
5.1.3
Coverage Studies
For each TBC transmitter, Txi, Atoll determines the selected criterion on each bin inside the Txi calculation area. In fact,
each bin within the Txi calculation area is considered as a potential (fixed or mobile) receiver.
Coverage study parameters to be set are:
5.1.3.1
The study conditions in order to determine the service area of each TBC transmitter,
The display settings to select how to colour service areas.
Service Area Determination

Atoll uses parameters entered in the Condition tab of the coverage study property dialog to predetermine areas where it
will display coverage.
We can distinguish seven cases as below. Let us assume that:
5.1.3.1.1
Each transmitter, Txi, belongs to a Hierarchical Cell Structure (HCS) layer, k, with a defined priority and a defined
reception threshold.
The maximum range option (available in the System tab of the Predictions property dialog) is inactive.
All Servers
For each HCS layer, k, the service area of Txi corresponds to the bins where:
Txi
Txi
MinimumThreshold P rec tt or L tot orTotal Losses
Txi
MaximumThreshold
Note:
5.1.3.1.2
The minimum threshold is either globally defined or specifically for each subcell (subcell
reception threshold)
Best Signal Level and a Margin

The service area of Txi corresponds to the bins where:
Txi
Txi
Txi
MinimumThreshold P rec ic or L total or L path MaximumThreshold

And
Txi
Txj
P rec ic Best P rec ic M
ji
M is the specified margin (dB).

Best function: considers the highest value.
Notes:
If the margin equals 0 dB, Atoll will consider bins where the signal level received from Txi is
the highest.
If the margin is set to 2 dB, Atoll will consider bins where the signal level received from Txi
is either the highest or 2dB lower than the highest.
If the margin is set to -2 dB, Atoll will consider bins where the signal level received from Txi
is 2dB higher than the signal levels from transmitters, which are 2nd best servers.
5.1.3.1.3
Second Best Signal Level and a Margin

Txi
Txi
Txi

And
Txi
P rec ic 2
132
nd
Best P Txj ic M
rec
ji
AT271_TRG_E6
Forsk 2009

2nd Best function: considers the second highest value.
Notes:
the second highest.
is either the second highest or 2dB lower than the second highest.
is 2dB higher than the signal levels from transmitters, which are 3rd best servers.
5.1.3.1.4
Best Signal Level per HCS Layer and a Margin

Txi
Txi
Txi
MaximumThreshold
And
Txi
Txj
P rec BCCH Best P rec BCCH M
ji

Notes:
the highest.
is 2dB higher than the signal levels from transmitters that are the 2nd best servers.
5.1.3.1.5
HCS Servers and a Margin

Txi
Txi
Txi
MaximumThreshold
And
Txi
Txj
ji
Txi
The received P rec tt exceeds the reception threshold defined per HCS layer
Notes:
the highest.
5.1.3.1.6
Highest Priority HCS Server and a Margin

In this case, the service area of Txi corresponds to the bins where:
Txi
Txi
Txi
MaximumThreshold
And
Txi
Txj
ji
And
Forsk 2009
AT271_TRG_E6
133

Txi belongs to the HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest) assumTxi
ing the received P rec tt exceeds the reception threshold defined per HCS layer.
Notes:
the highest.
5.1.3.1.7
In the case two layers have the same priority, the traffic is served by the transmitter for
which the difference between the received signal strength and the HCS threshold is the
highest. The way the competition is managed between layers with the same priority can be
modified. For more information, see the Administrator Manual.
Second Best Signal Level per HCS Layer and a Margin

Txi
Txi
Txi
MaximumThreshold
And
Txi
P rec BCCH 2
nd
Best P Txj BCCH M

rec
ji

Notes:
the second highest.
is 2dB higher than the signal levels from transmitters that are the 3rd best servers.
5.1.3.1.8
Best Idle Mode Reselection Criterion (C2)

Such type of coverage would is useful :
To compare Idle and Dedicated mode best servers for Voice traffic
Display the GPRS/EGPRS best server map (based on GSM idle mode)
The path loss criterion parameter C1 used for cell selection and reselection is defined by :
Txi
C1 = P rec ic MinimumThreshold
The path loss criterion (GSM03.22) is satisfied if C1 0 .
The reselection criterion C2 is used for cell reselection only and is defined by :
C2 = C1 + CELL_RESELECT_OFFSET
where CELL_RESELECT_OFFSET is the reselection value (in dB) defined for at the transmitter level.
Txi
Txi
Txi

And
C2
Txi
Txj
BCCH = Best C2 BCCH
j

On each bin, the best C2 value is kept. It corresponds to the best server in Idle Mode. Since the C2 value is an integer
value, so must be rounded.
134
AT271_TRG_E6
Forsk 2009
5.1.3.2
Coverage Display
5.1.3.2.1
Plot Resolution
Prediction plot resolution is independent of the matrix resolutions and can be defined on a per study basis. Prediction plots
are generated from multi-resolution path loss matrices using bilinear interpolation method (similar to the one used to evaluate site altitude).
5.1.3.2.2
Display Types
It is possible to display the transmitter service area with colours depending on any transmitter attribute or other criteria
such as:
Signal Level (in dBm, dBV, dBV/m)

Atoll calculates signal level received from the transmitter on each bin of each transmitter service area. A bin of a service
area is coloured if the signal level exceeds ( ) the defined minimum thresholds (bin colour depends on signal level).
Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many
layers as transmitter service areas. Each layer shows the different signal levels available in the transmitter service area.
Best Signal Level (in dBm, dBV, dBV/m)

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. When other serviceWhen other service areas overlap the studied one, Atoll chooses the highest value. A bin of a service area is coloured if
the signal level exceeds ( ) the defined thresholds (the bin colour depends on the signal level). Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as defined
thresholds. Each layer corresponds to an area where the signal level from the best server exceeds a defined minimum
threshold.
Path Loss (dB)

Atoll calculates path loss from the transmitter on each bin of each transmitter service area. A bin of a service area is
coloured if path loss exceeds ( ) the defined minimum thresholds (bin colour depends on path loss). Coverage consists
of several independent layers whose visibility in the workspace can be managed. There are as many layers as service
areas. Each layer shows the different path loss levels in the transmitter service area.
Total Losses (dB)

Atoll calculates total losses from the transmitter on each bin of each transmitter service area. A bin of a service area is
coloured if total losses exceed ( ) the defined minimum thresholds (bin colour depends on total losses). Coverage
consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as
service areas. Each layer shows the different total losses levels in the transmitter service area.
Best Server Path Loss (dB)

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. When other service
areas overlap the studied one, Atoll determines the best transmitter and evaluates path loss from the best transmitter. A
bin of a service area is coloured if the path loss exceeds ( ) the defined thresholds (bin colour depends on path loss).
layers as defined thresholds. Each layer corresponds to an area where the path loss from the best server exceeds a
defined minimum threshold.
Best Server Total Losses (dB)

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. Where service areas
overlap the studied one, Atoll determines the best transmitter and evaluates total losses from the best transmitter. A bin
of a service area is coloured if the total losses exceed ( ) the defined thresholds (bin colour depends on total losses).
layers as defined thresholds. Each layer corresponds to an area where the total losses from the best server exceed a
Number of Servers
Atoll evaluates how many service areas cover a bin in order to determine the number of servers. The bin colour depends
on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the number of servers exceeds ( ) a defined minimum threshold.
Cell Edge Coverage Probability (%)

On each bin of each transmitter service area, the coverage corresponds to the pixels where the signal level from this transmitter fulfils signal conditions defined in Conditions tab with different cell edge coverage probabilities. There is one coverage area per transmitter in the explorer.
Forsk 2009
AT271_TRG_E6
135
Best Cell Edge Coverage Probability (%)

On each bin of each transmitter service area, the coverage corresponds to the pixels where the best signal level received
fulfils signal conditions defined in Conditions tab. There is one coverage area per cell edge coverage probability in the
explorer.
Best C2 (dBm)
Atoll calculates C2 values received from transmitters on each bin of each transmitter service area. When other service
areas overlap the studied one, Atoll chooses the highest value. A bin of a service area is coloured if the C2 value exceeds
( ) the defined thresholds (the bin colour depends on the C2 value). Coverage consists of several independent layers
whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the best C2 value exceeds a defined minimum threshold.
5.2
Traffic Analysis
When starting a traffic analysis, Atoll distributes the traffic from maps to transmitters of each layer according to the compatibility criteria defined in the transmitter, services, mobility type, terminal type properties. Transmitters considered in traffic
analysis are the active and filtered transmitters that belong to the focus zone.
Notes:
If no focus zone exists in the .atl document, Atoll takes into account the computation zone.
For details of the average timeslot capacity calculation, see the Network Dimensioning
section (calculation of minimum reduction factor).
5.2.1
Traffic Distribution
5.2.1.1
Normal Cells (Nonconcentric, No HCS Layer)
5.2.1.1.1
Circuit Switched Services

A user with a given circuit switched service, c, a terminal, t, and a mobility type, m, will be distributed to the BCCH and
TCH subcells of a transmitter if:
5.2.1.1.2
The terminal, t, works on the frequency band used by the BCCH subcell,
The terminal, t, works on the frequency band used by the TCH subcell.
Packet Switched Services

A user with a given packet switched service, p, a terminal, t, and a mobility type, m, will be distributed to the BCCH and
TCH subcells of a transmitter if:
5.2.1.2
The transmitter is an GPRS/EGPRS station (option specified in the transmitter property dialog),
The terminal, t, is technologically compatible with the transmitter,
The terminal, t, works on the frequency band used by the TCH subcell.
Concentric Cells
In case of concentric cells, TCH_INNER TRX type has the highest priority to carry traffic.
5.2.1.2.1

A user with a given circuit switched service, c, a terminal, t, and a mobility type, m, will be distributed to the TCH_INNER,
BCCH and TCH subcells of a transmitter if:
5.2.1.2.2
The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells.

A user with a given packet switched service, p, a terminal, t, and a mobility type, m, will be distributed to the TCH_INNER,
BCCH and TCH subcells of a transmitter if:
5.2.1.3
The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells.
HCS Layers
For each HCS layer, k, you may specify the maximum mobile speed supported by the transmitters of the layer.
136
AT271_TRG_E6
Forsk 2009
5.2.1.3.1

A user with a given circuit switched service, c, a terminal, t, and a mobility type, m, will be distributed to the BCCH and
TCH subcells (and TCH_INNER in case of concentric cells) of a transmitter if:
5.2.1.3.2
The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells,
The users mobility, m, is less than the maximum speed supported by the layer, k.

A user with a given packet switched service, p, a terminal, t, and a mobility type, m, will be distributed to the BCCH and
TCH subcells (and TCH_INNER in case of concentric cells) of a transmitter if:
5.2.2
The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells,
The user mobility, m, is less than the maximum speed supported by the layer, k.
Calculation of the Traffic Demand per Subcell

Here we assume that:
Users considered for evaluating the traffic demand fulfil the compatibility criteria defined in the transmitter, services, mobility, terminal properties as explained above.
Atoll distributes traffic on subcell service areas, which are determined using the option Best signal level per HCS
layer with a 0dB margin and the subcell reception threshold as lower threshold.
Same traffic is distributed to the BCCH and TCH subcells.
5.2.2.1
Traffic Maps Based on Environments and User Profiles
5.2.2.1.1

Number of subscribers ( X up m ) for each TCH subcell (Txi, TCH), per user profile up with a given mobility m, is inferred as:
X up m Txi TCH = S up m Txi TCH D
Where Sup,m is the TCH service area containing the user profile up with the mobility m and D is the user profile density.
For each behaviour described in the user profile up, Atoll calculates the probability for the user to be connected with a
given service using a terminal t.

For a circuit switched service c, we have:
N call d
p up c t = --------------------3600
Where Ncall is the number of calls per hour and d is the average call duration (in seconds).
Then, Atoll evaluates the traffic demand, D up c t m , in Erlangs for the subcell (Txi, TCH) service area.
D up c t m Txi TCH = X up m Txi TCH p up c t

For a packet switched service p, we have:
N call V 8
p up p t = ------------------------------3600
Where Ncall is the number of calls per hour and V is the transmitted data volume per call (in Kbytes).
Then, Atoll evaluates the traffic demand, D up p t m , in kbits/s for the subcell (Txi, TCH) service area.
D up p t m Txi TCH = X up m Txi TCH p up p t
5.2.2.1.2
Concentric Cells
In case of concentric cells, Atoll distributes a part of traffic on the TCH_INNER service area (TCH_INNER is the highest
priority traffic carrier) and the remaining traffic on the outer ring served by the TCH subcell. The traffic spread over the
TCH_INNER subcell may overflow to the TCH subcell. In this case, the traffic demand is the same on the TCH_INNER
subcell but increases on the TCH subcell.
Forsk 2009
AT271_TRG_E6
137

Note:
Traffic overflowing from the TCH_INNER to the TCH is not uniformly spread over the TCH
service area. It is still located on the TCH_INNER service area.
Number of subscribers ( X up m ) for each TCH_INNER (Txi, TCH_INNER) and TCH (Txi, TCH) subcell, per user profile up
with a given mobility m, is inferred as:
X up m Txi,TCH_INNER = S up m Txi,TCH_INNER D
X up m Txi,TCH = S up m Txi,TCH S up m Txi,TCH_INNER D
S up m Txi,TCH_INNER and S up m Txi,TCH respectively refer to the TCH_INNER and TCH subcell service areas
containing the user profile up with the mobility m. D is the user profile density.
Figure 5.1: Representation of a Concentric Cell TXi

For each user of the user profile up using a circuit switched service c with a terminal t, Atoll calculates the probability
( p up c t ) of the user being connected. Calculations are detailed in "Circuit Switched Services" on page 136.
Then, Atoll evaluates the traffic demand, D up c t m , in Erlangs in the (Txi, TCH_INNER) and (Txi, TCH) subcell service
areas.
D up c t m Txi,TCH_INNER = X up m Txi,TCH_INNER p up c t
D up c t m Txi,TCH = X up m Txi,TCH p up c t + D up c t m Txi,TCH_INNER O max Txi,TCH_INNER
Where O max Txi,TCH_INNER is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.

For each user of the user profile up using a packet switched service p with a terminal t, probability of the user being
connected ( p up p t ) is calculated as explained in "Packet Switched Services" on page 136.
Atoll evaluates the traffic demand, D up p t m , in kbits/s in the (Txi, TCH_INNER) and (Txi, TCH) subcell service areas.
D up p t m Txi,TCH_INNER = X up m Txi,TCH_INNER p up p t
D up p t m Txi,TCH = X up m Txi,TCH p up p t + D up p t m Txi,TCH_INNER O max Txi,TCH_INNER
Where O max Txi,TCH_INNER is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.
5.2.2.1.3
HCS Layers
We assume two HCS layers: the micro layer has a higher priority than the macro layer. Txi belongs to the micro layer and
Txj to the macro.
Normal Cells
Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option Best signal level per HCS
macro
layer meaning that there is an overlap between HCS layers service areas. Let S overlapping Txj TCH denote this area
(TCH service area of the macro layer overlapped by the TCH service area of the micro layer). Traffic on the overlapping
area is distributed to the TCH subcell of the micro layer because it has a higher priority. On this area, traffic of the micro
layer may overflow to the macro layer. In this case, the traffic demand is the same on the TCH subcell of the micro layer
but increases on the TCH subcell of the macro layer.
Note:
138
Traffic overflowing to the macro layer is not uniformly spread over the TCH service area of
Txj. It is only located on the overlapping area.
AT271_TRG_E6
Forsk 2009
Figure 5.2: Representation of Micro and Macro Layers

Atoll evaluates the traffic demand on the micro layer (higher priority) as explained above. For further details, please refer
to formulas for normal cells. Then, it proceeds with the macro layer (lower priority).
macro
Number of subscribers ( X up m ) for each TCH subcell (Txj, TCH) of the macro layer, per user profile up with the mobility
m, is inferred as:
macro
macro
macro
X up m Txj TCH = S up m Txj TCH S up m overlapping Txj TCH D

macro
Where S up m Txj TCH is the TCH service area of Txj containing the user profile up with the mobility m and D is the
profile density.
For each user described in the user profile up with the circuit switched service c and the terminal t, the probability for the
user being connected ( p up c t ) is calculated as explained in "Circuit Switched Services" on page 136.
macro
Then, Atoll evaluates the traffic demand, D up c t m , in Erlangs in the subcell (Txj, TCH) service area.
macro
S upm overlapping Txj TCH

macro
macro
micro
- Omax Txi TCH
D up c t m Txj TCH = X up m Txj TCH p up c t + D up c t m Txi TCH -------------------------------------------------------------------------micro
S up m Txi TCH
For each user described in the user profile up with the packet switched service p and the terminal t, probability for the user
to be connected ( p up p t ) is calculated as explained in "Packet Switched Services" on page 136.
macro
Then, Atoll evaluates the traffic demand, D up p t m , in kbits/s in the subcell (Txj, TCH) service area.
macro
S upm overlapping Txj TCH

macro
macro
micro
- Omax Txi TCH
D up p t m Txj TCH = X up m Txj TCH p up p t + D up p t m Txi TCH -------------------------------------------------------------------------micro
S up m Txi TCH
Where O max Txi TCH is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro
micro
layer) and S up m Txi TCH is the TCH service area of Txi containing the user profile up with the mobility m.
Concentric Cells
Atoll evaluates the traffic demand on the micro layer (higher priority HCS layer) as explained above. For further details,
please refer to formulas given in case of concentric cells. Then, it proceeds with the macro layer (lower priority HCS layer).
The traffic capture is calculated with the option Best signal level per HCS layer. It means that there are overlapping areas
between HCS layers where traffic is spread according to the layer priority. On these areas, traffic of the higher priority layer
may overflow.
Forsk 2009
AT271_TRG_E6
139
Figure 5.3: Concentric Cells

The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area
macro
overlapped by the TCH service area of the micro layer S overlapping Txi TCH Txj,TCH_INNER and another overlapped
macro
by the TCH_INNER service area of the micro layer S overlapping Txi,TCH_INNER Txj,TCH_INNER .
Let us consider three areas, S1, S2 and S3.
macro
macro
S 1 = S up m Txj,TCH_INNER S up m overlapping Txi TCH Txj,TCH_INNER

macro
S 2 = S up m overlapping Txi,TCH_INNER Txj,TCH_INNER

macro
S 3 = S up m overlapping Txi TCH Txj,TCH_INNER S 2

macro
Where S up m Txj,TCH_INNER is the TCH_INNER subcell service area of Txj containing the user profile up with the
mobility m. We only consider the overlapping areas containing the user profile up with the mobility m.
macro
On S1, the number of subscribers per user profile up with a given mobility m ( X up m ) is inferred:
macro
X up m Txj,TCH_INNER = S 1 D
Where D is the user profile density.
The traffic spread over the TCH_INNER service area of the micro layer may overflow on the TCH subcell. The traffic overflowing to the TCH subcell is located on the TCH_INNER service area. On S2, the TCH subcell traffic coming from the
TCH_INNER subcell traffic overflow may overflow proportional to R2.
S2
R 2 = ----------------------------------------------------------------micro
S up m Txi,TCH_INNER
The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S3 proportional to R3.
S3
R 3 = ------------------------------------------------------------------------------------------------------------------micro
micro
S up m Txi,TCH S up m Txi,TCH_INNER
micro
micro
Where S up m Txi,TCH and S up m Txi,TCH_INNER are the TCH and TCH_INNER service areas of Txi respectively
containing the user profile up with the mobility m.
For each user described in the user profile up with a circuit switched service c and a terminal t, the probability for the user
being connected ( p up c t ) is calculated as explained in "Circuit Switched Services" on page 136. Then, Atoll evaluates
macro
the traffic demand, D up c t m , in Erlangs in the subcell (Txj, TCH_INNER) service area.
macro
macro
X up m Txj,TCH_INNER p up c t +
D up c t m Txj,TCH_INNER = R D micro
2
up c t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi,TCH +
micro
R 3 X up m Txi TCH p up c t O max Txi TCH
140
AT271_TRG_E6
Forsk 2009

For each user described in the user profile up with a packet switched service p and a terminal t, probability for the user to
be connected ( p up p t ) is calculated as explained in "Packet Switched Services" on page 136.
macro
Then, Atoll evaluates the traffic demand, D up p t m , stated in kbits/s in the subcell (Txj, TCH_INNER) service area.
macro
macro
D up p t m Txj,TCH_INNER
X up m Txj,TCH_INNER p up p t +
= R D micro
2
up p t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi,TCH +
micro
R 3 X up m Txi TCH p up p t O max Txi TCH

Where O max Txi TCH and O max Txi,TCH_INNER are the maximum rates of traffic overflow (stated in %) specified
for the TCH and TCH_INNER subcells of Txi respectively.
The area of the TCH ring of the macro layer is overlapped by the micro layer. There are two parts: an area overlapped by
macro
the TCH service area of the micro layer S overlapping Txi TCH Txj,TCH -- TCH_INNER and another one by the
macro
TCH_INNER service area of the micro layer S overlapping Txi,TCH_INNER Txj,TCH -- TCH_INNER .
macro
macro
macro
S' 1 = S up m Txj,TCH S up m Txj,TCH_INNER S up m overlapping Txi TCH Txj,TCH -- TCH_INNER

macro
S' 2 = S up m overlapping Txi,TCH_INNER Txj,TCH -- TCH_INNER

macro
S' 3 = S up m overlapping Txi TCH Txj,TCH -- TCH_INNER S' 2

macro
macro
Where S up m Txj,TCH and S up m Txj,TCH_INNER are the TCH and TCH_INNER subcell service areas of Txj
respectively. We only consider the overlapping areas containing the user profile up with the mobility m.
macro
On S1, the number of subscribers per user profile up with a given mobility m ( X up m ) is inferred:
macro
X up m Txj,TCH = S' 1 D
Where D is the user profile density.
The traffic spread over the TCH_INNER service area of the micro layer may overflow on the TCH subcell. The traffic overflowing on the TCH subcell is located on the TCH_INNER service area. On S2, the TCH subcell traffic coming from the
TCH_INNER subcell traffic overflow may overflow proportionally to R2.
S' 2
R' 2 = ----------------------------------------------------------------micro
S up m Txi,TCH_INNER
The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S3 proportional to R3.
S' 3
R' 3 = ------------------------------------------------------------------------------------------------------------------micro
micro
S up m Txi,TCH S up m Txi,TCH_INNER
micro
micro
Where S up m Txi,TCH and S up m Txi,TCH_INNER are the TCH and TCH_INNER service areas of Txi respectively
containing the user profile up with the mobility m.
For each user described in the user profile up with a circuit switched service c and a terminal t, the probability for the user
being connected ( p up c t ) is calculated as explained in "Circuit Switched Services" on page 136.
macro
Then, Atoll evaluates the traffic demand, D up c t m , in Erlangs in the subcell (Txj, TCH) service area.
macro
X up m Txj TCH p up c t +
macro
D up c t m Txj TCH =
macro
D up c t m Txj,TCH_INNER O max Txj,TCH_INNER +

micro
R' 2 D up c t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi,TCH +

micro
R' 3 X up m Txi TCH p up c t m O max Txi TCH

For each user described in the user profile up with a packet switched service p and a terminal t, the probability for the user
being connected ( p up p t ) is calculated as explained in "Packet Switched Services" on page 136.
macro
Then, Atoll evaluates the traffic demand, D up p t m , in kbits/s in the subcell (Txj, TCH) service area.
Forsk 2009
AT271_TRG_E6
141
macro
X up m Txj TCH p up p t +
macro
D up p t m Txj TCH =
macro
D up p t m Txj,TCH_INNER O max Txj,TCH_INNER +

micro
R' 2 D up p t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi,TCH +

micro
R' 3 X up m Txi TCH p up p t m O max Txi TCH

Where O max Txi,TCH is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro
layer), O max Txi,TCH_INNER the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txi (macro
layer), O max Txj,TCH_INNER the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txj (macro
micro
layer) and X up m Txi TCH the number of subscribers with the user profile up and mobility m on the TCH service area
of Txi (as explained in "Concentric Cells" on page 136).
5.2.2.2
Traffic Maps Based on Transmitters and Services

We assume that the traffic map is built from a coverage by transmitter prediction study calculated for the TCH subcells
with options:
HCS Servers and no margin if the network only consists of normal cells and concentric cells,
Highest Priority HCS Server and no margin in case of HCS layers.
When creating the traffic map, you have to specify the traffic demand per transmitter and per service (throughput for a
packet switched service and Erlangs for a circuit switched service) and the global distribution of terminals and mobility
types.
Let E c Txi TCH denote the Erlangs for the circuit switched service, c, on the TCH subcell of Txi.
Let T p Txi TCH denote the throughput of the packet switched service, p, on the TCH subcell of Txi.
We assume that 100% of users have the terminal, t, and the mobility type, m.
5.2.2.2.1

For each circuit switched service, c, Atoll evaluates the traffic demand, Dc,t,m, in Erlangs in the subcell (Txi, TCH) service
area.
D c t m Txi TCH = E c Txi TCH
For each packet switched service, p, Atoll evaluates the traffic demand, Dp,t,m, in kbits/s in the subcell (Txi, TCH) service
area.
D p t m Txi TCH = T p Txi TCH
5.2.2.2.2
Concentric Cells
In case of concentric cells, Atoll distributes a part of traffic on the TCH_INNER service area (TCH_INNER is the highest
priority traffic carrier) and the remaining traffic, on the ring served by the TCH subcell only. The traffic spread over the
TCH_INNER subcell may overflow to the TCH subcell. In this case, the traffic demand is the same on the TCH_INNER
subcell and rises on the TCH subcell.
Note:
Traffic overflowing from the TCH_INNER to the TCH is not uniformly spread over the TCH
service area. It is only located on the TCH_INNER service area.
For each circuit switched service, c, Atoll evaluates the traffic demand, Dc,t,m, in Erlangs in the subcell, (Txi, TCH_INNER)
and (Txi, TCH), service areas.
S Txi,TCH_INNER
D c t m Txi,TCH_INNER = ----------------------------------------------------- E c Txi TCH
S Txi TCH
and
D c t m Txi,TCH =
S Txi,TCH S Txi,TCH_INNER
----------------------------------------------------------------------------------------------- E c Txi TCH +
S Txi TCH
D c t m Txi,TCH_INNER O max Txi,TCH_INNER
For each packet switched service, p, Atoll evaluates the traffic demand, Dp,t,m, in kbits/s in the subcell, (Txi, TCH_INNER)
and (Txi, TCH), service areas.
S Txi,TCH_INNER
D p t m Txi,TCH_INNER = ----------------------------------------------------- T p Txi TCH
S Txi TCH
and
142
AT271_TRG_E6
Forsk 2009
D p t m Txi,TCH =
S Txi,TCH S Txi,TCH_INNER
----------------------------------------------------------------------------------------------- T p Txi TCH +
S Txi TCH
D p t m Txi,TCH_INNER O max Txi,TCH_INNER
Where O max Txi,TCH_INNER is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell,
S Txi,TCH and S Txi,TCH_INNER are the TCH and TCH_INNER service areas of Txi respectively.
5.2.2.2.3
HCS Layers
We assume we have two HCS layers: the micro layer has a higher priority and the macro layer has a lower one. Txi belongs
to the micro layer and Txj to the macro one.
Normal Cells
Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option HCS Servers. It means
macro
that there is an overlapping area between HCS layers. Let S overlapping Txj TCH denote the TCH service area of the
macro layer overlapped by the TCH service area of the micro layer. Traffic on the overlapping area is distributed to the
TCH subcell of the micro layer (higher priority layer). On this area, traffic of the micro layer may overflow to the macro layer.
In this case, the traffic demand is the same on the TCH subcell of the micro layer but rises on the TCH subcell of the macro
layer.
Note:
Traffic overflowing on the macro layer is not uniformly spread over the TCH service area of
Txj. It is only located on the overlapping area.
Atoll starts evaluating the traffic demand on the micro layer (highest priority HCS layer).
micro
For each circuit switched service, c, Atoll calculates the traffic demand, D c t m , in Erlangs in the subcell (Txi, TCH) service
area.
micro
D c t m Txi TCH = E c Txi TCH

micro
For each packet switched service, p, Atoll calculates the traffic demand, D p t m , in kbits/s in the subcell (Txi, TCH) service
area.
micro
D p t m Txi TCH = T p Txi TCH

Then, Atoll proceeds with the macro layer (lower priority HCS layer). For each circuit switched service, c, Atoll calculates
macro
the traffic demand, D c t m , in Erlangs in the subcell (Txj, TCH) service area.
macro
S overlapping Txj TCH

macro
micro
- O max Txi TCH
D c t m Txj TCH = E c Txj TCH + D c t m Txi TCH -----------------------------------------------------------micro
S
Txi TCH
macro
For each packet switched service, p, Atoll calculates the traffic demand, D p t m , in kbits/s in the subcell (Txj, TCH) service
area.
macro
S overlapping Txj TCH

macro
micro
- O max Txi TCH
D p t m Txj TCH = T p Txj TCH + D p t m Txi TCH -----------------------------------------------------------micro
S
Txi TCH
Where O max Txi TCH is the maximum rate of traffic overflow (in %) specified for the TCH subcell of Txi (micro cell) and
S
micro
Txi TCH the TCH service area of Txi.
Concentric Cells
Atoll evaluates the traffic demand on the micro layer as explained above in case of concentric cells and then proceeds with
the macro layer (lower priority layer).
The traffic capture is calculated with the option HCS Servers. It means that there is overlapping areas between HCS
layers where traffic is spread over according to the layer priority. On these areas, traffic of the higher priority layer may
overflow.
Forsk 2009
AT271_TRG_E6
143
Figure 5.4: Concentric Cells

The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area
macro
overlapped by the TCH service area of the micro layer S overlapping Txi TCH Txj,TCH_INNER and another overlapped
macro
by the TCH_INNER service area of the micro layer S overlapping Txi,TCH_INNER Txj,TCH_INNER .
S1 = S
macro
macro
Txj,TCH_INNER S overlapping Txi TCH Txj,TCH_INNER
macro
S 2 = S overlapping Txi,TCH_INNER Txj,TCH_INNER

macro
S 3 = S overlapping Txi TCH Txj,TCH_INNER S 2

Where S
macro
Txj,TCH_INNER is the TCH_INNER subcell service area of Txj.
The traffic specified for Txj in the map description ( E c Txj TCH ) is spread over S1 proportionally to R1.
S1
R 1 = ------------------------------------------map
S
Txj TCH
map
S
Txj TCH is the TCH service area of Txj in the traffic map with the option Best signal level of the highest priority
layer.
The traffic spread over the TCH_INNER service area of the micro layer may overflow to the TCH subcell. The traffic overflowing to the TCH subcell is located on the TCH_INNER service area. On S2, the TCH subcell traffic coming from the
S2
R 2 = ----------------------------------------------------------------micro
S
Txi,TCH_INNER
The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S3 proportional to R3.
S3
R 3 = ------------------------------------------------------------------------------------------------------------------micro
micro
S
Txi,TCH S
Txi,TCH_INNER
macro
For each circuit switched service, c, Atoll calculates the traffic demand, D c t m , in Erlangs in the subcell (Txj,
TCH_INNER) service area.
R 1 E c Txj TCH +
macro
D c t m Txj,TCH_INNER =
micro
R 2 D c t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi TCH +

micro
micro
S
Txi TCH S
Txi,TCH_INNER
E c Txi TCH O max Txi TCH
R 3 --------------------------------------------------------------------------------------------------------------------------micro
S
Txi TCH
macro
For each packet switched service, p, Atoll calculates the traffic demand, D p t m , in kbits/s in the subcell (Txj,
TCH_INNER) service area.
144
AT271_TRG_E6
Forsk 2009

R 1 T p Txj TCH +
macro
D p t m Txj,TCH_INNER =
micro
R 2 D p t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi TCH +

micro
micro
S
Txi TCH S
Txi,TCH_INNER
T p Txi TCH O max Txi TCH
R 3 --------------------------------------------------------------------------------------------------------------------------micro
S
Txi TCH
Where O max Txi TCH is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
O max Txi,TCH_INNER is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi
and S
micro
Txi TCH is the TCH subcell service area of Txi.
The area of the TCH ring of the macro layer is overlapped by the micro layer. There are two parts: an area overlapped by
macro
the TCH service area of the micro layer S overlapping Txi TCH Txj,TCH -- TCH_INNER and another overlapped by the
macro
TCH_INNER service area of the micro layer S overlapping Txi,TCH_INNER Txj,TCH -- TCH_INNER .
S' 1 = S
macro
Txj TCH S
macro
macro
Txj,TCH_INNER S overlapping Txi TCH Txj,TCH -- TCH_INNER
macro
S' 2 = S overlapping Txi,TCH_INNER Txj,TCH -- TCH_INNER

macro
S' 3 = S overlapping Txi TCH Txj,TCH -- TCH_INNER S' 2

macro
Where S
respectively.
Txj TCH and S
macro
Txj,TCH_INNER are the TCH and TCH_INNER subcell service areas of Txj
The traffic specified for Txj in the map description ( E c Txj TCH ) is spread over S1 proportional to R1.
S' 1
R' 1 = ------------------------------------------map
S
Txj TCH
map
S
Txj TCH is the TCH service area of Txj in the traffic map with the option Best signal level of the highest priority
layer.
The traffic spread over the TCH_INNER service area of the micro layer may overflow to the TCH subcell. The traffic overflowing to the TCH subcell is located on the TCH_INNER service area. On S2, the TCH subcell traffic coming from the
S' 2
R' 2 = ----------------------------------------------------------------micro
S
Txi,TCH_INNER
The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S3 proportional to R3.
S' 3
R' 3 = ------------------------------------------------------------------------------------------------------------------micro
micro
S
Txi,TCH S
Txi,TCH_INNER
macro
For each circuit switched service, c, Atoll calculates the traffic demand, D c t m , in Erlangs in the subcell (Txj, TCH) service
area.
R' 1 E c Txj TCH +
macro
macro
D c t m Txj
TCH =
D c t m Txj,TCH_INNER O max Txj,TCH_INNER +

micro
R' 2 D c t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi TCH +

micro
micro
S
Txi,TCH S
Txi,TCH_INNER
E c Txi TCH O max Txi TCH
R' 3 -----------------------------------------------------------------------------------------------------------------------micro
S
Txi,TCH
macro
For each packet switched service, p, Atoll calculates the traffic demand, D p t m , in kbits/s in the subcell (Txj, TCH) service
area.
R' 1 T p Txj TCH +
macro
macro
D p t m Txj TCH =
D c t m Txj,TCH_INNER O max Txj,TCH_INNER +

micro
R' 2 D p t m Txi,TCH_INNER O max Txi,TCH_INNER O max Txi TCH +

micro
micro
S
Txi,TCH S
Txi,TCH_INNER
T p Txi TCH O max Txi TCH
R' 3 -----------------------------------------------------------------------------------------------------------------------micro
S
Txi,TCH
Forsk 2009
AT271_TRG_E6
145

Where O max Txj,TCH_INNER is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell
of Txj, O max Txi TCH is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
O max Txi,TCH_INNER is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi,
micro
S
Txi,TCH is the TCH subcell service area of Txi and S
area of Txi.
5.3
micro
Txi,TCH_INNER is the TCH_INNER subcell service
Network Dimensioning
Atoll is capable of dimensioning a GSM GPRS EDGE network with a mixture of circuit and package switched services.
This section describes the technical details of Atolls dimensioning engine.
5.3.1
Dimensioning Models and Quality Graphs

In Atoll, a dimensioning model is an entity utilized by the dimensioning engine along with other inputs (traffic, limitations,
criteria, etc.) in the process of dimensioning. A dimensioning model defines the QoS KPIs to be taken into account when
dimensioning a network for both circuit and packet switched traffic. The user can define either to use Erlang B or Erlang
C queuing model for circuit switched traffic and can define which KPIs to consider when dimensioning the network for
packet switched traffic. The dimensioning engine will only utilize the quality curves of the KPI selected. The KPIs not
selected are supposed to be either already satisfactory or not relatively important.
5.3.1.1
Circuit Switched Traffic

The network dimensioning for circuit switched traffic is performed using the universally accepted and adopted Erlang B
and Erlang C formulas. The dimensioning criterion in these formulas is the Grade of Service or the allowed blocking probability of the circuit switched traffic.
In the Erlang B approach, this Grade of Service is defined as the percentage of incoming circuit switched calls that are
blocked due to lack of resources or timeslots. This formula implies a loss system. The blocked calls are supposed to be
lost and the caller has to reinitiate it.
In the Erlang C approach, the Grade of Service is the percentage of incoming calls that are placed in a waiting queue when
there are no resources available, until some resources or timeslots are liberated. This queuing system has no lost calls.
As the load on the system increases, the average waiting time in the queue also increases.
These formulas and their details are available in many books. For example, Wireless Communications Principles and
Practice by Theodore S. Rappaport, Prentice Hall.
Following the common practice, network dimensioning in Atoll is based on the principle that a voice or GSM call has priority
over data transmission. Therefore, as explained later in the network dimensioning steps, Atoll first performs network
dimensioning according to the circuit switched traffic present in the subcell in order to ensure the higher priority service
availability before performing the same for the packet switched traffic.
5.3.1.2
Packet Switched Traffic

Since packet switched traffic does not occupy an entire timeslot the whole time, it is much more complicated to study than
circuit switched traffic. Packet traffic is intermittent and bursty. Whenever there is packet data to be transferred, a Temporary Block Flow (TBF) is initiated for transferring these packets. Multiple TBFs can be multiplexed on the same timeslot.
This implies that there can be many packet switched service users that have the same timeslots assigned for packet data
transfer but at different intervals of time.
This multiplexing of a number of packet switched service users over the same timeslots incurs a certain reduction in the
throughput (data transfer rate) for each multiplexed user. This reduction in the throughput is more perceivable when the
system traffic load is high. The following parts describe the three most important Key Performance Indicators in GPRS/
EDGE networks and how they are modelled in Atoll.
5.3.1.2.1
Throughput
Throughput is defined as the amount of data delivered to the Logical Link Control Layer in a given unit of time. Each temporary block flow (TBF), and hence each user, has an associated measured throughput sample in a given network. Each
network will have a different throughput probability distribution depending on the load and network configuration. Instead
of using the precise probability distributions, it is more practical to compute the average and percentile throughput values.
In GPRS, the resources are shared between the users being served, and consequently, the throughput is reduced as the
number of active users increases. This reduction in user perceived throughput is modelled through a reduction factor. The
throughput experienced by a user accessing a particular service can be calculated as:
User throughput = Number of allocated timeslots x Timeslot capacity x Reduction Factor
Or
User throughput per allocated timeslot = Timeslot capacity x Reduction Factor
146
AT271_TRG_E6
Forsk 2009
Timeslot Capacity
The timeslot capacity is the average throughput per fully utilized timeslot. It represents the average throughput from the
network point of view. It mainly depends on the networks propagation conditions and criteria in the coverage area of a
transmitter (carrier power, carrier-to-interference distribution, etc.). It is a measure of how much data the network is able
to transfer with 1 data Erlang, or in other words, how efficiently the hardware resources are being utilized by the network.
It may also depend on the RLC protocol efficiency.
Atoll computes the average timeslot capacity during the traffic analysis and is used to determine the minimum throughput
reduction factor. But since this information is displayed in the network dimensioning results (only due to relevance), this
information has been considered as a part of the network dimensioning process in this document.
Timeslot Utilisation
Timeslot utilization takes into account the average number of timeslots that are available for packet switched traffic. It is a
measure of how much the network is loaded with data services. Networks with timeslot utilisation close to 100% are close
to saturation and the end-user performance is likely to be very poor.
In Atoll this parameter is termed as the Load (Traffic load for circuit switched traffic and packet switched traffic load for
packet switched traffic). It is described in more detail in the Network dimensioning steps section.
Reduction Factor
Reduction factor takes into account the user throughput reduction due to timeslot sharing among many users. The figure
below shows how the peak throughput available per timeslot is reduced by interference and sharing.Reduction factor is a
function of the number of timeslots assigned to a user (Nu), number of timeslots available in the system (Ns) and the average system packet switched traffic load (Lp) (utilization of resources in the system). Data Erlangs or data traffic is given by:
Data Erlangs = L P N S
Figure 5.5: Reduction of Throughput per Timeslot

More precisely, the reduction factor is a function of the ratio Ns/Nu (Np). Np models the equivalent timeslots that are available for the packet switched traffic in the system. For example, a 24-timeslot system with each user assigned 3 timeslots
per connection can be modelled by a single timeslot connection system with 8 timeslots in total.
The formula for reduction factor can be derived following the same hypotheses followed by Erlang in the derivation of the
blocking probability formulas (Erlang B and Erlang C).
Let X be a random variable that measures the reduction factor in a certain system state:
0
1
if n = 0
if 0 < n N P
N
------P- if n > N P
n
Where n is the instantaneous number of connections in the system. The throughput reduction factor is defined as:
PX= n
X ---------------------PX= 0
RF
n=0
Or,
RF =
PX= n
X -----------------------------n=0
PX= i
i=0
Here, P(X=n) is the probability function of having n connections in the system. Under the same assumptions as those of
the Erlang formulas, the probability function can be written as:
Forsk 2009
AT271_TRG_E6
147

n
LP NP
------------------------n!
P X = n = ---------------------------------------------------------------------------------------------N
P
LP NP
-----------------------+
i!
i=0
i = NP
if 0 n N P
LP NP
------------------------------- i NP
N ! NP
+1 P
n
LP NP
------------------------------- i NP
N P! N P
P X = n = ---------------------------------------------------------------------------------------------N
P
LP NP
-----------------------+
i!
i=0
i = NP
if n > N P
LP NP
------------------------------- i NP
N ! NP
+1 P
Hence the reduction factor can finally be written as:

NP
NP
LP NP
-------------------------------- -------
i NP i
N ! NP
i=1
i = NP + 1 P
RF = ---------------------------------------------------------------------------------------------------------------N
LP NP
-----------------------+
i!
P
LP NP
-----------------------+
i!
i=1
LP NP
------------------------------- i NP
N
!
N
P
+1 P
i = NP
This formula is not directly applicable in any software application due to the summations up to infinity. Atoll uses the following version of this formula that is exactly the same formula without the summation overflow problem.
NP
NP
NP + 1
NP
LP NP
L
------------------------- --------------------- ln 1 L P +
-----P-
N
!
n!
n
P
=1
n=1
RF = n---------------------------------------------------------------------------------------------------------------------------------N
n
n=1
NP
n
LP
LP NP
LP NP
------------------------- + ----------------------------- --------------N P!
1 LP
n!
The default quality curves for the Reduction Factor have been derived using the above formula. Each curve is for a fixed
number of timeslots available for packet switched traffic (Np) describing the reduction factor at different values of packet
switched traffic load (Lp). The figure below contains all the reduction factor quality curves in Atoll. The Maximum reduction
factor can be 1, implying a maximum throughput, and the minimum can be 0, implying a saturated system with no data
throughput.
Figure 5.6: Reduction Factor for Different Packet Switched Traffic Loads (Lp, X-axis)
Each curve in the above figure represents an equivalent number of packet switched timeslots, NP.
148
AT271_TRG_E6
Forsk 2009
5.3.1.2.2
Delay
Delay is the time required for an LLC PDU to be completely transferred from the SGSN to the MS, or vice versa. Its modelling in an RF planning tool is a difficult task. Currently, study on this subject is underway at Forsk. Models for different traffic
types (HTTP, FTP, SMTP, SMS, etc.) are being studied to search for a possible analytical solution for this problem.
As the delay is a function of the delays and the losses incurred at the packet level, the network parameters, such as the
packet queue length, and different protocol properties, such as the size of the LLC PDU, become important. It is also quite
dependent upon the radio access round trip time (RA RTT) and has a considerable impact on the application level performance viewed by the user.
The delay parameter is a user level parameter rather than being a network level quantity, like throughput per cell, timeslot
capacity, TBF blocking and reduction factor, hence it is difficult to model and is currently under study. Hence, no default
curve is presently available for delay in Atoll.
5.3.1.2.3
Blocking Probability
In GPRS, there is no blocking as in circuit switched connections. If a new temporary block flow (TBF) establishment is
requested and there are already M users per timeslot, M being the maximum limit of multiplexing per timeslot (Multiplexing
factor), the request is queued in the system to be established later when resources become available.
Supposing that M number of users can be multiplexed over a single timeslot (PDCH), we can have a maximum of M * Np
users in the system. This implies that if a new TBF is requested when there are already M * Np users active, it will be
blocked and placed in a queue. So the blocking probability is the probability of having M * Np + 1 users in the system or
more, meaning,
PX= n
for n = M N P + 1
as in this case n is always greater than Np, we have,

n
LP NP
------------------------------- i NP
N P! N P
P X = n = ---------------------------------------------------------------------------------------------N
P
LP NP
------------------------ +
i!
i=0
LP NP
------------------------------- i NP
N
!
N
P
P
+1
i = NP
So, the Blocking Probability can be given as:
BP =
LP NP
--------------------------------- 1 NP
N ! NP
n = M NP + 1 P
P X = n = -----------------------------------------------------------------------------------------------N
n = MN+1
LP NP
-----------------------+
i!
i=0
i = NP
LP NP
--------------------------------- 1 NP
N
!
N
P
+1 P
Eliminating the summations to infinity, the blocking probability can be stated in a simpler form:
M NP
LP
LP NP
-------------------------------------------- -------------- M N P NP 1 L
P
N P! N P
BP = -----------------------------------------------------------------------------------------N
P
i=0
NP
i
LP
LP NP LP NP
------------------------ + ----------------------------- --------------N P!
1 LP
i!
The above formula has been used to generate the default quality curves for blocking probability in Atoll.
These graphs are generated for a user multiplexing factor of 8 users per timeslot. Each curve represents an equivalent
number of packet switched timeslots, NP.
The curves depict the blocking probabilities for different number of available connections (Np) at different packet switched
traffic loads (Lp) for a fixed user multiplexing factor of 8. The figure below contains all the blocking probability curves for
packet switched traffic dimensioning in Atoll. The blocking probability increases with the packet switched traffic load, which
implies that as the packet switched traffic increases for a given number of timeslots, the system starts to get more and
more loaded, hence there is higher probability of having a temporary block flow placed in a waiting queue.
Forsk 2009
AT271_TRG_E6
149
Figure 5.7: Blocking Probability for Different Packet Switched Traffic Loads (Lp, X-axis)
Reference:
T. Halonen, J. Romero, J. Melero; GSM, GPRS and EDGE performance Evolution towards 3G/UMTS, John Wiley
and Sons Ltd.
5.3.2
Network Dimensioning Process

The network dimensioning process is described below in detail. As the whole dimensioning process is in fact a chain of
small processes that have there respective inputs and outputs, with outputs of a preceding one being the inputs to the
next, the best method is to detail each process individually in form of steps of the global dimensioning process.
5.3.2.1
Network Dimensioning Engine

During the dimensioning process, Atoll first computes the number of timeslots required to accommodate the circuit
switched traffic. Then it calculates the number of timeslots to add in order to satisfy the demand of packet switched traffic.
This is performed using the quality curves entered in the dimensioning model used. If the dimensioning model has been
indicated to take all three KPIs in to account (throughput reduction factor, delay and blocking probability), the number of
timeslots to be added is calculated such that:
1.
2.
3.
The throughput reduction factor is greater than the minimum throughput reduction factor,
Delay is less than the maximum permissible delay defined in the service properties, and
The blocking probability is less than the maximum allowable blocking probability defined in the service properties.
The figure below depicts a simplified flowchart of the dimensioning engine in Atoll.
Figure 5.8: Network Dimensioning Process

On the whole, following are the inputs and outputs of the network dimensioning process:
5.3.2.1.1
Inputs
150
Circuit switched traffic demand

Packet switched traffic demand
AT271_TRG_E6
Forsk 2009
5.3.2.1.2
Outputs
5.3.2.2
Timeslot configurations defined for each subcell

Target traffic overflow rate and Half-rate traffic ratio for each subcell
Service availability criteria: minimum required throughput per user, maximum permissible delay, maximum allowable blocking probability etc.
Dimensioning model parameters: Maximum number of TRXs per transmitter, dimensioning model for circuit
switched traffic, number of minimum dedicated packet switched timeslots per transmitter, maximum number of
TRXs added for packet switched services, KPIs to consider, and their quality curves.
Number of required TRXs per transmitter

Number of required shared, circuit switched and packet switched timeslots
Traffic load
Served circuit switched traffic
Served packet switched traffic
Effective rate of traffic overflow
Actual KPI values: throughput reduction factor, delay and blocking probability
Network Dimensioning Steps

This section describes the entire process step by step as it is actually performed in Atoll. Details of the calculations of the
parameters that are calculated during each step are described as well.
5.3.2.2.1
Step 1: Timeslots Required for CS Traffic

Atoll computes the number of timeslots required to accommodate the circuit switched traffic assigned to each subcell. Atoll
takes the circuit switched traffic demand (Erlangs), calculated in the traffic analysis and assigned to the current subcell,
and the maximum blocking probability defined for the circuit switched service, and computes the required number of timeslots to satisfy this demand using the Erlang B or Erlang C formula (as defined by the user).
If the user-defined target rate of traffic overflow per subcell, OTarget, is greater than the maximum blocking rate defined in
the services properties, it is going to be taken as the Grade of Service required for that subcell instead of the maximum
blocking rate of the service.
For the blocking probability GoS and circuit switched traffic demand TDC, Atoll determines the required number of timeslots TSreq. C for each subcell using formulas described below. In fact, Atoll searches for TSreq. C value until the defined
grade of service is reached.
For Erlang B, we have:
TS reqC
TD C
------------------------------- TS reqC !
GoS = ---------------------------------TS
reqC
TD C
-----------------k!
k=0
For Erlang C, we have:

TS reqC
TD C
GoS = ----------------------------------------------------------------------------------------------------------------------------------------------------TS
1
TD C
TS reqC
TD C
+ TS reqC ! 1 ------------------
TS
reqC
reqC
TD C
-----------------k!
k=0
Atoll considers the effect of half-rate circuit switched traffic by taking into account a user-defined percentage of half-rate
traffic. Atoll computes the effective equivalent number of full-rate timeslots that will be required to carry the total traffic with
the defined percentage of half-rate traffic.
If the number of timeslots required to accommodate the full-rate circuit switched traffic is TSreq. FR, and the percentage of
half-rate traffic within the subcell is defined by HR, then the effective number of equivalent full-rate circuit switched timeslots TSeff. that can carry this traffic mix is calculated by:
HR
TS eff = TS reqFR 1 ---------
2
Atoll employs this simplified approach to integrating half-rate circuit switched traffic, which provides approximately the
same results as obtained by using the half-rate traffic charts.
5.3.2.2.2
Step 2: TRXs Required for CS Traffic and Dedicated PS Timeslots

This stage of the network dimensioning process computes the number of TRXs required to carry the circuit switched traffic
demand through the number of required timeslots calculated above and the timeslot configuration defined by the user in
the network settings. Atoll distributes the number of required circuit switched timeslots calculated in Step 1 taking into
account the presence of dedicated packet switched timeslots in each TRX according to the timeslot configurations.
If a timeslot configuration defines a certain number of dedicated packet switched timeslots pre-allocated in certain TRXs,
those timeslots will not be considered capable of carrying circuit switched traffic and hence will not be allocated. For example, if 4 timeslots have been marked as packet switched timeslots in the first TRX and Atoll computes 8 timeslots for carry-
Forsk 2009
AT271_TRG_E6
151

ing a certain circuit switched traffic demand, then the number of TRXs to be allocated cannot be 1 even if there is no packet
switched traffic considered yet.
The total numbers of timeslots that carry circuit switched and packet switched traffic respectively are the sums of respective dedicated and shared timeslots:
TS P = TS S + TS P dedicated and TS C = TS S + TS C dedicated
5.3.2.2.3
Step 3: Effective CS Blocking, Effective CS Traffic Overflow and Served CS Traffic

In this step, the previously calculated number of required TRXs is used to compute the effective blocking rate for the circuit
switched traffic. This is performed by using the Erlang B or Erlang C formula with the circuit switched traffic demand and
the number of required TRXs as inputs and computing the Grade of Service (or blocking probability). It then calculates the
effective traffic overflow rate, Oeff..
In case of Erlang B formula, the effective rate of traffic overflow for the circuit switched traffic is the same as the circuit
switched blocking rate. While in case of the Erlang C model, the circuit switched traffic is supposed to be placed in an
infinite-length waiting queue. This implies that there is no overflow in this case.
From this data, it also computes the served circuit switched traffic. This is the difference of the circuit switched traffic
demand and the percentage of traffic that overflows from the subcell to other subcells calculated above. Hence, for an
effective traffic overflow rate of Oeff. and the circuit switched traffic demand of TDC, the served circuit switched traffic STC
is computed as:
ST C = TD C 1 O eff
5.3.2.2.4
Step 4: TRXs to Add for PS Traffic

This step is the core of the dimensioning process for packet switched services. First of all, Atoll computes the number of
TRXs to be added to carry the packet switched traffic demand. This is the number of TRXs that contain dedicated packet
switched and shared timeslots.
To determine this number of TRXs, Atoll calculates the equivalent average packet switched traffic demand in timeslots by
studying each pixel covered by the transmitter. This calculation is in fact performed in the traffic analysis process. Knowing
the traffic demand per pixel of the covered area in terms of kbps and the maximum attainable throughput per pixel (according to the C and/or C/I conditions and the coding scheme curves in the GPRS/EDGE equipment), Atoll calculates the average traffic demand in packet switched timeslots by:
TD P
Timeslots
pixel
Traffic demand per pixel (kbps)

---------------------------------------------------------------------------------Throughput per pixel (kbps)
The average timeslot capacity of a transmitter is calculated by dividing the packet switched traffic demand over the entire
coverage area (in kbps) by the packet switched traffic demand in timeslots calculated above.
With the number of timeslots required to serve the circuit switched traffic, the timeslots required for packet switched traffic
and their respective distributions according to the timeslot configurations being known, Atoll calculates the number of
timeslots available for carrying the packet switched traffic demand. These timeslots can be dedicated packet switched
timeslots and the shared ones. So, following the principle that shared timeslots are potential carriers of both traffic types,
TS P = TS S + TS P dedicated
TS C = TS S + TS C dedicated
The packet switched traffic load is calculated by the formula:
ST C TS C dedicated + TD P
Timeslots
L P = ---------------------------------------------------------------------------------------------TS P
The second important parameter for the calculation of Reduction Factor, Delay and Blocking Probability is the equivalent
number of available timeslots for packet switched traffic, i.e. NP. This is computed by dividing the total number of timeslots
available for carrying packet switched traffic by the number of downlink timeslots defined in the mobile terminal properties.
So, NP is calculated at this stage as:
TS P
N P = --------------------------TS Terminal
Where, TSTerminal is the number of timeslots that a terminal will use in packet switched calls. This is determined by taking
the lower of the maximum number of timeslots for packet switched service defined in the service properties and the maximum number of timeslots that a mobile terminal can use for packet switched services defined in the terminal type properties.
TS Terminal = min TS Max Service TS Max TerminalType
Here, the min(X,Y) function yields the lower value among X and Y as result.
Now, knowing the packet switched traffic load, LP, and the equivalent number of available timeslots, NP, Atoll finds out the
KPIs that have been selected before launching the dimensioning process using the quality curves stored in the dimensioning model.
This particular part of this step can be iterative if the KPIs to consider in dimensioning are not satisfied in the first try. If the
KPIs calculated above are within acceptable limits as defined by the user, it means that the dimensioning process has
152
AT271_TRG_E6
Forsk 2009

acceptable results. If these KPIs are not satisfied, then Atoll increases the number of TRXs calculated for carrying packet
switched traffic by 1 (each increment adding 8 more timeslots for carrying packet switched traffic as the least unit that can
be physically added or removed is a TRX) and resumes the computations from Step 3. It then recalculates the packet
switched traffic load, LP, and the equivalent number of available timeslots, NP. Then it recomputes the KPIs with these
new values of LP and NP. If the KPIs are within satisfactory limits the results are considered to be acceptable. Otherwise,
Atoll performs another iteration to find the best possible results.
The calculated values of all the KPIs are compared with the ones defined in the service properties. The values for maximum Delay and Blocking probability are defined directly in the properties but the minimum throughput reduction factor is
calculated by Atoll using the users inputs: minimum throughput per user and required availability. This calculation is in fact
performed during the traffic analysis process, but since it is relevant to the dimensioning procedure, it is displayed in a
column in the dimensioning results so that the user can easily compare the minimum requirement on the reduction factor
KPI with the resulting one.
Minimum Throughput Reduction Factor Calculation

The minimum throughput reduction factor is computed using the input data: minimum required throughput per user defined
in the service properties, the average throughput per timeslot deduced from the throughput curves stored in the GPRS/
EDGE equipment properties for each coding scheme, the number of downlink timeslots defined in the properties of the
mobile terminal and the required availability defined in the service properties.
It is at the stage of calculating the average timeslot capacity per transmitter that Atoll studies each covered pixel for carrier
power or carrier-to-interference ratio. According to the measured carrier power or carrier-to-interference ratio, Atoll
deduces the maximum throughput available on that pixel through the throughput vs. C or throughput vs. C/I curves of the
GPRS/EDGE equipment.
The throughput per timeslot per pixel TPTS, Pixel can be either a function of carrier power C, or carrier power C and the
carrier-to-interference ratio C/I, depending on the user-defined traffic analysis RF conditions criteria. Therefore,
TP TS Pixel = f C
Or
C
TP TS Pixel = f C and TP TS Pixel = f ----
i
The required availability parameter defines the percentage of pixels within the coverage area of the transmitter that must
satisfy the minimum throughput condition. This parameter renders user-manageable flexibility to the throughput requirement constraint.
To calculate the minimum throughput reduction factor for the transmitter, Atoll computes the minimum throughput reduction factor for each pixel using the formula:
TP user min
RF min Pixel = ------------------------------------------------------------TP TS Pixel TS Terminal
Once the minimum reduction factor for each pixel is known, Atoll calculates the global minimum reduction factor that is
satisfied by the percentage of covered pixels defined in the required availability. The following example may help in understanding the concept and calculation method.
Example: Let the total number of pixels, covered by a subcell S, be 1050. The reliability level set to 90%. This implies that
the required minimum throughput for the given service will be available at 90% of the pixels covered. This, in turn, implies
that there will be a certain limit on the reduction factor, i.e. if the actual reduction factor in that subcell becomes less than
a minimum required, the service will not be satisfactory.
Atoll computes the minimum reduction factor at each pixel using the formula mentioned above, and outputs the following
results:
RFmin
Number of pixels
0.3
189
0.36
57
0.5
20
0.6
200
0.72
473
0.9
23
0.98
87
So for a reliability level of 90%, the corresponding RFmin will be the one provided at least 90% of the pixels covered, i.e.
945 pixels. The corresponding value of the resulting RFmin in this example hence turns out to be 0.9, since this value
covers 962 pixels in total. Only 87 of the covered pixels imply an RFmin of 0.98. These will be the pixels that do not provide
satisfactory service.
This calculation is performed for each service type available in the subcell coverage area. The final minimum throughput
reduction factor is the highest one amongst all calculated for each service separately.
The minimum throughput reduction factor RFmin value is a minimum requirement that must be fulfilled by the network
dimensioning process when the Reduction Factor KPI is selected in the dimensioning model.
Forsk 2009
AT271_TRG_E6
153
Figure 5.9: Minimum Throughput Reduction Factor
5.3.2.2.5
Step 5: Served PS Traffic

Atoll calculates the served packet switched traffic using the number of timeslots available to carry the packet switched traffic demand. As the result of the above iterative step, Atoll always finds the best possible answer in terms of number of
timeslots required to carry the packet switched traffic demand unless the requirement exceeds the maximum limit on the
number of the packet switched traffic timeslots defined in the dimensioning model properties. Hence, there is no packet
traffic overflow unless the packet switched traffic demand requires more TRXs than the maximum allowed
5.3.2.2.6
Step 6: Total Traffic Load

This step calculates the final result of the dimensioning process, i.e. the total traffic load. The total traffic load L is calculated
as:
ST C + ST P
L = ----------------------------------------------------------------------------------------------TS C dedicated + TS P dedicated + TS S
Where,
5.4
STC is the served circuit switched traffic

STP is the served packet switched traffic
TSC, dedicated is the number of dedicated circuit switched timeslots
TSP, dedicated is the number of dedicated packet switched timeslots
TSS is the number of shared timeslots
Key Performance Indicators Calculation

This feature calculates the current values for all circuit switched and packet switched Key Performance Indicators as a
measure of the current performance of the network. It can be used to evaluate an already dimensioned network in which
recent traffic changes have been made in limited regions to infer the possible problematic areas and then to improve the
network dimensioning with respect to these changes.
The concept of this computation is the inverse of that of the dimensioning process. In this case, Atoll has the results of the
dimensioning process already committed and known. Atoll then computes the current values for all the KPIs knowing the
number of required TRXs, the respective numbers of shared and dedicated timeslots and the circuit switched and packet
switched traffic demands.
The computation algorithm utilizes the parameters set in the dimensioning model properties and the quality curves for the
throughput reduction factor, delay and the blocking probability.
The following conventional relations apply:
If,
TSC, dedicated is the number of timeslots dedicated to the circuit switched traffic,
TSP, dedicated is the number of timeslots dedicated to the packet switched traffic,
TSS is the number of shared timeslots for a transmitter,
Then, the number of timeslots available for the circuit switched traffic, TSC, is defined as:
TS C = TS S + TS C dedicated
And the number of timeslots available for the packet switched traffic, TSP, is given by:
TS P = TS S + TS P dedicated
154
AT271_TRG_E6
Forsk 2009
5.4.1
Circuit Switched Traffic

For each subcell, Atoll has already calculated the effective traffic overflow rate and the blocking rate during the dimensioning process. Also knowing the circuit switched traffic demand, TDC, and the number of timeslots available for circuit
switched traffic, TSC, the blocking probability can be easily computed using the Erlang formulas or tables.
5.4.1.1
Erlang B
Under the current conditions of circuit switched traffic demand, TDC, and the number of timeslots available for the circuit
switched traffic, TSC, the percentage of blocked circuit switched traffic can be computed through:
TS C
TD C
------------------------- TS C !
% of blocked traffic = ----------------------------TS
C
TD C
-----------------k!
k=0
In a network dimensioning based on Erlang B model, the circuit switched traffic overflow rate, OC, is the same as the
percentage of traffic blocked by the subcell calculated above.
5.4.1.2
Erlang C
Similarly, under the current conditions of circuit switched traffic demand, TDC, and the number of timeslots available for
the circuit switched traffic, TSC, the percentage of delayed circuit switched traffic can be computed through:
TS C
TD C
% of traffic delayed = -------------------------------------------------------------------------------------------------------------------------TS 1
TD C
TS C
TD C
+ TS C ! 1 ----------
TS
C
TD C
------------------k!
k=0
If the circuit switched traffic demand, TDC, is higher than the number of timeslots available to accommodate circuit
switched traffic, the column for this result will be empty signifying that there is a percentage of circuit switched traffic actually being rejected rather than just being delayed under the principle of Erlang C model.
The circuit switched traffic overflow rate, OC, will be 0 if the circuit switched traffic demand, TDC, is less than the number
of timeslots available for the circuit switched traffic, TSC.
If, on the other hand, the circuit switched traffic demand, TDC, is higher than the number of timeslots available to carry the
circuit switched traffic, TSC, then there will be a certain percentage of circuit switched traffic that will overflow from the
subcell. This circuit switched traffic overflow rate, OC, is calculated as:
TD C TS C
O C = ---------------------------TD C
5.4.1.3
Served Circuit Switched Traffic

The result of the above two processes will be a traffic overflow rate for the circuit switched traffic for each subcell, OC. The
served circuit switched traffic, STC, is calculated as:
ST C = TD C 1 O C
5.4.2
Packet Switched Traffic

Identifying the total traffic demand, TDT, (circuit switched traffic demand + packet switched traffic demand) as:
TD T = TD C + TD P
The following two cases can be considered.
5.4.2.1
Case 1: Total Traffic Demand > Dedicated + Shared Timeslots

In the case where the total number of timeslots available is less than the total traffic demand, there will be packet switched
data traffic that will be rejected by the subcell as it will not be able to accommodate it. The following results are expected
in this case:
5.4.2.1.1
Traffic Load
The traffic load will be 100%, as the subcell will have more traffic to carry than it can. This implies that the system will be
loaded to the maximum and even saturated. Hence the user level quality of service is bound to be very unsatisfactory.
Forsk 2009
AT271_TRG_E6
155
5.4.2.1.2
Packet Switched Traffic Overflow

In a 100% loaded, or even saturated subcell, the packet switched data calls will start being rejected because of shortage
of available resources. Hence there will be a perceptible packet switched traffic overflow in this subcell, OP. This overflow
rate is calculated as show below:
TS C dedicated + TS P dedicated + TS S ST C
O P = 1 ---------------------------------------------------------------------------------------------------------------------------- 100
TD P
5.4.2.1.3
Throughput Reduction Factor

The resulting throughput reduction factor for a 100% loaded or saturated subcell will be 0. Hence, the throughput perceived
by the packet switched service user will be 0, implying a very bad quality of service.
5.4.2.1.4
Delay
Again for a 100% loaded or saturated subcell, the delay at the packet switched service user end will be infinite as there is
no data transfer (throughput = 0).
5.4.2.1.5
All the data packets will be rejected by the system since it is saturated and has no free resources to allocate to incoming
data packets. Hence, the blocking probability will be 100%.
5.4.2.1.6
Served Packet Switched Traffic

With the packet switched data traffic overflowing from the subcell, there will be a part of that traffic that is not served. The
served packet switched data traffic, STP, is calculated on the same principle as the served circuit switched traffic:
ST P = TD P 1 O P
5.4.2.2
Case 2: Total Traffic Demand < Dedicated + Shared Timeslots

In the case where the total traffic demand is less than the number of timeslots available to carry the traffic, the subcell will
not be saturated and there will be some deducible values for all the data KPIs. In a normally loaded subcell, the packet
switched data traffic will have no overflow percentage. This is due to the fact that the packet switched data traffic is rather
placed in a waiting queue than be rejected.
Therefore, there will be a within limits packet switched traffic load, LP, calculated as under:
ST C TS C dedicated + TD P
Timeslots
L P = ---------------------------------------------------------------------------------------------TS P
The second parameter for computing the KPIs from the quality curves of the dimensioning model is the number of equivalent timeslots available for the packet switched data traffic, NP, which is calculated in the same manner as in the dimensioning process as well:
TS P
N P = --------------------------TS Terminal
These parameters calculated, now Atoll can compute the required KPIs through their respective quality curves.
5.4.2.2.1
Traffic Load
The traffic load is computed knowing the total traffic demand and the total number of timeslots available to carry the entire
traffic demand:
TD T
Traffic Load = ----------------------------------------------------------------------------------------------TS C dedicated + TS P dedicated + TS S
5.4.2.2.2
Packet Switched Traffic Overflow

In a normally loaded subcell, no packet switched data calls will be rejected. The packet switched traffic overflow will, therefore, be 0.
5.4.2.2.3
Throughput Reduction Factor

The resulting throughput reduction factor for a normally loaded subcell is calculated through the throughput reduction
factor quality curve for given packet switched traffic load, LP, and number of equivalent timeslots, NP.
5.4.2.2.4
Delay
The resulting delay the subcell is calculated through the delay quality curve for given packet switched traffic load, LP, and
number of equivalent timeslots, NP.
5.4.2.2.5
The resulting blocking probability for a normally loaded subcell is calculated through the blocking probability quality curve
for given packet switched traffic load, LP, and number of equivalent timeslots, NP.
156
AT271_TRG_E6
Forsk 2009
5.4.2.2.6
Served Packet Switched Traffic

As there is no overflow of the packet switched traffic demand from the subcell under consideration, the served packet
switched traffic will be the same as the packet switched traffic demand:
ST P = TD P
5.5
Neighbour Allocation
The intra-technology neighbour allocation algorithm takes into account all the TBC transmitters. It means that all the TBC
transmitters of the .atl document are potential neighbours.
The transmitters to be allocated will be called TBA transmitters. They must fulfil the following conditions:
They are active,

They satisfy the filter criteria applied to the Transmitters folder,
They are located inside the focus zone,
They belong to the folder on which allocation has been executed. This folder can be either the Transmitters folder
or a group of transmitters or a single transmitter.
Only TBA transmitters may be assigned neighbours.

Note:
5.5.1
Global Allocation for All Transmitters

We assume a reference transmitter A and a candidate neighbour, transmitter B.
When automatic allocation starts, Atoll checks following conditions:
1.
2.
The distance between both transmitters must be less than the user-definable maximum inter-site distance. If the
distance between the reference transmitter and the candidate neighbour is greater than this value, then the
candidate neighbour is discarded.
The calculation options,
Force co-site transmitters as neighbours: This option enables you to force transmitters located on the reference transmitter
site in the candidate neighbour list. This constraints can be weighted among the others and ranks the neighbours through
the importance field (see after).
Force adjacent transmitters as neighbours: This option enables you to force transmitters geographically adjacent to the
reference transmitter in the candidate neighbour list. This constraints can be weighted among the others and ranks the
neighbours through the importance field (see after).
Notes:
Forsk 2009
Adjacence criterion: Geographically adjacent transmitters are determined on the basis of

their Best Server coverages in 2G (GSM GPRS EGPRS) projects. More precisely, a
transmitter TXi is considered adjacent to another transmitter TXj if there exists at least one
pixel of TXi Best Server coverage area where TXj is the 2nd Best Server. The ranking of
the adjacent neighbour transmitter increases with the number of these pixels. The figure
below shows the above concept.
AT271_TRG_E6
157
When this option is checked, adjacent cells are sorted and listed from the most adjacent to
the least, depending on the above criterion. Adjacence is relative to the number of pixels
satisfying the criterion.
This criteria is only applicable to transmitters belonging to the same HCS layer. The
geographic adjacency criteria is not the same in 3G (UMTS HSPA, CDMA2000) projects.
Force neighbour symmetry: This option enables user to force the reciprocity of a neighbourhood link. Therefore, if the reference transmitter is a candidate neighbour of another transmitter, the later will be considered as candidate neighbour of the
reference transmitter.
Force exceptional pairs: This option enables you to force/forbid some neighbourhood relationships. Therefore, you may
force/forbid a transmitter to be candidate neighbour of the reference transmitter.
Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current neighbours
and carries out a new neighbour allocation. If not selected, the existing neighbours are kept.
There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability where:
3.
SA is the area where the received signal level from the transmitter A is greater than a minimum signal level. SA is
the coverage area of reference transmitter A restricted between two boundaries; the first boundary represents the
start of the handover area (best server area of A plus the handover margin named handover start) and the
second boundary shows the end of the handover area (best server area of A plus the margin called handover
end)
SB is the coverage area where the candidate transmitter B is the best server.
SA SB
Atoll calculates either the percentage of covered area ( ---------------------- 100 ) if the option Take into account Covered Area is
SA
selected, or the percentage of traffic covered on the overlapping area S A S B for the option Take into account Covered
Traffic. Then, it compares this value to the % minimum covered area (minimum percentage of covered area for the option
Take into account Covered Area or minimum percentage of covered traffic for the option Take into account Covered
Traffic). If this percentage is not exceeded, the candidate neighbour B is discarded.
The coverage condition can be weighted among the others and ranks the neighbours through the importance field (see
number 4 below).
Figure 5.10: Overlapping Zones

4.
The importance values are used by the allocation algorithm to rank the neighbours according to the allocation
reason, and to quantify the neighbour importance.
Atoll lists all neighbours and sorts them by importance value so as to eliminate some of them from the neighbour list if the
maximum number of neighbours to be allocated to each transmitter is exceeded. If we consider the case for which there
are 15 candidate neighbours and the maximum number of neighbours to be allocated to the reference transmitter is 8.
158
AT271_TRG_E6
Forsk 2009

Among these 15 candidate neighbours, only 8 (having the highest importances) will be allocated to the reference transmitter.
As indicated in the table below, the neighbour importance depends on the neighbourhood cause; this value goes from 0
to 100%.
Neighbourhood cause
When
Importance
value
Existing neighbour
Only if the Delete existing neighbours option is not selected

and in case of a new allocation
Existing
importance
Exceptional pair
Only if the Force exceptional pairs option is selected
100 %
Co-site transmitter
Only if the Force co-site transmitters as neighbours option is

selected
(IF) function
Adjacent transmitter
Only if the Force adjacent transmitters as neighbours option is

selected
(IF) function
Neighbourhood relationship that fulfils

coverage conditions
Only if the % minimum covered area is exceeded
(IF) function
Symmetric neighbourhood
relationship
Only if the Force neighbour symmetry option is selected
(IF) function
Except forced neighbour case (importance = 100%), priority assigned to each neighbourhood cause is now linked to the
(IF) Importance Function evaluation. The importance is evaluated through a function (IF), taking into account the following
3 factors:
Co-site factor (C) which is a Boolean factor,

Adjacency factor (A) which deals with the percentage of adjacency,
Overlapping factor (O) meaning the percentage of overlapping
The (IF) function is user-definable using the Min importance and Max importance fields.
Factor
Min
importance
Default value
Max
importance
Default value
Overlapping factor (O)

Adjacency factor (A)
Min(O)
1%
Max(O)
30%
Min(A)
30%
Max(A)
60%
Co-site factor (C)
Min(C)
60%
Max(C)
100%
The (IF) function is evaluated as follows:
Neighbourhood cause
(IF) function
(IF) function with default Min

and Max default values
no
Min(O) + Delta(O)(O)
1% + 29%(O)
no
yes
Min(A)+Delta(A){Max(O)(O)+(100%Max(O))(A)}
30% + 30%{30%(O) + 70%(A)}
yes
yes
Min(C)+Delta(C){Max(O)(O)+(100%Max(O))(A)}
60% + 40%{30%(O )+ 70%(A)}
Co-site
Adjacent
no
Where Delta(x) = Max(x) - Min(x)

Notes:
If there is no overlapping between the range of each factor, the neighbours will be ranked
by neighbourhood cause. Using the default values for minimum and maximum importance
fields, neighbours will be ranked in this order: first co-site neighbours, then adjacent
neighbours, and finally neighbours found on overlapping criterion.
If ranges of (IF) factors overlap each other, the neighbours may not be ranked by
neighbourhood cause.
The ranking between neighbours from the same category will depend on (A) and (O)
factors.
The default value of Min(O)= 1%, ensures that neighbours selected for symmetry will have
an importance greater than 0%. With a value of Min(O)= 0%, neighbours selected for
symmetry, will have an importance field greater than 0% only if there is some overlapping.
In the Results part, Atoll provides the list of neighbours, the number of neighbours and the maximum number of neighbours allowed for each cell. In addition, it indicates the importance (in %) of each neighbour and the allocation reason.
Therefore, a neighbour may be marked as exceptional pair, co-site, adjacent, coverage or symmetric. For neighbours
accepted for co-site, adjacency and coverage reasons, Atoll displays the percentage of area meeting the coverage conditions (or the percentage of covered traffic on this area) and the corresponding surface area (km2) (or the traffic covered
on the area in Erlangs), the percentage of area meeting the adjacency conditions and the corresponding surface area
(km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.
Forsk 2009
AT271_TRG_E6
159

Notes:
No prediction study is needed to perform an automatic neighbour allocation. When starting

an automatic neighbour allocation, Atoll automatically calculates the path loss matrices if
not found.
Atoll uses traffic map(s) selected in the default traffic analysis in order to determine the
percentage of traffic covered in the overlapping area.
The percentage of area (or the percentage of covered traffic) is calculated with the
resolution specified in the property dialog of the predictions folder (Default resolution
parameter).
When the option Force adjacent transmitters as neighbours is used, the margin
handover start is not taken into account. Atoll considers a fixed value of 0 dB.
A forbidden neighbour must not be listed as neighbour except if the neighbourhood

relationship already exists and the Delete existing neighbours option is unchecked when
you start the new allocation. In this case, Atoll displays a warning in the Event viewer
indicating that the constraint on the forbidden neighbour will be ignored by algorithm
because the neighbour already exists.
The force neighbour symmetry option enables the users to consider the reciprocity of a
neighbourhood link. This reciprocity is allowed only if the neighbour list is not already full.
Thus, if transmitter B is a neighbour of the transmitter A while transmitter A is not a
neighbour of the transmitter B, two cases are possible:
1st case: There is space in the transmitter B neighbour list: the transmitter A will be added
to the list. It will be the last one.
2nd case: The transmitter B neighbour list is full: Atoll will not include transmitter A in the list
and will cancel the link by deleting transmitter B from the transmitter A neighbour list.
5.5.2
When the options Force exceptional pairs and Force symmetry are selected, Atoll
considers the constraints between exceptional pairs in both directions so as to respect
symmetry condition. On the other hand, if neighbourhood relationship is forced in one
direction and forbidden in the other one, symmetry cannot be respected. In this case, Atoll
displays a warning in the Event viewer.
In the Results, Atoll displays only the transmitters for which it finds new neighbours.
Therefore, if a transmitter has already reached its maximum number of neighbours before
starting the new allocation, it will not appear in the Results table.
Allocation for a Group of Transmitters or One Transmitter

In this case, Atoll allocates neighbours to:
TBA transmitters,
Neighbours of TBA transmitters marked as exceptional pair, adjacent and symmetric,
Neighbours of TBA transmitters that satisfy coverage conditions.
Automatic neighbour allocation parameters are described in "Global Allocation for All Transmitters" on page 157.
5.6
Interference Prediction Studies
5.6.1
Coverage Studies
Two interference studies with predefined settings are available:
The coverage by C/I level study: This study provides you a global analysis of the network quality.
The interfered areas study: This study shows the areas where a transmitter is interfered by other ones.
In both cases, Atoll calculates C/I ratio on each calculation bin where conditions on signal level reception are satisfied.
Then, it either considers the bins where the calculated C/I exceeds a lower threshold in the coverage and colours these
bins depending on C/I value (coverage by C/I level study), or it considers the bins where the calculated C/I is lower than
a upper threshold in the coverage and colours them depending on colour of the interfered transmitter (interfered areas
study).
The user-defined thermal noise (N) value is used in the calculations if the corresponding calculation conditions are
selected in the conditions tab of an interference study. The thermal noise is considered to be a white guassian background
noise fixed at the user-defined value for the entire network or part of the network under consideration. This value is defined
in the document database at -121 dBm by default.
All the TBC transmitters are taken into account in these studies. Let us assume that each bin within each TBC transmitter
calculation area corresponds to a probe mobile receiver.
160
The study conditions in order to determine the coverage area of each TBC transmitter
The display settings to select how to colour coverage areas.
AT271_TRG_E6
Forsk 2009

Note:
5.6.1.1
For information on the common prediction studies (like coverage by transmitter, profile
study, ), please, refer to Common prediction studies part.

The areas, where Atoll will calculate C/I, depend on signal level reception conditions. Atoll uses the parameters entered
in the Conditions tab in order to determine service area of each TBC transmitter. We can distinguish four cases:
Here we presume that:
5.6.1.1.1
Each transmitter, Txi, belongs to a hierarchical cell structure (HCS) layer, k, with a defined priority.
All Servers
Txi
Minimum threshold P rec tt Maximum threshold
5.6.1.1.2

Txi

and
Txi
Txj

ji
where,
5.6.1.1.3
Best Signal Level of the Highest Priority HCS Layer and a Margin
In this case, the service area of Txi corresponds to the bins where:
Txi

and
Txi
Txj

ji
and
Txi belongs to the HCS layer with the highest priority
where,
5.6.1.1.4

Txi

and
Txi
P rec BCCH 2
nd
Txj
Best P rec BCCH M
ji
where,
Note:
When the maximum range option is selected, Atoll searches for interference on the bins:
- Where the respective criteria described above are checked, and
- Located within a specified distance from the transmitter (maximum range).
5.6.1.2
Carrier to Interference Ratio Calculation

Atoll works out carrier to interference ratio on each bin of transmitter service areas.
Forsk 2009
AT271_TRG_E6
161

In order to understand the difference between each frequency hopping mode from the mobile point of view, it is interesting
to consider the Mobile Station Allocation (MSA). MSA is characterised by the pair (Channel list, MAIO). When a non
hopping (NH) mode is used, channel list is a channel while it corresponds to the mobile allocation list (MAL) in case of
base band hopping (BBH) or synthesised frequency hopping (SFH). For BBH, channels of MAL belong to a unique TRX
type.
Examples:
For each example given below, we assume that.
In case of NH, we have:
TRX index
Channel list
MAIO
MSA
53
(53,*)
54
(54,*)
In case of BBH, assuming TRXs belong to the same TRX type, we have:
TRX index
Channel list
MAIO
MSA
53
([53,54,55],0)
54
([53,54,55],1)
55
([53,54,55],2)
TRX index
Channel list
MAIO
MSA
53 54 55 56
([53,54,55,56],2)
53 54 55 56
([53,54,55,56],3)
In case of SFH, we have:
Therefore, for a mobile station, BBH and SFH work in the same way.
Consider the following notations:
v is a victim transmitter (TBC transmitter with a service area),
MSAS(v) is the set of MSAs associated to v. The number of MSAS(v) depends on TRX type(s) to be analysed (option
available in study properties): you may study a given TRX type tt (There are as many MSA(v) as TRXs allocated to the
subcell (v,tt)) or all the TRX types (The number of MSA(v) corresponds to the number of TRXs allocated to v),
i is a potential interfering transmitter (TBC transmitters which calculation area intersects service area of v),
MSAS(i) is the set of MSAs related to potential interferers i,
INT(v) is the set of transmitters that interfere v.
Several MSAs, m, are related to a transmitter. Therefore, for each victim transmitter v with MSA m (m MSAS(v)), Atoll
C v m
- , received at the mobile; mobile is connected to a victim transmitter, v with
calculates carrier to interference ratio --------------- Iv m
v
a given m. C m is the carrier power level received from v on m and I m corresponds to the interference received from
interfering transmitters i on m.
Atoll studies the most interfered MSA. So, it considers:
C v m
C
- except if analysis is detailed (Detailed result option).
---- = Min --------------- I v
k Iv m
If the interference conditions for the prediction study are defined using the option C/(I+N), Atoll takes the total noise N tot
into account as well. The total noise is computed by adding the thermal noise N thermal (defined in the document database
at -121 dBm by default) to the noise figure NF (either defined at the terminal type properties level, if a terminal type is
defined for the study, or defined directly in the prediction study conditions). So,
N tot = N thermal + NF
Cv m
C
-
Thus, for computations based on C/(I+N), ------------------ = Min ----------------------------- I + N tot
k
I v m + N tot
v
Note:
The M Shadowing used in the computations of C/I is a function of C/I standard deviation and
not the Model standard deviation.
5.6.1.2.1
Carrier Power Level

v
C m = P rec m
162
AT271_TRG_E6
Forsk 2009
5.6.1.2.2
Interference Calculation
Potential interferers can be transmitters i (iv), using co-channels and/or adjacent channels. Therefore, we can write:
v
I m = I co m + I adj m + I IMPx3 G PC
v
Where I co m is the interference received at v on m due to co-channels, I adj m is the interference received at v on m
i
due to adjacent channels, I IMPx3 is the third order intermodulation interference, and G PC is the average power control
gain defined for the interfering transmitter i.
v
I co m is the interference received at v on m due to co-channels, given by:
I co m =
i INT v n MSAS i
v i
p m n
P rec n T i n
co
And, I adj m is the interference received at v on m due to adjacent channels, given by:
I adj m =
i INT v n MSAS i
v i
p m n
P rec n
------------------- T i n
F
adj
v i
p m n is the probability of having a co- or adjacent channel collision between MSAs n and m (when n and m contain coand adjacent channels). It depends on the used frequency hopping mode.
i
P rec n is the carrier power level received from i on n,

Ti(n) is occupancy of the MSA n.
i
T i n = L traffic n f act n
i
If Average is selected in the study properties, L traffic n is the traffic load defined for the MSA n of i. If Maximum option
i
is selected, L traffic n = 1 .
i
f act n is the activity factor defined for the MSA n of i. If the subcell (i,tt) supports DTX mode, it is a global value specified
in the study properties. Otherwise, the activity factor is 1.
Note:
Since BCCH carrier is always On Air, DTX and traffic load gains do not reduce BCCH
i
interfering energy. In other words, f act n = 1 and L traffic n = 1 on the BCCH for the
interference estimation.
I IMPx3 has three components, i.e., intermodulation interference from frequencies used by the interfering transmitter, interference due to spurious emissions from the interfering transmitter, and the intermodulation interference received at the
mobile terminal:
TX
SE
Term
I IMPx3 = I IMPx3 + I IMPx3 + I IMPx3

The above components are calculated as follows:
i
P rec
TX
I IMPx3 = ---------------v
L IMPx3
SE i
SE
I IMPx3 = P rec
3 P rec
Term
I IMPx3 = -------------------------Prot
2 F IMPx3
i
Where P rec is the carrier power level received from the interferer i, L IMPx3 is the third order intermodulation loss at the
SE i
Prot
victim transmitter v, P rec is the spurious emission power level received from the interferer i, and F IMPx3 is the third order
intermodulation protection factor for the terminal.
For a pair of frequencies, f 1 and f 2 , two third order intermodulation products are generated at frequencies f 3 = 2 f 1 f 2
and f 4 = 2 f 2 f 1 . If a transmitter uses f 3 or f 4 , it is interfered by transmitters using f 1 and f 2 . All interferer frequencies
are used to calculate intermodulation products. When several frequency pairs generate intermodulation products, the
IMPs are independenly calculated and added to the interference. If power received over different frequencies is not the
same for two frequencies (not the same power offset for example), the corresponding intermodulation frequencies are
Forsk 2009
AT271_TRG_E6
163

ignored. Frequency hopping is not considered to have any impact on the intermodulation products. IMPs for hopping and
non-hopping cases are considered to be the same.
Intermodulation products generated by the adjacent frequencies of the frequencies actually being used by an interferer
are not taken into account. Similarly, intermodulation interference received on the adjacent frequencies of the frequencies
used by the victim are also ignored.
5.6.1.2.3
Collision Probability for Non Hopping Mode

v i
We have: p m n = 1
5.6.1.2.4
Collision Probability for BBH and SFH Modes

MSA m of v can be defined as the pair ([f1,f2,.fn], MAIO) and MSA n of i as the pair ([f1,f2,.fn], MAIO) (where f and f
are channels).
v
Now, let us consider the occurrence, OCCUR f m f' n , such that a channel f of m can meet a channel f of n during hopping
sequence.
There is a collision if f and f are co- or adjacent channels. Then, we can define a collision as follows:
v
Collision = OCCUR f m f' n such that f m f' n =

v
( equals 0 if f m and f' n are co-channels or 1 if adjacent channels)

Therefore, we have:
n collision
v i
p m n = -------------------------n occurence
ncollision and noccurence respectively correspond to the number of collisions and the number of occurrences. They are
closely linked to the correlation between m and n. We can have two cases:
1st Case: MSAs m and n are Correlated

m and n must have identical HSN and synchronisation.
The number of occurrences depends on the MAL size, MAIO And MAIO.
Example:
Schematic view of hopping sequences

MSA m of v
([34 37 39], MAIO=0)
34 37 39
MSA n of i
([38 36 34], MAIO=2)
38 36 34
Here, the number of occurrences is 3; the number of co-channel collisions is 1 and the number of adjacent channel collisions is 1.
So, we have:
v i
v i
1
1
p m n co = --- and p m n adj = --3
3
2nd Case: MSAs m and n are Not Correlated

Condition specified above is not fulfilled.
Probability to have each pair is the same. All the occurrences are possible.
Example:
Schematic view of hopping sequences

MSA m of v
([34 37 39], MAIO=0)
34 37 39
MSA n of i
([38 36 34], MAIO=2)
38 36 34
Here, the number of occurrences is 9; the number of co-channel collisions is 1 and the number of adjacent channel collisions is 3.
So, we have:
v i
v i
1
1
p m n co = --- and p m n adj = --9
3
164
AT271_TRG_E6
Forsk 2009

Note:
5.6.1.3
Only the carrier power level is downgraded by the shadowing margin. The interference
level is not altered.
Coverage Area Determination

C
C
For each victim transmitter v, coverage area corresponds to bins where ---- or ------------ is between lower and upper
I v
I + N v
thresholds specified in study properties. There are two possibilities:
5.6.1.3.1
Interference Condition Satisfied by At Least One TRX

This criterion implies that the interference condition defined in the interference study properties dialog must be satisfied
by at least on TRX of the transmitter in order for the pixel under study to be included in the coverage area.
In this case, the coverage area of a transmitter Txi corresponds to the bins where:
C
Minimum threshold ----
I v
TRX j
where,
C
Maximum threshold or Minimum threshold ------------
I + N v
Maximum threshold
TRX j
TRXj is any TRX belonging to Txi.
This coverage area will include all the bins satisfying the above criteria even if they are only covered by the TRX with the
best C/I or C/(I+N) conditions.
5.6.1.3.2
Interference Condition Satisfied by The Worst TRX

This criterion implies that the interference condition defined in the interference study properties dialog must be satisfied
by the worst TRX of the transmitter in order for the pixel under study to be included in the coverage area.
In this case, the coverage area of a transmitter Txi corresponds to the bins where:
C
Minimum threshold ----
I v
TRX j
C
Maximum threshold or Minimum threshold ------------
I + N v
Maximum threshold
TRX j
where, TRXj is the TRX (belonging to Txi) with the worst C/I or C/(I+N) conditions at the bin.
This coverage area will include only the bins satisfying the above criteria, i.e. covered by the TRX with the worst C/I or C/
(I+N) conditions.
5.6.1.4
Coverage Area Display

You can display the transmitter coverage area depending on the C/I (or C/(I+N)) level, prefer a display depending on transmitter colour or on any other transmitter attribute.
5.6.1.4.1
C/I Level
Each bin of the transmitter coverage area is coloured if the calculated C/I (or C/(I+N)) level exceeds ( ) the specified minimum thresholds (bin colour depends on C/I (or C/(I+N)) level). Coverage consists of several independent layers whose
visibility in the workspace can be managed. There are as many layers as transmitter coverage areas. Each layer shows
the different C/I levels available in the transmitter coverage area.
5.6.1.4.2
Max C/I Level

Atoll compares calculated C/I (or C/(I+N)) levels received from transmitters on each bin of each transmitter coverage area
where coverage areas overlap the studied one and chooses the highest value. A bin of a coverage area is coloured if the
C/I (or C/(I+N)) level exceeds ( ) the specified thresholds (the bin colour depends on the C/I (or C/(I+N)) level). Coverage
defined thresholds. Each layer corresponds to an area where the highest received C/I level exceeds a defined minimum
threshold.
5.6.1.4.3
Min C/I Level

Atoll compares C/I (or C/(I+N)) levels received from transmitters on each bin of each transmitter coverage area where the
coverage areas overlap the studied one and chooses the lowest value. A bin of a coverage area is coloured if the C/I (or
C/(I+N)) level exceeds ( ) the specified thresholds (the bin colour depends on the C/I (or C/(I+N)) level). Coverage
defined thresholds. Each layer corresponds to an area where the lowest received C/I level exceeds a defined minimum
threshold.
5.6.1.4.4
Transmitter
Atoll colours each bin of each transmitter coverage area. The bin colour corresponds to the transmitter colour. Coverage
interfered transmitters.
Forsk 2009
AT271_TRG_E6
165
5.6.2
Point Analysis
Analysis provided in the Interference tab is based on path loss matrices. You can study interference on:
TBC transmitters for which path loss matrices have been computed,
calculation areas.
Atoll indicates the following at the receiver:
The carrier power level received from the victim transmitter v on the most interfered MAS m,
Either the overall interference received from interfering transmitters i on MAS m (both co-channel and adjacent
channel interferers are considered), or the co-channel interference received from co-channel interfering transmitters i on MAS m, or the adjacent channel interference received from adjacent channel interfering transmitters i on
MAS m (for further information about noise calculation, please refer to Signal to noise calculation: noise calculation
part)
The interference level received from each interfering transmitter i on m. Interferers are sorted in a descending
order w.r.t. carrier power level.
Notes:
Neither DTX nor traffic load of TRXs are taken into account to evaluate interference levels.
i
Therefore, we have T i n = L traffic n f act n = 1 .
5.7
Only carrier power level is downgraded by the shadowing margin. The interference level is
not altered.
GPRS EGPRS Coverage Studies

Atoll calculates a coverage area for all the TBC transmitters, assuming that each bin within a TBC EGPRS transmitter
calculation area corresponds to a probe mobile receiver.
5.7.1
The study conditions in order to determine the coverage area of each TBC transmitter,

Let us assume that:
Each transmitter, Txi, belongs to a HCS layer, k, with a defined priority and a reception threshold.
Each transmitter, Txi, is a GPRS/EGPRS station (Txi-EGPRS as notation).
GPRS/EGPRS equipment installed on each transmitter, Txi, does not support 8PSK modulation.
We can have the following four cases.
5.7.1.1
All Servers
For each HCS layer, k, the coverage area of Txi corresponds to Txi calculation area.
5.7.1.2

For each HCS layer, k, the coverage area of Txi corresponds to the bins where the signal level received from Txi,
Txi EGPRS
P rec
BCCH , is the highest one (Txi is the best server) or within a defined margin of the highest signal level (within
a margin of the best server).

Note:
If the margin equals 0, the coverage area of Txi corresponds to the bins where
Txi EGPRS
P rec
5.7.1.3
BCCH is the highest.

Txi EGPRS
P rec
BCCH , is the second highest one (Txi is the second best server) or within a defined margin of the second
highest signal level (within a margin of the second best server).

Note:
Txi EGPRS
P rec
166
BCCH is the second highest.
AT271_TRG_E6
Forsk 2009
5.7.1.4

Txi EGPRS
P rec
Txi EGPRS
a margin of the best server). In addition P rec
BCCH must exceed the reception threshold defined per HCS layer.
Notes:
Txi EGPRS
P rec
5.7.1.5
BCCH is the highest.

In this case, the coverage area of Txi corresponds to the bins where the signal level received from Txi,
Txi EGPRS
P rec
a margin of the best server). And, Txi belongs to the HCS layer with the highest priority. The highest priority is defined by
Txi EGPRS
the priority field (0: lowest) assuming the received P rec
BCCH exceeds the reception threshold defined per HCS
layer.
Note:
5.7.2
Calculation Options
GPRS/EGPRS studies can be based either on signal level (C), or on the carrier-to-interference ratio (C/I) with or without
considering thenoise (N).
If a reference terminal type is defined when launching a CQI study, Atoll will consider the noise figure defined for that terminal type when computing the total noise ( N tot ). If no terminal type is defined, the value defined in the coverage prediction
properties (8 dB, by default) is used.
Different GPRS/EGPRS equipment can be defined for a transmitter and for the reference terminal type. In this case, Atoll
performs an intersection for the coding schemes defined in the transmitter and the reference terminal type GPRS/EGPRS
equipment to determine the coding schemes that are common in the two. Hence, Atoll creates a temporary GPRS/EGPRS
equipment comprising only these common coding schemes and uses this equipment to eventually compute the coding
scheme, throughput and other values.
If no reference terminal type is defined or if the defined reference terminal type has no GPRS/EGPRS equipment assigned
to it, Atoll will perform the computations based on the GPRS/EGPRS equipment of the transmitter. Similarly, if a transmitter
has no GPRS/EGPRS equipment defined, Atoll will compute the coverage study considering only the GPRS/EGPRS
equipment defined for the reference terminal type. If there is no GPRS/EGPRS equipment defined for the transmitter and
for the reference terminal type, there will be no coverage for the transmitter.
5.7.2.1
Calculations Based on C
In this case, only the received signal level is taken into account. Atoll evaluates the signal level received from GPRS/
Txi EGPRS
EGPRS transmitters on TRXs (TRX) belonging to a selected TRX type (tt) or on all the TRXs, P rec
5.7.2.2
TRX .
Calculations Based on C/I Without Considering Thermal Noise

For GPRS/EGPRS studies based on the received signal level and C/I ratio, Atoll evaluates:
The signal level received from GPRS/EGPRS transmitters on TRXs (TRX) belonging to the selected TRX type (tt)
Txi EGPRS
or on all the TRXs, P rec
TRX .
And
The carrier-to-interference ratio received on TRXs (TRX) belonging to the selected TRX type (tt) or on all the
Txi EGPRS
P rec
TRX
TRXs, -------------------------------------------------- .
I
Forsk 2009
AT271_TRG_E6
167

Notes:
5.7.2.3
When GPRS/EGPRS prediction studies calculations are based on C/I, Atoll calculates the
carrier-to-interference ratio for all the EGPRS TBC transmitters but takes into account all
the TBC transmitters (GSM and GPRS/EGPRS) to evaluate the interference.
For further information on interference (I) calculation, please refer to Interference prediction
studies: Interference calculation part.
Calculations Based on C/I Considering Thermal Noise

For GPRS/EGPRS studies based on the received signal level and C/I ratio considering the effect of thermal noise, Atoll
evaluates:
The received signal-level-to-thermal-noise ratio from GPRS/EGPRS transmitters on TRXs (TRX) belonging to the
Txi EGPRS
P rec
TRX
selected TRX type (tt) or on all the TRXs, -------------------------------------------------- .
N
And
The carrier-to-interference-and-noise ratio received on TRXs (TRX) belonging to the selected TRX type (tt) or on
Txi EGPRS
P rec
TRX
all the TRXs, -------------------------------------------------- .
I+N
Where, N is the thermal noise whose value is defined in the document database at -121 dBm by default.
Notes:
5.7.2.4
When GPRS/EGPRS prediction studies calculations are based on C/I, Atoll calculates the
carrier-to-interference ratio for all the EGPRS TBC transmitters but takes into account all
the TBC transmitters (GSM and GPRS/EGPRS) to evaluate the interference.
For further information on interference (I) calculation, please refer to Interference prediction
studies: Interference calculation part.
Ideal Link Adaptation (ILA)

Ideal link adaptation implies that the coding scheme selected will be the one that provides the maximum throughput.
5.7.3
Coverage Study Scenarios

With the above options, there are many different possible scenarios of calculations. These scenarios are described below.
5.7.3.1
GPRS/EGPRS Studies Based on C Without ILA
5.7.3.1.1
Coding Scheme Based on C Without ILA

Atoll calculates the signal level received from Txi on each bin of the Txi coverage area. Then, selects a coding scheme,
cs, from among the coding schemes available in the GPRS/EGPRS equipment, such that:
For each TRX type, tt, cs = Lowest CS
Txi EGPRS
P rec
TRX Reception Threshold CS
Where, cs is the resulting coding scheme, CS is the set of all available coding schemes, and Reception Threshold CS
are the values of reception thresholds for the coding schemes available in the GPRS/EGPRS equipment, defined in the
Reception Thresholds column of the properties dialog.
There can be more than one coding schemes whose reception thresholds are less than the received signal level. cs is the
coding scheme with the lowest coding scheme number.
A Power Backoff, in dBs, can be defined for each subcell using an 8PSK modulation (EDGE) based codec equipment.
This power backoff is taken in to account when selecting the codec mode available in the coverage area of the subcell,
such that:
Txi EGPRS
For each TRX type, tt, P rec
5.7.3.1.2
Txi EGPRS
TRX P Backoff
Throughput Based on C Without ILA

Txi EGPRS
Atoll reads the throughput value for the received signal level, P rec
TRX , from the Throughput=f(C) graph asso-
ciated to the coding scheme, cs, determined above.
168
AT271_TRG_E6
Forsk 2009
5.7.3.2
GPRS/EGPRS Studies Based on C With ILA
5.7.3.2.1
Coding Scheme Based on C With ILA

With Ideal Link Adaptation active, Atoll selects the coding scheme that provides the highest throughput at the received
signal level value for a bin.
Atoll calculates the signal level received from Txi on each bin of the Txi coverage area. Then, selects a coding scheme,
cs, from among the coding schemes available in the GPRS/EGPRS equipment, such that:
For each TRX type, tt, cs = Lowest CS
Where,
cs
is
the
resulting
Txi EGPRS
TP = Highest TP=f C=P rec
coding
Txi EGPRS
Highest TP=f C=P rec
TRX
scheme,
CS
is
the
TRX
set
of
all
available
coding
schemes,
and
is the highest throughput provided by any coding scheme at the received signal
level, from the Throughput = f(C) graphs defined in the properties dialog.
If there are more than one coding schemes providing the highest throughput at the bin, cs will be the one with the lowest
coding scheme number.
such that:
Txi EGPRS
For each TRX type, tt, TP = Highest TP=f C=P rec
5.7.3.2.2
Txi EGPRS
TRX - P Backoff
TRX
Throughput Based on C With ILA

The throughput at the given bin and for each TRX type, tt, is simply the throughput computed earlier for the coding scheme,
cs, determined above. This throughput is computed from the Throughput = f(C) graphs for the value of the received signal
Txi EGPRS
level at the bin, P rec
TRX .
5.7.3.3
GPRS/EGPRS Studies Based on C/I Without ILA and Thermal Noise
5.7.3.3.1
Coding Scheme Based on C/I Without ILA and Thermal Noise

Atoll calculates signal level and C/I level received from Txi on each bin of the Txi coverage area. Then, selects two coding
schemes from among the coding schemes available in the GPRS/EGPRS equipment, such that:
For each TRX type, tt, cs C = Lowest CS
And, cs C I = Lowest CS
Txi EGPRS
P rec
Txi EGPRS
P rec
TRX
C
----------------------------------------------------- -- I Threshold CS
I
Where, csC is the coding scheme determined from the signal level, csC/I is the coding scheme determined from the C/I
level, and CS is the set of all available coding schemes.
Reception Threshold CS are the values of reception thresholds for the coding schemes available in the GPRS/EGPRS
C
equipment, defined in the Reception Thresholds column of the properties dialog. And, ---- Threshold
are the values of
I
CS
C/I thresholds for the coding schemes available in the GPRS/EGPRS equipment, defined in the C/I Thresholds column of
the properties dialog.
There can be more than one coding schemes whose reception thresholds are less than the received signal level, and
whose C/I thresholds are less than the received C/I level. cs is the coding scheme with the lowest coding scheme number.
The resulting coding scheme, cs, is the coding scheme with the lowest coding scheme number from among csC and csC/I.
cs = Lowest cs C cs C I
such that:
Txi EGPRS
For each TRX type, tt, P rec
Txi EGPRS
TRX P Backoff
And
Txi EGPRS
Txi EGPRS
P rec
TRX P Backoff
TRX C
---------------------------------------------------------------------------------------------------------- ---- Threshold
I
CS
I
Forsk 2009
AT271_TRG_E6
169
5.7.3.3.2
Throughput Based on Worst Case Between C and C/I Without ILA

Atoll determines two throughput values, TPC from the Throughput = f(C) graph corresponding to the coding scheme csC
determined above, and TPC/I from Throughput = f(C/I) graph corresponding to the coding scheme csC/I determined above.
The resulting throughput TP is the lower of the two values, TPC and TPC/I.
TP = Lowest TP C TP C I
5.7.3.4
GPRS/EGPRS Studies Based on C/I With ILA and Without Thermal Noise
5.7.3.4.1
Coding Scheme Based on C/I With ILA and Without Thermal Noise
With Ideal Link Adaptation active, Atoll selects the coding scheme that provides the highest throughput at the received
signal level and C/I value for a bin.
Atoll calculates signal level and C/I level received from Txi on each bin of the Txi coverage area. Then, selects two coding
For each TRX type, tt, cs C = Lowest CS
And, cs C I = Lowest CS
Txi EGPRS
TP = Highest TP=f C=P rec
TRX
Txi EGPRS
C P rec

TP = Highest TP=f ---- = ----------------------------------- TRX
I
I
Where, csC is the coding scheme determined from the signal level, csC/I is the coding scheme determined from the C/I
level, and CS is the set of all available coding schemes.
Txi EGPRS
Highest TP=f C=P rec
TRX is the highest throughput provided by any coding scheme at the received signal
Txi EGPRS
C P rec
level, from the Throughput = f(C) graphs defined in the properties dialog. And, Highest TP=f ---- = -------------------------------- TRX
I
I
is the highest throughput provided by any coding scheme at the received C/I level, from the Throughput = f(C/I) graphs
defined in the properties dialog.
If there are more than one coding schemes providing the highest throughput at the bin, csC and csC/I will be the ones with
the lowest coding scheme numbers.
The resulting coding scheme, cs, is the coding scheme with the lowest coding scheme number from among csC and csC/I.
cs = Lowest cs C cs C I
such that:
Txi EGPRS
For each TRX type, tt, TP = Highest TP=f C=P rec
Txi EGPRS
TRX - P Backoff
TRX
And
Txi EGPRS
Txi EGPRS
TRX - P Backoff
TRX
C P rec
TP = Highest TP=f ---- = -------------------------------------------------------------------------------------------------------------

I
I
5.7.3.4.2
Throughput Based on Worst Case Between C and C/I With ILA

Atoll determines two throughput values, TPC from the Throughput = f(C) graph corresponding to the coding scheme csC
determined above, and TPC/I from Throughput = f(C/I) graph corresponding to the coding scheme csC/I determined above.
The resulting throughput TP is the lower of the two values, TPC and TPC/I.
TP = Lowest TP C TP C I
5.7.3.5
GPRS/EGPRS Studies Based on C/I Without ILA and With Thermal Noise
5.7.3.5.1
Coding Scheme Based on C/I Without ILA and With Thermal Noise
The reception thresholds given for signal level C are internally converted to C/N thresholds (where N is the thermal noise
defined in the document database at -121 dBm by default) in order to be indexed by C/(I+N) values. C/I thresholds are
also indexed by the C/(I+N) value.
Atoll calculates the C/N and C/(I+N) level received from Txi on each bin of the Txi coverage area. Then, selects two coding
For each TRX type, tt, cs C N = Lowest CS
170
Txi EGPRS
P rec
TRX
C----------------------------------------------------- ----------Threshold
I + N
CS
N
AT271_TRG_E6
Forsk 2009
And, cs C I + N = Lowest CS
Txi EGPRS
P rec
TRX
C----------------------------------------------------- ----------Threshold
I + N
CS
I+N
Where, csC/N is the coding scheme determined from the C/N level, csC/(I+N) is the coding scheme determined from the C/
(I+N) level, and CS is the set of all available coding schemes.
C ----------Threshold
I + N
CS are the values of C/(I+N) thresholds for the coding schemes available in the GPRS/EGPRS equipment, determined from the C/I threshold values defined in the C/I Thresholds column of the properties dialog.
There can be more than one coding schemes whose C/(I+N) thresholds are less than the received C/N level, whose C/
(I+N) thresholds are less than the received C/(I+N) level. cs is the coding scheme with the lowest coding scheme number.
The resulting coding scheme, cs, is the coding scheme with the highest coding scheme number from among csC/N and
csC/(I+N).
cs = Highest cs C N cs C I + N
such that:
Txi EGPRS
Txi EGPRS
P rec
TRX P Backoff
TRX
C
For each TRX type, tt, ----------------------------------------------------------------------------------------------------------- ------------ Threshold
I + N
CS
N
And
Txi EGPRS
Txi EGPRS
P rec
TRX P Backoff
TRX C
---------------------------------------------------------------------------------------------------------- ------------ Threshold
I + N
CS
I+N
5.7.3.5.2
Throughput Based on Interpolation Between C/N and C/(I+N) Without ILA

Atoll determines two throughput values, TPC/N from the Throughput = f(C) graph corresponding to the coding scheme csC/
determined above, and TPC/(I+N) from Throughput = f(C/I) graph corresponding to the coding scheme csC/(I+N) determined above.
N
The Throughput = f(C) graph is internally converted to Throughput = f(C/N) graph, in order to be indexed with the C/(I+N)
value. The Throughput = f(C/I) graph is also indexed with the C/(I+N) value. The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The throughput interpolation method consists in
interpolating TPC/N and TPC/(I+N) according to the respective weights of I and N values.
The resulting throughput TP is given by:
TP = TP C N + 1 TP C I + N
Where,
pN
= --------------------pI + N
pN is the thermal noise power (value in Watts)
p(I+N) is the interferences + thermal noise power (value in Watts)
TPC/N is the throughput obtained from the C/N graph
TPC/(I+N) is the throughput obtained from the C/I+N graph
5.7.3.6
GPRS/EGPRS Studies Based on C/I With ILA and Thermal Noise
5.7.3.6.1
Coding Scheme Based on C/I With ILA and Thermal Noise

The reception thresholds given for signal level C are internally converted to C/N thresholds (where N is the thermal noise
defined in the document database at -121 dBm by default) in order to be indexed by C/(I+N) values. C/I thresholds are
also indexed by the C/(I+N) value.
With Ideal Link Adaptation active, Atoll selects the coding scheme that provides the highest throughput at the received C/
N and C/(I+N) values for a bin.
Atoll calculates the C/N and C/(I+N) level received from Txi on each bin of the Txi coverage area. Then, selects two coding
For each TRX type, tt, cs C N = Highest CS
Forsk 2009
Txi EGPRS
C P rec
TP = Highest TP=f ------------ = ----------------------------------- TRX

N
I + N
AT271_TRG_E6
171
And, cs C I + N = Highest CS
Txi EGPRS
C P rec
TP = Highest TP=f ------------ = ----------------------------------- TRX

I
+
N
I
+
N
Where, csC/N is the coding scheme determined from the C/N level, csC/(I+N) is the coding scheme determined from the C/
(I+N) level, and CS is the set of all available coding schemes.
Txi EGPRS
C P rec
Highest TP=f ------------ = -------------------------------- TRX is the highest throughput provided by any coding scheme at the received C/
N
I
+
N
N level, from the Throughput = f(C/(I+N)) graphs converted from the f(C/I) graphs defined in the properties dialog. And,
Txi EGPRS
C P rec
Highest TP=f ------------ = -------------------------------- TRX is the highest throughput provided by any coding scheme at the received C/
I+N
I + N
(I+N) level, from the Throughput = f(C/(I+N)) graphs converted from the f(C/I) graphs defined in the properties dialog.
If there are more than one coding schemes providing the highest throughput at the bin, csC/N and csC/(I+N) will be the ones
with the highest coding scheme numbers.
The resulting coding scheme, cs, is the coding scheme with the highest coding scheme number from among csC/N and
csC/(I+N).
cs = Highest cs C N cs C I + N
such that:
Txi EGPRS
Txi EGPRS
P rec
TRX P Backoff
TRX
C
For each TRX type, tt, ----------------------------------------------------------------------------------------------------------- ------------ Threshold
I + N
CS
N
Txi EGPRS
Txi EGPRS
P rec
TRX P Backoff
TRX
C
And ----------------------------------------------------------------------------------------------------------- ------------ Threshold
I + N
CS
I+N
5.7.3.6.2
Throughput Based on Interpolation Between C/N and C/(I+N) With ILA

Atoll determines two throughput values, TPC/N from the Throughput = f(C) graph corresponding to the coding scheme csC/
determined above, and TPC/(I+N) from Throughput = f(C/I) graph corresponding to the coding scheme csC/(I+N) determined above.
The Throughput = f(C) graph is internally converted to Throughput = f(C/N) graph, in order to be indexed with the C/(I+N)
value. The Throughput = f(C/I) graph is also indexed with the C/(I+N) value. The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The throughput interpolation method consists in
interpolating TPC/N and TPC/(I+N) according to the respective weights of I and N values.
The resulting throughput TP is given by:
TP = TP C N + 1 TP C I + N
Where,
pN
= --------------------pI + N
pN is the thermal noise power (value in Watts)
p(I+N) is the interferences + thermal noise power (value in Watts)
TPC/N is the throughput obtained from the C/N graph
TPC/(I+N) is the throughput obtained from the C/I+N graph
5.7.4
Coverage Display
Coverage area can be displayed with colours depending on:
5.7.4.1
GPRS/EGPRS Coding Schemes Study Display Types
5.7.4.1.1
Coding Schemes
Only the bins with a coding scheme assigned are coloured. The bin colour depends on the assigned coding scheme.
Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many
layers as transmitter coverage areas. Each layer shows the coding schemes available in the transmitter coverage area.
172
AT271_TRG_E6
Forsk 2009
5.7.4.1.2
Best Coding Schemes

On each bin, Atoll chooses the highest coding scheme available from the TRXs of different transmitters covering that bin.
Only the bins with a coding scheme assigned are coloured. The bin colour depends on the assigned coding scheme.
Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many
layers as possible coding schemes. Each layer shows the areas where a given coding scheme can be used.
5.7.4.2
RLC/MAC and Application Throughput/Timeslot Studies Display Types
5.7.4.2.1
Relation Between RLC/MAC and Application Throughputs

Application throughput per timeslot is deduced from the RLC/MAC (or gross) throughput per timeslot by the equation:
SF
TP Application = TP RLC MAC ---------- TP Offset
100
Where, TP Application = Application throughput,
TP RLC MAC = RLC/MAC level throughput,
TP Offset = Throughput offset (kbps) accounting for headers, guard-bits etc.,
SF = Throughput scaling factor (%) accounting for coding, redundance etc.
Note:
5.7.4.2.2
The default value of the throughput offset is 0 kbps. If you are getting the same coverage
plots for the RLC/MAC and application level throughputs, try assigning a higher value to
this offset to visualise the difference.
Throughput/Timeslot
A bin of a coverage area is coloured if the calculated throughput exceeds the defined minimum threshold. The bin colour
depends on throughput. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as transmitter coverage areas. Each layer shows the throughputs that a transmitter
can provide on one timeslot.
5.7.4.2.3
Best Throughput/Timeslot
On each bin, Atoll chooses the highest throughput available from the TRXs of different transmitters covering that bin. A
bin of a coverage area is coloured if the best throughput exceeds the defined minimum threshold. The bin colour depends
on throughput. Coverage consists of several independent layers whose visibility in the map window can be managed.
There are as many layers as defined thresholds. Each layer shows the areas where a throughput can be provided on one
timeslot.
5.7.4.2.4
Average Throughput/Timeslot
On each bin, Atoll calculates the average throughput available from the TRXs of different transmitters covering that bin. A
bin of a coverage area is coloured if the average throughput exceeds the defined minimum threshold. The bin colour
depends on throughput. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as defined thresholds. Each layer shows the areas where a throughput can be
provided on one timeslot.
5.7.4.2.5
Block Error Rate Computation

TP
Block error rate is computed according to the formula BLER = -----------------TP MAX
Where, TP = Throughput per timeslot calculated for a bin, and TPMAX = Maximum throughput per timeslot deduced from
the GPRS/EGPRS equipment assigned to the terminal type (or transmitter, if no terminal type has been selected for the
study) for the coding scheme calculated for a bin.
Notes:
5.7.4.2.6
BLER can be displayed for RLC/MAC and application throughput according to the option
selected.
If TP > TPMAX, then BLER = 0.
BLER Percentage
Atoll calculates BLER percentage by considering throughput/timeslot per bin (computed as described earlier) and the
maximum throughput/timeslot possible (deduced from the GPRS/EGPRS equipment graphs). A bin of a coverage area is
coloured if the calculated BLER percentage exceeds the defined minimum threshold. The bin colour depends on the
BLER. Coverage consists of several independent layers whose visibility in the map window can be managed. There are
as many layers as transmitter coverage areas. Each layer shows the BLERs that a transmitter experiences on one timeslot.
Forsk 2009
AT271_TRG_E6
173
5.7.4.2.7
Maximum BLER Percentage

On each bin, Atoll chooses the maximum BLER percentage from the BLER values corresponding to TRXs of different
transmitters covering that bin. A bin of a coverage area is coloured if the maximum BLER exceeds the defined minimum
threshold. The bin colour depends on the BLER. Coverage consists of several independent layers whose visibility in the
map window can be managed. There are as many layers as defined thresholds. Each layer shows the areas where a BLER
is observed on one timeslot.
5.8
Circuit Quality Indicators Studies

Atoll calculates a coverage area for all the TBC transmitters. Let us assume that each bin within a TBC transmitter calculation area corresponds to a probe mobile receiver. Coverage study parameters to be set are:
5.8.1
The study conditions in order to determine the coverage area of each TBC transmitter,
Circuit Quality Indicators

Atoll has the following circuit quality indicators included by default:
FER or Frame Erasure Rate

- The number of frames in error divided by the total number of frames. These frames are usually discarded, in
which case this can be called the Frame Erasure Rate.
BER or Bit Error Rate
- BER is a measurement of the raw bit error rate in reception before the decoding process begins. Any factor
that impacts the decoding performance, such as frequency hopping, will impact the correlation between BER
and FER, or the perceived end-user voice quality.
MOS or Mean Opinion Score
- Voice quality can be quantified using mean opinion score (MOS). MOS values can only be measured in a test
laboratory environment. MOS values range from 1 (bad) to 5 (excellent). Different voice codecs have slightly
different FER to MOS correlation since the smaller the voice codec bit rate is, the more sensitive it becomes
to frame erasures.
There are various codec modes defined for AMR depending on the FR and HR channel modes that it supports. Codec
modes defined in Atoll include the basic EFR, FR and HR codec modes along with the AMR codec modes:
AMR FR
- AMR TCH/AFS12.2
- AMR TCH/AFS10.2
- AMR TCH/AFS7.95
- AMR TCH/AFS7.4
- AMR TCH/AFS6.7
- AMR TCH/AFS5.9
- AMR TCH/AFS5.15
- AMR TCH/AFS4.75
AMR HR
- AMR TCH/AHS7.95
- AMR TCH/AHS7.4
- AMR TCH/AHS6.7
- AMR TCH/AHS5.9
- AMR TCH/AHS5.15
- AMR TCH/AHS4.75
EFR
FR
HR
A codec equipment should contain codec mode adaptation thresholds and quality graphs for the relevant circuit quality
indicators in order to be considered in circuit quality indicators coverage studies. The default codec equipment in Atoll
include default FER, BER and MOS quality graphs with respect to the carrier to interference ratio, and codec mode adaptation thresholds (computed from the FER vs. C/I graphs for all codec modes at 5% FER.).
Note:
Depending on the reference CQI, codec mode that provides the lowest BER or FER, or the
highest MOS are selected during computations.
References:
The above graphs are based on:
[1] T. Halonen, J. Romero, J. Melero; GSM, GPRS and EDGE performance Evolution towards 3G/UMTS, John Wiley
and Sons Ltd.
[2] J. Wigard, P. Mogensen; A simple mapping from C/I to FER and BER for a GSM type of air interface.
[3] 3GPP Specifications TR 26.975 V6.0.0; Performance characterization of the Adaptive Multi-Rate (AMR) speech
codec (Release 6)
174
AT271_TRG_E6
Forsk 2009
Figure 5.11: FER vs. C/I Graphs
Figure 5.12: BER vs. C/I Graphs
Figure 5.13: MOS vs. C/I Graphs
5.8.2

We can have four different cases for coverage area determination. Let us assume that:
Forsk 2009
Each transmitter, Txi, belongs to a HCS layer, k, with a defined priority.

Each transmitter, Txi, is has a codec equipment assigned (Txi as notation).
AT271_TRG_E6
175
5.8.2.1
All Servers
For each HCS layer, k, the coverage area of Txi corresponds to Txi calculation area.
5.8.2.2

Txi
P rec BCCH , is the highest one (Txi is the best server) or within a defined margin of the highest signal level (within a
margin of the best server).
Note:
Txi
P rec BCCH is the highest.
5.8.2.3

Txi
P rec BCCH , is the second highest one (Txi is the second best server) or within a defined margin of the second highest
signal level (within a margin of the second best server).
Note:
Txi
P rec BCCH is the second highest.
5.8.2.4

Txi
P rec BCCH , is the highest one (Txi is the best server) or within a defined margin of the highest signal level (within a
Txi
margin of the best server). In addition P rec BCCH must exceed the reception threshold defined per HCS layer.
Notes:
Txi
P rec BCCH is the highest.
5.8.2.5

Txi
In this case, the coverage area of Txi corresponds to the bins where P rec BCCH is the highest one (Txi is the best
server) or within a defined margin of the highest signal level (within a margin of the best server). And, Txi belongs to the
HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest) assuming the received
Txi
P rec BCCH exceeds the reception threshold defined per HCS layer.
Note:
5.8.3
Calculation Options
CQI studies can be based either on the signal-to-noise level (C/N) or on the signal-to-noise-plus-interference ratio (C/
(I+N)).
If a reference terminal type is defined when launching a CQI study, Atoll will consider the noise figure defined for that terminal type when computing the total noise (N). If no terminal type is defined, the value defined in the coverage prediction
properties (8 dB, by default) is used.
Different codec equipment can be defined for a transmitter and for the reference terminal type. In this case, Atoll performs
an intersection for the codec modes defined in the transmitter and the reference terminal type codec equipment to determine the codec modes that are common in the two. Hence, Atoll creates a temporary codec equipment comprising only
these common codec modes and uses this codec equipment to eventually compute the CQI values.
If no reference terminal type is defined or if the defined reference terminal type has no codec equipment assigned to it,
Atoll will perform the computations based on the codec equipment of the transmitter. Similarly, if a transmitter has no codec
equipment defined, Atoll will compute the coverage study considering only the codec equipment defined for the reference
176
AT271_TRG_E6
Forsk 2009

terminal type. If there is no codec equipment defined for the transmitter and for the reference terminal type, there will be
no coverage for the transmitter.
For the case where more than one codec modes, compatible with the transmitter - reference terminal type pair, satisfy the
C or C/I conditions at a bin under study, Atoll chooses the codec mode for CQI determination according to their selection
priorities. The table below depicts the selection priorities for all codec modes modeled in Atoll. These selection priorities
are based on the chronological order of their development and on their C/I - MOS performance.
5.8.3.1
Codec Mode
Selection priority
FR
HR
EFR
AMR TCH/AFS4.75
AMR TCH/AFS5.15
AMR TCH/AFS5.9
AMR TCH/AFS6.7
AMR TCH/AFS7.4
AMR TCH/AFS7.95
AMR TCH/AFS10.2
10
AMR TCH/AFS12.2
11
AMR TCH/AHS4.75
12
AMR TCH/AHS5.15
13
AMR TCH/AHS5.9
14
AMR TCH/AHS6.7
15
AMR TCH/AHS7.4
16
AMR TCH/AHS7.95
17
Calculations Based on C/N

Txi
In this case, only signal level received and the total noise are taken into account. Atoll evaluates P rec TRX , the signal
level received from transmitter Txi on TRXs (TRX) belonging to the selected TRX type (tt), or on all the TRXs, on each bin
of Txi coverage area and converts it into C/N values using the value of total noise computed as follows:
N tot = N + NF
Where, N is the thermal noise (defined in the document database at -121 dBm by default), and NF is the receiver noise
figure (either defined at the terminal type properties level, if a reference terminal type is defined for the study, or defined
in the coverage study properties).
The computed total noise N tot is then compared to the codec equipment reference noise N Ref . If the values are the same,
the defined graphs are used as is, unless the entry is downshifted by the difference N tot N Ref .
5.8.3.2
Calculations Based on C/(I+N)

For circuit quality indicator studies based on the signal-to-noise-and-interference ratio (C/(I+N)), Atoll evaluates:
Txi
P rec TRX
----------------------------- : The signal-level-to-noise ratio received from transmitter Txi on TRXs (TRX) belonging to the selected
N tot
TRX type (tt) or on all the TRXs.
And
Txi
P rec TRX
----------------------------- : The carrier-to-interference-and-noise ratio received from transmitter Txi on TRXs (TRX) belonging
I + N tot
to the selected TRX type (tt) or on all the TRXs.
The value of total noise computed as follows:

N tot = N + NF
Where, N is the thermal noise (defined in the document database at -121 dBm by default), and NF is the receiver noise
figure (either defined at the terminal type properties level, if a terminal type is defined for the study, or defined in the coverage study properties).
The computed total noise N tot is then compared to the codec equipment reference noise N Ref . If the values are the same,
the defined graphs are used as is, unless the entry is downshifted by the difference N tot N Ref .
Note:
Forsk 2009
AT271_TRG_E6
177
5.8.3.3
Atoll calculates the carrier-to-interference ratio for all the TBC transmitters but takes into
account all the transmitters (even the ones with no codec equipment assigned) to evaluate
the interference.
Ideal Link Adaptation (ILA)

Ideal link adaptation for circuit quality indicator studies is defined at the codec equipment level. If the ideal link adaptation
option is checked, Atoll will select the codec mode, for the transmitter under study, according to the codec quality graphs
(CQI = f(C/N) and CQI = f(C/I)) related to the defined reference CQI, which may be different from the CQI being calculated.
Otherwise, Atoll will use the adaptation thresholds defined in the Adaptation Thresholds tab to determine the codec mode
to be used in the studies.
5.8.4
Calculation Scenarios
With the above options, there are many different possible scenarios of calculations. These scenarios are described below.
5.8.4.1
CQI Study Based on C/N Without ILA

Atoll calculates signal level received from Txi on each bin of Txi coverage area and converts it into C/N values as described
earlier. Then, Atoll filters all the codec modes that satisfy the C/N values and are common between the transmitter and the
terminal type codec equipment. It then determines the codec mode for the bin, such that:
For each TRX type, tt, cm = Highest Priority CM
Txi
P rec TRX
------------------------------ Adaptation Threshold CM
N tot
Where, cm is the codec mode with the highest priority among the set of codec modes CM having their adaptation threshTxi
P rec TRX
olds less than the received C/N level, ----------------------------- .
N tot
From the CQI=f(C/N) graph associated to the selected codec mode cm, Atoll evaluates the CQI for which the study was
Txi
P rec TRX
performed corresponding to ----------------------------- for the selected codec mode.
N tot
5.8.4.2
CQI Study Based on C/N With ILA

Ideal link adaptation is used by a codec equipment according to a defined reference CQI (MOS by default).
Atoll calculates signal level received from Txi on each bin of Txi coverage area and converts it into C/N values as described
earlier. Then, Atoll filters all the codec modes that satisfy the C/N criterion (defined by the CQI = f(C/N) graphs for the reference CQI) and are common between the transmitter and the terminal type codec equipment.
The selected codec mode among these filtered codec modes will be,
Or, cm = Highest Priority CM
, for MOS
Txi
C P rec TRX
CQI Ref = Highest CQI=f ---- = ------------------------------
N tot
, for BER and FER

Txi
TRX
C rec

CQI Ref = Lowest CQI=f ---- = ------------------------------
N tot
Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI
Txi
P rec TRX
gives the highest or the lowest value at the received C/N level, ----------------------------- .
N tot
If more than one codec mode graphs give the same value for reference CQI, then Atoll selects the codec mode with the
highest priority.
From the CQI = f(C/N) graph associated to the selected codec mode cm, Atoll evaluates the CQI for which the study was
Txi
P rec TRX
performed corresponding to ----------------------------- for the selected codec mode.
N tot
5.8.4.3
CQI Study Based on C/(I+N) Without ILA

Atoll calculates the C/I level received from the transmitter on each bin of Txi coverage area, for each TRX and converts it
into C/(I+N). Then, Atoll filters all the codec modes that satisfy the C/(I+N) values and are common between the transmitter
and the terminal type codec equipment.
It then determines the codec mode for the bin, such that:
178
AT271_TRG_E6
Forsk 2009
Txi
P rec TRX
------------------------------ Adaptation Threshold CM
I + N tot
Where, cm is the codec mode with the highest priority among the set of codec modes CM having their adaptation threshTxi
P rec TRX
olds less than the received C/(I+N) level, ----------------------------- .
I + N tot
From the CQI = f(C/I) graph associated to the selected codec mode cm (indexed with the C/(I+N) values), Atoll evaluates
Txi
P rec TRX
the CQI for which the study was performed corresponding to ----------------------------- for the selected codec mode.
I + N tot
5.8.4.4
CQI Study Based on C/(I+N) With ILA

Ideal link adaptation is used by a codec equipment according to a defined reference CQI (MOS by default).
Atoll calculates the C/I level received from the transmitter on each bin of Txi coverage area, for each TRX and converts it
into C/(I+N). Then, Atoll filters all the codec modes that satisfy the C/(I+N) criteria (defined by the CQI = f(C/I) graphs for
the reference CQI) and are common between the transmitter and the terminal type codec equipment.
The selected codec mode among these filtered codec modes will be,
Or, cm = Highest Priority CM
, for MOS
Txi
P
TRX
C rec

CQI Ref = Highest CQI=f ---- = ------------------------------
I + N tot
, for BER and FER

Txi
C P rec TRX
CQI Ref = Lowest CQI=f ---- = ------------------------------
I
I
+
N
tot
Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI
Txi
P rec TRX
gives the highest or the lowest value at the received C/(I+N) level, ----------------------------- .
I + N tot
If more than one codec mode graphs give the same value for reference CQI, then Atoll selects the codec mode with the
highest priority.
From the CQI = f(C/I) graph associated to the selected codec mode cm (indexed with the C/(I+N) values), Atoll evaluates
Txi
P rec TRX
the CQI for which the study was performed corresponding to ----------------------------- for the selected codec mode.
I + N tot
5.8.5
Coverage Display
Coverage area can be displayed with colours depending on:
5.8.5.1
Circuit Quality Indicators Study Display Types
5.8.5.1.1
FER/BER/MOS
Only the bins with a CQI assigned are coloured. The bin colour depends on the assigned CQI value. Coverage consists
of several independent layers whose visibility in the map window can be managed. There are as many layers as transmitter
coverage areas. Each layer shows the CQI values available in the transmitter coverage area.
5.8.5.1.2
Max FER/Max BER/Max MOS

On each bin, Atoll chooses the maximum CQI value available from the TRXs of different transmitters covering that bin.
Only the bins where the CQI values exceeds a defined threshold are coloured. The bin colour depends on the assigned
CQI value. Coverage consists of several independent layers whose visibility in the map window can be managed. There
are as many layers as the number of thresholds defined. Each layer shows the areas where a given CQI value is available.
Forsk 2009
AT271_TRG_E6
179
180
AT271_TRG_E6
Forsk 2009
Chapter 6
UMTS HSPA Networks
This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations,
simulations and prediction studies available in UMTS HSPA projects.
Atoll
182
AT271_TRG_E6
Forsk 2009
Chapter 6: UMTS HSPA Networks
UMTS HSPA Networks
6.1
6.1.1
Three criteria can be studied in point analysis (Profile tab) and in common coverage studies. Study criteria are detailed in
the table below:
Study criteria
Formulas
Signal level ( P rec ) in dBm
Signal level received from a transmitter on a carrier (cell)

P rec ic = EIRP ic L path M Shadowing model L Indoor + G term L term
Path loss ( L path ) in dBm

Total losses ( L total ) in dBm
Tx
L total = L path + L Tx + L term + L indoor + M Shadowing model G Tx + G term
where,
ic is a carrier number,
L ant
Tx
L Indoor are the indoor losses. These losses are defined for each clutter class.They are taken into account when the option
L term are the receiver losses,
G term is the receiver antenna gain,
G Tx is the transmitter antenna gain,
L Tx is the transmitter loss ( L Tx = L total DL ). For information on calculating transmitter loss, see "UMTS HSPA,
CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, and TD-SCDMA Documents" on page 124.
Notes:
EIRP ic = P pilot ic + G Tx L Tx ( P pilot ic is the cell pilot power).
It is possible to analyse all the carriers. In this case, Atoll takes the highest pilot power of
cells to calculate the signal level received from a transmitter.
Atoll considers that G term and L term equal zero.
6.1.2
Point Analysis
6.1.2.1
Profile Tab
Atoll displays either the signal level received from the selected transmitter on a carrier ( P rec ic ), or the highest signal
level received from the selected transmitter on all the carriers.
Note:
losses, L total . Path loss and total losses are the same on any carrier.
6.1.2.2
Reception Tab
Forsk 2009
AT271_TRG_E6
183

For each transmitter, Atoll displays either the signal level received on a carrier, ( P rec ic ), or the highest signal level
received on all the carriers.
Note:
6.1.3
Coverage Studies
For each TBC transmitter, Txi, Atoll determines the selected criterion on each pixel inside the Txi calculation area. In fact,
each pixel within the Txi calculation area is considered as a potential (fixed or mobile) receiver.
6.1.3.1

We can distinguish three cases:
6.1.3.1.1
All Servers
Txi
Txi
Txi
6.1.3.1.2

Txi
Txi
Txi

And
Txi
Txj
ji

Notes:
the highest.
6.1.3.1.3

Txi
Txi
Txi

And
Txi
P rec ic 2
nd
Best P Txj ic M
rec
ji

184
AT271_TRG_E6
Forsk 2009

Notes:
the second highest.
6.1.3.2
Coverage Display
6.1.3.2.1
Plot Resolution
6.1.3.2.2
Display Types
such as:

Atoll calculates signal level received from the transmitter on each pixel of each transmitter service area. A pixel of a service
area is coloured if the signal level exceeds ( ) the defined minimum thresholds (pixel colour depends on signal level).

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where other service
areas overlap the studied one, Atoll chooses the highest value. A pixel of a service area is coloured if the signal level
exceeds ( ) the defined thresholds (the pixel colour depends on the signal level). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer
corresponds to an area where the signal level from the best server exceeds a defined minimum threshold.
Path Loss (dB)

Atoll calculates path loss from the transmitter on each pixel of each transmitter service area. A pixel of a service area is
coloured if path loss exceeds ( ) the defined minimum thresholds (pixel colour depends on path loss). Coverage consists
Total Losses (dB)

Atoll calculates total losses from the transmitter on each pixel of each transmitter service area. A pixel of a service area is
coloured if total losses exceed ( ) the defined minimum thresholds (pixel colour depends on total losses). Coverage

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where other service
pixel of a service area is coloured if the path loss exceeds ( ) the defined thresholds (pixel colour depends on path loss).

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where service
areas overlap the studied one, Atoll determines the best transmitter and evaluates total losses from the best transmitter.
A pixel of a service area is coloured if the total losses exceed ( ) the defined thresholds (pixel colour depends on total
losses). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are
as many layers as defined thresholds. Each layer corresponds to an area where the total losses from the best server
exceed a defined minimum threshold.
Forsk 2009
AT271_TRG_E6
185
Number of Servers
Atoll evaluates how many service areas cover a pixel in order to determine the number of servers. The pixel colour
depends on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can
be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the number of
servers exceeds ( ) a defined minimum threshold.

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the signal level from this
transmitter fulfils signal conditions defined in Conditions tab with different Cell edge coverage probabilities. There is one
coverage area per transmitter in the explorer.

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the best signal level received
explorer.
6.2
Definitions and Formulas

Input parameters and formulas used in simulations and predictions (coverage predictions and point analysis) are detailed
in the tables below.
6.2.1
Inputs
This table lists simulation and prediction inputs (calculation options, quality targets, active set management conditions,
etc.).
Name
Value
Unit
Description
F ortho
Clutter parameter or global parameter (default value)
None
Orthogonality factor
Tx
Site equipment parameter
None
MUD factor
F MUD
Terminal parameter - HSDPA properties
None
MUD factor
ic
Frequency band parameter
None
Carrier number
AS_Th Txi ic
Cell parameter
None
Threshold for macro diversity

specified for a transmitter on a given
carrier ic
E
-----c-
Mobility parameter
I 0 threshold
None
Ec/Io target on downlink for the best

server
Global parameter
None
Pilot RSCP threshold for compressed

mode activation
Global parameter
None
Ec/Io threshold for compressed mode

activation
E
-----b-
(Bearer, Mobility) parameter
N t req
None
Eb/Nt target on downlink
Global parameter
None
Downlink Eb/Nt target increase due to

compressed mode activation
E
-----b-
(Bearer, Mobility) parameter
N t req
None
Eb/Nt target on uplink
Global parameter
None
Uplink Eb/Nt target increase due to

compressed mode activation
F MUD
Term
req
Q pilot
CM activation
RSCP pilot
CM activation
Q pilot
DL
DL
Q req
DL
Q req
UL
UL
Q req
UL
Q req
CE UL
NI
Site parameter
None
Number of channel elements available

for a site on uplink
CE DL
NI
Site parameter
None

for a site on downlink
CE UL
NI
Simulation result
None
Number of channel elements of a site

consumed by users on uplink
CE DL
NI
Simulation result
None

consumed by users on downlink
N max
N max
186
AT271_TRG_E6
Forsk 2009
Overhead C E UL
None
Number of channel elements used by

the cell for common channels on
uplink
Overhead C E DL
None

downlink
TCH C E UL
(Bearer, site equipment) parameter
None
Number of channel elements used for

traffic channels on uplink
TCH C E DL
(Bearer, site equipment) parameter
None

traffic channels on downlink
Simulation constraint
None
Maximum number of 512 bit-length

OVSF codes available per cell (512)
Simulation result
None
Number of 512 bit-length OVSF codes

used by the cell
None
Number of 256 bit-length OVSF codes

used by the cell for common channels
Codes
N max Txi ic
N
N
Codes
Txi ic
Overhead C odes
Site equipment parameter - Corresponds to N
Overhead C E DL
Codes HS PDSCH
Txi ic
Cell parameter (for HSDPA only)
None
Maximum number of 16 bit-length

OVSF codes available per cell for HSPDSCH
Codes HS PDSCH
Txi ic
None
Minimum number of 16 bit-length

OVSF codes available per cell for HSPDSCH
NF term
Terminal parameter
None
Terminal Noise Figure
NF Tx
Transmitter parameter (user-defined or calculated from transmitter

equipment characteristics)
None
Transmitter Noise Figure
1.38 10-23
J/K
Boltzman constant
N max
N min
293
Ambient temperature
3.84 MHz
Hz
Spreading Bandwidth
RF ic ic adj
Network parameter
If not defined, it is assumed that there is no inter-carrier interference
None
Interference reduction factor between

two adjacent carriers ic and ic adj
Simulation constraint (global parameter or cell parameter)
Maximum uplink load factor
Maximum percentage of used power
Tx
NF Tx K T W
Thermal noise at transmitter
Term
NF Term K T W
Thermal noise at terminal
bps
Chip rate
UL
X max
DL
%Power max
N0
N0
Rc
W 10
UL
Equipment parameter
None
Uplink rake receiver efficiency factor
DL
Terminal parameter
None
Downlink rake receiver efficiency

factor
R nominal
Bearer parameter
kbps
Bearer downlink nominal bit rate
F spreading Active user
Bearer parameter
None
Downlink spreading factor for active

users
DL
Bearer parameter
None
Downlink spreading factor for inactive

users
f rake efficiency
f rake efficiency
DL
DL
F spreading Inactive user
DL
Bearer parameter
None
ratio between DPCCH and DPCH

transmission duration on downlink
DPCCH and DPCH respectively refer
to the Dedicated Physical Control
Channel and Dedicated Physical
Channel
DL
Cell parameter
kbps
Maximum connection rate per user on

downlink
R nominal
Bearer parameter
kbps
Bearer uplink nominal bit rate
UL
Service parameter
kbps
Uplink activity factor on E-DPCCH

channels
rc
R max
UL
f act ADPCH
Forsk 2009
AT271_TRG_E6
187

DL
f act ADPCH
Service parameter
kbps
Downlink Activity factor on A-DPCH

channel
UL
Bearer parameter
None
ratio between the DPCCH and DPCH

powers transmitted on uplink
DPCCH and DPCH respectively refer
to the Dedicated Physical Control
Channel and Dedicated Physical
Channel
UL
Cell parameter
kbps
Maximum connection rate per user on

uplink
W --------------------DL
R nominal
None
Service downlink processing gain
W
---------------------UL
R nominal
None
Service uplink processing gain
DL
HSDPA study result
kbps
User application throughput on

downlink
DL
HSDPA study result - HSDPA Bearer parameter
kbps
RLC peak rate supported by the

HSDPA bearer
DL
HSDPA study result
kbps
RLC peak throughput supported by

the HSDPA bearer
T RLC Av
HSDPA study result
kbps
Average RLC throughput supported

by the HSDPA bearer
R MAC
DL
HSDPA study result
kbps
MAC rate supported by the HSDPA

bearer
DL
HSDPA study result
kbps
MAC throughput supported by the

HSDPA bearer
T application
UL
HSUPA study result
kbps
User application throughput on uplink
UL
HSUPA study result - HSUPA Bearer parameter
kbps
RLC peak rate supported by the

HSUPA bearer
T RLC Min
UL
HSUPA study result
kbps
Minimum RLC throughput supported

by the HSUPA bearer
R MAC
UL
HSUPA study result
kbps
MAC rate supported by the HSUPA

bearer
Service parameter (for HSDPA only)
kbps
Throughput offset
SF Rate
Service parameter (for HSDPA only)
Scaling factor
P max Txi
Transmitter parameter
Maximum shared power

Available only if the inter-carrier power
sharing option is activated
P SCH Txi ic
Cell parameter
Cell synchronisation channel power
P OtherCCH Txi ic
Cell parameter
P pilot Txi ic
Cell parameter
Cell pilot power
P HSDPA Txi ic
Cell parameter (user-defined or simulation result) (for HSDPA only)

P HS PDSCH Txi ic + n HS SCCH P HS SCCH Txi ic
Available cell HSDPA power

HSDPA: High Speed Downlink Packet
Access
P HS PDSCH Txi ic
Simulation result (for HSDPA only)
Cell HS-PDSCH power

HS-PDSCH: High Speed Physical
Downlink Shared Channel
P HS SCCH Txi ic
Cell HS-SCCH power

HS-SCCH: High Speed Shared
Control Channel
n HS SCCH
Cell parameter (user-defined or simulation result) (for HSDPA only)
P Headroom Txi ic
Cell headroom power
P max Txi ic
Cell parameter
Maximum Cell power
P tch Txi ic
Simulation result
R99 traffic channel power transmitted

on carrier ic
rc
R max
DL
Gp
UL
Gp
T application
R RLC peak
T RLC peak
DL
T MAC
R RLC peak
188
AT271_TRG_E6
Cell other common channels (except

CPICH and SCH) powera
number of HS-SCCH channels

managed by the cell
Forsk 2009

min
Bearer parameter
Minimum power allowed on traffic data

channel
P tch
max
Bearer parameter
Maximum power allowed on traffic

data channel
P HSUPA Txi ic
Cell parameter
Cell HSUPA power

HSUPA: High Speed Uplink Packet
Access
Transmitter R99 transmitted power on

carrier ic
P tch
Cell parameter or simulation result

P pilot Txi ic + P SCH Txi ic + P OtherCCH Txi ic +
P tx R99 Txi ic
P tch Txi ic +
tch(ic) used for

R99 users
DL
P tch Txi ic f act ADPCH
tch(ic) used for

HSUPA users
P tx Txi ic
Simulation result
P tx R99 Txi ic + P HSDPA Txi ic + P HSUPA Txi ic
Transmitter total transmitted power on

carrier ic
P term R99
Simulation result
Terminal power transmitted to obtain

the R99 radio bearer
P term HSUPA
Simulation result
Terminal power transmitted to obtain

the HSUPA radio bearer
P term
P term R99 f act ADPCH + P term HSUPA for HSPA users
Total power transmitted by the

terminal
Simulation result
UL
P term R99 for R99 users

P term
min
Terminal parameter
Minimum terminal power allowed
P term
max
Terminal parameter
Maximum terminal power allowed
BTS
BTS parameter
Percentage of BTS signal correctly

transmitted
term
Terminal parameter
Percentage of terminal signal correctly

transmitted
Clutter parameter
Percentage of pilot finger - percentage

of signal received by the terminal pilot
finger
G Tx
Antenna parameter
None
Transmitter antenna gain
G Term
Terminal parameter
None
Terminal gain
G Div
DL
Bearer parameter - Depends on the transmitter Tx diversity
None
Gain due to transmit diversity
G Div
UL
Bearer parameter - Depends on the transmitter Rx diversity
None
Gain due to receive diversity
L Tx

None
Transmitter lossb
L body
Service parameter
None
Body loss
L Term
Terminal parameter
None
Terminal loss
L indoor
Clutter parameter
L path
Propagation model result
None
Path loss
M Shadowing model
Result calculated from cell edge coverage probability and model

standard deviation
None
Model Shadowing margin

Only used in prediction studies
M Shadowing Ec Io
Result calculated from cell edge coverage probability and Ec/Io

standard deviation
None
Ec/Io Shadowing margin

DL
None
n=2 or 3
M Shadowing Eb Nt
DL
M Shadowing Eb Nt
UL
Forsk 2009
npaths
DL
G macro diversity
Indoor loss
DL gain due to availability of several

pilot signals at the mobile c.
Result calculated from cell edge coverage probability and DL Eb/Nt

standard deviation
None
DL Eb/Nt Shadowing margin

Result calculated from cell edge coverage probability and UL Eb/Nt

standard deviation
None
UL Eb/Nt Shadowing margin

AT271_TRG_E6
189

UL
npaths
UL
G macro diversity
UL
M Shadowing Eb Nt
n=2 or 3
Global parameter (default value)
UL
None
UL quality gain due to signal diversity

in soft handoffd.
None
Random shadowing error drawn

during Monte-Carlo simulation
Only used in simulations
None
Transmitter-terminal total loss
P pilot Txi ic
----------------------------------LT
Chip power received at terminal
DL
P tch Txi ic
-------------------------------LT
Bit power received at terminal on

carrier ic
DL
P tx Txi ic
-----------------------------LT
Total power received at terminal from

a transmitter on carrier ic

traffic channels of a transmitter on
carrier ic
P term
-------------LT
Bit power received at transmitter on

carrier ic used by terminal
P term R99
--------------------------LT

carrier ic used by terminal

DPDCH from a terminal on carrier ic
E Shadowing
Simulation result
In prediction studiese
For Ec/Io calculation
L path L Tx L term L body L indoor M Shadowing Ec Io
--------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
For DL Eb/Nt calculation
L path L Tx L term L body L indoor M Shadowing Eb Nt
DL
-----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
LT
For UL Eb/Nt calculation

UL
-----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
In simulations
L path L Tx L term L body L indoor E Shadowing
-------------------------------------------------------------------------------------------------------------------------------G Tx G term
P c Txi ic
P b Txi ic
P tot Txi ic
DL
P traf Txi ic
tch ic
UL
P b ic
UL
P b R99 ic
UL
P tch Txi ic
-------------------------------LT
UL
P b DPDCH ic
UL
P b R99 ic 1 r c
a.
For the calculation of interference, P OtherCCH Txi ic also includes the MBMS SCCPCH channel power
when the optional MBMS feature is activated. You must modify the data structure for activating the optional MBMS feature.
For more information, see the Administrator Manual.
b.
L Tx = L total UL on uplink and L Tx = L total DL on downlink. For information on calculating transmitter
losses on uplink and downlink, see "UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, and TD-SCDMA
Documents" on page 124.
c.
npaths
M Shadowing Ec Io corresponds to the shadowing margin evaluated from the shadowing error probability
density function (n paths) in case of downlink Ec/Io modelling.

d.
npaths
M Shadowing Eb Nt
UL
corresponds to the shadowing margin evaluated from the shadowing error probability
density function (n paths) in case of uplink soft handoff modelling.

e.
In uplink prediction studies, only carrier power level is downgraded by the shadowing margin
( M Shadowing Eb Nt ). In downlink prediction studies, carrier power level and intra-cell interference are downgraded by
UL
the shadowing model ( M Shadowing Eb Nt

M Shadowing Eb Nt
6.2.2
DL
DL
or M Shadowing Ec Io ) while extra-cell interference level is not. Therefore,
or M Shadowing Ec Io is set to 1 in downlink extra-cell interference calculation.
Ec/Io Calculation
This table details the pilot quality ( Q pilot or Ec Io ) calculations.
Name
190
Value
AT271_TRG_E6
Unit
Description
Forsk 2009
DL
I intra txi ic
P SCH txi ic
DL
DL
P tot txi ic BTS P tot txi ic --------------------------------
L
Downlink intra-cell interference at

terminal on carrier ic
Downlink extra-cell interference at

Downlink inter-carrier interference at

Total received noise at terminal on

carrier ic
None
Quality level at terminal on pilot for

carrier ic
DL
I extra ic
DL
P tot txj ic
txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
DL
Without Pilot:
DL
DL
Term
I intra txi ic + I extra ic + I inter carrier ic + N 0

1 BTS P c txi ic
DL
I 0 ic
DL
DL
Term
Total noise: P DL txi ic + I extra ic + I inter carrier ic + N 0

tot
BTS P c txi ic
------------------------------------------------------DL
I 0 ic
Ec
Q pilot txi ic ------
I0
6.2.3
DL Eb/Nt Calculation
Eb
DL
This table details calculations of downlink traffic channel quality ( Q tch or ------- ). When the optional MBMS feature is
Nt DL
activated, the MBMS Eb/Nt is also calculated in the same manner. You must modify the data structure for activating the
optional MBMS feature. For more information, see the Administrator Manual.
Name
DL
I intra txi ic
Value
Unit
Description
P SCH txi ic
DL
DL
P tot txi ic BTS F ortho P tot txi ic --------------------------------
L




carrier ic
DL
I extra ic
DL
P tot txj ic
txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
DL
N tot ic
DL
DL
DL
Term
I intra ic + I extra ic + I inter carrier ic + N 0

Without useful signal:
DL
Eb
DL
Q tch txi ic ------
N t DL
BTS P b txi ic
DL
--------------------------------------------------------------------------------------------------------------- G DL
Div G p
DL
DL
N tot ic 1 F ortho BTS P b txi ic
None
DL
BTS P b txi ic
DL
- G DL
Total Noise: -----------------------------------------------Div G p
DL
N tot ic
DL
ic
DL
G SHO
req
P tch txi ic
DL
f rake efficiency
DL
Q tch tx k
ic
Quality level at terminal on a traffic

channel from one transmitter on
carrier ic a
None
Quality level at terminal using carrier

ic due to combination of all
transmitters of the active set (Macrodiversity conditions).
None
Soft handover gain on downlink
Required transmitter traffic channel

power to achieve Eb/Nt target at
txk ActiveSet
DL
Q ic
--------------------------------------------------------DL
Q tch BestServer ic
DL
Q req
-------------------- P tch txi ic
DL
Q ic
a.
Calculation option may be selected in the Global parameters tab. The chosen option will be taken into account
only in simulations. In point analysis and coverage studies, Atoll uses the option Total noise to evaluate DL and UL Eb/Nt.
Forsk 2009
AT271_TRG_E6
191
6.2.4
UL Eb/Nt Calculation
Eb
UL
This table details calculations of uplink traffic channel quality ( Q tch or ------- ).
Nt UL
Name
UL intra
I tot
txi
UL extra
I tot
Value
Pb
UL
ic
term
txi
txi ic
term
txj j i
Pb
ic
Description
Total power received at transmitter

from intra-cell terminals using carrier
ic

from extra-cell terminals using carrier
ic
Uplink inter-carrier interference at

Total received interference at

transmitter on carrier ic
Total noise at transmitter on carrier ic

(Uplink interference)
UL
P b ic
UL
UL
I inter carrier txi
ic
Unit
ic adj
term
txj j
--------------------------------------
RF ic ic adj
UL
I tot txi ic
UL extra
I tot
UL intra
Tx
txi ic + 1 F MUD term I tot
UL
UL
N tot txi ic
UL
txi ic + I inter carrier txi ic
tx
I tot txi ic + N 0

UL
Eb
UL
Q tch txi ic ------
N t UL
term P b DPDCH ic
UL
--------------------------------------------------------------------------------------------------------------- G UL
Div G p
UL
Tx
UL
N tot txi ic 1 F MUD term P b ic
None
UL
term P b DPDCH ic
UL
- G UL
Total noise: --------------------------------------------------------Div G p
UL
N tot txi ic
Quality level at transmitter on a traffic

channel for carrier ic a
UL
No HO: Q tch txi ic
UL
Softer HO: f rake efficiency
UL
Q tch tx k ic
tx k ActiveSet
samesite
Soft, softer/soft HO (No MRC):

UL
UL
UL
Max Q tch tx k ic G macro diversity

tx ActiveSet
k
ic
Softer/soft HO (MRC):
Quality level at site using carrier ic due

to combination of all transmitters of
the active set located at the same site
and taking into account increasing of
None
the quality due to macro-diversity
(macro-diversity gain).
Max UL
UL
UL
f
tx
ic
tx
ic
rake efficiency
tch
k
tch
l
tx ,tx ActiveSet
k l
txk samesite
tx k
UL
In simulations G macro diversity = 1 .
tx l othersite
UL
G macro diversity
UL
G SHO
req
P term ic
UL
Q ic
---------------------------------------------------------UL
Q tch BestServer ic
None
Soft handover gain on uplink
Required terminal power to achieve

Eb/Nt target at transmitter on carrier
ic
UL
Q req
-------------------- P term
UL
Q ic
a.
6.3
Active Set Management

The mobiles active set (AS) is the list of the transmitters to which the mobile is connected. The active set may consist of
one or more transmitters; depending on whether the service supports soft handover and on the terminal active set size.
The terminal frequency bands are taken into account and transmitters in the mobiles active set must use a frequency band
supported by the terminal. Finally, the quality of the pilot (EcIo) is what determines whether or not a transmitter can belong
192
AT271_TRG_E6
Forsk 2009

to the active set. The active set management is detailed hereafter. Cells entering a mobiles active set must satisfy the
following conditions:
The best server (first cell entering active set)

The pilot quality from the best serving cell must exceed the Ec/Io threshold. Best server cell is the one with the
highest pilot quality.
Other cells in the active set

-
Must use the same carrier as the best server,

The pilot quality difference between other candidate cells and the best server must be less than the AS
threshold specified for the best server,
Other candidate cells must belong to the neighbour list of the best server if it is located on a site where the
equipment imposes this restriction (the restricted to neighbours option selected in the equipment properties).
6.4
Simulations
The simulation process consists of two steps:
1. Obtaining a realistic user distribution
Atoll generates a user distribution using a Monte-Carlo algorithm, which requires traffic maps and data as input. The resulting user distribution complies with the traffic database and maps provided to the algorithm.
Each user is assigned a service, a mobility type, and an activity status by random trial, according to a probability law that
uses the traffic database.
The user activity status is an important output of the random trial and has direct consequences on the next step of the
simulation and on the network interferences. A user may be either active or inactive. Both active and inactive users
consume radio resources and create interference.
Then, Atoll randomly assigns a shadowing error to each user using the probability distribution that describes the shadowing effect.
Finally, another random trial determines user positions in their respective traffic zone and whether they are indoors or
outdoors (according to the clutter weighting and the indoor ratio per clutter class defined for the traffic maps).
2. Power control simulation
6.4.1
Generating a Realistic User Distribution

During the simulation, a first random trial is performed to determine the number of users and their activity status. Four activity status are modelled:
Active UL: the user is active on UL and inactive on DL
Active DL: the user is active on DL and inactive on UL
Active UL+DL: the user is active on UL and on DL
Inactive: the user is inactive on UL and on DL
The determination of the number of users and the activity status allocation depend on the type of traffic cartography used.
Note:
Atoll follows a Poisson distribution to determine the total number of users attempting a
connection in each simulation. In order for Atoll to use a constant total number of users
attempting a connection, the following lines must be added to the Atoll.ini file:
[CDMA]
RandomTotalUsers=0
6.4.1.1
Simulations Based on Raster Traffic and Vector Traffic Maps

Raster traffic maps are based on environments. Each pixel of the map is assigned an environment class which contains
a list of user profiles with an associated mobility type and a given density (number of subscribers with the same profile per
km).
Vector traffic maps are directly based on user profiles. Each polygon and line of the map is assigned a density of subscribers with given user profile and mobility type. If the map is composed of points, each point is assigned a number of subscribers with given user profile and mobility type.
The user profile models the behaviour of the different subscriber categories. Each user profile contains a list of services
and their associated parameters describing how these services are accessed by the subscriber.
From environment (or polygon) surface (S) and user profile density (D), a number of subscribers (X) per user profile is
inferred.
X = SD
Notes:
Forsk 2009
AT271_TRG_E6
193
In case of vector traffic map composed of lines, the number of subscribers (X) per user
profile is calculated from the line length (L) and the user profile density (D) (nb of
subscribers per km) as follows: X = L D
The number of subscribers (X) is an input when the vector traffic map is composed of
points.
For each behaviour described in a user profile, according to the service, frequency use and exchange volume, Atoll calculates the probability for the user being active in uplink and in downlink at an instant t.
6.4.1.1.1
Circuit Switched Service (i)

User profile parameters for circuit switched services are:
The used terminal (equipment used for the service (from the Terminals table)),
The average number of calls per hour N call ,
The average duration of a call (seconds) d .
The number of users and their distribution per activity status is determined as follows:
1.
Calculation of the service usage duration per hour ( p 0 : probability of a connection):
N call d
p o = --------------------3600
2.
Calculation of the number of users trying to access the service i ( n i ):
ni = X p0
Next, we can take into account activity periods during the connection in order to determine the activity status of each user.
3.
Calculation of activity probabilities:

UL
DL
Probability of being inactive on UL and DL: p inactive = 1 f act 1 f act

UL
DL
DL
UL
Probability of being active on UL only: p UL = f act 1 f act

Probability of being active on DL only: p DL = f act 1 f act
UL
DL
Probability of being active both on UL and DL: p UL + DL = f act f act

UL
DL
Where, f act and f act are respectively the UL and DL activity factors defined for the circuit switched service i.
4.
Calculation of number of users per activity status:

inactive
Number of inactive users on UL and DL: n i
= n i p inactive
Number of users active on UL and inactive on DL: n i UL = n i p UL

Number of users active on DL and inactive on UL: n i DL = n i p DL
Number of users active on UL and DL both: n i UL + DL = n i p UL + DL
Therefore, a user when he is connected can have four different activity status: either active on both links, or inactive on
both links, or active on UL only, or active on DL only.
6.4.1.1.2
Packet Switched Service (j)

User profile parameters for packet switched services are:
The used terminal (equipment used for the service (from the Terminals table)),
The average number of packet sessions per hour N sess ,
The volume (in kbytes) which is transferred on the downlink V DL and the uplink V UL during a session.
A packet session consists of several packet calls separated by a reading time. Each packet call is defined by its size and
may be divided in packets of fixed size (1500 Bytes) separated by an inter arrival time.
In Atoll, a packet session is described by following parameters:
UL
N packet c all : Average number of packet calls on the uplink during a session,
DL
N packet c all : Average number of packet calls on the downlink during a session,
UL
T packet call : Average time (millisecond) between two packets calls on the uplink ,
DL
T packet call : Average time (millisecond) between two packets calls on the downlink ,
194
AT271_TRG_E6
Forsk 2009

UL
T packet : Average time (millisecond) between two packets on the uplink ,

DL
T packet : Average time (millisecond) between two packets on the downlink ,

UL
S packet : Packet size (Bytes) on uplink,

DL
S packet : Packet size (Bytes) on downlink.
Figure 6.1: Description of a Packet Session

The number of users and their distribution per activity status is determined as follows:
1.
Calculation of the average packet call size (kBytes):
V UL
V DL
UL
DL
S packet c all = ------------------------------------------- and S packet c all = ------------------------------------------UL
UL
DL
DL
N packet c all f eff
N packet c all f eff
UL
DL
Where f eff and f eff are the UL and DL efficiency factors defined for the packet switched service j.
Note:
UL
2.
DL
In case of HSDPA and HSUPA services, f eff and f eff are set to 1.
Calculation of the average number of packets per packet call:

UL
DL
S packet c all
S packet c all
UL
- + 1 and N DL
- + 1
N packet = int ----------------------------------packet = int ----------------------------------UL
S packet 1024
S DL
packet 1024
Note:
3.
1kBytes = 1024Bytes.
Calculation of the average duration of inactivity within a packet call (s):

UL
UL
DL
DL
N packet 1 T packet
N packet 1 T packet
UL
DL
D Inactivity packet call = -------------------------------------------------------------- and D Inactivity packet call = -------------------------------------------------------------1000
1000
4.
Calculation of the average duration of inactivity in a session (s):

UL
UL
UL
DL
DL
DL
D Inactivity session = N packet c all D Inactivity packet call and D Inactivity session = N packet c all D Inactivity packet call
5.
Calculation of the average duration of activity in a session (s):

UL
UL
DL
DL
N packet S packet 8
N packet S packet 8
UL
UL
DL
- and D DL
D Activity session = N packet c all ----------------------------------------------------Activity session = N packet c all ----------------------------------------------------UL
DL
R average 1000
R average 1000
UL
DL
Where R average and R average are the uplink and downlink average requested rates defined for the service j.
Therefore, the average duration of a connection (in s) is:
UL
UL
UL
DL
DL
DL
D Connection = D Activity session + D Inactivity session and D Connection = D Activity session + D Inactivity session
6.
Calculation of the service usage duration per hour (probability of a connection):
N sess
N sess
UL
UL
DL
DL
p Connection = -------------- D Connection and p Connection = -------------- D Connection
3600
3600
Forsk 2009
AT271_TRG_E6
195

7.
Calculation of the probability of being connected:

UL
DL
p Connected = 1 1 p Connection 1 p Connection

Therefore, the number of users who want to get the service j is:
n j = X p Connected
As you can see on the picture above, we have to consider three possible cases when a user is connected:
1st case: At a given time, packets are downloaded and uploaded.
In this case, the probability of being connected is:

UL
DL
p Connection p Connection
UL + DL
p Connected = ---------------------------------------------------------------p Connected
2nd case: At a given time, packet are uploaded (no packet is downloaded).
Here, the probability of being connected is:

UL
DL
p Connection 1 p Connection
UL
p Connected = -----------------------------------------------------------------------------p Connected
3rd case: At a given time, packet are downloaded (no packet is uploaded).
In this case, the probability of being connected is:

DL
UL
DL
p Connected = -----------------------------------------------------------------------------p Connected
Now, we have to take into account activity periods during the connection in order to determine the activity status of each
user.
8.
f
UL
Calculation of the probability of being active:

UL
DL
D Activity session
D Activity session
DL
= ----------------------------------------------------------------------------------------------------- and f
= ----------------------------------------------------------------------------------------------------UL
UL
DL
DL
D Inactivity session + D Activity session
D Inactivity session + D Activity session
Therefore, we have:
1st case: At a given time, packets are downloaded and uploaded.
The user can be active on UL and inactive on DL; this probability is:
1
p UL = f
UL
1 f
DL
UL + DL
p Connected
The user can be active on DL and inactive on UL; this probability is:
1
p DL = f
DL
1 f
UL
UL + DL
p Connected
The user can be active on both links; this probability is:

1
p UL + DL = f
UL
DL
UL + DL
p Connected
The user can be inactive on both links; this probability is:

1
p inactive = 1 f
UL
1 f
DL
UL + DL
p Connected
2nd case: At a given time, packet are uploaded (no packet is downloaded).
The user can be active on UL and inactive on DL; this probability is:
2
p UL = f
UL
UL
p Connected

2
p inactive = 1 f
UL
UL
p Connected
3rd case: At a given time, packet are downloaded (no packet is uploaded).
The user can be active on DL and inactive on UL; this probability is:
3
p DL = f
DL
DL
p Connected

3
p inactive = 1 f
9.
DL
DL
p Connected
Calculation of number of users per activity status

inactive
Number of inactive users on UL and DL: n j
196
= n j p inactive + p inactive + p inactive
AT271_TRG_E6
Forsk 2009

1
Number of users active on UL and inactive on DL: n j UL = n j p UL + p UL

Number of users active on DL and inactive on UL: n j DL = n j p DL + p DL
1
Number of users active on UL and DL: n j UL + DL = n j p UL + DL

Note:
6.4.1.2
The user distribution per service and the activity status distribution between the users are
average distributions. And the service and the activity status of each user are randomly
drawn in each simulation. Therefore, if you compute several simulations at once, the
average number of users per service and average numbers of inactive, active on UL, active
on DL and active on UL and DL users, respectively, will correspond to calculated
distributions. But if you check each simulation, the user distribution between services as
well as the activity status distribution between users is different in each of them.
Simulations Based on Traffic Map per Service and per Transmitter

Traffic map per service and per transmitter is based on live traffic data from OMC (Operation and Maintenance Centre).
Live traffic is spread over the best pilot coverage area of each transmitter and each coverage area is assigned either the
total throughput demand or the number of active users or Erlangs.
For each transmitter Txi and each service:
If you are creating a Map based on Transmitters and Services (Throughputs), enter the throughput demands
in the uplink and downlink for each sector and for each listed service. Atoll calculates the number of active users
on UL and DL in the Txi cell using the service (NUL and NDL) as follows:
UL
DL
Rt
Rt
N UL = ---------------------and N DL = ---------------------UL
DL
R average
R average
UL
is the kbits per second transmitted on UL in the Txi cell to supply the service.
DL
is the kbits per second transmitted on DL in the Txi cell to supply the service.
Rt
Rt
DL
R average is the downlink average requested rate defined for the service,
UL
. R average is the uplink average requested rate defined for the service.
If you are creating a Map based on Transmitters and Services (# Active Users), enter the active users in the
uplink and downlink for each sector and for each listed service (NUL and NDL).
For each transmitter Txi and each circuit-switched service:
6.4.1.2.1
If you are creating a Map based on Transmitters and Services (Erlangs), enter Erlangs, i.e., the total number
of users, active and inactive, for each sector and for each circuit-switched service (n).

There can be two cases for circuit-switched service:
1. Number of active users on uplink and downlink (NUL and NDL),
2. Erlangs, i.e, a total number of users (n).
Active Users on Uplink and Downlink

NUL and NDL values include:
Users active on uplink and inactive on downlink (ni(UL)),

Users active on downlink and inactive on uplink (ni(DL)),
And users active on both links (ni(UL+DL)).
They are calculated as follows:

Calculation of activity probabilities
UL
DL
DL
UL

UL
DL
Forsk 2009
AT271_TRG_E6
197

UL
DL
Probability of being inactive both on UL and DL: p inactive = 1 f act 1 f act

UL
DL
active
Calculation of the total number of active users, n i
, attempting to access the circuit switched service i
We have:
active
= N UL
active
= N DL
p UL + p UL + DL n i
p DL + p UL + DL n i
Therefore, we have:
N UL p UL + DL N DL p UL + DL
n i UL + DL = min ------------------------------------- -------------------------------------
P UL + p UL + DL P DL + p UL + DL
n i UL = N UL n i UL + DL
n i DL = N DL n i UL + DL
And
active
ni
= n j UL + n j DL + n j UL + DL
inactive
Calculation of the number of inactive users, n i
, attempting to access the circuit switched service i
The number of inactive users on uplink and downlink is calculated from the total number of active users as follows:
active
ni
= ------------------------------ p inactive
1 p inactive
inactive
ni
Erlangs
n i is the total number of users trying to access the service i. This figure includes both active and inactive users. They are
determined as follows:
UL
DL
Probability of being inactive on UL and DL: p inactive = 1 f act 1 f act

UL
DL
DL
UL

UL
DL

UL
DL
inactive
Number of inactive users on UL and DL: n i
= n i p inactive
Number of users active on UL only: n i UL = n i p UL

Number of users active on DL only: n i DL = n i p DL
Number of users active on UL and DL both: n i UL + DL = n i p UL + DL
6.4.1.2.2

Here, Atoll considers all the users as active. Activity probabilities are not calculated.
NUL and NDL values include:
Users active on uplink and inactive on downlink (ni(UL)),

Users active on downlink and inactive on uplink (ni(DL)),
And users active on both links (ni(UL+DL)).

If N UL N DL
198
AT271_TRG_E6
Forsk 2009

n j UL + DL = N UL
n j UL = 0
n j DL = N DL N UL
If N UL N DL
n j UL + DL = N DL
n j DL = 0
n j UL = N UL N DL
Therefore, we have:
n j = n j UL + n j DL + n j UL + DL
where n j is the total number of users in the Txi cell attempting to access the packet switched service j.
Note:
6.4.2
The activity status distribution between users is an average distribution. In fact, in each
simulation, the activity status of each user is randomly drawn. Therefore, if you compute
several simulations at once, average numbers of inactive, active on UL, active on DL and
active on UL and DL users correspond to the calculated distribution. But if you check each
simulation, the activity status distribution between users is different in each of them.
Power Control Simulation

The power control algorithm simulates the way a UMTS network regulates itself by using uplink and downlink power
controls in order to minimize interference and maximize capacity.
HSDPA users are linked to the A-DPCH radio bearer (an R99 radio bearer). Therefore, the network uses a A-DPCH power
control on UL and DL and then it performs fast link adaptation on DL in order to select an HSDPA radio bearer. For HSUPA
users, the network first uses a E-DPCCH/A-DPCH power control on UL and DL, checks that there is an HSDPA connection
on downlink and then carries out noise rise scheduling in order to select an HSUPA radio bearer on uplink. Atoll simulates
these network regulation mechanisms with an iterative algorithm and calculates, for each user distribution, network parameters such as cell power, mobile terminal power, active set and handoff status for each terminal. During each iteration of
the algorithm, all the users (R99, HSDPA, and HSUPA service users) selected during the user distribution generation (1st
step) attempt to connect one by one to network transmitters. The process is repeated until the network is balanced, i.e.,
until the convergence criteria (on UL and DL) are satisfied.
As shown in Figure 6.2: on page 200, the simulation algorithm is divided in three parts. All users (R99, HSDPA, and
HSUPA service users) are evaluated by the R99 part of the algorithm. HSDPA and HSUPA users active on DL, unless
they have been rejected during the R99 part of the algorithm, are then evaluated by the HSDPA part of the algorithm.
Finally, HSUPA users active on uplink, unless they have been rejected during the R99 or HSDPA parts of the algorithm,
are then evaluated by the HSUPA part of the algorithm.
Forsk 2009
AT271_TRG_E6
199

Initialisation
R99 part
Mi Best Server Determination
Mi Active Set Determination
For HSDPA users, this part of

the algorithm is performed for
the A-DPCH bearer (R99 bearer)
For HSUPA users, this part is
performed for the E-DPCCH/ADPCH bearer (R99 bearer)
UL Power Control
For each R99, HSDPA

and HSUPA mobile, Mi
DL Power Control
UL and DL Interference Update
Congestion and Radio Resource Control

HSDPA part
For each HSDPA and
HSUPA mobile, Mi
Fast Link Adaptation
Mobile Scheduling
Radio Resource Control

HSUPA part
Admission Control
For each HSUPA
mobile, Mi
Noise Rise Scheduling
Radio Resource Control
Convergence Study
Figure 6.2: UMTS HSPA Power Control Algorithm

The steps of this algorithm are detailed below.
6.4.2.1
Algorithm Initialization
The total power transmitted by the base station txi on the carrier ic m ,
P Tx txi ic m , is initialised to
P pilot txi ic m + P SCH txi ic m + P otherCCH txi ic m + P HSDPA txi ic m + P HSUPA txi ic . Uplink received powers by
UL intra
the base station txi on carrier ic m , I tot
UL extra
txi ic m , I tot
UL
txi ic m and I inter carrier txi ic m are initialised to 0 W (i.e.
no connected mobile).
UL
I tot txi ic m
UL
X k txi ic m = --------------------------------- = 0
UL
N tot txi ic m
6.4.2.2
R99 Part of the Algorithm

UL
The algorithm is detailed for any iteration k. Xk is the value of the X (variable) at the iteration k. In the algorithm, all Q req
DL
and Q req thresholds depend on user mobility and are defined in Service and Mobility parameters tables. All variables are
described in Definitions and formulas part.
Here, the rate downgrading is not taken into account.
For each mobile Mb
200
AT271_TRG_E6
Forsk 2009
Determination of Mbs Best Server (SBS(Mb))

The algorithm used in this step depends on the type of network you plan, a single frequency band network (with f1 as
frequency band) or a dual-band network (with f1 and f2 as frequency bands). For dual-band networks, you can activate
the dual-band dedicated carrier selection mode in the atoll.ini file and model different configurations of dual-band terminals.
To activate the dual-band dedicated carrier selection mode, add the following lines in the Atoll.ini file:
[CDMA]
MultiBandSimu=1
Dual-band terminals can have the following configurations:
Configuration 1: The terminal can work on f1 and f2 without any priority (select "All" as main frequency band in the
terminal property dialogue).
Configuration 2: The terminal can work on f1 and f2 but f1 has a higher priority (select "f1" as main frequency band
and "f2" as secondary frequency band in the terminal property dialogue).
1. Default mode (the dual-band dedicated carrier selection mode is not active)
This part describes the way the algorithm works when the dual-band dedicated carrier selection mode is not active. It
applies to single frequency band networks as well as dual-band networks.
For each station txi containing Mb in its calculation area and using the main frequency band supported by the Mbs terminal
(i.e. either f1 for a single frequency band network, or f1 or f2 for a dual-band terminal with the configuration 1, or f1 for a
dual-band terminal with the configuration 2).
Determination of BestCarrier k txi M b .
If a given carrier is specified for the service requested by Mb and if it is used by txi
BestCarrier k txi M b is the carrier specified for the service.
Else the carrier selection mode defined for txi is considered.
If carrier selection mode is UL min noise
For each carrier ic used by txi, we calculate current loading factor:
UL
I tot txi ic
UL
UL
- + X
X k txi ic = ----------------------------UL
N tot txi ic
EndFor
UL
BestCarrier k txi M b is the carrier with the lowest X k txi ic

Else if carrier selection mode is DL min power
BestCarrier k txi M b is the carrier with the lowest P tx txi ic k
Else if carrier selection mode is Random
BestCarrier k txi M b is randomly selected
Else if carrier selection mode is "Sequential"
UL
UL
BestCarrier k txi M b is the first carrier so that X k txi ic X max

Calculation of
Q pilot txi BestCarrier
k
BTS P c txi M b BestCarrier

= -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Term
DL
DL
DL
P tot txi BestCarrier k txi M b + I extra BestCarrier k txi M b + I inter carrier BestCarrier k txi M b + N 0
If user selects without Pilot
Q pilot txi M b BestCarrier
k

= -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------DL
DL
I intra BestCarrier k txi M b + I extra BestCarrier k txi M b +

DL
Term
I inter carrier BestCarrier k txi M b + N 0
1 BTS P c txi M b BestCarrier
Rejection of station txi if the pilot is not received

pilot
If Q pilot txi M b BestCarrier Q req Mobility M b then txi is rejected by Mb

k
Forsk 2009
AT271_TRG_E6
201

max
If Q pilot txi M b BestCarrier Q pilot M b

k
Admission control (If simulation respects a loading factor constraint and Mb was not connected in previous iteration).
UL
UL
If X k txi BestCarrier txi M b X max , then txi is rejected by Mb

Else
max
Q pilot M b = Q pilot txi M b BestCarrier

k
Tx BS M b = txi
Endif
EndFor
If no TxBS has been selected and Mbs terminal can work on one frequency band only, Mb has failed to be connected to
the network and is rejected.
If no TxBS has been selected and Mbs terminal can work on another frequency band.
Determination of BestCarrier k txi M b for each station txi containing Mb in its calculation area and using another
frequency band supported by the Mbs terminal (i.e. f1 or f2 for a dual-band terminal with the configuration 1, or f2 for a
dual-band terminal with the configuration 2)
If carrier selection mode for txi is UL min noise
UL
I tot txi ic
UL
UL
- + X
X k txi ic = ----------------------------UL
N tot txi ic
EndFor
UL

UL
UL

Calculation of
Q pilot txi BestCarrier
k

= -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Term
DL
DL
DL
P tot txi BestCarrier k txi M b + I extra BestCarrier k txi M b + I inter carrier BestCarrier k txi M b + N 0
Q pilot txi M b BestCarrier
k

= -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------DL
DL
I intra BestCarrier k txi M b + I extra BestCarrier k txi M b +

DL
Term
I inter carrier BestCarrier k txi M b + N 0
1 BTS P c txi M b BestCarrier

pilot
If Q pilot txi M b BestCarrier Q req Mobility M b then txi is rejected by Mb

k
max

k
Admission control (If simulation respects a loading factor constraint and Mb was not connected in previous iteration).
UL
UL
If X k txi BestCarrier txi M b X max , then txi is rejected by Mb

Else
max

k
202
AT271_TRG_E6
Forsk 2009

Tx BS M b = txi
Endif
EndFor
If no TxBS has been selected, Mb has failed to be connected to the network and is rejected.
Then we will consider BestCell k M b as the best serving cell and ic = BestCarrier k Tx BS M b M b .
2. Dual-band dedicated carrier selection mode
This part describes the way the algorithm works when the dual-band dedicated carrier selection mode is active. It applies
to dual-band networks when dual-band terminals have the first congiguration (i.e. they can work on two frequency bands
without any priority).
For each transmitter txi containing Mb in its calculation area and using one of the frequency bands supported by the Mbs
terminal (i.e. either f1 or f2)
BTS P c txi M b ic
Calculation of Q pilot txi ic Mb = --------------------------------------------------------------------------------------------------------------------------------Term
k
DL
DL
DL
P tot txi ic + I extra ic + I inter carrier ic + N 0
BTS P c txi M b ic
Q pilot txi ic Mb = ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Term
k
DL
DL
DL
1 BTS P c txi M b ic
Rejection of bad candidate cells when the pilot is not received and when the uplink load factor is exceeded during the
admission load control (If simulation respects a loading factor constraint and Mb was not connected in previous iteration)
pilot
If Q pilot txi M b ic Q req Mobility M b then (txi,ic) is rejected by Mb

k
UL
UL
If X k txi ic X max , then (txi,ic) is rejected by Mb

Else
Keep (txi,ic) as good candidate cell
If no good candidate cell has been selected, Mb has failed to be connected to the network and is rejected.
Compare the pilot quality received from good candidate cells
The site containing the cell (txi, ic) with the highest Q pilot txi M b ic is the best site, N BS M b .
k
For each cell (txi, ic) on the best site N BS M b and using one of the frequency bands supported by the Mbs terminal (i.e.
either f1 or f2)
Rejection of bad candidate cells when the pilot is not received and when the uplink load factor is exceeded during the
admission load control (If simulation respects a loading factor constraint and Mb was not connected in previous iteration)
pilot
If Q pilot txi M b ic Q req Mobility M b then (txi,ic) is rejected by Mb

k
UL
UL
If X k txi ic X max , then (txi,ic) is rejected by Mb

Determination of the best cell (txi, ic) on the best site N BS M b
If a given carrier is specified for the service requested by Mb
BestCell k Ni M b is the transmitter associated to the carrier specified for the service
Else the carrier selection mode defined for Ni is considered.
For each pair (txi,ic) on Ni, we calculate the current loading factor:
UL
I tot txi ic
UL
UL
- + X
X k txi ic = ----------------------------UL
N tot txi ic
EndFor
UL
BestCell k Ni M b is the cell with the lowest X k txi ic

BestCell k Ni M b is the cell with the lowest P tx txi ic k
BestCell k Ni M b is randomly selected
Forsk 2009
AT271_TRG_E6
203

Note:
The Sequential carrier selection mode is not implemented with the dual-band dedicated
carrier selection algorithm. It only works with the default algorithm.
Endif
max
BestCell k M b is the best serving cell and its pilot quality is Q pilot M b
k
In the following lines, we will consider ic as the carrier used by the best serving cell
Active Set Determination

For each station txi containing Mb in its calculation area, using ic , and, if neighbours are used, neighbour of
BestCell k M b
BTS P c txi M b ic
Calculation of Q pilot txi M b ic = --------------------------------------------------------------------------------------------------------------------------------DL
DL
DL
Term
k
P tot txi ic + I extra ic + I inter carrier ic + N 0
BTS P c txi M b ic
Q pilot txi M b ic = -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Term
k
DL
DL
DL
1 BTS P c txi M b ic
Rejection of txi from the active set if difference with the best server is too high
max
If Q pilot M b Q pilot txi M b ic AS_Th BestCell k M b then txi is rejected

k
Else txi is included in the Mb active set

Rejection of a station if the mobile active set is full
Station with the lowest Q pilot in the active set is rejected
k
EndFor
Uplink Power Control

R99 req
Calculation of the terminal power required by Mb to obtain the R99 radio bearer: P term
M b ic k
For each cell (txi,ic) of the Mb active set

Calculation of quality level on Mb traffic channel at (txi,ic), with the minimum power allowed on traffic channel for the Mb
service
req
P term R99 M b ic k 1
UL
P b R99 txi M b ic = -------------------------------------------------------L T txi M b
UL
UL
UL
UL
UL
P b DPDCH txi M b ic = P b R99 txi M b ic 1 r c

UL
P b DPCCH txi M b ic = P b R99 txi M b ic r c

UL
UL
UL
UL
UL
P b R99 txi M b ic = P b DPCCH txi M b ic + P b DPDCH txi M b ic if the user is active,

P b R99 txi M b ic = P b DPCCH txi M b ic if the user is inactive,
UL
term P b DPDCH txi M b ic k

UL
UL
- G UL
Q tch txi M b ic k = ------------------------------------------------------------------------------------------------------------------------------------------------------p Service M b G div
UL
Tx
UL
N tot txi ic 1 F MUD term P b R99 txi M b ic k 1
If user selects "Total noise",
UL
term P b DPDCH txi M b ic k

UL
UL
UL
Q tch txi M b ic k = -------------------------------------------------------------------------------- G p Service M b G div
UL
N tot txi ic
End For
If (Mb is in not in handoff)
UL
UL
Q k M b = Q tch txi M b ic k
Else if (Mi is in softer handoff)
204
AT271_TRG_E6
Forsk 2009

UL
UL
Q k M b = f rake efficiency
UL
Q tch txi M b ic k
txi ActiveSet
Else if (Mb is in soft, or softer/soft without MRC)

UL
UL
UL
Q k M b = Max Q tch txi M b ic k G macro diversity 2 links

txi ActiveSet
Else if (Mb is in soft/soft)

UL
UL
UL
Q k M b = Max Q tch txi M b ic k G macro diversity 3 links

txi ActiveSet
Else if (Mb is in softer/soft with MRC)
UL
UL
UL
UL
UL
Q k M b = Max f rake efficiency
Q tch ic Q tch ic G macro diversity 2 links
other site
txi ActiveSet
samesite
End If
UL
Q req Service M b Mobility M b

req
req
P term R99 M b ic k = ------------------------------------------------------------------------------------------- P term R99 M b ic k 1
UL
Qk Mb
If compressed mode is operated,
Note:
Compressed mode is operated if:

- Mi and Sj support compressed mode,
And
Resulting
- Either Q pilot
CM activation
txi M b ic Q pilot
CM activation
- Or P c txi M b ic RSCP pilot

UL
if the Ec/Io Active option is selected,
if the RSCP Active option is selected.

UL
Q req Service M b Mobility M b Q req Service M b Mobility M b

req
- P req
P term R99 M b ic k = -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------term R99 M b ic k 1
UL
Qk Mb
req
min
req
max
req
min
If P term R99 M b ic k P term M b then P term R99 M b ic k = P term txi M b

If P term R99 M b ic k P term M b then Mb cannot select any cell and its active set is cleared
UL
UL
If R nominal M b R max txi ic then Mb cannot be connected

Endif
Downlink Power Control

If (mobile does not use a packet switched service that is inactive on the downlink)
For each cell (txi,ic) in Mb active set
Calculation of quality level on (txi,ic) traffic channel at Mb with the minimum power allowed on traffic channel for the Mb
service
min
P tch Service M b
DL
P b txi M b ic = ---------------------------------------------------L T txi M b
DL
BTS P b txi M b ic k
DL
DL
- G DL
Q tch txi M b ic k = ----------------------------------------------------------------------------------------------------------------------------------p Service M b G div
DL
DL
N tot ic 1 F ortho BTS P b txi M b ic k 1
If the user selects the option "Total noise"
DL
BTS P b txi M b ic k
DL
DL
DL
Q tch txi M b ic k = ------------------------------------------------------------- G p Service M b G div
DL
N tot ic
End For
Forsk 2009
AT271_TRG_E6
205

DL
DL
DL
Q tch txi M b ic k
txi ActiveSet
Do
Calculation of the required power for DL traffic channel between (txi,ic) and Mb:
DL
Q req Service M b Mobility M b

req
min
P tch txi M b ic k = ------------------------------------------------------------------------------------------- P tch Service M b
DL
Qk Mb
If compressed mode is operated.
DL
DL
Q req Service M b Mobility M b Q req Service M b Mobility M b

req
- P min
P tch txi M b ic k = -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------tch Service M b
DL
Qk Mb
Note:
Compressed mode is operated if:

- Mi and Sj support compressed mode,
And
Resulting
- Either Q pilot
CM activation
txi M b ic Q pilot
CM activation
- Or P c txi M b ic RSCP pilot

req
max
if the Ec/Io Active option is selected,
if the RSCP Active option is selected.

max
If P tch txi M b ic k P tch Service M b then txi ic is set to P tch

DL
max
Recalculation of a decreased Q req (a part of the required quality is managed by the cells set to P tch )
req
P tch Service M b
DL
P b txi M b ic = --------------------------------------------------L T txi M b
DL
BTS P b txi M b ic
DL
DL
- G DL
Q tch txi M b ic k = -------------------------------------------------------------------------------------------------------------------------p Service M b G div
DL
DL
N tot ic 1 F ortho BTS P b txi M b ic
DL
DL
DL
If the user is inactive, then his contribution to interference in the calculation of N tot ic is P b txi M b ic r c .
EndFor
DL
DL
DL
Q tch txi M b ic k
txi ActiveSet
DL
DL
While Q k M b Q req Service M b Mobility M b and Mb active set is not empty

DL
DL
If R nominal M b R max txi ic then Mb cannot be connected

Endif
Uplink and Downlink Interference Update

Update of interference on active mobiles only (old contributions of mobiles and stations are replaced by the new ones).
For each cell (txi,ic)
UL
Update of N tot txi ic

EndFor
For each mobile Mi
DL
Update of N tot ic
EndFor
EndFor
Control of Radio Resource Limits (OVSF Codes, Cell Power, Channel Elements)
P tx txi ic k
DL
While ------------------------------ %Power max
P max
206
AT271_TRG_E6
Forsk 2009

Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
While N
Codes
Codes
txi ic k N max txi ic
EndFor
For each site (Node B) Ni
While N
CE DL
CE DL
N i k N max
Ni
While N
CE UL
CE UL
N i k N max
Ni
EndFor
Uplink Load Factor Control

For each cell (txi,ic) with X
UL
UL
txi ic X max
EndFor
While at least one cell with X
6.4.2.3
UL
UL
txi ic X max exists
HSDPA Part of the Algorithm

HSDPA and HSUPA users active on DL, unless they have been rejected during the R99 part of the algorithm, are then
evaluated by the HSDPA part of the algorithm.
6.4.2.3.1
HSDPA Power Allocation

The total transmitted power of the cell ( P tx ic ) is the sum of the R99 transmitted power, the HSUPA power and the
HSDPA power.
P tx ic = P tx R99 ic + P HSDPA ic + P HSUPA ic
In case of a static HSDPA power allocation strategy, Atoll checks in the simulation that:
DL
P tx ic P max ic %Power max

where:
DL
%Power max is the maximum DL load allowed.

Therefore, if the maximum DL load is set to 100%, we have:
P tx ic P max ic
In case of dynamic HSDPA power allocation strategy, Atoll checks in the simulation that:
DL
P tx R99 ic + P HSUPA ic P max ic %Power max

And it calculates the HSDPA power as follows:
P HSDPA ic = P max ic P Headroom ic P tx R99 ic P HSUPA ic
6.4.2.3.2
Number of HS-SCCH Channels and Maximum Number of HSDPA Users

The number of HS-SCCH channels ( n HS SCCH ) is the maximum number of HS-SCCH channels that the cell can manage.
Each HSDPA user consumes one HS-SCCH channel. Therefore, at a time (over a transmission time interval), the number
of connected HSDPA users cannot exceed the number of HS-SCCH channels per cell.
The maximum number of HSDPA users ( n max ) corresponds to the maximum number of HSDPA users that the cell can
support.
Let us assume there are 20 HSDPA users in the cell. All of them are active on DL and they are connected to the A-DCH
R99 bearer. Finally, the number of HS-SCCH channels and the maximum number of HSDPA users respectively equal 4
and 15.
In the HSDPA part, HSDPA users are ranked by the scheduler and treated as described in the figure below.
Forsk 2009
AT271_TRG_E6
207
Figure 6.3: Connection status of HSDPA users
6.4.2.3.3
The first four users may be simultaneously served if there are enough HSDPA power and OVSF codes available
in order for them to obtain a HSDPA bearer. In this case, they will be connected. Else, they will be delayed.
The next eleven ones will be delayed since there are no longer HS-SCCH channels available. Their connection
status will be "HSDPA Delayed".
Finally, the last five users will be rejected beacuse the maximum number of HSDPA user has been fixed to 15.
Their connection status will be "HSDPA Scheduler Saturation".
HSDPA Bearer Allocation Process

Let us focus on the first four users mentionned in the example of the previous paragraph, "Number of HS-SCCH Channels
and Maximum Number of HSDPA Users" on page 207. Once scheduled, a new fast link adaptation is carried out on these
users in order to determine if they can obtain a HSDPA bearer. They are processed in the order defined by the scheduler
and the cell HSDPA power is shared between them as explained below.
In the bearer allocation process shown below, the 4 HSDPA users are represented by Mj, with j = 1 to 4. And, the initial
values of their respective HSDPA powers is 0, i.e. PHSDPA(B(MX)) = 0, where X = 0 to 4. These power values are assigned
one by one by the scheduler, so that with their allocated values, looped back to the starting point, are used in successive
steps.
For the user, Mj, with j varying from 1 to 4:
Figure 6.4: Bearer Allocation Process for Scheduled Users
208
AT271_TRG_E6
Forsk 2009
6.4.2.3.4
Fast Link Adaptation Modelling

Fast link adaptation (or Adaptive Modulation and Coding) is used in HSDPA. The power on the HS-DSCH channel is transmitted at a constant power while the modulation, the coding and the number of codes are changed to adapt to the radio
conditions variations. Based on the reported channel quality indicator (CQI), the node-B may change every 2ms the modulation (QPSK and optionally 16QAM), the coding and the number of codes during a communication.
Atoll calculates for each user either the best pilot quality (CPICH Ec/Nt) or the best HS-PDSCH quality (HS-PDSCH Ec/
Nt); this depends on the option selected in Global parameters (HSDPA part): CQI based on CPICH quality or CQI based
on HS-PDSCH quality (CQI means channel quality indicator). Then, it determines the HS-PDSCH CQI, calculates the best
bearer that can be used and selects the suitable bearer so as to comply with cell and terminal user equipment HSDPA
capabilities. Once the bearer selected, Atoll finds the highest downlink rate that can be provided to the user and may calculate the application throughput.
CQI Based on CPICH Quality

When the option CQI based on CPICH quality is selected, Atoll proceeds as follows.
1.
CPICH Quality Calculation
Ec
Let us assume the following notation: ------- ic
corresponds to the CPICH quality.
Nt
pilot
Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total
noise.
Therefore, we have:
BTS P c ic
i
Ec
------- ic
for the total noise option,
= --------------------------------------------- Nt
pilot
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
for the without useful signal option.
= -------------------------------------------------------------------------------------- Nt
pilot
DL
N tot ic 1 BTS P c ic
i
With
DL
DL
DL
term
DL
N tot ic = I intra ic + I extra ic + I inter carrier ic + N 0
DL
DL
DL
P SCH ic
P SCH ic
DL
term
I intra ic = P tot ic + BTS 1 F MUD 1 P tot ic ----------------------- BTS P tot ic ----------------------
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
txj j
I inter carrier ic = -------------------------------------RF ic ic adj
icadj is a carrier adjacent to ic.

RF ic ic adj is the interference reduction factor, defined between ic and icadj and set to a value different from 0.
P pilot ic
P c ic = ----------------------i
LT
i

L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
BTS ,
term
and N 0
are defined in "Inputs" on page 186.
Note:
Atoll performs intra-cell interference computations based on the total power. You can
instruct Atoll to use maximum power by adding the following lines in the Atoll.ini file:
[CDMA]
PmaxInIntraItf = 1
In this case, Atoll considers the following formula:
P max ic P SCH ic
P max ic P SCH ic
P max ic
DL
term
I intra ic = ---------------------- + BTS 1 F MUD 1 ------------------------------------------------------ BTS ------------------------------------------------------
LT
LT
LT
Forsk 2009
AT271_TRG_E6
209

2.
CPICH CQI Determination
Let us assume the following notation: CQI pilot corresponds to the CPICH CQI. CQI pilot is read in the table
Ec
. This table is defined for the terminal reception equipment and the selected mobility.
CQI pilot = f ------- ic
Nt
pilot
3.
HS-PDSCH Quality Calculation
Atoll proceeds as follows:

1st step: Atoll calculates the HS-PDSCH power ( P HS PDSCH ).
P HSDPA ic is the power available for HSDPA on the carrier ic. This parameter is either a simulation output, or a userdefined cell input.
P HSDPA ic = P HS PDSCH ic + n HS SCCH P HS SCCH ic
Therefore, we have:
P HS PDSCH ic = P HSDPA ic n HS SCCH P HS SCCH ic
n HS SCCH is the number of HS-SCCH channels and P HS SCCH ic is the HS-SCCH power on carrier ic. It is either fixed
by the user (when the option HS-SCCH Power Dynamic Allocationin the cell property dialog is unchecked) or dynamically
calculated (when the option HS-SCCH Power Dynamic Allocation is selected).
req
Ec
In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ------- ic
). It
Nt
HS SCCH
is specified in mobility properties.

We have:
BTS P c ic
i
Ec
------- ic
= -----------------------------------for the total noise option,
Nt
HS SCCH
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
= -------------------------------------------------------------------------------------------------------------------------------------- for the without useful signal option.
Nt
HS SCCH
DL
term
N tot ic 1 F ortho 1 F MUD BTS P c ic
i
With
DL
DL
DL
DL
term
DL
DL
DL
P SCH ic
P SCH ic
DL
term
I intra ic = P tot ic + BTS 1 F MUD 1 F ortho P tot ic ----------------------- BTS P tot ic ----------------------
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
j
I inter carrier ic = txj
-------------------------------------RF ic ic adj

P HS SCCH ic
P c ic = -------------------------------------i
LT
i
and
L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
term
term
BTS , F ortho , F MUD and N 0
Therefore,
req
DL
Ec
------- ic
N tot ic
HS SCCH
Nt
P HS SCCH ic = --------------------------------------------------------------------------- L T for the total noise option,

i
BTS
And
210
AT271_TRG_E6
Forsk 2009

req
DL
Ec
------- ic
N tot ic
Nt
HS SCCH
- L T for the without useful signal option.

P HS SCCH ic = --------------------------------------------------------------------------------------------------------------------------------------------------------req
i
term
Ec- ic
BTS 1 + 1 F ortho 1 F MUD ----- HS SCCH
Nt
2nd step: Then, Atoll evaluates the HS-PDSCH quality

Ec
corresponds to the HS-PDSCH quality.
Nt
HS PDSCH
Therefore, we have:
BTS P c ic
i
Ec
------- ic
= ----------------------------------- Nt
HS PDSCH
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
Nt
HS PDSCH
P c ic
DL
term
i
---------------N tot ic 1 F ortho 1 F MUD BTS
n
Here, Atoll works on the assumption that five HS-PDSCH channels are used (n=5).
With
DL
DL
DL
DL
term
DL
DL
DL
P SCH ic
P SCH ic
DL
term
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
txj j

P HS PDSCH ic
P c ic = ----------------------------------------i
LT
i
And
L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
term
term
Note:
[CDMA]
PmaxInIntraItf = 1
P max ic P SCH ic
P max ic P SCH ic
P max ic
DL
term
I intra ic = ----------------------+ BTS 1 F MUD 1 Fortho ------------------------------------------------------ BTS ------------------------------------------------------
LT
LT
LT
4.
HS-PDSCH CQI Determination
The best bearer that can be used depends on the HS-PDSCH CQI. Let us assume the following notation:
CQI HS PDSCH corresponds to the HS-PDSCH CQI. Atoll calculates CQI HS PDSCH as follows:
CQI HS PDSCH = CQI pilot P pilot + P HS PDSCH
5.
HSDPA Bearer Selection
Knowing the HS-PDSCH CQI, Atoll finds the best bearer that can be used in the table Best Bearer=f(HS-PDSCH CQI).
This table is defined for the terminal reception equipment and the specified mobility.
Forsk 2009
AT271_TRG_E6
211

Then, Atoll checks if best bearer characteristics are compliant with cell and user equipment category capabilities. If they
are compliant, Atoll selects the best bearer. Otherwise, it searches for so as to find the suitable bearer.
Bearer characteristics are provided in the HSDPA Bearer table. We assume the best bearer index is 23. Characteristics
of this bearer are:
Transport block size: 9719 bits

Number of HS-PDSCH channels used: 7
16QAM modulation used: Yes
RLC Peak Rate: 4.48 Mb/s
Figure 6.5: HSDPA Radio Bearers Table

We assume that the terminal user equipment category is 3. Its capabilities are:
Maximum transport block size: 7298 bits

Maximum number of HS-PDSCH channels: 5
16QAM modulation supported: Yes
Minimum number of TTI between two TTI used: 2
Figure 6.6: HSDPA UE Categories Table

HSDPA cell capabilities are:
Maximum number of HS-PDSCH channels: 15.
The bearer 23 cannot be selected because:
212
AT271_TRG_E6
Forsk 2009
The number of HS-PDSCH channels (7) exceeds the maximum number of HS-PDSCH channels the terminal can
use (5),
And the transport block size (9719 bits) exceeds the maximum transport block size (7298 bits) the terminal can
carried.
In the Bearer table, Atoll searches a suitable bearer and selects the bearer index 22.
6.
The number of HS-PDSCH channels (5) does not exceed the maximum number of HS-PDSCH channels the terminal can use (5) and the maximum number of HS-PDSCH channels available at the cell level (15),
The transport block size (7168 bits) does not exceed the maximum transport block size (7298 bits) the terminal
can carried.
16QAM modulation is supported by the terminal.
HS-PDSCH Quality Update
Once the bearer selected, Atoll exactly knows the number of HS-PDSCH channels. Therefore, when the method Without
useful signal is used, it may recalculate the HS-PDSCH quality with the real number of HS-PDSCH channels (A default
value (5) was taken into account in the first HS-PDSCH quality calculation).
CQI Based on HS-PDSCH Quality

When the option CQI based on HS-PDSCH quality is selected, Atoll proceeds as follows.
1.
HS-PDSCH Quality Calculation
Atoll proceeds as follows:

1st step: Atoll calculates the HS-PDSCH power ( P HS PDSCH ).
P HSDPA ic is the power available for HSDPA on the carrier ic. This parameter is either a simulation output, or a userdefined cell input.
P HSDPA ic = P HS PDSCH ic + n HS SCCH P HS SCCH ic
Therefore, we have:
P HS PDSCH ic = P HSDPA ic n HS SCCH P HS SCCH ic
n HS SCCH is the number of HS-SCCH channels and P HS SCCH ic is the HS-SCCH power on carrier ic. It is either fixed
by the user (when the option HS-SCCH Power Dynamic Allocationin the cell property dialog is unchecked) or dynamically
calculated (when the option HS-SCCH Power Dynamic Allocation is selected).
req
Ec
In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ------- ic
). It
Nt
HS SCCH
is specified in mobility properties.

We have:
BTS P c ic
i
Ec
------- ic
Nt
HS SCCH
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
Nt
HS SCCH
DL
term
i
With
DL
DL
DL
term
DL
DL
DL
DL
P SCH ic
P SCH ic
DL
term
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
txj j

Forsk 2009
AT271_TRG_E6
213

P HS SCCH ic
P c ic = -------------------------------------i
LT
i
And
L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
term
term
Therefore,
req
DL
Ec
------- ic
N tot ic
HS SCCH
Nt
P HS SCCH ic = --------------------------------------------------------------------------- L T for the total noise option,

i
BTS
And
req
DL
Ec
------- ic
N tot ic
Nt
HS SCCH
- L T for the without useful signal option.

P HS SCCH ic = --------------------------------------------------------------------------------------------------------------------------------------------------------req
i
Ec
term
- ic
BTS 1 + 1 F ortho 1 F MUD ----- HS SCCH
Nt
2nd step: Then, Atoll evaluates the HS-PDSCH quality

Ec
corresponds to the HS-PDSCH quality.
Nt
HS PDSCH
Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total
noise.
We have:
BTS P c ic
i
Ec
------- ic
Nt
HS PDSCH
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
Nt
HS PDSCH
P c ic
DL
term
i
N tot ic 1 F ortho 1 F MUD BTS ----------------n
Here, Atoll works on the assumption that five HS-PDSCH channels are used (n=5). Then, it calculates the HS-PDSCH
CQI and the bearer to be used. Once the bearer selected, Atoll exactly knows the number of HS-PDSCH channels and
recalculates the HS-PDSCH quality with the real number of HS-PDSCH channels.
With
DL
DL
DL
DL
term
DL
DL
DL
P SCH ic
P SCH ic
DL
term
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
j
I inter carrier ic = txj
-------------------------------------RF ic ic adj

P HS PDSCH ic
P c ic = ----------------------------------------i
LT
i
And
L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
term
term
214
AT271_TRG_E6
Forsk 2009

Note:
[CDMA]
PmaxInIntraItf = 1
P max ic P SCH ic
P max ic P SCH ic
P max ic
DL
term
I intra ic = ----------------------+ BTS 1 F MUD 1 F ortho ------------------------------------------------------ BTS ------------------------------------------------------
L
LT
LT
T
2.
HS-PDSCH CQI Determination
Let us assume the following notation: CQI HS PDSCH corresponds to the HS-PDSCH CQI. CQI HS PDSCH is read in
Ec
. This table is defined for the terminal reception equipment and the
the table CQI HS PDSCH = f ------- ic
Nt
HS PDSCH
specified mobility.
3.
HSDPA Bearer Selection
The bearer is selected as described in "HSDPA Bearer Selection" on page 211.
6.4.2.3.5
Scheduling Algorithms
Three scheduling algorithms are available, Max C/I, Round Robin and Proportional Fair. Impact they have on the simulation result is described in the tables below.
Let 16 HSDPA users in the cell. All of them are active on DL and connected to the A-DCH R99 bearer. The number of HSSCCH channels and the maximum number of HSDPA users have been respectively set to 4 and 15.
Forsk 2009
Max C/I: 15 HSDPA users (where 15 corresponds to the maximum number of HSDPA users defined) enters the
scheduler in the same order as in the simulation. Then, they are sorted in descending order by the channel quality
indicator (CQI), i.e. in a best bearer descending order.
Mobiles
Simulation Rank
Best Bearer
(kbps)
DL Obtained Rate
(kbps)
Connection
Status
M1
2400
2400
Connected
M2
15
2400
1440
Connected
M3
2080
160
Connected
M4
2080
3.4
Delayed
M5
10
2080
3.4
Delayed
M6
12
2080
3.4
Delayed
M7
13
2080
3.4
Delayed
M8
14
2080
3.4
Delayed
M9
1920
3.4
Delayed
M10
1600
3.4
Delayed
M11
1600
3.4
Delayed
M12
1600
3.4
Delayed
M13
1600
3.4
Delayed
M14
1600
3.4
Delayed
M15
11
1440
3.4
Delayed
M16
16
2080
Scheduler Saturation
Round Robin: HSDPA users are taken into account in the same order than the one in the simulation (random
order).
Mobiles
Simulation Rank
Best Bearer
(kbps)
DL Obtained Rate
(kbps)
Connection
Status
M1
1600
1600
Connected
M2
2400
960
Connected
M3
1600
3.4
Delayed
M4
1600
3.4
Delayed
M5
1600
3.4
Delayed
M6
1600
3.4
Delayed
AT271_TRG_E6
215
M7
1920
3.4
Delayed
M8
2080
3.4
Delayed
M9
2080
3.4
Delayed
M10
10
2080
3.4
Delayed
M11
11
1440
3.4
Delayed
M12
12
2080
3.4
Delayed
M13
13
2080
3.4
Delayed
M14
14
2080
3.4
Delayed
M15
15
2400
3.4
Delayed
M16
16
2080
Proportional Fair: 15 HSDPA users (where 15 corresponds to the maximum number of HSDPA users defined)
enters the scheduler in the same order as in the simulation. Then, they are sorted in an ascending order according
to a new random parameter which corresponds to a combination of the user rank in the simulation and the channel
quality indicator (CQI).
For a user i, the random parameter RP i is calculated as follows:
Simu
RP i = 50 R i
CQI
+ 50 R i
Where,
Simu
is the user rank in the simulation.
Ri
CQI
Ri
is the user rank according to the CQI.
Note:
Mobiles
6.4.2.4
You can change the default weights by editing the atoll.ini file. For more information, see
the Administrator Manual.
Simulation
CQI Rank
Rank
RP
Best Bearer DL Obtained Rate

(kbps)
(kbps)
Connection Status
M1
150
2400
2400
Connected
M2
10
550
1600
960
Connected
M3
550
2080
160
Connected
M4
650
2080
3.4
Delayed
M5
11
700
1600
3.4
Delayed
M6
10
750
2080
3.4
Delayed
M7
12
800
1600
3.4
Delayed
M8
800
1920
3.4
Delayed
M9
15
850
2400
3.4
Delayed
M10
13
900
1600
3.4
Delayed
M11
12
900
2080
3.4
Delayed
M12
14
1000
1600
3.4
Delayed
M13
13
1000
2080
3.4
Delayed
M14
14
1100
2080
3.4
Delayed
M15
11
15
1300
1440
3.4
Delayed
M16
16
2080
HSUPA Part of the Algorithm

Finally, HSUPA users active on uplink, unless they have been rejected during the R99 or HSDPA parts of the algorithm,
are then evaluated by the HSUPA part of the algorithm.
Let us assume there are eight HSUPA users within an HSUPA cell and all of them are active on UL. The first four have
been connected to an HSDPA bearer, the last one has been rejected and the remaining three have been delayed in the
HSDPA part. Finally, the maximum number of HSUPA users within each cell has been set to six.
216
AT271_TRG_E6
Forsk 2009
6.4.2.4.1
Mobiles
Simulation Rank
HSDPA
Connection
Status
M1
Connected
M2
Connected
M3
Connected
M4
Connected
M5
Delayed
M6
Delayed
M7
Delayed
M8
Rejected
HSUPA Users
Atoll processes all the HSUPA users who are connected to an HSDPA bearer or were delayed in the HSDPA part. It
considers each user in the order established during the generation of the user distribution without exceeding the maximum
number of HSUPA users within each cell.
In this case, Atoll will consider the first six HSUPA users only and will reject the seventh user in order not to exceed the
maximum number of HSUPA users allowed in the cell (his connection status is "HSUPA scheduler saturation").
6.4.2.4.2
Admission Control
During admission control, Atoll selects a list of HSUPA bearers that are compatible with the user equipment capabilities
for each HSUPA user.
Let us focus on one HSUPA user with category 3 user equipment and a 50km/h mobility.
HSUPA user equipment categories are provided in the HSUPA User Equipment Categories table. The capabilities of the
category 3 user equipment are:
Maximum Number of E-DPDCH codes: 2

TTI 2 ms: No so it supports 10 ms TTI
Minimum Spreading Factor: 4
Maximum Block Size for a 2ms TTI: no value
Maximum Block Size for a 10ms TTI: 14484 bits
Figure 6.7: HSUPA UE Categories Table

HSUPA bearer characteristics are provided in the HSUPA Bearer table. An HSUPA bearer is described with following characteristics:
Radio Bearer Index: The bearer index number.

TTI Duration (ms): The TTI duration in ms. The TTI can be 2 or 10 ms.
Transport Block Size (Bits): The transport block size in bits.
Number of E-DPDCH Codes: The number of E-DPDCH channels used.
Minimum Spreading Factor: The smallest spreading factor used.
RLC Peak Rate (bps): The RLC peak rate represents the peak rate without coding (redundancy, overhead,
addressing, etc.).
Atoll considers an HSUPA bearer as compatible with the category 3 user equipment if:
The TTI duration used by the bearer is supported by the user equipment (10 ms).
The transport block size does not exceed the maximum transport block size supported by the user equipment
(14484 bits):
The number of E-DPDCH channels required by the bearer does not exceed the maximum number of E-DPDCH
channels that the terminal can use (2).
The minimum spreading factor used by the bearer is not less than the smallest spreading factor supported by the
terminal (4).
The HSUPA bearers compatible with category 3 user equiment are framed in red:
Forsk 2009
AT271_TRG_E6
217
Figure 6.8: HSUPA Radio Bearers Table

Then, during admission control, Atoll checks that the lowest compatible bearer in terms of the required E-DPDCH EcNt
does not require a terminal power higher than the maximum terminal power allowed.
Atoll uses the HSUPA Bearer Selection table. Among the compatible HSUPA bearers, Atoll chooses the one with the
lowest required Ec/Nt threshold.
Here, this is the index 1 HSUPA bearer; the required Ec/Nt threshold to obtain this bearer is -21.7dB.
Ec req
req
Then, from the required Ec/Nt threshold, -------
, Atoll calculates the required terminal power, P term HSUPA .
Nt E DPDCH
req
Ec
req
UL
P term HSUPA = -------
L T N tot
Nt E DPDCH
With
UL
UL intra
tx
N tot ic = 1 F MUD term I tot
UL extra
ic + I tot
UL
tx
ic + I inter carrier ic + N 0

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
tx
UL intra
term , F MUD , I tot
UL extra
, I tot
UL
tx
, I inter carrier and N 0 are defined in "Inputs" on page 186.
Figure 6.9: HSUPA Bearer SelectionTable

req
Atoll rejects the user if the terminal power required to obtain the lowest compatible HSUPA bearer ( P term HSUPA )
exceeds the maximum terminal power (his connection status is "HSUPA Admission Rejection").
At the end of this step, the number of non-rejected HSUPA users is n HSUPA . All of them will be connected to an HSUPA
bearer at the end.
6.4.2.4.3
HSUPA Bearer Allocation Process

Noise Rise Scheduling and Determination of the Obtained HSUPA Bearer
The obtained HSUPA radio bearer is the bearer that the HSUPA user obtains after noise rise scheduling and radio
resource control. The noise rise scheduling algorithm attempts to evenly share the remaining cell load between the users
admitted in admission control; in terms of HSUPA, each user is allocated a right to produce interference.
UL
The remaining cell load factor on uplink ( X txi ic ) depends on the maximum load factor allowed on uplink and how
much uplink load is produced by the served R99 traffic. It can be expressed as follows:
218
AT271_TRG_E6
Forsk 2009

X
UL
UL
UL
txi ic = X max txi ic X R99 txi ic
Then, Atoll evenly shares the remaining cell load factor between the HSUPA users admitted during the previous step.
UL
UL
X txi ic
X user txi ic = ---------------------------------n HSUPA
Ec max
From this value, Atoll calculates the maximum E-DPDCH EcNt allowed ( -------
). For further information on the
Nt E DPDCH
calculation, see "Uplink Load Factor Due to One User" on page 229.
max
1
Ec
-------
= ----------------------------------------------- for the Without useful signal option
Nt E DPDCH
UL
F txi ic
-------------------------------------- 1
UL
X user txi ic
UL
max
X user
Ec
-------
= ----------------- for the Total noise option
Nt E DPDCH
UL
F
Then, it selects an HSUPA bearer. The allocation depends on the maximum E-DPDCH EcNt allowed and on UE capabilities. Atoll selects the best HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer
UL
R RLC peak Index HSUPABearer

( Index HSUPABearer ) with the highest potential throughput ( ------------------------------------------------------------------------------------ ) where:
N Rtx Index HSUPABearer
req
Ec max
Ec
-------
-------
Nt E DPDCH Nt E DPDCH
And P term HSUPA P term
max
req
Ec req
When several HSUPA bearers are available, Atoll selects the one with the lowest -------
.
Nt E DPDCH
If we keep the same example as previously, we have six HSUPA users in the cell. The remaining cell load factor is shared
between the users. If the remaining cell load factor is 0.6, the UL load factor alloted to each user is 0.1. Lets take the cell
UL reuse factor equal to 1.5. Atoll calculates the maximum E-DPDCH EcNt allowed (the Without useful signal option is
selected).
Ec max
We have: -------
= -11.5 dB
Nt E DPDCH
Here, the obtained HSUPA bearer is the index 5 HSUPA bearer. It provides a potential throughput of 128 kbps and
requires E-DPDCH EcNt of -13 dB (lower than -11.5 dB) and a terminal power lower than the maximum terminal power
allowed.
.
HSUPA Bearers
Index
Required Ec/Nt
Threshold (dB)
Nb of
Retransmissions
RLC Peak Rate

(kbps)
Potential
Throughput
(kbps)
-21.7
32
16
-19
64
32
-16.1
128
64
-13.9
192
96
-13
256
128
-10.1
512
256
-8
768
384
UL
After selecting the bearer, Atoll reads the corresponding RLC peak rate, R RLC peak Index HSUPABearer and checks that
each user has obtained the average requested rate (defined in the properties of the service). A user is considered as
"happy" if the RLC peak rate provided by the HSUPA bearer exceeds the average requested rate and "unhappy" if not.
Atoll collects the unused load of "happy" users and redistributes it among the "unhappy" users. This process is repeated
until there is no more available load.
UL
The unused load ( X user unused ) of an "happy" user corresponds the difference of load between the user load allowing
Ec max
the maximum E-DPDCH EcNt ( -------
) and the user load required to reach the required E-DPDCH EcNt
Nt E DPDCH
Ec req
( -------
). It is determined as follows:
Nt E DPDCH
UL
UL
UL
X user unused = X user X user req
Forsk 2009
AT271_TRG_E6
219

Where
UL
X user is the UL load factor alloted to the user after equally sharing the remaining cell UL load between the HSUPA users.
Ec req
UL
).
X user req is the UL load factor required in order to reach the required E-DPDCH EcNt ( -------
Nt E DPDCH
Determination of the Requested HSUPA Bearer

The requested HSUPA radio bearer is selected from the HSUPA bearers compatible with the user equipment. Atoll determines the HSUPA bearer the user would obtain by considering the entire remaining load of the cell. The HSUPA user is
treated as if he is the only user in the cell. Therefore, if we go on with the previous example, the maximum E-DPDCH
EcNt allowed is equal to -1.8 dB and the requested HSUPA bearer is the index 7 HSUPA bearer. It requires E-DPDCH
EcNt of -8 dB (lower than -1.8 dB) and a terminal power lower than the maximum terminal power allowed.
6.4.2.5
Convergence Criteria
The convergence criteria are evaluated for each iteration, and can be written as follow:
DL
max P tx ic k P tx ic k 1
max N DL
user ic k N user ic k 1
Stations
Stations
- 100 int --------------------------------------------------------------------------------------------------- DL = max int -------------------------------------------------------------------------------------- 100
DL

P tx ic k
N user ic k
UL
UL
max I UL
max N UL
tot ic k I tot ic k 1
user ic k N user ic k 1
Stations
Stations
- 100 int --------------------------------------------------------------------------------------------------- UL = max int -------------------------------------------------------------------------------------- 100
UL
UL

I tot ic k
N user ic k

Atoll stops the algorithm if:

1st case: Between two successive iterations, UL and DL are lower ( ) than their respective thresholds (defined when
creating a simulation).
The simulation has reached convergence.
Example: Let us assume that the maximum number of iterations is 100, UL and DL convergence thresholds are set to 5.
If UL 5 and DL 5 between the 4th and the 5th iteration, Atoll stops the algorithm after the 5th iteration. Convergence
has been reached.
2nd case: After 30 iterations, UL and/or DL are still higher than their respective thresholds and from the 30th iteration,
UL and/or DL do not decrease during the next 15 successive iterations.
The simulation has not reached convergence (specific divergence symbol).
Examples: Let us assume that the maximum number of iterations is 100, UL and DL convergence thresholds are set to 5.
1. After the 30th iteration, UL and/or DL equal 100 and do not decrease during the next 15 successive iterations:
Atoll stops the algorithm at the 46th iteration. Convergence has not been reached.
2. After the 30th iteration, UL and/or DL equal 80, they start decreasing slowly until the 40th iteration (without going
under the thresholds) and then, do not change during 15 successive iterations: Atoll stops the algorithm at the 56th
iteration without reaching convergence.
3rd case: After the last iteration.
If UL and/or DL are still strictly higher than their respective thresholds, the simulation has not reached convergence
(specific divergence symbol).
If UL and DL are lower than their respective thresholds, the simulation has reached convergence.
6.4.3
Results
6.4.3.1
R99 Related Results

This table contains some R99 specific simulation results provided in the Cells and Mobiles tabs of the simulation property
dialog.
220
AT271_TRG_E6
Forsk 2009
Name
DL
I intra txi
Value
DL
SCH txi ic
P txi ic P
DL
--------------------------------P tot txi ic F ortho BTS tot
LT
txi
ic
Unit
Description
None



DL
1 F ortho BTS P b txi ic
DL
I extra ic
DL
P tot txj ic
txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
DL
DL
I tot ic
DL
N tot ic
UL intra
I tot
DL
DL
Total effective interference at terminal

on carrier ic (after unscrambling)
I tot ic + N 0
DL
Term

carrier ic
Pb

ic

ic



None
Cell uplink load factor on carrier ic
I intra ic + I extra ic + I inter carrier ic
UL
txi ic
UL extra
I tot
txi
term
txi
ic
UL
P b ic
term
txj j i
Pb
UL
UL
I inter carrier txi
ic
ic
ic adj
term
txj j
--------------------------------------
RF ic ic adj
UL
I tot txi ic
UL extra
I tot
UL intra
Tx
UL
UL
N tot txi ic
UL
txi ic +I inter carrier txi ic
tx
I tot txi ic + N 0
UL
txi ic
I tot txi ic
----------------------------UL
N tot txi ic
UL
txi ic
I tot txi ic
------------------------------------------------------------------------------------------UL intra
Tx
I tot
txi ic 1 F MUD term
None
Cell uplink reuse factor on carrier ic
UL
txi ic
1
-----------------------------UL
F txi ic
None
Cell uplink reuse efficiency factor on

carrier ic
None
Downlink load factor on carrier ic
None
Downlink reuse factor on a carrier ic
UL
UL
Simulation result available per cell

DL
I extra ic
DL
+ I inter carrier ic L T
--------------------------------------------------------------------------------------- + 1 F ortho BTS
DL
P Tx txi ic
-----------------------------------------------------------------------------------------------------------------------------------------1
------------- + 1 F ortho BTS
tch
DL
CI req
DL
txi ic
DL
Q req
DL
with CI req = ----------DL
Gp
Simulation result available per mobile
DL
I tot ic
-------------------DL
N tot ic
DL
DL
txi ic
I tot ic
------------------------------DL
I intra txi ic
NR
DL
txi ic
10 log 1 X
DL
txi ic
dB
Noise rise on downlink
NR
UL
txi ic
10 log 1 X
UL
txi ic
dB
Noise rise on uplink
Forsk 2009
AT271_TRG_E6
221
6.4.3.2
HSPA Related Results
6.4.3.2.1
Statistics Tab
In the Statistics tab, Atoll displays as results:
The number of non-connected users: This figure includes rejected and delayed users.
The number of R99 bearer users connected to a cell (result of the R99 part). This figure includes R99 users as
well as HSDPA and HSUPA users since all of them request a R99 bearer.
- The number of R99 bearer users per frequency band.
- The number of R99 bearer users per activity status.
DL
UL
The downlink and uplink rates ( R R99 and R R99 ) generated by their connection to R99 bearers. Only active
users are considered.
DL
R R99 =
DL
UL
R nominal R99 Bearer and R R99 =
Active
users
UL
R nominal R99 Bearer
Active
users
DL
UL
R nominal R99 Bearer is the downlink nominal rate of the user R99 radio bearer and R nominal R99 Bearer is the uplink
nominal rate of the user R99 radio bearer.
The number of connected users with an HSDPA bearer (result of the HSDPA part) and the downlink rate they generate. Both HSDPA and HSUPA users are considered since they both request an HSDPA bearer. On the other
DL
hand, only active users are taken into consideration in the downlink rate calculation ( R HSDPA ).
DL
R HSDPA =
DL
R RLC peak Index HSDPABearer
Active
users
DL
R RLC peak Index HSDPABearer is the RLC peak rate provided by the HSUPA bearer.
The number of connected HSUPA users (result of the HSUPA part). Only HSUPA users are considered.
At the end of the R99 part, HSDPA and HSUPA users can be:
-
Either connected and in this case, they obtain the requested R99 bearer (ADPCH-UL64 for HSDPA users and
ADPCH-EDPCCH for HSUPA users),
Or rejected exactly for the same reasons as R99 users.
Then, only connected HSDPA and HSUPA users are considered in the HSDPA part. At the end of the HSDPA
part, they can be:
-
Either connected if they obtain an HSDPA bearer,

Or rejected if the maximum number of HSDPA users per cell is exceeded,
Or delayed in case of lack of resources (HSDPA power, HS-SCCH power, HS-SCCH channels, OVSF codes).
In the HSUPA part, Atoll processes all the HSUPA users who are connected to an HSDPA bearer or were delayed
in the previous step. At the end, they can be:
-
Either connected if they obtain an HSUPA bearer,

Or rejected if the maximum number of HSUPA users per cell is exceeded or if the terminal power required to
obtain the lowest compatible HSUPA bearer exceeds the maximum terminal power.
UL
In addition, Atoll indicates the uplink rate the active HSUPA users connected with an HSUPA bearer ( R HSUPA )
generate:
UL
R HSUPA =
UL
Active
users
UL
R RLC peak Index HSUPABearer is the RLC peak rate provided by the HSUPA bearer.
6.4.3.2.2
Mobiles Tab
In the Mobiles tab, Atoll indicates for each user:
UL
DL
The uplink and downlink total requested rates in kbps (respectively, R requested M b and R requested M b )
For an R99 user, the DL and UL total requested rates correspond to the DL and UL nominal rates of the R99 bearer associated to the service.
For an HSDPA user, the uplink requested rate corresponds to the nominal rate of ADPCH R99 radio bearer and the downlink requested rate is the sum of the ADPCH radio bearer nominal rate and the RLC peak rate that the selected HSDPA
radio bearer can provide. Here, the HSDPA user is treated as if he is the only user in the cell and then, Atoll determines
the HSDPA bearer he would obtain by considering the entire HSDPA power available of the cell.
For an HSUPA user, the uplink requested rate is equal to the sum of the ADPCH-EDPCCH radio bearer nominal rate and
the RLC peak rate of the requested HSUPA radio bearer. The requested HSUPA radio bearer is selected from the HSUPA
222
AT271_TRG_E6
Forsk 2009

bearers compatible with the user equipment. Here, the HSUPA user is treated as if he is the only user in the cell and then,
Atoll determines the HSUPA bearer the user would obtain by considering the entire remaining load of the cell. The downlink requested rate is the sum of the ADPCH-EDPCCH radio bearer nominal rate and the RLC peak rate that the requested
HSDPA radio bearer can provide. The requested HSDPA radio bearer is determined as explained in the previous paragraph.
UL
DL
The uplink and downlink total obtained rates in kbps (respectively, R obtained M b and R obtained M b )
For an R99 user, the obtained rate is the same as the requested rate if he is connected without being downgraded. Otherwise, the obtained rate is lower (it corresponds to the nominal rate of the selected R99 bearer). If the user is rejected, the
obtained rate is zero.
For an HSDPA user connected to an HSDPA bearer, the uplink obtained rate equals the requested one and the downlink
obtained rate corresponds to the instantaneous rate; this is the sum of the A-DPCH radio bearer nominal rate and the RLC
peak rate provided by the selected HSDPA radio bearer after scheduling and radio resource control. If the HSDPA user is
delayed (he is only connected to an R99 radio bearer), uplink and downlink obtained rates correspond to the uplink and
downlink nominal rates of ADPCH radio bearer. Finally, if the HSDPA user is rejected either in the R99 part or in the
HSDPA part (i.e., because the HSDPA scheduler is saturated), the uplink and downlink obtained rates are zero.
For a connected HSUPA user, on uplink, if the user is connected to an HSUPA bearer, the obtained uplink rate is the sum
of the ADPCH-EDPCCH radio bearer nominal rate and the RLC peak rate provided by the selected HSUPA radio bearer
after noise rise scheduling. On downlink, if the user is connected to an HSDPA bearer, the obtained downlink rate corresponds to the instantaneous rate. The instantaneous rate is the sum of the ADPCH-EDPCCH radio bearer nominal rate
and the RLC peak rate provided by the selected HSDPA radio bearer after scheduling and radio resource control. If the
user is delayed, the obtained downlink rate corresponds to the downlink nominal rate of ADPCH-EDPCCH radio bearer.
If the HSUPA user is rejected, the obtained uplink and downlink rates are "0."
The mobile total power ( P term )

UL
P term = P term R99 f act EDPCCH + P term HSUPA for HSUPA users
And
P term = P term R99 in case of a R99 user
DL
The HSDPA application throughput in kbps ( T application M b )
This is the net HSDPA throughput without coding (redundancy, overhead, addressing, etc.).
DL
R obtained M b 1 BLER HSDPA SF Rate R

DL
T application M b = ---------------------------------------------------------------------------------------------------------------------------------TTI
Where:
BLER HSDPA is read in the quality graph defined for the triplet reception equipment-selected bearer-mobility (HSDPA
Quality Graphs tab in the Reception equipment properties). This graph describes the variation of BLER as a function of
the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll calculates the corresponding BLER.
SF Rate and R respectively represent the scaling factor between the application throughput and the RLC (Radio Link
Control) throughput and the throughput offset. These two parameters model the header information and other supplementary data that does not appear at the application level. They are defined in the service properties.
TTI is the minimum number of TTI (Transmission Time Interval) between two TTI used; it is defined in the terminal user
equipment category properties.
The number of OVSF codes
This is the number of 512-bit length OVSF codes consumed by the HSDPA or the HSUPA user for ADPCH and HSPDSCH channels.
The required HSDPA power in dBm
It corresponds to the HSDPA power required to provide the HSDPA or HSUPA user with the downlink requested rate.
If the HSDPA bearer allocated to the user is the best one, the required HSDPA power corresponds to the available HSDPA
power of the cell. On the other hand, if the HSDPA has been downgraded in order to be compliant with cell and UE capabilities, the required HSDPA power will be lower than the available HSDPA power of the cell.
The served HSDPA power in dBm
This is the HSDPA power required to provide the HSDPA or HSUPA user with the downlink obtained rate.
The No. of HSUPA Retransmissions (Required)
The number of retransmissions performed by the requested HSUPA radio bearer.
The No. of HSUPA Retransmissions (Obtained)
The number of retransmissions performed by the obtained HSUPA radio bearer.
UL
The HSUPA application throughput in kbps ( T application M b )
This is the net HSUPA throughput without coding (redundancy, overhead, addressing, etc.).
Forsk 2009
AT271_TRG_E6
223

UL
R obtained M b 1 BLER HSUPA SF Rate R

UL
T application M b = ---------------------------------------------------------------------------------------------------------------------------------N Rtx
Where:
BLER HSUPA is the residual BLER after N Rtx retransmissions. It is read in the quality graph defined for the quartet reception equipment-selected bearer-number of retransmissions-mobility (HSUPA Quality Graphs tab in the Reception equipment properties). This graph describes the variation of BLER as a function of the measured quality (E-DPDCH Ec/Nt).
Knowing the E-DPDCH Ec/Nt, Atoll calculates the corresponding BLER.
N Rtx is the number of retransmissions for the obtained HSUPA bearer. This figure is read in the HSUPA Bearer Selection
table.
The following columns appear if, when creating the simulation, you select "Detailed information about mobiles":
The uplink and downlink requested RLC peak rates (kbps)
For an R99 user, the uplink and downlink requested RLC peak rates are 0.
For an HSDPA user, the uplink RLC peak rate is 0 and the downlink requested RLC pear rate is the rate that the selected
HSDPA radio bearer can provide. Here, the HSDPA user is treated as if he is the only user in the cell and then, Atoll determines the HSDPA bearer he would obtain by considering the entire HSDPA power available of the cell.
For an HSUPA user, the requested uplink RLC peak rate is the rate of the requested HSUPA radio bearer. The requested
HSUPA radio bearer is selected from the HSUPA bearers compatible with the user equipment. Here, the HSUPA user is
treated as if he is the only user in the cell and then, Atoll determines the HSUPA bearer the user would obtain by considering the entire remaining load of the cell. If the user is connected to an HSDPA bearer in the downlink, the downlink
requested RLC peak rate is the rate that the requested HSDPA radio bearer can provide. The requested HSDPA radio
bearer is determined as explained in the previous paragraph.
The uplink and downlink obtained RLC peak rate (kbps)
For an R99 user, the uplink and downlink obtained RLC peak rates are 0.
For an HSDPA user connected to an HSDPA bearer, the uplink obtained RLC peak rate is 0, and the downlink obtained
RLC peak rate is the rate provided by the selected HSDPA radio bearer after scheduling and radio resource control.
For a connected HSUPA user, on uplink, if the user is connected to an HSUPA bearer, the obtained uplink RLC peak rate
is the rate provided by the selected HSUPA radio bearer after noise rise scheduling. On downlink, if the user is connected
to an HSDPA bearer, the downlink obtained RLC peak rate is the rate provided by the selected HSDPA radio bearer after
scheduling and radio resource control.
6.4.3.2.3
Cells Tab
In the Cells tab, Atoll gives:
The available HSDPA power in dBm ( P HSDPA ic )
This is:
-
Either a fixed value in case of a static HSDPA power allocation strategy,

Or a simulation result when the option "HSDPA Power Dynamic Allocation" is selected. We have:
P HSDPA ic = P max ic P Headroom ic P tx R99 ic P HSUPA ic
with P tx R99 ic = P pilot ic + P SCH ic + P OtherCCH ic +
tch used for

R99 users
P tch ic +
DL
P tch ic f act ADPCH
tch used for

HSPA users
The number of HSDPA users in the cell
They are the connected and delayed HSDPA bearer users. This figure includes HSDPA and HSUPA users.
The number of simultaneous HSDPA users in the cell ( n M )

b
It corresponds to the number of connected HSDPA bearer users that the cell supports at a time, i.e. within one transmission time interval. At the end of the HSDPA part of the simulation, all these users are connected to the cell, with a connection with the R99 radio bearer (ADPCH-UL64 for HSDPA users and ADPCH-EDPCCH bearer for HSUPA users) and a
HSDPA bearer.
DL
The instantaneous HSDPA rate in kbps ( R Inst cell )
This is the number of kilobits per second that the cell supports on downlink to provide simultaneous connected HSDPA
bearer users with a HSDPA bearer.
DL
R Inst cell =
DL
R obtained M b
M b cell
224
AT271_TRG_E6
Forsk 2009

DL
The instantaneous HSDPA MAC Throughput in kbps ( T MAC cell )
DL
T MAC cell =
M b cell
S block M b
------------------------------------------T TTI TTI M b
Where,
S block M b is the transport block size (in kbits) of the HSDPA bearer selected by the user; it is defined for each HSDPA
bearer in the HSDPA Radio Bearers table.
TTI M b is the minimum number of TTI (Transmission Time Interval) between two TTI used; it is defined in the terminal
user equipment category properties.
3
T TTI is the TTI duration, i.e. 2 10 s (2000 TTI in one second). This value is specified by the 3GPP.
DL
The average instantaneous HSDPA rate in kbps ( R Av Inst cell )
DL
R obtained M b
M cell
DL
b
R Av Inst cell = -------------------------------------------------------nM
b
DL
The HSDPA application throughput in kbps ( T application cell )
DL
Either T application cell =
DL
T application M b if the scheduling algorithm is Round Robin or Proportional Fair,
M b cell
DL
DL
Or T application cell = T application M b maxC I if the scheduling algorithm is Max C/I.

M b maxC I is the user with the highest C I in the cell.
The minimum HSDPA RLC peak rate in kbps (
DL
min R obtained M b )
M b cell
It corresponds to the lowest of RLC peak rates obtained by HSDPA bearer users connected to the cell.
DL
max R obtained M b )
The maximum HSDPA RLC peak rate in kbps (
M b cell
It corresponds to the highest of RLC peak rates obtained by HSDPA bearer users connected to the cell.
The number of HSUPA users in the cell ( n M ):

c
They are the connected HSUPA users.
UL
The HSUPA application throughput in kbps ( T application cell )
UL
T application cell =
UL
T application M b
M b cell
UL
The uplink cell load factor due to HSUPA traffic ( X HSUPA cell ):
UL
I tot cell HSUPA

UL
X HSUPA cell = -------------------------------------------UL
N tot cell
Where
UL
I tot cell HSUPA is the total interference at transmitter received from HSUPA bearer users.
6.4.3.2.4
Sites Tab
In the Sites tab, Atoll displays:
DL
The instantaneous HSDPA rate carried by the site in kbps ( R Inst site )
DL
R Inst site =
DL
R Inst cell
cell site
DL
DL
The instantaneous HSDPA MAC Throughput carried by the site in kbps ( T MAC site in kbps)
T MAC site =
DL
T MAC cell
cell site
Forsk 2009
AT271_TRG_E6
225
UL
The HSUPA rate carried by the site in kbps ( R
site =
UL
site )
UL
R obtained M c
M c site
6.4.4
Appendices
6.4.4.1
Admission Control in the R99 Part

During admission control of the R99 part of the simulation, Atoll calculates the uplink load factor of a considered cell
assuming the mobile concerned is connected with it. Here, activity status assigned to users is not taken into account. So
even if the mobile is not active on UL, it can be rejected due to cell load saturation. To calculate the cell UL load factor,
either Atoll takes into account the mobile power determined during power control if mobile was connected in previous iteration, or it estimates a load rise due to the mobile and adds it to the current load. The load rise ( X
follows:
X
UL
UL
) is calculated as
1
= -------------------------------------------------W
1 + ---------------------------------------UL
UL
Q req R nominal
6.4.4.2
Resources Management
6.4.4.2.1
OVSF Codes Management

OVSF codes are managed on the downlink during the simulation since this resource is downlink limited only. Atoll checks
the availability of this resource during the simulation, first in the R99 part and then in the HSDPA part. It determines the
number of codes that will be consumed by each cell.
OVSF codes form a binary tree. Codes of longer lengths are generated from codes of a shorter length. Length-k OVSF
codes are generated from length-k/2 OVSF codes. Therefore, if one channel needs 1 length-k/2 OVSF code, it is equivalent to use 2 length-k OVSF codes, or 4 length-2k OVSF codes and so on.
512 512-bit-length codes per cell are available in UMTS HSPA projects.
In the R99 part, during the resource control, Atoll determines the number of 512 bit-length codes that will be consumed for
each cell.
If the cell supports HSUPA, Atoll allocates codes for the DL channels used for HSUPA:
A 128 bit-length code for the E-HICH and E-RGCH channels (i.e. four 512 bit-length OVSF codes), for each cell.
Therefore, Atoll will take four 512-bit-length codes,
A 256 bit-length code for the E-AGCH channel (i.e. two 512 bit-length OVSF codes), for each cell. Therefore, Atoll
will take two 512-bit-length codes,
If the cell supports HSDPA, Atoll reserves for potential HSDPA users:
Codes HS PDSCH
The minimum number of HS-PDSCH codes defined for the cell, N min
. They are 16-bit length OVSF

Codes HS PDSCH
codes (i.e. thirty-two 512 bit-length OVSF codes). Therefore, Atoll will take 32 N min
512-bit-
length codes,
A 128 bit-length code per HS-SCCH channel (i.e. four 512 bit-length OVSF codes), for each cell. Therefore, Atoll
will take 4 n HS SCCH 512-bit-length codes,
Then, it allocates to the cell OVSF codes to support R99 bearers required by R99 and HSDPA users:
A 256 bit-length code per common channel (i.e. two 512 bit-length OVSF codes), for each cell. Therefore, Atoll will
take 2 N
512-bit-length codes,
A code per cell-receiver link, for TCH (traffic channels). The length of code to be allocated, Code_Length, depends
on the user activity. We have:
Overhead Codes
DL
Either Code_Length = F spreading Active user when the user is active,

DL
Or Code_Length = F spreading Inactive user if the user is inactive.

The number of 512 bit-length OVSF codes needed N
as follows:
N
226
Codes-TCH
Codes-TCH
is calculated from the length of the code to be allocated
512
= -----------------------------------Code_Length
AT271_TRG_E6
Forsk 2009
Figure 6.10: OVSF Code Tree Indices (Not OVSF Code Numbers)
The OVSF code allocation follows the Buddy algorithm, which guarantees that:
If a k-length OVSF code is used, all of its children with lengths 2k, 4k, , cannot be used as they will not be orthogonal.
If a k-length OVSF code is used, all of its ancestors with lengths k/2, k/4, , cannot be used as they will not be
orthogonal.
Example: We consider a user with a service requiring the UDD64 R99 radio bearer. This user is active on DL while
connected to a cell (which does not support HSDPA). The spreading factor for active users has been set to 64 and site
equipment requires four overhead downlink channel elements per cell. Atoll will consume four 256 bit-length OVSF codes
for common channels (i.e. eight 512 bit-length OVSF codes) and a 64 bit-length OVSF code for traffic channels (i.e. eight
additional 512 bit-length OVSF codes).
Notes:
In the R99 part, the OVSF code allocation follows the mobile connection order (mobile
order in the Mobiles tab).
The OVSF code and channel element management is differently dealt with in case of
softer handover. Atoll allocates OVSF codes for each cell-mobile link while it globally
assigns channel elements to a site.
In the HSDPA part, each HSDPA user is assigned a HSDPA bearer (Fast link adaptation). Therefore, Atoll allocates to the
cell:
Codes HS PDSCH
A number of 16-bit length OVSF codes per cell-HSDPA receiver, for HS-PDSCH N
. This
number depends on the HSDPA bearer assigned to the user; this is the number of HS-PDSCH channels required
by the HSDPA bearer. Therefore, Atoll will take 32 N
user connected to the cell
Codes HS PDSCH
512-bit-length codes for each HSDPA
Notes:
In the HSDPA part, Atoll sorts HSDPA mobiles according the selected scheduling
technique. Then, the OVSF code allocation follows this order.
When HSDPA users (at least one) are connected to the cell, Atoll gives the cell back the
Codes HS PDSCH
minimum number of OVSF codes reserved for HS-PDSCH ( N min
). On the
other hand, if no HSDPA user is connected, Atoll still keeps these codes and the codes for
HS-SCCH too. This is the same with HSUPA users. Even if no HSUPA user is connected
to the cell, Atoll still keeps the codes for E-HICH, E-RGCH and E-AGCH channels.
6.4.4.2.2
Channel Elements Management

Channel elements are controlled in the simulation. Atoll checks the availability of this resource on uplink and downlink.
In the R99 part, during the resource control, Atoll determines the number of channel elements required by each site on
uplink and downlink.
On uplink, Atoll consumes N
Overhead CE UL
TCH CE UL
CE UL
j channel elements for each cell j on a site NI. This figure includes:
channel elements for control channels,
per cell-receiver link, for TCH (traffic channels).
Therefore, the number of channel elements required on uplink at the site level, N
N
CE UL
NI =
CE UL
CE UL
N I , is:
j NI
Forsk 2009
AT271_TRG_E6
227

On downlink, Atoll consumes N
Overhead CE DL
N
channels),
TCH CE DL
CE DL
channel elements for control channels (Pilot channel, Synchronisation channel, common
per cell-receiver link, for TCH (traffic channels).
Therefore, the number of channel elements required on downlink at the site level, N
N
CE DL
NI =
CE DL
CE DL
N I , is:
j NI
Note:
6.4.4.3
In case of softer handover (the mobile has several links with co-site cells), Atoll allocates
channel elements for the best serving cell-mobile link only.
Downlink Load Factor Calculation

Atoll calculates a downlink load factor for each cell (available in the Cells tab of any simulation result) and each connected
mobile (available in the Mobiles tab of any given simulation result).
6.4.4.3.1
Downlink Load Factor per Cell

Approach for downlink load factor evaluation is highly inspired by the downlink load factor defined in the book WCDMA
for UMTS by Harry Holma and Antti Toskala.
DL
Q req
Let CI req = ----------- be the required quality.
DL
Gp
DL
Gp
DL
and Q req are the processing gain on downlink and the Eb/Nt target on downlink respectively.
In case of soft-handoff, required quality is limited to the effective contribution of the transmitter.
DL
P tx ic = P pilot ic + P SCH ic + P otherCCH ic +
Ptch ic
tch
DL
ortho
nonOrtho
P tx ic = P CCH ic + P CCH
ic +
Ptch ic
tch
where
ortho
P CCH ic = P pilot ic + P otherCCH ic

nonOrtho
P CCH
ic = P SCH ic
At mobile level, we have a required power, Ptch:

term
P tch ic = CI req I extra ic + I inter carrier ic + I intra ic + N 0
DL
With r = 1 when the user is active on the downlink and r = r c
LT r
when the user is inactive. In case of an HSDPA bearer
DL
user, r = f act ADPCH .
DL
nonOrtho
P tx ic P CCH
ic P tch ic
I extra ic + I inter carrier ic + 1 F ortho BTS --------------------------------------------------------------------------------------------LT
P tch ic = CI req
nonOrtho
ic
P CCH
term
+ ---------------------------------- + N0
LT
L r
T
DL
I extra ic + I inter carrier ic L T r + 1 F ortho BTS P tx ic r
+
nonOrtho
term
F ortho BTS P CCH
ic r + N 0
LT r
P tch ic = --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------1
---------------------- + 1 F ortho BTS
CI req r
I intra ic is the total power received at the receiver from the cell with which it is connected.
I extra ic is the total power received at the receiver from other cells.
I inter carrier ic is the inter-carrier interference received at the terminal.
We have:
228
AT271_TRG_E6
Forsk 2009

DL
I extra ic + I inter carrier ic L T r + 1 F ortho BTS P tx ic r
+
nonOrtho
term
F
ic
r
+
N
r
ortho
BTS
CCH
0
T
DL
ortho
nonOrtho
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------P tx ic = P CCH ic + P CCH
ic +
1
---------------------- + 1 F ortho BTS
tch
CI req r
DL
P tx ic
I extra ic + I inter carrier ic L T r

------------------------------------------------------------------------------------------------ P DL
tx ic
DL
P
ic
tx
= P ortho ic + P nonOrtho ic + --------------------------------------------------------------------------------------------------------------------------- +
CCH
CCH
1
+
--------------------1
F
tch
ortho
BTS
CI req r
DL
1 F ortho BTS P tx ic r
------------------------------------------------------------------------------------ +
1
tch ---------------------- + 1 F ortho BTS
CI req r
nonOrtho
term
F ortho BTS P CCH

ic r + N 0
LT r
---------------------------------------------------------------------------------------------------------------------------1
---------------------- + 1 F ortho BTS
tch
CI req r

------------------------------------------------------------------------------------------------ + 1 F ortho BTS r
DL
ic
DL
tx
----------------------------------------------------------------------------------------------------------------------------------------------------------- P DL
P tx ic
tx ic
1
---------------------- + 1 F ortho BTS
tch
r
CI
req
ortho
nonOrtho
= P CCH ic + P CCH
nonOrtho
ic +
F ortho BTS P CCH
term
ic r + N 0
LT r
---------------------------------------------------------------------------------------------------------------------------1
---------------------- + 1 F
tch
ortho
CI req r
nonOrtho
BTS
term
F ortho BTS P CCH

ic r + N 0
LT r
---------------------------------------------------------------------------------------------------------------------------1
---------------------- + 1 F ortho BTS
tch
CI req r
DL
P tx ic = -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- I extra ic + I inter carrier ic L T r
------------------------------------------------------------------------------------------------- + 1 F ortho BTS r
DL
P tx ic
-----------------------------------------------------------------------------------------------------------------------------------------------------------
1
1
---------------------- + 1 F ortho BTS
tch
CI req r
ortho
nonOrtho
P CCH ic + P CCH
ic +
Therefore, the downlink load factor can be expressed as:
DL

------------------------------------------------------------------------------------------------ + 1 F ortho BTS r
DL
P tx ic
-----------------------------------------------------------------------------------------------------------------------------------------------------------=
1
---------------------- + 1 F ortho BTS
tch
CI req r
The downlink load factor represents the signal degradation in relation to the reference interference (thermal noise plus
synchronisation channel power).
6.4.4.3.2
Downlink Load Factor per Mobile

Atoll evaluates the downlink load factor for any connected mobile as follows:
X
6.4.4.4
DL
DL
I tot ic
= -------------------DL
N tot ic
Uplink Load Factor Due to One User

UL
This part details how Atoll calculates the contribution of one user to the UL load factor ( X k ).
In this calculation, we assume that the cell UL reuse factor ( F
UL
txi ic ) is constant.
The result depends on the option used to calculate Nt (Without useful signal or Total noise that you may select in Global
parameters).
Without Useful Signal Option

UL
P b k req
W
UL
- -----------------------------------------------------------------------------------------------------------------------Q req k = ----------------------------UL
UL
tx
R nominal k I intra P b k req + I extra + I inter carrier + N 0
UL
P b k req
W
UL
- ------------------------------------------------------------------------------Q req k = ----------------------------UL
UL
tx
UL
R nominal k I intra F P b k req + N 0
Forsk 2009
AT271_TRG_E6
229

UL
UL
R nominal k
R nominal k
UL
UL
UL
UL
tx
P b k req 1 + Q req k ------------------------------ = Q req k ------------------------------ I intra F + N 0
W
W
UL
UL
R nominal k
R nominal k
UL
UL
UL
tx
Q req k ------------------------------ I intra F
Q req k ------------------------------ N 0
W
W
= -------------------------------------------------------------------------------------------- + -----------------------------------------------------------------------UL
UL
R nominal k
R nominal k
UL
UL
1 + Q req k -----------------------------1 + Q req k -----------------------------W
W
UL
P b k req
UL
req
R nominal k
Ec
UL
We note ------- k
= Q req k ----------------------------- Nt E DPDCH
W
UL
tx
I intra F
N0
UL
P b k req = ------------------------------------------------------------- + ------------------------------------------------------------

1
1
- + 1 ---------------------------------------------- + 1
--------------------------------------------req
Ec req
Ec
- k
- k
----- -----
Nt
Nt
E DPDCH
Pb
As I intra =
E DPDCH
UL
k req , we have:
UL
I intra = I intra F
tx
N0
- + N 0 ------------------------------------------------------------- ------------------------------------------------------------
tx
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
------------------------------------------------------------
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
= -----------------------------------------------------------------------------------------------UL
1
1F
-------------------------------------------------------------
K
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
K
I intra
tx
UL
N0 F
I intra = ----------------------------------------------------------------------------------------------1
-------------------------------------------------------------------------------------- 1
UL
1
F
-------------------------------------------------------------
K
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
UL
UL
I intra + I extra + I inter carrier
I intra F
1
- = ------------------------------------------ = -------------------------------------= ----------------------------------------------------------------------------------UL
tx
tx
tx
I intra F + N 0
N0
I intra + I extra + I inter carrier + N 0
1 + --------------------------UL
I intra F
Therefore, we have:
X
UL
= F
UL
------------------------------------------------------------
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
So, we can conclude that the contribution of one user to the UL load is defined as:
UL
X k
230
= F
UL
1
-------------------------------------------------------------
1
- + 1
--------------------------------------------req
Ec
- k
-----
Nt E DPDCH
AT271_TRG_E6
Forsk 2009
Total Noise Option

UL
P b k req
W
UL
- ----------------------------------------------------------------------------------Q req k = ----------------------------UL
tx
R nominal k I intra + I extra + I inter carrier + N 0
UL
P b k req
W
UL
- -----------------------------------------Q req k = ----------------------------UL
UL
tx
R nominal k I intra F + N 0
UL
R nominal k
UL
UL
UL
tx
P b k req = Q req k ------------------------------ I intra F + N 0
W
UL
req
R nominal k
Ec
UL
We note ------- k
= Q req k ----------------------------- Nt E DPDCH
W
req
Ec
UL
UL
tx
P b k req = ------- k
I intra F + N 0
Nt E DPDCH
Pb
UL
As I intra =
k req , we have:
I intra = I intra F
UL
tx
+ N0
Ec
req
- k
-----Nt E DPDCH
K
tx
N0
req
Ec
------- k
Nt E DPDCH
K
I intra = -------------------------------------------------------------------UL
1F
UL
UL
I intra + I extra + I inter carrier
I intra F
1
- = ------------------------------------------ = -------------------------------------= ----------------------------------------------------------------------------------tx
UL
tx
tx
I intra F + N 0
N0
I intra + I extra + I inter carrier + N 0
1 + --------------------------UL
I intra F
Therefore, we have:
X
UL
= F
UL
Ec
req
- k
-----Nt E DPDCH
K
So, we can conclude that the contribution of one user to the UL load is defined as:
UL
X k
6.4.4.5
= F
UL
req
Ec
------- k
Nt E DPDCH
Inter-carrier Power Sharing Modelling

Inter-carrier power sharing enables the network to dynamically allocate available power from R99-only and HSDPA carriers among HSDPA carriers.
In this part, we will consider the most common scenario, a network consisting of a R99-only carrier (c1) and an HSDPA
carrier with dynamic power allocation (c2) (c2 does not support HSUPA).
As explained in The User Manual, the maximum power of the HSDPA cell must be set to the same value as the maximum
shared power in order to use power sharing efficiently. In this case, the HSDPA cell can use 100% of the available power,
i.e, all of the R99-only cells unused power can be allocated to the HSDPA cell.
Lets take the following example to measure the impact of the inter-carrier power sharing.
1st case: Inter-carrier power sharing is not activated
On c1, we have: P max Tx c 1 = 43dBm and P tx R99 Tx c 1 = 39.1dBm .

On c2, we have: P max Tx c 2 = 43dBm , P tx R99 Tx c 2 = 36.1dBm and P Headroom Tx c 2 = 0dB .
Therefore, P HSDPA Tx c 2 = P max Tx c 2 P tx R99 Tx c 2 P Headroom Tx c 2 = 42dBm
2nd case: Inter-carrier power sharing is activated and P max Tx = 46dBm
On c1, we have: P max Tx c 1 = 43dBm and P tx R99 Tx c 1 = 39.1dBm .

On c2, we have: P max Tx c 2 = 46dBm , P tx R99 Tx c 2 = 36.1dBm and P Headroom Tx c 2 = 0dB .
Therefore, P HSDPA Tx c 2 = P max Tx P tx R99 Tx c 1 P tx R99 Tx c 2 P Headroom Tx c 2 = 44.4dBm
Forsk 2009
AT271_TRG_E6
231
6.5
UMTS HSPA Prediction Studies
6.5.1
Point Analysis
6.5.1.1
AS Analysis Tab
Let us suppose a receiver with a terminal, a service and a mobility type. This receiver does not create any interference.
You can make the prediction for a specific carrier or for all carriers of the main frequency band for the selected terminal.
If you have selected a dual-band terminal, you can make the coverage prediction on a specific carrier or on all carriers of
any frequency band for the selected terminal, or for all the carriers of all the frequency bands. The analysis is based on
the following parameters:
The uplink load factor and the downlink total power of cells,
The cell HSDPA power in case of a HSDPA user,
The cell UL reuse factor, the cell UL load factor due to HSUPA and the maximum cell UL load factor for HSUPA
users.
These parameters can be results of a given simulation, average values calculated from a group of simulations, or userdefined cell inputs. In the last case, when no value is defined in the Cells table, Atoll uses the following default values:
Total transmitted power = 50% of the maximum power (i.e, 40 dBm if the maximum power is set to 43 dBm)
Uplink load factor = 50%.
Uplink reuse factor = 1
Uplink load factor due to HSUPA = 0%
Maximum uplink load factor = 75%
On the other hand, no default value is used for the HSDPA power; this parameter must be defined by the user.
6.5.1.1.1
Bar Graph and Pilot Sub-Menu

We can consider the following cases:
1st case: Analysis based on a specific carrier
The carrier that can be used by transmitters is fixed. In this case, for each transmitter i containing the receiver in its calculation area and using the selected carrier, Atoll calculates the pilot quality at the receiver on this carrier. Then, it determines
the best serving transmitter using the selected carrier ic.
2nd case: Analysis based on all carriers
Atoll determines the best carrier for each transmitter i which contains the receiver in its calculation area and uses a
frequency band supported by the receivers terminal. The best carrier selection depends on the option selected for the site
equipment (UL minimum noise, DL minimum power, random, sequential). Then, Atoll calculates the pilot quality at the
receiver from these transmitters on their best carriers (ic) and defines the best server (on its best carrier).
3rd case: Analysis based on all carriers of any frequency band (for dual-band terminals with priority defined on frequency
bands only)
The frequency band that can be used is fixed. Atoll determines the best carrier for each transmitter i containing the receiver
in its calculation area and using the selected frequency band. The best carrier selection depends on the option selected
for the site equipment (UL minimum noise, DL minimum power, random, sequential). Then, Atoll calculates the pilot quality
at the receiver from these transmitters on their best carriers (ic) and defines the best server (on its best carrier).
Ec/Io (or Q pilot ic ) Evaluation

Let us assume that ic is either the best carrier or the selected carrier of a transmitter i containing the receiver in its radius
calculation and icadj is another carrier adjacent to ic. An interference reduction factor, RF ic ic adj , is defined between ic
and icadj and set to a value different from 0.
Two ways may be used to calculate Io.
Option Total noise: Atoll considers the noise generated by all the transmitters and the thermal noise.
Option Without pilot: Atoll considers the total noise deducting the pilot signal.
Calculation option may be selected in Global parameters.
Therefore, we have:
BTS P c i ic
Q pilot i ic = ------------------------------------------------DL
I 0 ic
With,
DL
DL
DL
term
DL
I 0 ic = P tot i ic + I extra ic + I inter carrier ic + N 0
And
DL
DL
DL
DL
term
I 0 ic = I intra ic + I extra ic + I inter carrier ic + N 0
232
AT271_TRG_E6
1 BTS P c i ic for the without pilot option.
Forsk 2009

1st step: P c i ic calculation for each cell (i,ic)
P c i ic is the pilot power of a transmitter i on carrier ic at the receiver.
P pilot i ic
P c i ic = --------------------------LT
I
L T is the total loss between transmitter i and receiver.

I
L Tx L path L term L body L Indoor M Shadowing Ec Io

L T = -------------------------------------------------------------------------------------------------------------------------------------------------I
G Tx G term
DL
DL
DL
2nd step: P tot j ic , P tot i ic and P tot j ic adj calculations

We have:
DL
I extra ic =
DL
P tot j ic
txj j i
P SCH ic
DL
DL
DL
I intra ic = P tot i ic BTS P tot i ic ----------------------
L
T
And
Ptot j icadj
DL
DL
txj j
I inter carrier ic = ------------------------------------------RF ic ic adj
DL
For each transmitter of the network, P tot ic is the total power received at the receiver from the transmitter on the best
carrier ic of the transmitter i.
P Tx ic
DL
P tot ic = -----------------LT
P Tx ic is the total power transmitted by the transmitter on the best carrier. Total power transmitted by each cell is either
a simulation result (provided in Simulation properties (Cells tab)) or a value user-defined in Cell properties.
DL
For each transmitter of the network, P tot ic adj is the total power received at the receiver from the transmitter on the
carrier icadj. This carrier is adjacent to ic.
P Tx ic adj
DL
P tot ic adj = ------------------------LT
P Tx ic adj is the total power transmitted by the transmitter on the carrier icadj. Total power transmitted by each cell is either
a simulation result (provided in Simulation properties (Cells tab)) or a value user-defined in Cell properties.
term
3rd step: N 0
term
N0
calculation
= NF term K T W
DL
4th step: I 0 ic and Q pilot i ic evaluation using formulas described above

DL
5th step: G macro diversity calculation

DL
The macro-diversity gain, G macro diversity , models the decrease in shadowing margin due to the fact there are several
available pilot signals at the mobile.
DL
npaths

npaths
M Shadowing Ec Io is the shadowing margin when the mobile receives n pilot signals (not necessarily from transmitters
belonging to the mobile active set).
Note:
This parameter is determined from cell edge coverage probability and Ec/Io standard
deviation. When the Ec/Io standard deviation is set to 0, the macro-diversity gain equals 0.
6th step: Determination of active-set

Atoll takes the transmitter i with the highest Q pilot i ic and calculates the best pilot quality received with a fixed cell edge
Resulting
coverage probability, Q pilot
Forsk 2009
ic .
AT271_TRG_E6
233

Resulting
Q pilot
DL
ic = G macro diversity max Q pilot i ic
Resulting
If Q pilot
req
Resulting
Q pilot , it means pilot quality at the receiver exceeds Q pilot
ic x% of time (x is the fixed cell edge cover-
age probability). The cell whose Q pilot i ic is the highest one enters the active set as best server ( Q pilot BS ic ) and
the best carrier (icBS) of the best server, BS, will be the carrier used by other transmitters of the active set (when active
set size is greater than 1). Pilot is available.
Resulting
If Q pilot
req
Q pilot , no cell (i,ic) can enter the active set. Pilot is unavailable.
Then, pilot qualities at the receiver from transmitters i (except the best server) on the best carrier of the best server, icBS,
are recalculated to determine the entire receiver active set (when active set size is greater than 1). Same formulas and
DL
calculation method are used to update I 0 ic BS value and determine Q pilot i ic BS .

We have:
BTS P c i ic
Q pilot i ic = ------------------------------------------------DL
I 0 ic
With,
DL
DL
DL
DL
term
I 0 ic = P tot i ic + I extra ic + I inter carrier ic + N 0
And
DL
DL
DL
term
DL
I 0 ic = I intra ic + I extra ic + I inter carrier ic + N 0
1 BTS P c i ic for the without pilot option.
Other cells (i,icBS) in the active set must satisfy the following criteria:
Q pilot i ic BS Q pilot BS ic BS AS_threshold i BS ic BS
i ic BS neighbour list i BS ic BS (optionally)
Number of Cells in Active Set

This is a user-specified input in the Terminal properties. It corresponds to the active set size.
Thermal Noise
This parameter is calculated as described above (3rd step).
Io (Best Server)
Io (Best server) is the total noise received at the receiver on icBS. The notation Best server refers to the best server of
active set. This is relevant when using the calculation option Without pilot. In this case, it informs that the pilot signal of
the best server (BS,icBS) is deducted from the total noise.
Downlink Macro-Diversity Gain

This parameter is calculated as described above (5th step).
6.5.1.1.2
Downlink Sub-Menu
The Downlink sub-menu may contain R99-related results and HSDPA-related results when a HSPA user is modelled.
R99-related Results
Atoll calculates the traffic channel quality from each cell (k,icBS) of the receivers active set at the receiver. No power
control is performed as in simulations. Here, Atoll determines the downlink traffic channel quality at the receiver for the
maximum allowed traffic channel power per transmitter. Then, after combination, the total downlink traffic channel quality
is evaluated and compared with the specified target quality.
Eb/Nt Target
DL
Eb/Nt target ( Q req ) is a user-defined parameter for a given service, mobility and reception equipment. This parameter is
available in the R99 Radio Bearers table.
234
AT271_TRG_E6
Forsk 2009

Notes:
Compressed mode is operated when:

- A mobile supporting compressed mode is connected to a cell located on a site with a
compressed-mode-capable equipment
And
- Either the received Ec/Io is lower than the Ec/Io activation threshold (Global parameters):
Resulting
Q pilot
CM activation
Q pilot
- Or the pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):
CM activation
P c RSCP pilot
When compressed mode is activated, the downlink Eb/Nt target is increased by the value
DL
user-defined for the DL Eb/Nt target increase field (Global parameters), Q req .
Required transmitter power on traffic channels

req
The calculation of the required transmitter power on traffic channels ( P tch ) may be divided into three steps.
DL
1st step: Q max k ic BS evaluation for each cell

DL
Let us assume the following notation: Eb/Nt max corresponds to Q max

Therefore, for each cell (k,icBS), we have:
DL
BTS P b max k ic BS
DL
DL
- G DL
Q max k ic BS = ------------------------------------------------------------p G Div
DL
N tot ic BS
max
P tch
DL
DL
DL
DL
DL
term
With P b max k ic BS = ------------ and N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
LT
k
Where
max
P tch is the maximum power allowed on traffic channel. This parameter is user-defined in the R99 Radio Bearers table.
DL
N tot ic BS is the total noise at the receiver on the best carrier of the best server.
DL
I intra ic BS is the intra-cell interference at the receiver on the best carrier of the best server.
P SCH k ic BS
DL
I intra ic BS = P DL k ic BTS F ortho P DL k ic ----------------------------------- tot
tot
BS
BS
L
T
DL
I extra ic BS
DL
is the extra-cell interference at the receiver on the best carrier of the best server.
I extra ic BS =
Ptot j icBS
DL
j j k
DL
I inter carrier ic BS
is the inter-carrier interference at the receiver on the best carrier of the best server.
Ptot j icadj
DL
DL
txj j
I inter carrier ic BS = ------------------------------------------RF ic BS ic adj
icadj is a carrier adjacent to icBS.

RF ic BS ic adj is the interference reduction factor, defined between ic and icadj and set to a value different from 0.
2nd step: Calculation of the total traffic channel quality
DL
Q MAX is the traffic channel quality at the receiver on icBS after signal combination of all the transmitters k of the active set.
On downlink, if there is no handoff, we have:
DL
DL
Q MAX ic BS = Q max k ic BS
For any other handoff status, we have:
DL
DL
Q MAX ic BS = f rake efficiency
Qmax k icBS
DL
Forsk 2009
AT271_TRG_E6
235

Where
DL
f rake efficiency is the downlink rake efficiency factor defined in Terminal properties.
req
3rd step: P tch calculation

DL
Q req
req
- P max
P tch = ------------------------------tch
DL
Q MAX ic BS
Notes:

compressed-mode-capable equipment
And
- Either the received Ec/Io is lower than the Ec/Io activation threshold (Global parameters):
Resulting
Q pilot
CM activation
Q pilot
- Or the pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):
CM activation
P c RSCP pilot
When compressed mode is activated, the downlink Eb/Nt target is increased by the value
DL
user-defined for the DL Eb/Nt target increase field (Global parameters), Q req . In this
DL
DL
Q req Q req
req
- P max
case, we have: P tch = --------------------------------tch
DL
Q MAX ic BS
Eb/Nt Max for Each Cell of Active Set

For each cell (k,icBS), we have:
DL
BTS P b max k ic BS
DL
DL
- G DL
Q max k ic BS = ------------------------------------------------------------p G Div
DL
N tot ic BS
max
P tch
DL
With P b max k ic BS = -----------LT
k
DL
DL
DL
DL
term
N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
max
req
P SCH k ic BS
P tch P tch
DL
I intra ic BS = P DL k ic BTS F ortho P DL k ic ------------------------------------ 1 BTS max (-----------------------------,0)
tot
tot
BS
BS
L
L
T
DL
I extra ic BS =
Tk
Ptot j icBS
DL
j j k
Ptot j icadj
DL
DL
txj j
Where
req
P tch is the required transmitter power on traffic channels.
Eb/Nt Max
DL
Q MAX is the traffic channel quality at the receiver on icBS after signal combination of all the transmitters k of the active set.
DL
DL
DL
DL
Qmax k icBS
DL
Where
DL
236
AT271_TRG_E6
Forsk 2009

DL
DL
DL
DL
DL
Therefore, the service on the downlink traffic channel is available if Q MAX ic BS Q req (or Q MAX ic BS Q req Q req
when compressed mode is activated).
Effective Eb/Nt
DL
Q eff is the effective traffic channel quality at the receiver on icBS.

DL
DL
DL
DL
DL
DL
DL
Q eff = min Q MAX Q req (or Q eff = min Q MAX Q req Q req when compressed mode is activated).
Downlink Soft Handover Gain

DL
G SHO corresponds to the DL soft handover gain.

DL
Q MAX ic BS
DL
G SHO = ----------------------------------------------------DL
max Q max k ic BS
DL
DL
max Q max k ic BS corresponds to the highest Q max k icBS value.
HSDPA-related Results
Atoll determines the best HSDPA bearer that the user can obtain. The HSDPA user is processed as if he is the only user
in the cell i.e. he uses the entire HSDPA power available in the cell.
For further information on the fast link adaptation modelling, see "Fast Link Adaptation Modelling" on page 209.
HS-PDSCH Ec/Nt
Atoll calculates the best HS-PDSCH quality (HS-PDSCH Ec/Nt). The way of calculating it depends on the selected option
in the transmitters global parameters (HSDPA part): CQI based on CPICH quality or CQI based on HS-PDSCH quality.
For further details on the HS-PDSCH quality calculation, see either "HS-PDSCH Quality Calculation" on page 210 if the
selected option is "CQI based on CPICH quality" or "HS-PDSCH Quality Calculation" on page 213 if the selected option
is "CQI based on HS-PDSCH quality".
HS-SCCH Ec/Nt
When the HS-SCCH power allocation strategy is dynamic, this parameter corresponds to the HS-SCCH Ec/Nt threshold
defined for the selected mobility type.
When the HS-SCCH power allocation strategy is static, the HS-SCCH Ec/Nt is calculated from the fixed HS-SCCH power.
We have:
BTS P c ic
i
Ec
------- ic
= ----------------------------------- Nt
HS SCCH
DL
N tot ic
And
BTS P c ic
i
Ec
------- ic
Nt
HS SCCH
DL
term
i
With
DL
DL
DL
term
DL
DL
DL
DL
P SCH ic
P SCH ic
DL
term
LT
LT
txi
txi
txi
DL
I extra ic =
DL
P tot ic
txj j i
Ptot icadj
DL
DL
txj j

P HS SCCH ic
P c ic = -------------------------------------i
LT
i
Forsk 2009
AT271_TRG_E6
237

And
L T = -------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
term
term
CQI
It corresponds to the HS-PDSCH CQI. The way of calculating it depends on the selected option in the transmitters global
parameters (HSDPA part): CQI based on CPICH quality or CQI based on HS-PDSCH quality.
For further details on the HS-PDSCH quality calculation, see either "HS-PDSCH CQI Determination" on page 211 if the
selected option is "CQI based on CPICH quality" or "HS-PDSCH CQI Determination" on page 215 if the selected option is
"CQI based on HS-PDSCH quality".
RLC Peak Rate

Knowing the HS-PDSCH CQI, Atoll calculates the best HSDPA bearer that can be used and selects the suitable bearer
so as to comply with cell and terminal user equipment HSDPA capabilities. Once the bearer selected, Atoll reads the RLC
peak rate that can be provided to the user.
For further details of the HSDPA bearer selection, see "HSDPA Bearer Selection" on page 211.
6.5.1.1.3
Uplink Sub-Menu
The Uplink sub-menu may contain R99-related results and HSUPA-related results when a HSPA user is modelled.
R99-related Results
For each cell (k,icBS) in the receivers active set, Atoll calculates uplink traffic channel quality from receiver. No power
control is performed as in simulations. Here, Atoll determines the uplink traffic channel quality at the cell for the maximum
terminal power allowed. Then, the total uplink traffic channel quality is evaluated with respect to the receiver handover
status. From this value, Atoll calculates the terminal power required to obtain the R99 bearer and compares it to the maximum terminal power allowed.
Max Terminal Power

max
Max terminal power ( P term ) is an input user-defined for each terminal. It corresponds to the terminals maximum power.
Required Terminal Power

req
The calculation of the terminal power required to obtain a R99 bearer ( P term R99 ) may be divided into three steps.
UL
1st step: Q max k ic BS evaluation for each cell

For each cell (k,icBS) in the receivers active set, we have:
UL
term P b max k ic BS
UL
UL
- G UL
Q max k ic BS = -------------------------------------------------------------p G Div
UL
N tot k ic BS
max
UL
P term 1 r c
UL
With P b max k ic BS = -----------------------------------------LT
k
UL
N tot k ic BS is the total noise at the transmitter on the best carrier of the best server. This value is calculated from the
cell uplink load factor X
UL
k ic BS .
tx
N0
UL
N tot k ic BS = ----------------------------------------UL
1 X k ic BS
tx
N 0 is the transmitter thermal noise.

UL
Q MAX ic BS is the traffic channel quality at the transmitter on icBS after signal combination of all the transmitters k of the
active set.
UL
UL
If there is no handoff (1/1): Q MAX ic BS = Q max k ic BS

For soft handoff (2/2):
UL
UL
UL
Q MAX ic BS = G macro diversity 2 links max Q max k ic BS
238
AT271_TRG_E6
Forsk 2009

UL
G macro diversity 2 links is the uplink macro-diversity gain. This parameter is determined from the fixed cell edge coverage
probability and the uplink Eb/Nt standard deviation. When the option Shadowing taken into account is not selected
(Prediction properties), Atoll considers the uplink macro-diversity gain defined by the user in Global parameters.
UL
UL
max Q max k ic BS corresponds to the highest Q max k icBS value.

For soft-soft handoffs (3/3):
UL
UL
UL

UL
For softer and softer-softer handoffs (1/2 and 1/3):
UL
UL
Qmax k icBS
UL
For softer-soft handoffs (2/3), there are two possibilities. If the MRC option is selected (option available in Global parameters), we have:
UL
UL
UL
Q MAX ic BS = G macro diversity 2 links max f rake efficiency
UL
UL
Q max k ic BS Q max
k on the same site
k on the same site
k ic BS
Else,
UL
UL
UL

req
3rd step: P term R99 calculation

req
P term R99 is the required terminal power.

UL
Q req
req
- P max
P term R99 = ------------------------------term
UL
Q MAX ic BS
UL
Q req is the uplink traffic quality target defined by the user for a given service and mobility. This parameter is available in
the R99 Radio Bearers table.
Notes:

compressed-mode-capable equipment, and
- The received Ec/Io is lower than the Ec/Io activation threshold (Global parameters):
Resulting
Q pilot
CM activation
Q pilot
- The pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):
CM activation
P c RSCP pilot
When compressed mode is activated, the uplink Eb/Nt target is increased by the value
UL
user-defined for the UL Eb/Nt target increase field (Global parameters), Q req . In this
UL
UL
Q req Q req
req
- P max
case, we have: P term R99 = --------------------------------term
UL
Q MAX ic BS
req
max
Therefore, the service on the uplink traffic channel is available if P term R99 P term .
Eb/Nt Max
For each cell (k,icBS) in the receivers active set, we have:
UL
term P b max k ic BS
UL
UL
- G UL
Q max k ic BS = -------------------------------------------------------------p G Div
UL
N tot k ic BS
max
UL
P term 1 r c
UL
With P b max k ic BS = -----------------------------------------LT
k
Forsk 2009
AT271_TRG_E6
239

UL
N tot k ic BS is the total noise at the transmitter on the best carrier of the best server. This value is calculated from the
cell uplink load factor X
UL
k ic BS .
tx
max
req
N0
P term P term R99
UL
- + 1 term max (---------------------------------------------N tot k ic BS = ----------------------------------------- , 0)
UL
LT
1 X k ic BS
k
tx

UL
Q MAX ic BS is the traffic channel quality at the transmitter on icBS after signal combination of all the transmitters k of the
active set.
UL
UL
If there is no handoff (1/1): Q MAX ic BS = Q max k ic BS

For soft handoff (2/2):
UL
UL
UL

UL
UL
UL
max Q max k ic BS corresponds to the highest Q max k ic BS value.

For soft-soft handoffs (3/3):
UL
UL
UL

UL
For softer and softer-softer handoffs (1/2 and 1/3):
UL
UL
Qmax k icBS
UL
For softer-soft handoffs (2/3), there are two possibilities. If the MRC option is selected (option available in Global parameters), we have:
UL
UL
UL
UL
UL
Q max k ic BS Q max
k on the same site
k on the same site
k ic BS
Else,
UL
UL
UL
Effective Eb/Nt
UL
Q eff is the effective traffic channel quality at the transmitter on icBS.

UL
UL
UL
UL
UL
UL
UL
Q eff = min Q MAX Q req (or Q eff = min Q MAX Q req Q req when compressed mode is activated).
Uplink Soft Handover Gain

UL
G SHO corresponds to the uplink soft handover gain.

UL
Q MAX ic BS
UL
G SHO = ----------------------------------------------------UL
max Q max k ic BS
UL
UL
max Q max k ic BS corresponds to the highest Q max k ic BS value.
HSUPA-related Results
Atoll determines the best HSUPA bearer that the user can obtain. The HSUPA user is processed as if he is the only user
in the cell i.e. he uses the entire remaining load of the cell.
For further information on the HSUPA bearer selection, see "HSUPA Bearer Allocation Process" on page 218.
240
AT271_TRG_E6
Forsk 2009
Required E-DPDCH Ec/Nt

Ec req
It corresponds to the E-DPDCH Ec/Nt required to obtain the HSUPA bearer ( -------
). This value is defined for a
Nt E DPDCH
HSUPA bearer ( Index HSUPABearer ) and a certain number of retransmissions ( N Rtx ) in the HSUPA Bearer Selection
table.
Required Terminal Power

req
Ec
req
From -------
, Atoll calculates the terminal power required to obtain the HSUPA bearer, P term HSUPA .
Nt E DPDCH
Ec req
req
UL
P term HSUPA = -------
L T N tot
Nt E DPDCH
With
UL
UL intra
tx
N tot ic = 1 F MUD term I tot
UL extra
ic + I tot
UL
tx
ic + I inter carrier ic + N 0

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
tx
UL intra
term , F MUD , I tot
UL extra
, I tot
UL
tx
, I inter carrier and N 0 are defined in "Inputs" on page 186.
RLC Peak Rate

Atoll selects the best HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer with the highest poUL

tential throughput ( ------------------------------------------------------------------------------------ ) where:
N Rtx Index HSUPABearer
req
Ec max
Ec
-------
-------
Nt E DPDCH Nt E DPDCH
And P term HSUPA P term
max
req
With
max
Ec
-------
: the maximum E-DPDCH EcNt allowed.
Nt E DPDCH
max
P term : the maximum terminal power allowed.

UL
After selecting the HSUPA bearer, Atoll reads the corresponding RLC peak rate, R RLC peak Index HSUPABearer .
Application Throughput
UL
Atoll displays the provided application throughput ( T application ). The application throughput represents the net throughput
after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:
UL
R RLC p eak 1 BLER HSUPA SF Rate R

UL
T application M b = ------------------------------------------------------------------------------------------------------------------------N Rtx
Where:
Knowing the E-DPDCH Ec/Nt, Atoll finds the corresponding BLER.
6.5.2
Coverage Studies
Let us assume each pixel on the map corresponds to a probe receiver with a terminal, a mobility type and a service. This
receiver does not create any interference. You can make the coverage prediction for a specific carrier or for all carriers of
the main frequency band for the selected terminal. If you have selected a dual-band terminal, you can make the coverage
prediction on a specific carrier or on all carriers of any frequency band for the selected terminal, or for all the carriers of all
the frequency bands. Coverage predictions are based on parameters that can be either simulation results, or user-defined
cell inputs.
Forsk 2009
AT271_TRG_E6
241
6.5.2.1
Pilot Reception Analysis

For further details of calculation formulas and methods, please refer to Definitions and formulas part, and Point analysis
AS analysis tab Pilot sub-menu part.
We consider the following cases:
1st case: Analysis Based on a Specific Carrier
The carrier that can be used by transmitters is fixed. In this case, for each transmitter i containing the receiver in its calculation area and using the selected carrier, Atoll calculates the pilot quality at the receiver on this carrier icgiven. Then, it
determines the best serving transmitter BS using the carrier icgiven ( Q pilot ic given ) and calculates the best pilot quality
BS
received with a fixed cell edge coverage probability,
Resulting
Q pilot
ic given
Atoll displays the best pilot quality received with a fixed cell edge coverage probability.
2nd case: Analysis Based on All Carriers
Atoll proceeds as in point predictions. It determines the best carrier of each transmitter i containing the receiver in its calculation area and using a frequency band supported by the receivers terminal. The best carrier selection depends on the
option selected for the site equipment (UL minimum noise, DL minimum power, random, sequential) and is based on the
UL load percentage and the downlink total power of cells (simulation results or cell properties). Atoll calculates the pilot
quality at the receiver from these transmitters on their best carriers and determines the best serving transmitter BS on its
best carrier icBS ( Q pilot ic BS ). Then, it calculates the best pilot quality received with a fixed cell edge coverage probaBS
bility,
Resulting
Q pilot
ic BS
3rd case: Analysis based on all carriers of any frequency band (for dual-band terminals with priority defined on frequency
bands only)
The frequency band that can be used is fixed. Atoll determines the best carrier of each transmitter i containing the receiver
in its calculation area and using the selected frequency band. The best carrier selection depends on the option selected
for the site equipment (UL minimum noise, DL minimum power, random, sequential) and is based on the UL load percentage and the downlink total power of cells (simulation results or cell properties). Then, Atoll calculates the pilot quality at
the receiver from these transmitters on their best carriers and determines the best serving transmitter BS on its best carrier
icBS ( Q pilot ic BS ). Then, it calculates the best pilot quality received with a fixed cell edge coverage probability,
BS
Resulting
Q pilot
ic BS .
6.5.2.1.1
Prediction Study Inputs

The Pilot Reception Analysis depends on the downlink total transmitted power of cells. This parameter can be either a
simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the total
transmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power is set
to 43 dBm).
6.5.2.1.2
Study Display Options

Single colour
Resulting
Atoll displays a coverage if Q pilot
req
ic Q pilot . Coverage consists of a single layer with a unique colour
req
( ic = ic BS or ic given ). Q pilot is a target value defined in the Mobility table by the user.
Colour per transmitter

Resulting
req
ic Q pilot ( ic = ic BS or ic given ). Coverage consists of several layers with asso-
ciated colours. There is a layer per transmitter with no intersection between layers. Layer colour is the colour assigned to
the best serving transmitter BS.
Colour per mobility

In this case, receiver is not completely defined and no mobility is assigned.
Coverage consists of several layers with a layer per user-defined mobility defined in Mobility sub-folder. For each layer,
Resulting
area is covered if Q pilot
req
ic Q pilot ( ic = ic BS or ic given ). Each layer is assigned a colour and displayed with inter-
sections between layers.
Colour per probability

This display option is available only if analysis is based on all simulations in a group (i.e. if you select a group of simulations
and the All option in the Simulation tab of prediction properties). Coverage consists of several layers with a layer per
242
AT271_TRG_E6
Forsk 2009

user-defined probability level defined in the Display tab (Prediction properties). For each layer, area is covered if
Resulting
Q pilot
req
ic Q pilot ( ic = ic BS or ic given ) in the required number of simulations. Each layer is assigned a colour and
displayed with intersections between layers.
Colour per cell edge coverage probability

Coverage consists of several layers with a layer per user-defined cell edge coverage probability, p, defined in the Display
Resulting
tab (Prediction properties). For each layer, area is covered if Q pilot
req
ic p Q pilot ( ic = ic BS or ic given ). Each layer
is assigned a colour and displayed with intersections between layers.
Colour per quality level (Ec/Io)

Coverage consists of several layers with a layer per user-defined quality threshold defined in the Display tab (Prediction
Resulting
properties). For each layer, area is covered if Q pilot
ic Q pilot threshold ( ic = ic BS or ic given ). Each layer is
assigned a colour and displayed with intersections between layers.
Colour per quality margin (Ec/Io margin)

Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction propResulting
erties). For each layer, area is covered if Q pilot
req
ic Q pilot Q pilot m arg in ( ic = ic BS or ic given ). Each layer is
6.5.2.2
Downlink Service Area Analysis

As in point predictions, Atoll calculates traffic channel quality at the receiver for each cell (k,ic) (with ic=icBS or icgiven) in
the receivers active set. No power control is performed as in simulations. Here, Atoll determines downlink traffic channel
quality at the receiver for a maximum allowed traffic channel power for transmitters. Then, the total downlink traffic channel
DL
quality ( Q MAX ic ) is evaluated after recombination.

Note:
Best server and active set determination is performed as in point prediction (AS analysis).
Atoll displays traffic channel quality at the receiver for transmitters in active set on the carrier ic ( ic BS or ic given ).
For further details of calculation formulas and methods, see "Downlink Sub-Menu" on page 234.
6.5.2.2.1

The Downlink Service Area Analysis depends on the downlink total transmitted power of cells. This parameter can be
either a simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the
total transmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power
is set to 43 dBm).
6.5.2.2.2

Single colour
DL
DL
DL
DL
DL
Atoll displays a coverage with a unique colour if Q MAX ic Q req (or Q MAX ic Q req Q req if compressed mode is
activated).
DL
Q req is a user-defined parameter for a service and mobility. It is available in the R99 Radio Bearers table.
DL
Q req is the DL Eb/Nt target increase; this parameter is user-defined in the Global parameters.

DL
DL
DL
DL
DL
Atoll displays a coverage if Q MAX ic Q req (or Q MAX ic Q req Q req if compressed mode is activated). Coverage
consists of several layers with associated colours. There is a layer per transmitter with no intersection between layers.
Layer colour is the colour assigned to best serving transmitter.
Colour per mobility

In this case, receiver is not completely defined and no mobility is assigned. Coverage consists of several layers with a
DL
DL
layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if Q MAX ic Q req (or
DL
DL
DL
Q MAX ic Q req Q req if compressed mode is activated). Each layer is assigned a colour and displayed with intersections between layers.
Forsk 2009
AT271_TRG_E6
243
Colour per service3

In this case, receiver is not completely defined and no service is assigned. Coverage consists of several layers with a layer
DL
DL
per user-defined service defined in Services sub-folder. For each layer, area is covered if Q MAX ic Q req (or
DL
DL
DL

DL
DL
Q MAX ic Q req in the required number of simulations. Each layer is assigned a colour and displayed with intersections
between layers.

DL
DL
DL
DL
DL
tab (Prediction properties). For each layer, area is covered if Q MAX ic p Q req (or Q MAX ic Q req Q req if
compressed mode is activated). Each layer is assigned a colour and displayed with intersections between layers.
Colour per maximum quality level (max Eb/Nt)

DL
properties). For each layer, area is covered if Q MAX ic Threshold . Each layer is assigned a colour and displayed with
intersections between layers.
Colour per effective quality level (Effective Eb/Nt)

DL
properties). For each layer, area is covered if Q eff ic Threshold . Each layer is assigned a colour and displayed with
DL
DL
DL
DL
DL
DL
DL
intersections between layers. Q eff ic = min Q MAX ic Q req (or Q eff ic = min Q MAX ic Q req Q req when
compressed mode is activated).
Colour per quality margin (Eb/Nt margin)

Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction propDL
DL
DL
DL
DL
erties). For each layer, area is covered if Q MAX ic Q req M arg in (or Q MAX ic Q req Q req M arg in when
3.
Service oriented studies (effective service area, DL or UL service area analysis) with per service display are based
on a calculation and display optimisation method. Atoll considers that a calculation pixel covered for the upper service of
the list is automatically covered for the lower services. Sometimes, this optimisation is not possible. In such a case, the
composite coverage is reliable but single service coverage layers might be incorrect. Atoll detects when this optimisation
may involve errors. In this case, it advises the user to perform a study for each service to get a reliable service coverage.
Different cases where optimisation method does not work correctly are detailed below:
- Any uplink or downlink quality study must be performed for each service if handover is not available for all the services
and the service order is different from the one found when considering handover for all the services. A pixel could be
covered by a service requiring a high quality target, due to handover (Eb/Nt combination), while it would not be covered
by a service requiring a lower quality target but not allowing handover.
UL
For uplink, services are sorted according to a decreasing quality indicator ( I Q ):

UL
Q req
UL
I Q = ----------UL
Gp
DL
For downlink, services are sorted according to a decreasing quality indicator ( I Q ):

DL
Q req
DL
I Q = ----------------------------DL
max
G p P tch
- Effective service area study must be performed for each service if uplink and downlink orders of services (explained
above) are not the same.
244
AT271_TRG_E6
Forsk 2009
Colour per required power

req
Atoll calculates the downlink required power, P tch ic , as follows:

DL
Q req
req
- P max
P tch ic = -----------------------tch
DL
Q MAX ic
Where
DL
Q req is the Eb/Nt target on downlink. This parameter, available in the R99 Radio Bearers table, is user-defined for given
service and mobility.
max
P tch is a user-defined input for each bearer related to a service. It corresponds to the maximum allowable traffic channel
power for a transmitter.
DL
DL
Q req Q req
req
- P max
When compressed mode is activated, we have: P tch ic = --------------------------------tch .
DL
Q MAX ic
Coverage consists of several layers with a layer per user-defined required power threshold defined in the Display tab
req
(Prediction properties). For each layer, area is covered if P tch ic Threshold . Each layer is assigned a colour and
Colour per required power margin

Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction propreq
max
erties). For each layer, area is covered if P tch ic P tch M arg in . Each layer is assigned a colour and displayed with
6.5.2.3
Uplink Service Area Analysis

As in point prediction, Atoll calculates uplink traffic channel quality from receiver for each cell (k,ic) (with ic=icBS or icgiven)
in receiver active set. No power control simulation is performed. Atoll determines uplink traffic channel quality at the transUL
mitter for the maximum terminal power allowed. Then, the total uplink traffic channel quality ( Q MAX ic ) is evaluated with
respect to receiver handover status.
Note:
Atoll displays traffic channel quality at transmitters in active set on the carrier ic ( ic BS or ic given ) received from the
receiver.
For further details of calculations formulas and methods, see "Uplink Sub-Menu" on page 238.
6.5.2.3.1

The Uplink Service Area Analysis depends on the UL load factor of cells. This parameter can be either a simulation output,
or a user-defined cell input. In the last case, when no value is defined in the Cells table for the uplink load factor, Atoll uses
50% as default value.
6.5.2.3.2

Single colour
UL
UL
UL
UL
UL
colour is unique.
UL
Q req is a user-defined parameter for a service and mobility. It is available in the R99 Radio Bearers table.
UL
Q req is the UL Eb/Nt target increase; this parameter is user-defined in the Global parameters.

UL
UL
UL
UL
UL
consists of several layers with associated colours. There is a layer per transmitter with no intersection between layers.
Layer colour is the colour assigned to best server transmitter.
Forsk 2009
AT271_TRG_E6
245
Colour per mobility

UL
UL
layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if Q MAX ic Q req (or
UL
UL
UL
Colour per service

UL
UL
per user-defined service defined in Services sub-folder. For each layer, area is covered if Q MAX ic Q req (or
UL
UL
UL

UL
UL
UL
UL
UL
Q MAX ic Q req (or Q MAX ic Q req Q req if compressed mode is activated) in the required number of simulations.
Each layer is assigned a colour and displayed with intersections between layers.
Colour per maximum quality level (Max Eb/Nt)

UL

UL
properties). For each layer, area is covered if Q effective ic Threshold . Each layer is assigned a colour and displayed
with intersections between layers.
UL
UL
UL
UL
UL
UL
UL
Q eff ic = min Q MAX ic Q req (or Q eff ic = min Q MAX ic Q req Q req when compressed mode is activated).

Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction propUL
UL
UL
UL
UL
erties). For each layer, area is covered if Q MAX ic Q req M arg in (or Q MAX ic Q req Q req M arg in if

Coverage consists of several layers with a layer per user-defined power threshold defined in the Display tab (Prediction
req
properties). For each layer, area is covered if P term R99 ic Threshold . Each layer is assigned a colour and displayed

Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction propreq
max
erties). For each layer, area is covered if P term R99 ic P term M arg in . Each layer is assigned a colour and displayed
Colour per soft handover gain

Coverage consists of several layers with a layer per soft handover gain value defined in the Display tab (Prediction propUL
erties). For each layer, area is covered if G SHO Threshold . Each layer is assigned a colour and displayed with intersections between layers.
6.5.2.4
Downlink Total Noise Analysis

Atoll determines the downlink total noise generated by cells.
246
AT271_TRG_E6
Forsk 2009
Ptot icadj
DL
DL
N tot ic =
j
- + N0
Ptot ic + txj-------------------------------------RF ic ic adj
DL
term
txj j
DL
Downlink noise rise, NR DL ic , is calculated from the downlink total noise, N tot , as follows:
term
N0
-
NR DL ic = 10 log ------------ N DL
tot
6.5.2.4.1
Study Inputs
The Downlink Total Noise Analysis depends on the downlink total transmitted power of cells. This parameter can be either
a simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the total
transmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power is set
to 43 dBm).
6.5.2.4.2
Analysis on All Carriers

If all the carriers are selected, Atoll determines DL total noise for all the carriers. Then, allows the user to choose different
colours.
Colour per minimum noise level

Coverage consists of several layers with a layer per user-defined noise level defined in the Display tab (Prediction propDL
erties). For each layer, area is covered if minN tot ic Threshold . Each layer is assigned a colour and displayed with
ic
Colour per maximum noise level

erties). For each layer, area is covered if maxN tot ic Threshold . Each layer is assigned a colour and displayed with
ic
Colour per average noise level

erties). For each layer, area is covered if averageN tot ic Threshold . Each layer is assigned a colour and displayed
ic
Colour per minimum noise rise

Atoll displays bins where minNR DL ic Threshold . Coverage consists of several areas with an area per user-defined
ic
noise rise threshold defined in the Display tab. Each area is assigned a colour with intersections between areas.
Colour per maximum noise rise

Atoll displays bins where maxNR DL ic Threshold . Coverage consists of several areas with an area per user-defined
ic
Colour per average noise rise

Atoll displays bins where averageNR DL ic Threshold . Coverage consists of several areas with an area per useric
defined noise rise threshold defined in the Display tab. Each area is assigned a colour with intersections between areas.
6.5.2.4.3
Analysis on a Specific Carrier

When only one carrier is analysed, Atoll determines DL total noise or DL noise rise on this carrier. In this case, the
displayed coverage is the same for any selected display per noise level (average, minimum, maximum) or any display per
noise rise (average, minimum, maximum).
Colour per noise level

erties). For each layer, area is covered if N tot ic Threshold . Each layer is assigned a colour and displayed with intersections between layers.
Forsk 2009
AT271_TRG_E6
247
Colour per noise rise

Atoll displays bins where NR DL ic Threshold . Coverage consists of several areas with an area per user-defined noise
rise threshold defined in the Display tab. Each area is assigned a colour with intersections between areas.
6.5.2.5
HSDPA Prediction Study

When calculating the HSDPA coverage prediction, either you can take all the possible HSDPA radio bearers into consideration, or you can study a certain HSDPA radio bearer. Then, available display options depend on what you have
selected.
When considering all the HSDPA radio bearers, you can set display parameters:
To analyse the uplink and downlink A-DPCH qualities on the map,

To analyse the HS-SCCH quality/power,
To model fast link adaptation for a single HSDPA user or for a defined number of HSDPA users.
When studying a certain HSDPA radio bearer, you can display areas where a certain RLC peak rate is available with different cell edge coverage probabilities (i.e. the probability of having a certain RLC peak rate).
Let us assume each pixel on the map corresponds to one or several users with HSDPA capable terminal, mobility and
HSDPA service. Each user may be using a specific carrier or all of them. Moreover, he does not create any interference.
Note that the HSDPA service area is limited by the pilot quality and the A-DPCH quality.
6.5.2.5.1

Parameters used as input for the HSDPA prediction study are:
The cell HSDPA power,

The downlink total transmitted power of the cell,
The number of HSDPA users within the cell if the study is calculated for several users.
These parameters can be either simulation outputs, or user-defined cell inputs. In the last case, when no value is defined
in the Cells table for the total transmitted power and the number of HSDPA users, Atoll uses the following default values:
Total transmitted power = 50% of the maximum power (i.e, 40 dBm if the maximum power is set to 43 dBm)
Number of HSDPA users = 1
On the other hand, no default value is used for the HSDPA power; this parameter must be defined by the user.
6.5.2.5.2

When considering all the HSDPA radio bearers, several display options are available in the study properties dialogue.
They can be regrouped in four categories according to the objective of the study:
To analyse the uplink and downlink A-DPCH qualities on the map,

To analyse the HS-SCCH quality/power,
To model fast link adaptation for a single HSDPA user
To model fast link adaptation for a defined number of HSDPA users.
When studying a certain HSDPA radio bearer, only one display option is available. It allows you to display where a certain
RLC peak rate is available with different cell edge coverage probabilities.
Analysis of UL And DL A-DPCH Qualities
Colour per Max A-DPCH Eb/Nt DL

DL
Atoll displays the A-DPCH quality at the receiver ( Q MAX ic ) for the best server on the carrier ic ( ic BS or ic given ). No
power control is performed as in simulations. Here, Atoll determines downlink traffic channel quality at the receiver for a
maximum traffic channel power allowed for the best server.
For further details of calculation formulas and methods, please refer to Prediction studies: Point analysis AS analysis tab
Downlink sub-menu part.
DL
Colour per Max A-DPCH Eb/Nt UL

UL
Atoll displays the A-DPCH quality at the best server ( Q MAX ic ) on the carrier ic ( ic BS or ic given ). No power control is
performed as in simulations. Here, Atoll determines uplink traffic channel quality at the receiver for a maximum terminal
power allowed.
For further details of calculations formulas and methods, please refer to Point analysis AS analysis tab Uplink submenu part.
UL
248
AT271_TRG_E6
Forsk 2009
Analysis of The HS-SCCH Quality/Power
Colour per HS-SCCH Power
This display option is relevant in case of dynamic HS-SCCH power allocation only. In this case, Atoll displays on each pixel
the HS-SCCH power per HS-SCCH channel. Coverage consists of several layers with a layer per threshold. For each
layer, area is covered if P HS SCCH ic Threshold . Each layer is assigned a colour and displayed with intersections
between layers.
Colour per HS-SCCH Ec/Nt
This display option is relevant in case of static HS-SCCH power allocation only. In this case, Atoll displays on each pixel
the HS-SCCH quality per HS-SCCH channel. Coverage consists of several layers with a layer per threshold. For each
Ec
layer, area is covered if ------- ic
Threshold . Each layer is assigned a colour and displayed with intersections
Nt
HS SCCH
between layers.
Fast Link Adaptation Modelling For A Single User

When you calculate the study with the following display options, Atoll considers one HSDPA user on each pixel and determines the best HSDPA bearer that the user can obtain. On each pixel, the HSDPA user is processed as if he is the only
user in the cell i.e. he uses the entire HSDPA power available in the cell.
For further information on the fast link adaptation modelling, see "Fast Link Adaptation Modelling" on page 209.
Colour per HS-PDSCH Ec/Nt
Atoll displays on each pixel the HS-PDSCH quality. Coverage consists of several layers with a layer per threshold. For
Ec
each layer, area is covered if ------- ic
Threshold . Each layer is assigned a colour and displayed with inter Nt
HS PDSCH
sections between layers.
Colour per CQI
Atoll displays either the CPICH CQI (see the calculation detail in "CPICH CQI Determination" on page 210) when the
selected option in Global parameters (HSDPA part) is CQI based on CPICH quality, or the HS-PDSCH CQI (see the calculation detail in the section 10.7.1.2.2) when considering the CQI based on HS-PDSCH quality option.
Coverage consists of several layers with a layer per CQI threshold ( CQI threshold ). For each layer, area is covered if
CQI CQI threshold . Each layer is assigned a colour and displayed with intersections between layers.
Colour per MAC Rate

DL
Atoll displays the MAC rate ( R MAC ) provided on each pixel. The MAC rate is calculated as follows:
S block
DL
R MAC = --------------T TTI
Where,
S block is the transport block size (in kbits) of the selected HSDPA bearer; it is defined for each HSDPA bearer in the
related table.
3
DL
Coverage consists of several layers with a layer per possible MAC rate ( R MAC ). For each layer, area is covered if the
MAC rate exceeds the user-defined thresholds. Each layer is assigned a colour and displayed with intersections between
layers.
Colour per MAC Throughput

DL
Atoll displays the MAC throughput ( T MAC ) provided on each pixel. The MAC throughput is calculated as follows:
S block
DL
T MAC = -----------------------------T TTI TTI
Where,
S block is the transport block size (in kbits) of the selected HSDPA bearer; it is defined for each HSDPA bearer in the
HSDPA Radio Bearers table.
3
Forsk 2009
AT271_TRG_E6
249

DL
Coverage consists of several layers with a layer per possible MAC throughput ( T MAC ). For each layer, area is covered if
the MAC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed with intersections
between layers.
Colour per RLC Peak Rate
After selecting the bearer, Atoll reads the corresponding RLC peak rate. This is the highest rate that the bearer can provide
on each pixel.
DL
Coverage consists of several layers with a layer per possible RLC peak rate ( R RLC p eak ). For each layer, area is covered
if the RLC peak rate can be provided. Each layer is assigned a colour and displayed with intersections between layers.
Colour per RLC Peak Throughput

DL
Atoll displays the RLC peak throughput ( T RLC p eak ) provided on each pixel. The RLC peak throughput is calculated as
follows:
DL
R RLC p eak
DL
T RLC p eak = --------------------------TTI
Where TTI is the minimum number of TTI (Transmission Time Interval) between two TTI used; it is defined in the terminal user equipment category properties.
DL
Coverage consists of several layers with a layer per possible RLC peak throughput ( T RLC p eak ). For each layer, area is
covered if the RLC peak throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed
Colour per Average RLC Throughput

DL
Atoll displays the average RLC throughput ( T RLC Av ) provided on each pixel. The average RLC throughput is calculated
as follows:
DL
R RLC p eak 1 BLER HSDPA

DL
T RLC Av = --------------------------------------------------------------------------------TTI
Where,
the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll finds the corresponding BLER.
DL
Coverage consists of several layers with a layer per possible average RLC throughput ( T RLC Av ). For each layer, area
is covered if the average RLC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and
Colour per Application Throughput

DL
Atoll displays the application throughput ( T application ) provided on each pixel. The application throughput represents the
net throughput after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:
DL
R RLC p eak 1 BLER HSDPA SF Rate R

DL
T application = ------------------------------------------------------------------------------------------------------------------------TTI
Where:
the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll finds the corresponding BLER.
DL
Coverage consists of several layers with a layer per possible application throughput ( T application ). For each layer, area is
covered if the application throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed
250
AT271_TRG_E6
Forsk 2009
Fast Link Adaptation Modelling For Several Users

When you calculate the study with the following display options, Atoll considers several HSDPA users per pixel and determines the best HSDPA bearer that each user can obtain. In this case, the cell HSDPA power is shared between HSDPA
users. When the coverage prediction is not based on a simulation, the number of HSDPA users is taken from the cell properties. The displayed results of the coverage prediction will be an average result for one user.
For further information on the HSDPA bearer allocation process when there are several users, see "HSDPA Bearer Allocation Process" on page 208 For further information on the fast link adaptation modelling, see "Fast Link Adaptation
Modelling" on page 209.
Colour per MAC Throughput Per Mobile

DL
Atoll displays the average MAC throughput per mobile ( T MAC average ) provided on each pixel. The average MAC
throughput per mobile is calculated as follows:
n HSDPA
DL
DL
T MAC x
x=1
T MAC average = --------------------------------------n HSDPA
Where,
n HSDPA is the number of HSDPA users within the cell.
DL
T MAC x is the MAC throughput of each HSDPA user. For further information on the calculation of the MAC throughput,
see "Colour per MAC Throughput" on page 249.
DL
Coverage consists of several layers with a layer per possible average MAC throughput per mobile ( T MAC average ). For
each layer, area is covered if the average MAC throughput per mobile exceeds the user-defined thresholds. Each layer is
Colour per RLC Throughput Per Mobile

DL
Atoll displays the average RLC throughput per mobile ( T RLC average ) provided on each pixel. The average RLC throughput per mobile is calculated as follows:
n HSDPA
DL
DL
T RLC p eak x
x=1
T RLC average = ---------------------------------------------------n HSDPA
Where,
DL
T RLC p eak x is the RLC peak throughput of each HSDPA user. For further information on the calculation of the RLC peak
throughput, see "Colour per RLC Peak Throughput" on page 250.
DL
Coverage consists of several layers with a layer per possible average RLC throughput per mobile ( T RLC average ). For
each layer, area is covered if the average RLC throughput per mobile exceeds the user-defined thresholds. Each layer is
Colour per ApplicationThroughput Per Mobile

DL
Atoll displays the average application throughput per mobile ( T application average ) provided on each pixel. The average
application throughput per mobile is calculated as follows:
n HSDPA
DL
DL
T application x
x=1
T application average = ----------------------------------------------------n HSDPA
Where,
DL
T application x is the application throughput of each HSDPA user. For further information on the calculation of the application throughput, see "Colour per Application Throughput" on page 250.
Coverage consists of several layers with a layer per possible average application throughput per mobile
DL
( T application average ). For each layer, area is covered if the average application throughput per mobile exceeds the userdefined thresholds. Each layer is assigned a colour and displayed with intersections between layers.
Forsk 2009
AT271_TRG_E6
251
Probability of Having a Certain RLC Peak Rate

This result can be obtained only if you have selected an HSDPA radio bearer in the Condition tab.
Colour per Cell Edge Coverage Probability
Atoll shows areas where the selected HSDPA radio bearer is available with different cell edge coverage probabilities.
Coverage consists of several layers with a layer per cell edge coverage probability defined in the Display tab. For each
layer, area is covered if the selected HSDPA radio bearer is available. Each layer is assigned a colour and displayed with
6.5.2.6
HSUPA Prediction Study

A dedicated HSUPA study is available with different calculation and display options. Atoll determines on each pixel the
best HSUPA bearer that can be obtained; it can consider either a single HSUPA user or several ones on each pixel. For
further information on the HSUPA bearer selection, see "HSUPA Bearer Allocation Process" on page 218. By caclulating
this study with suitable display options, it is possible:
To analyse the power required by the selected terminal,

To analyse the required E-DPDCH quality,
To analyse rates and throughputs.
Let us assume each pixel on the map corresponds to one or several users with HSUPA capable terminal, mobility and
HSUPA service. Each user may be using a specific carrier or all of them. Moreover, he does not create any interference.
6.5.2.6.1

Parameters used as input for the HSUPA prediction study are:
The cell UL load factor,

The cell UL reuse factor,
The cell UL load factor due to HSUPA,
The maximum cell UL load factor,
The number of HSUPA users within the cell if the study is calculated for several users.
These parameters can be either simulation outputs, or user-defined cell inputs. In the last case, When no value is defined
in the Cells table, Atoll uses the following default values:
6.5.2.6.2
Uplink load factor = 50%

Uplink reuse factor = 1
Uplink load factor due to HSUPA = 0%
Maximum uplink load factor = 75%
Number of HSUPA users = 1
Calculation Options
Atoll can calculate the HSUPA coverage prediction in one of two ways:
6.5.2.6.3
HSUPA resources can be dedictated to a single user: On each pixel, the HSUPA user is processed as if he is
the only user in the cell i.e he will use the entire remaining load after allocating capacity to all R99 users.
HSUPA resources can be shared by HSUPA users defined or calculated per cell: Atoll considers several
HSUPA users per pixel. After allocating capacity to all R99 users, the remaining load of the cell will be shared
equally between all the HSUPA users. When the coverage prediction is not based on a simulation, the number of
HSUPA users is taken from the cell properties. The displayed results of the coverage prediction will be an average
result for one user.
Display Options
The following display options are available in the prediction property dialogue.
Colour per Required E-DPDCH Ec/Nt

Atoll displays on each pixel the E-DPDCH Ec/Nt required to obtain the selected HSUPA bearer. Coverage consists of
Ec req
several layers with a layer per threshold. For each layer, area is covered if -------
Threshold . Each layer is
Nt E DPDCH
Colour per Required Terminal Power

Atoll displays on each pixel the terminal power required to obtain the selected HSUPA bearer. The required terminal power
is calculated from the required E-DPDCH Ec/Nt. Coverage consists of several layers with a layer per threshold. For each
req
layer, area is covered if P term Threshold . Each layer is assigned a colour and displayed with intersections between
layers.
Colour per MAC Rate

UL
Atoll displays the MAC rate ( R MAC ) provided on each pixel. The MAC rate is calculated as follows:
252
AT271_TRG_E6
Forsk 2009

UL
S block
UL
R MAC = --------------T TTI
Where,
UL
S block is the transport block size (in kbits) for the selected HSUPA bearer; it is defined for each HSUPA bearer in the
HSUPA Radio Bearers table.
T TTI is the duration of one TTI for the selected HSUPA bearer; it is defined for each HSUPA bearer in the HSUPA Radio
Bearers table.
UL
Coverage consists of several layers with a layer per possible MAC rate ( R MAC ). For each layer, area is covered if the
MAC rate exceeds the user-defined thresholds. Each layer is assigned a colour and displayed with intersections between
layers.
Colour per RLC Peak Rate

After selecting the HSUPA bearer, Atoll reads the corresponding RLC peak rate. This is the highest rate that the selected
HSUPA bearer can provide on each pixel.
UL
Coverage consists of several layers with a layer per possible RLC peak rate ( R RLC p eak ). For each layer, area is covered
if the RLC peak rate can be provided. Each layer is assigned a colour and displayed with intersections between layers.
Colour per Minimum RLC Throughput

UL
Atoll displays the minimum RLC throughput ( T RLC Min ) provided on each pixel. The minimum RLC throughput is calculated as follows:
UL
R RLC p eak 1 BLER HSUPA

UL
T RLC Min = --------------------------------------------------------------------------------N Rtx
Where,
N Rtx is the number of retransmissions for the selected HSUPA bearer. This figure is read in the HSUPA Bearer Selection
table.
DL
Coverage consists of several layers with a layer per possible average RLC throughput ( T RLC Av ). For each layer, area
is covered if the minimum RLC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and
Colour per Application Throughput

UL
Atoll displays the application throughput ( T application ) provided on each pixel. The application throughput represents the
net throughput after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:
UL
R RLC p eak 1 BLER HSUPA SF Rate R

UL
T application M b = ------------------------------------------------------------------------------------------------------------------------N Rtx
Where:
N Rtx is the number of retransmissions for the selected HSUPA bearer. This figure is read in the HSUPA Bearer Selection
table.
UL
Coverage consists of several layers with a layer per possible application throughput ( T application ). For each layer, area is
covered if the application throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed
Forsk 2009
AT271_TRG_E6
253
6.6
Automatic Neighbour Allocation

Atoll permits the automatic allocation of intra-technology neighbours in the current network. Two allocation algorithms are
available, one dedicated to intra-carrier neighbours and the other for inter-carrier neighbours.
The intra-technology neighbour allocation algorithms take into account all the cells of TBC transmitters. It means that all
the cells of TBC transmitters of your .atl document are potential neighbours.
The cells to be allocated will be called TBA cells. They must fulfil following conditions:
They are active,

Only TBA cells may be assigned neighbours.

Note:
6.6.1

We assume a reference cell A and a candidate neighbour, cell B. When automatic allocation starts, Atoll checks the following conditions:
1.
2.
The distance between both cells must be less than the user-definable maximum inter-site distance. If the distance
between the reference cell and the candidate neighbour is greater than this value, then the candidate neighbour
is discarded.
Carriers: This option enables you to select the carrier(s) on which you want to run the allocation. You may choose one or
more carriers. Atoll will allocate neighbours to cells using the selected carriers.
Force co-site cells as neighbours: This option enables you to force cells located on the reference cell site in the candidate
neighbour list. This constraints can be weighted among the others and ranks the neighbours through the importance field
(see after).
Force adjacent cells as neighbours (only for intra-carrier neighbours): This option enables you to force cells geographically
adjacent to the reference cell in the candidate neighbour list.This constraints can be weighted among the others and ranks
the neighbours through the importance field (see after).
Notes:
Adjacence criterion: Let CellA be a candidate neighbour cell of CellB. CellA is considered
adjacent to CellB if there exists at least one pixel in the CellB Best Server coverage area
where CellA is Best Server (if several cells have the same best server value) or CellA is the
second best server that enters the Active Set (respecting the HO margin of the allocation).
the least, depending on the above criterion. Adjacence is relative to the number of pixels
Force neighbour symmetry: This option enables user to force the reciprocity of a neighbourhood link. Therefore, if the reference cell is a candidate neighbour of another cell, this one will be considered as candidate neighbour of the reference cell.
force/forbid a cell to be candidate neighbour of the reference cell.
254
AT271_TRG_E6
Forsk 2009

There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability:
3.
Intra-carrier neighbours: intra-carrier handover is a soft handover.
The reference cell A and the candidate cell B are located inside a continuous layer of cells with carrier c1 (c1 is the selected
carrier on which you run the allocation).
SA is the area where the cell A is the best serving cell. It means that the cell A is the first one in the active set.
-
The pilot signal received from the cell A is greater than the minimum pilot signal level.
The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0).
The pilot quality from A is the best.
SB is the area where the cell B can enter the active set.
-
The pilot signal received from the cell B is greater than the minimum pilot signal level.
The pilot quality from B is greater than the pilot quality from A minus the Ec/I0 margin. The Ec/I0 margin has
the same meaning as the AS-threshold defined in the Cell properties. So, it should logically have the same
value.
Figure 6.11: Overlapping Zone for Intra-carrier Neighbours
Inter-carrier neighbours: inter-frequency handover is a hard handover. It is needed in a multi-carrier W-CDMA network:
-
To balance loading between carriers and layers (1st case),
To make a coverage reason handover from micro cell frequency to macro cells (2nd case).
1st case: the reference cell A is located inside a continuous layer of cells with carrier c1 (c1 is the selected carrier on which
you run the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.
SA is the area where the cell A is not the best serving cell of its layer but can enter the active set.
-
The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0).
The pilot quality from A is not the highest one. It is strictly lower than the best pilot quality received and greater
than the best pilot quality minus the Ec/I0 margin.
SB is the area where the cell B is the best serving cell of its layer.
-
Forsk 2009
The pilot quality from B exceeds a user-definable minimum value (minimum Ec/I0).
The pilot quality from B is the highest one.
AT271_TRG_E6
255
Figure 6.12: Overlapping Zone for Inter-carrier Neighbours - 1st Case

2nd case: the reference cell A is located on the border of a layer with carrier c1 (c1 is the selected carrier on which you run
the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.
SA is the area where the pilot quality from the cell A starts significantly decreasing but the cell A is still the best serving cell
of its layer (since it is on the border).
-
The pilot quality from A is the highest one
The pilot quality from A is lower than a user-definable minimum value (minimum Ec/I0) plus the Ec/I0 margin.
SB is the area where the cell B is the best serving cell of its layer.
-
The pilot quality from B exceeds a user-definable minimum value (minimum Ec/I0).
The pilot quality from B is the highest one.
Figure 6.13: Overlapping Zone for Inter-carrier Neighbours - 2nd Case

Note:
Two ways enable you to determine the I0 value:

1 - A reduction factor (% of maximum powers contributing to I0) may be applied to cell
maximum powers (defined in Cell properties) to customize their contribution to I0. Thus, I0
represents the sum of effective powers received from the other cells. The entered
percentage is a kind of downlink load factor estimation.
If the % of maximum powers contributing to I0 is too low, i.e. if % P max P pilot , Atoll
takes into account the pilot powers to evaluate the I0 value.
2 - Atoll takes into account load parameters defined per cell (such as the total downlink
power used). I0 represents the sum of total transmitted powers.
256
AT271_TRG_E6
Forsk 2009

SA SB
Atoll calculates the percentage of covered area ( ---------------------- 100 ) and compares this value to the % minimum covered
SA
area. If this percentage is not exceeded, the candidate neighbour B is discarded.
after).
4.
are 15 candidate neighbours and the maximum number of neighbours to be allocated to the reference cell is 8. Among
these 15 candidate neighbours, only 8 (having the highest importances) will be allocated to the reference cell. Note that
specific maximum numbers of neighbours (maximum number of intra-carrier neighbours, maximum number of inter-carrier
neighbours) can be defined at the cell level (property dialog or cell table). If defined there, this value is taken into account
instead of the default one available in the dialog.
to 100%.
Neighbourhood cause
When
Importance
value
Existing neighbour

Existing
importance
Exceptional pair
100 %
Co-site transmitter
Only if the Force co-site cells as neighbours option is selected
(IF) function
Only if the Force adjacent cells as neighbours option is

selected
(IF) function

coverage conditions
(IF) function
relationship
(IF) function
(IF) Importance Function evaluation. The importance is evaluated through a function (IF), taking into account 3 factors:
The co-site factor (C) which is a Boolean factor,

The adjacency factor (A) which deals with the percentage of adjacency,
The overlapping factor (O) meaning the percentage of overlapping
Factor
Min
importance
Default value
Max
importance
Default value

Min(O)
1%
Max(O)
30%
Min(A)
30%
Max(A)
60%
Co-site factor (C)
Min(C)
60%
Max(C)
100%
Neighbourhood cause
(IF) function

no
1% + 29%(O)
no
yes
30% + 30%{30%(O) + 70%(A)}
yes
yes
60% + 40%{30%(O) + 70%(A)}
Co-site
Adjacent
no
Where Delta(x)= Max(x) - Min(x)

Notes:
Forsk 2009
factors.
AT271_TRG_E6
257
accepted for co-site, adjacency and coverage reasons, Atoll displays the percentage of area meeting the coverage conditions and the corresponding surface area (km2), the percentage of area meeting the adjacency conditions and the corresponding surface area (km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.
Notes:
No simulation or prediction study is needed to perform an automatic neighbour allocation.

When starting an automatic neighbour allocation, Atoll automatically calculates the path
loss matrices if not found.
Even if no specific terminal, mobility or service is selected in the automatic allocation, it is

interesting to know that the algorithm works such as finding the maximum number of
neighbours by selection the multi-service traffic data as follows:
Service: selection of the one with the lowest body loss.
Mobility: no impact on the allocation, no specific selection.
Terminal: selection of the one with the greatest (Gain - Loss) value, and, if equal, the one
with the lowest noise figure.
The neighbour lists may be optionally used in the power control simulations to determine
the mobile's active set.
The percentage of area is calculated with the resolution specified in the properties dialog of
the predictions folder (default resolution parameter).

Thus, if the cell B is a neighbour of the cell A while the cell A is not a neighbour of the cell
B, two cases are possible:
1st case: There is space in the cell B neighbour list: the cell A will be added to the list. It will
be the last one.
2nd case: The cell B neighbour list is full: Atoll will not include cell A in the list and will
cancel the link by deleting cell B from the cell A neighbour list.
6.6.2
In the Results, Atoll displays only the cells for which it finds new neighbours. Therefore, if a
TBA cell has already reached its maximum number of neighbours before starting the new
allocation, it will not appear in the Results table.

Atoll allocates neighbours to:
TBA cells,
Neighbours of TBA cells marked as exceptional pair, adjacent and symmetric,
Neighbours of TBA cells that satisfy coverage conditions.
6.7
Primary Scrambling Code Allocation

Downlink primary scrambling codes enable you to distinguish cells from one another (cell identification).
By default, there are 512 primary scrambling codes numbered (0...511).
The cells to which Atoll allocates scrambling codes are referred to as the TBA cells (cells to be allocated). TBA cells fulfil
following conditions:
-
258
They are active,

AT271_TRG_E6
Forsk 2009

-

They belong to the folder on which allocation has been executed. This folder can be either the Transmitters
folder or a group of transmitters or a single transmitter.
Note:
6.7.1
Automatic Allocation Description
6.7.1.1
Options and Constraints

The scrambling code allocation algorithm can take into account following constraints and options:
1.
Neighbourhood between cells,
You may consider:
First order neighbours: The neighbours of TBA cells listed in the Intra-technology neighbours table,
Second order neighbours: The neighbours of neighbours,
Third order neighbours: The neighbours neighbours neighbours.
Notes:
2.
In the context of the primary scrambling code allocation, the term "neighbours" refers to
intra-carrier neighbours.
Atoll can take into account inter-technology neighbour relations as constraints to allocate
different scrambling codes to the UMTS neighbours of a GSM transmitter. In order to
consider inter-technology neighbour relations in the scrambling code allocation, you must
make the Transmitters folder of the GSM .atl document accessible in the UMTS .atl
document. For information on making links between GSM and UMTS .atl documents, see
the User Manual.
Atoll considers symmetry relationship between a cell, its first order neighbours, its second
order neighbours and its third order neighbours.
Cells fulfilling a criterion on Ec/Io (option Additional Overlapping Conditions),
For a reference cell A, Atoll considers all the cells B that can enter the active set on the area where the reference cell
is the best server (area where (Ec/Io)A exceeds the minimum Ec/Io and is the highest one and (Ec/Io)B is within a Ec/Io
margin of (Ec/Io)A).
Note:
3.
Atoll takes the total downlink power used by the cell into account in order to evaluate Io. Io
equals the sum of total transmitted powers. In case this parameter is not specified in the
cell properties, Atoll uses 50% of the maximum power.
Reuse distance,
Notes:
4.
5.
Reuse distance is a constraint on the allocation of scrambling codes. A code cannot be

reused at a cell that is not at least as far away as the reuse distance from the cell allocated
with the particular code.
Scrambling code reuse distance can be defined at cell level. If this value is not defined,
then Atoll will use the default reuse distance defined in the Scrambling Code Automatic
Allocation dialog.
Exceptional pairs,
Domains of scrambling codes,
Note:
6.
7.
8.
Forsk 2009
When no domain is assigned to cells, Atoll considers the 512 primary scrambling codes
available.
The number of primary scrambling codes per cluster. In Atoll, we call "cluster", a group of scrambling codes as
defined in 3GPP specifications. 3GPP specifications define 64 clusters consisting of 8 scrambling codes (in this
case, clusters are numbererd from 0 to 63). However, it is possible to define another value (e.g. if you set the
number of codes per cluster to 4, scrambling codes will be distributed in 128 clusters).
The carrier on which the allocation is run: It can be a given carrier or all of them. In this case, either Atoll
independently plans scrambling codes for the different carriers, or it allocates the same primary scrambling code
to each carrier of a transmitter if the option "Allocate carriers identically" is selected.
The possibility to use a maximum of codes from the defined domains (option "Use a Maximum of Codes"): Atoll
will try to spread the scrambling code spectrum the most.
AT271_TRG_E6
259

9.
The "Delete All Codes" option: When selecting this option, Atoll deletes all the current scrambling codes and
carries out a new scrambling code allocation. If not selected, the existing scrambling codes are kept.
In addition, it depends on the selected allocation strategy. Allocation strategies can be:
Clustered allocation: The purpose of this strategy is to choose for a group of mutually constrained cells, scrambling
codes among a minimum number of clusters. In this case, Atoll will preferentially allocate all the codes within the
same cluster.
Distributed per cell allocation: This strategy consists in using as many clusters as possible. Atoll will preferentially
allocate codes from different clusters.
One cluster per site allocation: This strategy allocates one cluster to each site, then, one code from the cluster to
each cell of each site. When all the clusters have been allocated and there are still sites remaining to be allocated,
Atoll reuses the clusters as far as possible at another site.
Distributed per site allocation: This strategy allocates a group of adjacent clusters to each site, then, one cluster
to each transmitter on the site according to its azimuth and finally, one code from the cluster to each cell of each
transmitter. The number of adjacent clusters per group depends on the number of transmitters per site you have
in your network; this information is required to start allocation based on this strategy. When all the groups of adjacent clusters have been allocated and there are still sites remaining to be allocated, Atoll reuses the groups of
adjacent clusters as far as possible at another site.
In the Results table, Atoll only displays scrambling codes allocated to TBA cells.
6.7.1.2
Allocation Process
For each TBA cell, Atoll lists all cells which have constraints with the cell. They are referred to as near cells. The near cells
of a TBA cell may be:
Its neighbour cells: the neighbours listed in the Intra-technology neighbours table (options Existing neighbours
and "First Order"),
The neighbours of its neighbours (options Existing neighbours and Second Order),
The third order neighbours (options Existing neighbours and Third Order),
The cells that fulfil Ec/Io condition (option Additional Overlapping Conditions),
The cells with distance from the TBA cell less than the reuse distance,
The cells that make exceptional pairs with the TBA cell.
Additional constraints are considered when:
The cell and its near cells are neighbours of a same GSM transmitter (only if the Transmitters folder of the GSM
.atl document is accessible in the UMTS .atl document),
The neighbour cells cannot share the same cluster (for the "Distributed per site" allocation strategy only).
These constraints have a certain weight taken into account to determine the TBA cell priority during the allocation process
and the cost of the scrambling code plan. During the allocation, Atoll tries to assign different scrambling codes to the TBA
cell and its near cells. If it respects all the constraints, the cost of the scrambling code plan is 0. When a cell has too many
constraints and there are not anymore scrambling codes available, Atoll breaks the constraint with the lowest cost so as
to generate the scrambling code plan with the lowest cost. For information on the cost generated by each constraint, see
"Cell Priority" on page 261.
6.7.1.2.1
Single Carrier Network

The allocation process depends on the selected strategy. Algorithm works as follows:
Strategies: Clustered and Distributed per Cell

Atoll processes TBA cells according to their priority. It allocates scrambling codes starting with the highest priority cell and
its near cells, and continuing with the lowest priority cells not allocated yet and their near cells. For information on calculating cell priority, see "Cell Priority" on page 261.
Strategy: One Cluster per Site

All sites which have constraints with the studied site are referred to as near sites.
Atoll assigns a cluster to each site, starting with the highest priority site and its near sites, and continuing with the lowest
priority sites not allocated yet and their near sites. When all the clusters have been allocated and there are still sites
remaining to be allocated, Atoll reuses the clusters at another site. When the Reuse Distance option is selected, the algorithm reuses the clusters as soon as the reuse distance is exceeded. Otherwise, when the option is not selected, the algorithm tries to assign reused clusters as spaced out as possible.
Then, Atoll allocates a primary scrambling code from the cluster to each cell located on the sites (codes belong to the
assigned clusters). It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not
allocated yet and their near cells.
For information on calculating site priority, see "Site Priority" on page 264. For information on calculating cell priority, see
Strategy: Distributed per Site

Atoll assigns a group of adjacent clusters to each site, starting with the highest priority site and its near sites, and continuing
with the lowest priority sites not allocated yet and their near sites. When all the groups of adjacent clusters have been
260
AT271_TRG_E6
Forsk 2009

allocated and there are still sites remaining to be allocated, Atoll reuses the groups of adjacent clusters at another site.
When the Reuse Distance option is selected, the algorithm reuses the groups of adjacent clusters as soon as the reuse
distance is exceeded. Otherwise, when the option is not selected, the algorithm tries to assign reused groups of adjacent
clusters as spaced out as possible. Then, Atoll assigns each cluster of the group to each transmitter of the site according
to the transmitter azimuth and selected neighbourhood constraints (options "Neighbours in Other Clusters" and "Secondary Neighbours in Other Clusters"). Then, Atoll allocates a primary scrambling code to each cell located on the transmitters
(codes belong to the assigned clusters). It starts with the highest priority cell and its near cells and goes on with the lowest
priority cells not allocated yet and their near cells.
Determination of Groups of Adjacent Clusters

In order to determine the groups of adjacent clusters to be used, Atoll proceeds as follows: It defines theoretical groups of
adjacent clusters, independently of the defined domain, considering the 512 primary scrambling codes available and the
specified number of codes per cluster (if this one is set to 8, 64 clusters are supposed to be available). It starts the division
in group from the cluster 0 (hard coded) and takes into account the maximum number of transmitters per site user-specified
in order to determine the number of clusters in each group and then, the number of possible groups.
Let us assume that the number of codes per cluster is set to 8 and the maximum number of transmitters per site in the
network is 3. In this case, we have the following theoretical groups:
Group 1
Group 2
Group 3
Group 4
...
Group 21
Cluster 0
Cluster 1
Cluster 2
Cluster 3
Cluster 4
Cluster 5
Cluster 6
Cluster 7
Cluster 8
Cluster 9
Cluster 10
Cluster 11
...
Cluster 61
Cluster 62
Cluster 63
If no domain is assigned to cells, Atoll can use all these groups for the allocation. On the other hand, if a domain is used,
the tool compares adjacent clusters really available in the assigned domain to the theoretical groups and only keeps adjacent clusters mapping the theoretical groups.
Let us assume that we have a domain consisted of 12 clusters: clusters 1 to 8 and clusters 12 to 15.
Therefore, Atoll will be able to use the following groups of adjacent clusters:
Group 2 with cluster 3, 4 and 5,

Group 3 with cluster 6, 7 and 8,
Group 6 with cluster 12, 13 and 14.
The clusters 1, 2 and 15 will not be used.
If a domain does not contain any adjacent clusters, the user is warned through the 'Event Viewer'.
6.7.1.2.2
Multi-Carrier Network
In case you have a multi-carrier network and you run the scrambling code allocation on all the carriers, the allocation process depends on the allocation strategy as detailed above and in addition, wether the option "Allocate Carriers Identically"
is selected or not.
When the option is not selected, algorithm works for each strategy, as explained above. On the other hand, when the
option is selected, allocation order changes. It is no longer based on the cell priority but depends on the transmitter priority.
All transmitters which have constraints with the studied transmitter will be referred to as near transmitters.
In case of a "Per cell" strategy (Clustered and Distributed per cell), Atoll starts scrambling code allocation with the highest
priority transmitter and its near transmitters and continues with the lowest priority transmitters not allocated yet and their
near transmitters. The same scrambling code is assigned to each cell of the transmitter.
In case of the "One cluster per site" strategy, Atoll assigns a cluster to each site and then, allocates a scrambling code to
each transmitter. It starts with the highest priority transmitter and its near transmitters and continues with the lowest priority
transmitters not allocated yet and their near transmitters. The same scrambling code is assigned to each cell of the transmitter.
In case of the "Distributed per site" strategy, Atoll assigns a group of adjacent clusters to each site, then a cluster to each
transmitter and finally, allocates a scrambling code to each transmitter. It starts with the highest priority transmitter and its
near transmitters and continues with the lowest priority transmitters not allocated yet and their near transmitters. The same
scrambling code is assigned to each cell of the transmitter.
For information on calculating transmitter priority, see "Transmitter Priority" on page 264.
Note:
When cells, transmitters or sites have the same priority, processing is based on an
alphanumeric order.
6.7.1.3
Priority Determination
6.7.1.3.1
Cell Priority
Scrambling code allocation algorithm in Atoll allots priorities to cells before performing the actual allocation. Priorities
assigned to cells depend upon how much constrained each cell is and the cost defined for each constraint. A cell without
Forsk 2009
AT271_TRG_E6
261

any constraint has a default cost, C , equal to 0. The higher the cost on a cell, the higher the priority it has for the scrambling
code allocation process.
There are six criteria employed to determine the cell priority:
Scrambling Code Domain Criterion
The cost due to the domain constraint, C i Dom , depends on the number of scrambling codes available for the allocation.
The domain constraint is mandatory and cannot be broken.
When no domain is assigned to cells, 512 scrambling codes are available and we have:
C i Dom = 0
When domains of scrambling codes are assigned to cells, each unavailable scrambling code generates a cost. The higher
the number of codes available in the domain, the less will be the cost due to this criterion. The cost is given as:
C i Dom = 512 Number of scrambling codes in the domain
Distance Criterion
The constraint level of any cell i depends on the number of cells (j) present within a radius of "reuse distance" from its
centre. The total cost due to the distance constraint is given as:
Cj Dist i
C i Dist =
Each cell j within the reuse distance generates a cost given as:
C j Dist i = w d ij c dis tan ce
Where
w d ij is a weight depending on the distance between i and j. This weight is inversely proportional to the inter-cell distance.
For a reuse distance of 2000m, the weight for an inter-cell distance of 1500m is 0.25, the weight for co-site cells is 1 and
the weight for two cells spaced out 2100m apart is 0.
c dis tan ce is the cost of the distance constraint. This value can be defined in the Constraint Cost dialogue.
Exceptional Pair Criterion
The constraint level of any cell i depends on the number of exceptional pairs (j) for that cell. The total cost due to exceptional pair constraint is given as:
C i EP =
cEP i j
j
Where
c EP is the cost of the exceptional pair constraint. This value can be defined in the Constraint Cost dialogue.
Neighbourhood Criterion
The constraint level of any cell i depends on the number of its neighbour cells j, the number of second order neighbours k
and the number of third order neighbours l.
Lets consider the following neighbour schema:
Figure 6.14: Neighbourhood Constraints

The total cost due to the neighbour constraint is given as:
Ci N =
Cj N1 i + Cj j N1 i + Ck N2 i + Ck k N2 i + Cl N3 i + Cl l N3 i
j
Each first order neighbour cell j generates a cost given as:

C j N1 i = I j c N1
Where
I j is the importance of the neighbour cell j.
c N1 is the cost of the first order neighbour constraint. This value can be defined in the Constraint Cost dialogue.
262
AT271_TRG_E6
Forsk 2009

Because two first order neighbours must not have the same scrambling code, Atoll considers the cost created by two first
order neighbours to be each other.
C j N1 i + C j N1 i
C j j N1 i = ---------------------------------------------------------2
Each second order neighbour cell k generates a cost given as:
C k N2 i = Max ( C j N1 i C k N1 j , C j N1 i C k N1 j ) c N2
Where
c N2 is the cost of the second order neighbour constraint. This value can be defined in the Constraint Cost dialogue.
Because two second order neighbours must not have the same scrambling code, Atoll considers the cost created by two
second order neighbours to be each other.
C k N2 i + C k N2 i
C k k N2 i = -----------------------------------------------------------2
Each third order neighbour cell l generates a cost given as:
C N1 i C k N1 j C l N1 k C j N1 i C k N1 j C l N1 k
C l N3 i = Max j
c N3
C j N1 i C k N1 j C l N1 k C j N1 i C k N1 j C l N1 k
Where
c N3 is the cost of the third order neighbour constraint. This value can be defined in the Constraint Cost dialogue.
Because two third order neighbours must not have the same scrambling code, Atoll considers the cost created by two third
C l N3 i + C l N3 i
C l l N3 i = ---------------------------------------------------------2
Note:
Atoll considers the highest cost of both links when a neighbour relation is symmetric and the
importance value is different.
In this case, we have:

C j N1 i = Max I i j I j i c N1
And
C k N2 i = Max (C j N1 i C k N1 j ,C j N1 k C i N1 j ) c N2
GSM Neighbour Criterion
This criterion is considered when the co-planning mode is activated (i.e. the Transmitters folder of the GSM .atl document
is made accessible in the UMTS .atl document) and inter-technology neighbours have been allocated. If the cell i is neighbour of a GSM transmitter, the cell constraint level depends on how many cells j are neighbours of the same GSM transmitter. The total cost due to GSM neighbour constraint is given as:
C i N 2G =
cN
2G
j Tx 2G
Where
cN
2G
is the cost of the GSM neighbour constraint. This value can be defined in the Constraint Cost dialogue.
Cluster Criterion
When the "Distributed per Site" allocation strategy is used, you can consider additional constraints on allocated clusters
(one cell, its first order neighbours and its second order neighbours must be assigned scrambling codes from different clusters). In this case, the constraint level of any cell i depends on the number of first and second order neighbours, j and k.
The total cost due to the cluster constraint is given as:
C i Cluster =
Cj N1 i cCluster + Ck N2 i cCluster
j
Where
c Cluster is the cost of the cluster constraint. This value can be defined in the Constraint Cost dialogue.
Therefore, the total cost due to constraints on any cell i is defined as:
C i = C i Dom + C i U
With
Forsk 2009
AT271_TRG_E6
263

C i U = C i Dist + C i EP + C i N + C i N 2G + C i Cluster
6.7.1.3.2
Transmitter Priority
In case you have a multi-carrier network and you run scrambling code allocation on "all" the carriers with the option "allocate carriers identically", algorithm in atoll allots priorities to transmitters. Priorities assigned to transmitters depend on how
much constrained each transmitter is and the cost defined for each constraint. The higher the cost on a transmitter, the
higher the priority it has for the scrambling code allocation process.
Let us consider a transmitter Tx with two cells using carriers 0 and 1. The cost due to constraints on the transmitter is given
as:
C Tx = C Tx Dom + C Tx U
With C Tx U =
Max C U and C Dom = 512 Number of scrambling codes in the domain

i
Tx
i Tx
Here, the domain available for the transmitter is the intersection of domains assigned to cells of the transmitter. The
domain constraint is mandatory and cannot be broken.
6.7.1.3.3
Site Priority
In case of "Per Site" allocation strategies (One cluster per site and Distributed per site), algorithm in Atoll allots priorities
to sites. Priorities assigned to sites depend on how much constrained each site is and the cost defined for each constraint.
The higher the cost on a site, the higher the priority it has for the scrambling code allocation process.
Let us consider a site S with three transmitters; each of them has two cells using carriers 0 and 1. The cost due to
constraints on the site is given as:
C S = C S U + C S Dom
With C S U =

Tx
S
Tx S
Here, the domain considered for the site is the intersection of domains available for transmitters of the site. The domain
constraint is mandatory and cannot be broken.
6.7.2
Allocation Examples
6.7.2.1
Allocation Strategies and Use a Maximum of Codes

In order to understand the differences between the different allocation strategies and the behaviour of algorithm when
using a maximum of codes or not, let us consider the following sample scenario:
Figure 6.15: Primary Scrambling Codes Allocation

Let Site0, Site1, Site2 and Site3 be four sites with 3 cells using carrier 0 whom scrambling codes have to be allocated out
of three clusters consisted of 8 primary scrambling codes. This implies that the domain of scrambling codes for the four
sites is from 0 to 23 (cluster 0 to cluster 2). The reuse distance is supposed to be less than the inter-site distance. Only
co-site neighbours exist.
The following section lists the results of each combination of options with explanation where necessary.
264
AT271_TRG_E6
Forsk 2009
6.7.2.1.1
Strategy: Clustered
Since the restrictions of neighbourhood only apply to co-sites with the same importance and sites distances are greater
than reuse distances, every cell has the same priority. Then, scrambling code allocation to cells is performed in an alphanumeric order.
Without Use a Maximum of Codes
With Use a Maximum of Codes
Atoll starts allocating the codes from the start of cluster 0 at As it is possible to use a maximum of codes, Atoll starts
each site.
allocation at the start of a different cluster at each site.
When a cluster is reused, and there are non allocated
codes left in the cluster, Atoll first allocates those codes
before reusing the already used ones.
6.7.2.1.2
Strategy: Distributed
than reuse distances, every cell has the same priority. Then, scrambling code allocation to cells is performed in an alphanumeric order.
Atoll allocates codes from different clusters to each cell of Atoll allocates codes from different clusters to each sites
the same site. Under given constraints of neighbourhood cells. As it is possible to use a maximum of codes, Atoll alloand reuse distance, same codes can be allocated to each cates the codes so that there is least repetition of codes.
sites cells.
Forsk 2009
AT271_TRG_E6
265
6.7.2.1.3
Strategy: One Cluster per Site

than reuse distances, every site has the same priority. Then, cluster allocation to sites is performed in an alphanumeric
order.
In this strategy, a cluster of codes is limited to be used at When it is possible to use a maximum of codes, Atoll can
just one site at a time unless all codes and clusters have allocate different codes from a reused cluster at another
been allocated and there are still sites remaining to be allo- site.
cated. In this case Atoll reuses the clusters as far as possible at another site.
6.7.2.1.4

than reuse distances, every site has the same priority. Then, the group of adjacent clusters allocation to sites is performed
in an alphanumeric order.
In this strategy, a group of adjacent clusters is limited to be When it is possible to use a maximum of codes, Atoll can
used at just one site at a time unless all codes and groups allocate different codes from a reused group of adjacent
of adjacent clusters have been allocated and there are still cluster at another site.
sites remaining to be allocated. In this case (here only one
group of adjacent clusters (clusters 0, 1 and 2) is available),
Atoll reuses the group at another site.
6.7.2.2
Allocate Carriers Identically

In order to understand the behaviour of algorithm when using the option "Allocate Carriers Identically" or not, let us
consider the following sample scenario:
Let Site0, Site1, Site2 and Site3 be four sites with 3 cells using carrier 0 and 3 cells using carrier 1. Scrambling codes have
to be allocated out of 3 clusters consisted of 8 primary scrambling codes. This implies that the domain of scrambling codes
for the five sites is from 0 to 23 (cluster 0 to cluster 2). The reuse distance is supposed to be less than the inter-site
266
AT271_TRG_E6
Forsk 2009

distance. Only co-site neighbours exist. Allocation algorithm will be based on the "One Cluster per Site" strategy and the
option "Use a Maximum of Codes" is selected.
Without Allocate Carriers Identically
With Allocate Carriers Identically
Atoll allocates one cluster at each site as detailed in the In this case, Atoll allocates one cluster at each site and
previous section. Then, it allocates a code from the cluster then, one code to each transmitter so as to use a maximum
to each cell of the site so as to use a maximum of codes. of codes. Then, the same code is given to each cell of the
transmitter.
In both cases (with and without Allocate Carriers Identically), every site has the same priority. Then, cluster allocation to
sites is performed in an alphanumeric order.
6.8
Automatic GSM-UMTS Neighbour Allocation
6.8.1
Overview
It is possible to automatically calculate and allocate neighbours between GSM and UMTS networks. In Atoll, it is called
inter-technology neighbour allocation.
Inter-technology handover is used in two cases:
When the UMTS coverage is not continuous. In this case, the UMTS coverage is extended by UMTS-GSM
handover into the GSM network,
And in order to balance traffic and service distribution between both networks.
Note that the automatic inter-technology neighbour allocation algorithm takes into account both cases.
In order to be able to use the inter-technology neighbour allocation algorithm, you must have:
An .atl document containing the GSM network, GSM.atl, and another one describing the UMTS network,
UMTS.atl,
An existing link on the Transmitters folder of GSM.atl into UMTS.atl.
The external neighbour allocation algorithm takes into account all the GSM TBC transmitters. It means that all the TBC
transmitters of GSM.atl are potential neighbours. The cells to be allocated will be called TBA cells which, being cells of
UMTS.atl, satisfy following conditions:
They are active,

They satisfy the filter criteria applied to Transmitters folder,
They belong to the folder for which allocation has been executed. This folder can be either the Transmitters folder
or a group of transmitters subfolder.
Only UMTS TBA cells may be assigned neighbours.
6.8.2

The allocation algorithm takes into account criteria listed below:
The inter-transmitter distance,

The maximum number of neighbours fixed,
Allocation options,
The selected allocation strategy,
Two allocation strategies are available: the first one is based on distance and the second one on coverage overlapping.
We assume we have a UMTS reference cell, A, and a GSM candidate neighbour, transmitter B.
Forsk 2009
AT271_TRG_E6
267
6.8.2.1
Algorithm Based on Distance

1.
The distance between the UMTS reference cell and the GSM neighbour must be less than the user-definable
maximum inter-site distance. If the distance between the UMTS reference cell and the GSM neighbour is greater
than this value, then the candidate neighbour is discarded.
Candidate neighbours are sorted in descending order with respect to distance.

Note:
2.
Transmitter azimuths are taken into account to evaluate the inter-transmitter distance (for
further information on inter-transmitter distance calculation, please refer to "Calculation of
Inter-Transmitter Distance" on page 270)
Force co-site cells as neighbours: It enables you to automatically include GSM transmitters located on the same site as
the reference UMTS cell in the candidate neighbour list. This option is automatically selected.
force/forbid a GSM transmitter to be candidate neighbour of the reference UMTS cell.
and carries out a new neighbour allocation. If not selected, existing neighbours are kept.
3.
Atoll lists all candidate neighbours and sorts them by priority so as to eliminate some of them from the neighbour
list if the maximum number of neighbours to be allocated to each cell is exceeded. The candidate neighbour
priority depends on the neighbourhood cause. Priority assigned to each neighbourhood cause is listed in the table
below (1 is a higher priority than 2 and so on).
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

distance conditions
Only if the Max inter-site distance is not exceeded
If there are 15 candidate neighbours and the maximum number of neighbours to be allocated to the reference cell is 8.
Among 15 candidate neighbours, only 8 (those with the highest priority) will be allocated to the reference cell.
In the Results part, Atoll provides the list of neighbours, the number of neighbours and the maximum number of neighbours allowed for each cell. In addition, it indicates the allocation cause for each neighbour. Therefore, a neighbour may
be marked as exceptional pair or co-site. If the neighbour is not forced but fulfils distance conditions, Atoll displays the
distance from the reference cell. Finally, if cells have previous allocations in the list, neighbours are marked as existing.
6.8.2.2
Algorithm Based on Coverage Overlapping

1.
The distance between the UMTS reference cell and the GSM neighbour must be less than the user-definable
maximum inter-site distance. If the distance between the UMTS reference cell and the GSM neighbour is greater
Note:
2.
Here, real inter-transmitter distance is considered.
the reference UMTS cell in the candidate neighbour list. This option is automatically selected.
force/forbid a GSM transmitter to be candidate neighbour of the reference UMTS cell.
3.
There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability.
Four different cases may be considered for SA:
268
AT271_TRG_E6
Forsk 2009

-
1st case: SA is the area where the cell A is the best serving cell of the UMTS network.
- The pilot signal received from A is greater than the minimum pilot signal level,
- The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0) and is the highest one.
In this case, the Ec/I0 margin must be equal to 0dB and the max Ec/Io option disabled.
2nd case: SA represents the area where the pilot quality from the cell A strats decreasing but the cell A is still
the best serving cell of the UMTS network.
The Ec/I0 margin must be equal to 0dB, the max Ec/Io option selected and a maximum Ec/Io user-defined.
-
The pilot signal received from A is greater than the minimum pilot signal level,
The pilot quality from A exceeds the minimum Ec/I0 but is lower than the maximum Ec/Io.
The pilot quality from A is the highest one.
3rd case: SA represents the area where the cell A is not the best serving cell but can enter the active set.
Here, the Ec/I0 margin has to be different from 0dB and the max Ec/Io option disabled.
-
The pilot quality from A is within a margin from the best Ec/Io, where the best Ec/Io exceeds the minimum
Ec/Io.
4th case: SA represents the area where:

- The pilot quality from A is within a margin from the best Ec/Io (where the best Ec/Io exceeds the minimum
Ec/Io) and lower than the maximum Ec/Io.
In this case, the margin must be different from 0dB, the max Ec/Io option selected and a maximum Ec/Io userdefined.
Two different cases may be considered for SB:

-
1st case: SB is the area where the cell B is the best serving cell of the GSM network.
In this case, the margin must be set to 0dB.
-
The signal level received from B on the BCCH TRX type exceeds the user-defined minimum threshold and
is the highest one.
2nd case: The margin is different from 0dB and SB is the area where:
- The signal level received from B on the BCCH TRX type exceeds the user-defined minimum threshold and
is within a margin from the best BCCH signal level.
SA SB
SA
Candidate neighbours fulfilling coverage conditions are sorted in descending order with respect to percentage of covered
area.
Guidelines for the automatic allocation

When the automatic allocation is based on coverage overlapping, we recommend you to perform two successive automatic allocations:
- A first allocation in order to find handovers due to non-continuous UMTS coverage. In this case, you have to select the
max Ec/I0 option and define a high enough value.
- A second allocation in order to complete the previous list with handovers motivated for reasons of traffic and service
distribution. Here, the max Ec/I0 option must be disabled.
4.
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

coverage conditions
Therefore, among 15 candidate neighbours, only 8 (those with the highest priority) will be allocated to the reference cell.
Forsk 2009
AT271_TRG_E6
269

be marked as exceptional pair or co-site. If the neighbour is not forced but fulfils coverage conditions, Atoll displays the
percentage of covered area and the overlap area (km2) in brackets. Finally, if cells have previous allocations in the list,
neighbours are marked as existing.
Notes:

not found.
The percentage of covered area is calculated with the resolution specified in the properties
dialog of the predictions folder (default resolution parameter).

6.8.2.3
Appendices
6.8.2.3.1
Delete Existing Neighbours Option

As explained above, Atoll keeps the existing inter-technology neighbours when the Delete existing neighbours option is
not checked. We assume that we have an existing allocation of inter-technology neighbours.
A new TBA cell i is created in UMTS.atl. Therefore, if you start a new allocation without selecting the Delete existing neighbours option, Atoll determines the neighbour list of the cell i,
If you change some allocation criteria (e.g. increase the maximum number of neighbours or create a new GSM TBC transmitter) and start a new allocation without selecting the Delete existing neighbours option, it examines the neighbour list of
TBA cells and checks allocation criteria if there is space in their neighbour lists. A new GSM TBC transmitter can enter the
TBA cell neighbour list if allocation criteria are satisfied. It will be the first one in the neighbour list.
6.8.2.3.2
Calculation of Inter-Transmitter Distance

When allocation algorithm is based on distance, Atoll takes into account the real distance and azimuths of antennas in
order to calculate the effective inter-transmitter distance.
Dist CellA CellB = D 1 + x cos x cos
where x = 0.5% so that the maximum D variation not to exceed 1% and D is stated in m.
Figure 6.16: Inter-Transmitter Distance Computation

The formula above implies that two cells facing each other will have a smaller effective distance than the real physical
distance. It is this effective distance that will be taken into account rather than the real distance.
Note:
270
This formula is not used when allocation algorithm is based on coverage overlapping. In
this case, real inter-transmitter distance is considered.
AT271_TRG_E6
Forsk 2009
Chapter 7
IS-95 cdmaOne and
CDMA2000 Networks
simulations and prediction studies available in IS-95 cdmaOne and CDMA2000 projects.
Atoll
272
AT271_TRG_E6
Forsk 2009
Chapter 7: IS-95 cdmaOne and CDMA2000 Networks
IS-95 cdmaOne and CDMA2000 Networks
7.1
7.1.1
Three criteria can be studied in point analysis (Profile tab) and in common coverage studies. Study criteria are detailed in
the table below:
Study criteria
Formulas
Signal level ( P rec ) in dBm

P rec ic = EIRP ic L path M Shadowing model L Indoor + G term L term
Path loss ( L path ) in dBm

Total losses ( L total ) in dBm
Tx
L total = L path + L Tx + L term + L indoor + M Shadowing model G Tx + G term
where,
ic is a carrier number,
L ant
Tx
L term are the receiver losses,
G term is the receiver antenna gain,
G Tx is the transmitter antenna gain,
L Tx is the transmitter loss ( L Tx = L total DL ). For information on calculating transmitter loss, "UMTS HSPA, CDMA2000
1xRTT 1xEV-DO, IS-95 cdmaOne, and TD-SCDMA Documents" on page 124.
Notes:
For IS-95 cdmaOne and CDMA2000 1xRTT systems, EIRP ic = P pilot ic + G Tx L Tx

(where, P pilot ic is the cell pilot power).
For CDMA2000 1xEV-DO systems, EIRP ic = P max ic + G Tx L Tx (where P max ic

is the maximum cell power).
It is also possible to analyse all the carriers at once. In this case, Atoll displays the best
signal level received from a transmitter. Therefore, if the network consists of 1xRTT and
1xEV-DO carriers, Atoll takes the highest power of both cells for each transmitter (i.e. the
highest value between the pilot power of the 1xRTT cell and the maximum power of the
1xEV-DO cell) to calculate the received signal level.
Atoll considers that G term and L term equal zero.
7.1.2
Point Analysis
7.1.2.1
Profile Tab
Atoll displays either the signal level received from the selected transmitter on a carrier ( P rec ic ), or the highest signal
level received from the selected transmitter on all the carriers.
Forsk 2009
AT271_TRG_E6
273

Note:
7.1.2.2
Reception Tab
For each transmitter, Atoll displays either the signal level received on a carrier, ( P rec ic ), or the highest signal level
received on all the carriers.
Note:
7.1.3
Coverage Studies
For each TBC transmitter, Txi, Atoll determines the selected criterion on each bin inside the Txi calculation area. In fact,
each bin within the Txi calculation area is considered as a potential (fixed or mobile) receiver.
7.1.3.1

We can distinguish three cases:
7.1.3.1.1
All Servers
Txi
Txi
Txi
Minimum threshold P rec ic or L total or L path Maximum threshold
7.1.3.1.2

Txi
Txi
Txi

And
Txi
Txj

ji

Notes:
the highest.
7.1.3.1.3

274
AT271_TRG_E6
Forsk 2009

Txi
Txi
Txi

And
Txi
P rec ic 2
nd
Txj
Best P rec ic M
ji

Notes:
the second highest.
7.1.3.2
Coverage Display
7.1.3.2.1
Plot Resolution
7.1.3.2.2
Display Types
such as:

Atoll calculates signal level received from the transmitter on each bin of each transmitter service area. A bin of a service
area is coloured if the signal level exceeds ( ) the defined minimum thresholds (bin colour depends on signal level).

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. Where other service
areas overlap the studied one, Atoll chooses the highest value. A bin of a service area is coloured if the signal level
exceeds ( ) the defined thresholds (the bin colour depends on the signal level). Coverage consists of several independent
layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer
corresponds to an area where the signal level from the best server exceeds a defined minimum threshold.
Path Loss (dB)

Atoll calculates path loss from the transmitter on each bin of each transmitter service area. A bin of a service area is
coloured if path loss exceeds ( ) the defined minimum thresholds (bin colour depends on path loss). Coverage consists
Total Losses (dB)

Atoll calculates total losses from the transmitter on each bin of each transmitter service area. A bin of a service area is
coloured if total losses exceed ( ) the defined minimum thresholds (bin colour depends on total losses). Coverage

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. Where other service
bin of a service area is coloured if the path loss exceeds ( ) the defined thresholds (bin colour depends on path loss).
Forsk 2009
AT271_TRG_E6
275

Atoll calculates signal levels received from transmitters on each bin of each transmitter service area. Where service areas
overlap the studied one, Atoll determines the best transmitter and evaluates total losses from the best transmitter. A bin
of a service area is coloured if the total losses exceed ( ) the defined thresholds (bin colour depends on total losses).
layers as defined thresholds. Each layer corresponds to an area where the total losses from the best server exceed a
Number of Servers
Atoll evaluates how many service areas cover a bin in order to determine the number of servers. The bin colour depends
on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the number of servers exceeds ( ) a defined minimum threshold.

On each bin of each transmitter service area, the coverage corresponds to the pixels where the signal level from this transmitter fulfils signal conditions defined in Conditions tab with different Cell edge coverage probabilities. There is one coverage area per transmitter in the explorer.

On each bin of each transmitter service area, the coverage corresponds to the pixels where the best signal level received
explorer.
7.2
7.2.1
Parameters Used for IS-95 cdmaOne and CDMA2000 1xRTT

Modelling
7.2.1.1
Inputs
etc.)
Name
Value
Unit
Description
F ortho
None
F MUD
Tx
None
MUD factor
ic
Global parameter
None
Carrier number
req
Q pilot txi ic + Q pilot
min
req
req
None
Active set upper threshold

(used to determine the best server in
the active set)
min
min
None
Active set lower threshold

(used to determine other members of
the active set)
req
Min. Ec/I0 - Cell parameter
None
Minimum Ec/I0 required from the cell

to be the best server in the active set
min
T_Drop - Cell parameter
None

not to be rejected from the active set
req
Delta Min. Ec/I0 - Mobility parameter
None
Variation of the minimum Ec/I0

required from the cell to be the best
server in the active set
min
Delta T_Drop - Mobility parameter
None

required from the cell not to be
rejected from the active set
None
Eb/Nt target for FCH channel on

downlink
Q pilot
Q pilot
Q pilot txi ic
Q pilot txi ic
Q pilot
Q pilot
DL
Q req FCH
276
E
-----b-
N t req
FCH DL
(Service, Mobility) parameter
AT271_TRG_E6
Forsk 2009

E
-----b-
N t req
SCH DL
DL
Q req SCH
None
Eb/Nt target for SCH channel on

downlink
None
Eb/Nt target for FCH channel on

uplink
None
Eb/Nt target for SCH channel on

uplink
Site parameter
None

for a site on uplink
(Service, Mobility, SCH rate multiple) parameter

UL
Q req FCH
E
-----b-
N t req
FCH UL
(Service, Terminal) parameter

E
-----b-
N t req
SCH UL
UL
Q req SCH
(Service, Terminal, SCH rate multiple) parameter

CE U L
N max N I
CE D L
NI
Site parameter
None

for a site on downlink
CE U L
NI
Simulation result
None

consumed by users on uplink
CE D L
NI
Simulation result
None

consumed by users on downlink
N max
Overhead C E UL
None

uplink
Overhead C E DL
None

downlink
FCH C E UL
(Terminal, site equipment) parameter
None

FCH on uplink
FCH C E DL
None

FCH on downlink
None
Maximum number of Walsh codes

available per cell (128)
Simulation result
None
Number of Walsh codes used by the

cell
NF term
Terminal parameter
None
NF Tx

None
1.38 10-23
J/K
Boltzman constant
293
Ambient temperature
1.23 MHz
Hz
Spreading Bandwidth
RF ic ic adj
Network parameter
None

Maximum percentage of used power
Tx
NF Tx K T W
Term
NF Term K T W
Rc
bps
Chip rate
f rake efficiency
UL
Equipment parameter
None
DL
Terminal parameter
None
Downlink rake receiver efficiency

factor
Frate SCH
Simulation result
None
SCH rate factor (drawn following the

SCH probabilities of the service)
R FCH
DL
Terminal parameter
bps
Downlink FCH nominal rate
DL
R FCH Frate SCH
bps
Downlink SCH bit rate
Simulation result
None
SCH rate factor (drawn following the

SCH probabilities of the service)
Codes
N max txi ic
N
Codes
txi ic
UL
X max
DL
%Power max
N0
N0
f rake efficiency
DL
R SCH
UL
Frate SCH
Forsk 2009
DL
DL
AT271_TRG_E6
277

R FCH
UL
Terminal parameter
UL
bps
Uplink FCH nominal rate
R FCH Frate SCH
bps
Uplink SCH bit rate
W
-------------DL
R FCH
None
Downlink service processing gain on

FCH
W
-------------DL
R SCH
None
Downlink service processing gain on

SCH
W ------------UL
R FCH
None
Uplink service processing gain on

FCH
W
-------------UL
R SCH
None

SCH
DL
Service parameter
None
Downlink activity factor on FCH
AF FCH
UL
Service parameter
None
Uplink activity factor on FCH
P Sync txi ic
Cell parameter
Cell synchronisation channel power
P paging txi ic
Cell parameter
Cell other common channels (except

CPICH and SCH) power
P pilot txi ic
Cell parameter
Cell pilot power
P max txi ic
Cell parameter
Maximum cell power
M pooling txi ic
Cell parameter
dB
Maximum amount of power reserved

for pooling
P FCH
min
Service parameter
Minimum power allowed for FCH
P FCH
max
Service parameter
Maximum power allowed for FCH
P SCH
min
Service parameter
Minimum power allowed for SCH
P SCH
max
Service parameter
Maximum power allowed for SCH
P FCH txi ic tch
Simulation result including the term AFFCH Serv
Cell FCH power for a traffic channel

on carrier ic
Total FCH power on carrier ic
Simulation result
Transmitter SCH power for a traffic

channel on carrier ic
Total SCH power on carrier ic
Transmitter total transmitted power on

carrier ic
R SCH
UL
FCH DL
Gp
SCH DL
Gp
FCH UL
Gp
SCH UL
Gp
AF FCH
P FCH txi ic
UL
DL
P FCH txi ic tch
tch FCH ic
P SCH txi ic tch

P SCH txi ic
P SCH ic tch
tch SCH ic
P tx txi ic
+ P FCH txi ic
P term
min
Terminal parameter
max
Terminal parameter
P term
FCH
Simulation result including the term AFFCH Serv
Terminal FCH power transmitted in

carrier ic
P term ic
SCH
Simulation result
Terminal SCH power transmitted on

carrier ic
BTS
BTS parameter

transmitted
term
Terminal parameter

transmitted
Clutter parameter

finger
P term ic
278
P pilot txi ic + P Sync txi ic + P paging txi ic + P SCH txi ic
UL
AT271_TRG_E6
Forsk 2009

G Tx
Antenna parameter
None
G Term
Terminal parameter
None
Terminal gain
L Tx

None
Transmitter lossa
L body
Service parameter
None
Body loss
L Term
Terminal parameter
None
Terminal loss
L indoor
Clutter parameter
L path
None
Path loss
Terminal parameter
None
Number of fingers
Terminal parameter
Pilot power percentage
M Shadowing model

standard deviation
None

M Shadowing Ec Io
Result calculated from cell edge coverage probability and Ec/I0

standard deviation
None
Ec/I0 Shadowing margin

DL
Indoor loss
npaths
DL
G macro diversity
None
n=2 or 3
M Shadowing Eb Nt
DL
M Shadowing Eb Nt
UL
pilot signals at the mobile b.

standard deviation
None
DL Eb/Nt Shadowing margin


standard deviation
None

UL
UL
G macro diversity
npaths
UL
M Shadowing Eb Nt
n=2 or 3
E Shadowing
UL
None

in soft handoffc.
None

None
P pilot txi ic
--------------------------------LT
Chip power received at terminal
Simulation result
In prediction studiesd
For Ec/I0 calculation
--------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
LT

DL
-----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
-----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
In simulations
-------------------------------------------------------------------------------------------------------------------------------G Tx G term
P c txi ic
FCH DL
txi ic tch
P FCH txi ic tch

---------------------------------------------LT
Bit received power at terminal for FCH

on carrier ic
SCH DL
txi ic tch
P SCH txi ic tch

---------------------------------------------LT
Bit received power at terminal for SCH

on carrier ic
Bit received power at terminal for

FCH+SCH on carrier ic
Total received power at terminal from


traffic channels of a transmitter on
carrier ic
Pb
Pb
DL
P b txi ic tch
FCH DL
Pb
DL
Forsk 2009
txi ic tch
P tx txi ic
---------------------------LT
DL
P tot txi ic
P traf txi ic
SCH DL
txi ic tch + P b
tch ic
P FCH txi ic + P SCH txi ic

--------------------------------------------------------------------------LT
AT271_TRG_E6
279

FCH
FCH UL
ic
P term
-------------LT
SCH UL
ic
P term
-------------LT
Pb
Pb
Bit received power at transmitter for

FCH on carrier ic

SCH on carrier ic
SCH
UL
Pb
ic

SCH+FCH on carrier ic
UL
P b ic
UL
UL
P b ic + P c ic = ------------------1 p
Total power transmitted by the

UL
p P tot ic
Chip received power at transmitter
FCH UL
P b ic
SCH UL
ic + P b
UL
P tot ic
UL
P c ic
a.
L Tx = L total UL on uplink and
L Tx = L total DL on downlink. For information on calculating transmitter
losses on uplink and downlink, see "UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, IS-95 cdmaOne, and TD-SCDMA
Documents" on page 124.
b.
npaths
density function (n paths) in case of downlink Ec/I0 modelling.

c.
npaths
M Shadowing Eb Nt
UL

d.
UL
the shadowing model ( M Shadowing Eb Nt

M Shadowing Eb Nt
7.2.1.2
DL
DL
or M Shadowing Ec Io ) while extra-cell interference level is not. Therefore,
or M Shadowing Ec Io is set to 1 in downlink extra-cell interference calculation.
Ec/I0 Calculation
This table details the pilot quality ( Q pilot or Ec Io ) calculations.
Name
Value
DL
DL
I intra txi ic
P tot txi ic
DL
I extra ic
DL
P tot txj ic
Unit
Description




carrier ic
None
Quality level at terminal on pilot for

carrier ic
txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
DL
I 0 ic
Ec
Q pilot txi ic ------
I0
7.2.1.3
DL
DL
DL
Term

BTS P c txi ic
------------------------------------------------------DL
I 0 ic
DL Eb/Nt Calculation
Eb
DL
This table details calculations of downlink traffic channel quality ( Q tch (tch could be FCH or SCH) or ------- ).
Nt DL
Name
DL
I intra txi ic
DL
I extra ic
Value
Unit
Description
1 BTS F ortho P DL txi ic

tot



DL
P tot txj ic
txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
280
AT271_TRG_E6
Forsk 2009

DL
DL
N tot ic
DL
Term
DL

carrier ic

FCH DL
Pb
txi
DL
Q FCH txi
BTS
ic tch
DL
--------------------------------------------------------------------------------------------------------------- G FCH
p
DL
DL
E b DL
ic ------
N t FCH
None
FCH channel on carrier ic a
FCH DL
BTS P b
txi ic tch
DL
- G FCH
Total noise: --------------------------------------------------------------------------p
DL
N tot ic
DL
f rake efficiency
DL
Q FCH ic
DL
Q FCH tx k ic

channel from one transmitter for a
None
txk ActiveSet FCH
Quality level at terminal for FCH using

carrier ic due to combination of all

SCH DL
Pb
txi
DL
Q SCH txi
BTS
ic tch
DL
--------------------------------------------------------------------------------------------------------------- G SCH
p
DL
DL
E b DL
ic ------
N t SCH
None
SCH channel on carrier icb
SCH DL
BTS P b
txi ic tch
DL
- G SCH
Total noise: --------------------------------------------------------------------------p
DL
N tot ic
DL
f rake efficiency
DL
Q SCH ic
DL
Q SCH tx k ic
txk ActiveSet SCH
Quality level at terminal for SCH using

carrier ic due to combination of all
None
DL
Q FCH ic
------------------------------------------------------------DL
Q FCH BestServer ic
DL
G SHO FCH
None
Downlink soft handover gain for FCH

None
Downlink soft handover gain for SCH

Required transmitter FCH traffic

channel power to achieve Eb/Nt target
at terminal on carrier ic
Required transmitter SCH traffic

channel power to achieve Eb/Nt target
Required transmitter traffic channel

power on carrier ic
DL
Q SCH ic
------------------------------------------------------------DL
Q SCH BestServer ic
DL
G SHO SCH
DL
Q req FCH
--------------------------- P FCH txi ic
DL
Q FCH ic
req
P FCH txi ic
DL
Q req SCH
--------------------------- P SCH txi ic
DL
Q SCH ic
req
P SCH txi ic
req
req
P tch txi ic
req
P FCH txi ic + P SCH txi ic

channel from one transmitter for a
a.
b.
7.2.1.4
Eb
UL
This table details calculations of uplink traffic channel quality ( Q tch (tch could be FCH or SCH) or ------- ).
Nt UL
Name
UL intra
I tot
Value
Pb
UL
txi ic
UL extra
I tot
txi
ic
UL
term
txj j i
Pb
ic

ic

ic


UL
P b ic + P c ic
UL
UL
I inter carrier txi
Description
UL
ic + P c ic
term
txi
Unit
UL
ic adj + P c ic adj
term
txj
j
-----------------------------------------------------------------------------
RF ic ic adj
UL
I tot txi ic
Forsk 2009
UL extra
I tot
Tx
UL intra
UL
AT271_TRG_E6
281

UL
UL
N tot txi ic
tx
I tot txi ic + N 0


FCH UL
Eb
UL
Q FCH txi ic ------
N t UL
term P b
ic
UL
--------------------------------------------------------------------------------------------------------------- G FCH
p
UL
Tx
UL
None

channel for the FCH channel on
carrier ica
FCH UL
term P b
ic
UL
- G FCH
Total noise: ----------------------------------------------------p
UL
N tot txi ic
SCH UL
Eb
UL
Q SCH txi ic ------
N t UL
term P b
ic
UL
--------------------------------------------------------------------------------------------------------------- G SCH
p
UL
Tx
UL
None

channel for the SCH channel on
carrier icb
SCH UL
term P b
ic
UL
- G SCH
Total noise: -----------------------------------------------------p
UL
N tot txi ic
UL
No HO: Q tch txi ic
UL
Softer HO: f rake efficiency
UL
Q tch tx k ic
tx k ActiveSet
samesite
Soft, Softer/Soft HO (No MRC):

UL
UL

tx ActiveSet
k
UL
Q tch ic
Softer/Soft HO (MRC):

and taking into account increase of the
None quality due to macro-diversity (macrodiversity gain).
tch could be FCH or SCH
Max UL
UL
UL
f rake efficiency Q tch tx k ic Q tch tx l ic
tx ,tx ActiveSet
k l
txk samesite
tx k
UL
In simulations, G macro diversity = 1 .
tx othersite
l
UL
G macro diversity
UL
UL
G SHO FCH
Q FCH ic
------------------------------------------------------------UL
Q FCH BestServer ic
None
Uplink soft handover gain for FCH

None
Uplink soft handover gain for SCH


Eb/Nt target at transmitter for FCH on
carrier ic

Eb/Nt target at transmitter for SCH on
carrier ic
Required terminal power on carrier ic
UL
UL
G SHO SCH
Q SCH ic
------------------------------------------------------------UL
Q SCH BestServer ic
UL
FCH req
ic
Q req FCH
--------------------------- P FCH
term ic
UL
Q FCH ic
SCH req
ic
Q req SCH
--------------------------- P SCH
term ic
UL
Q SCH ic
P term
UL
P term
req
P term ic
FCH req
P term
SCH req
ic + P term
ic
a.
b.
7.2.1.5
Simulation Results
This table contains some simulation results provided in the Cells and Mobiles tabs of the simulation property dialog.
Name
DL
I intra txi ic
Value
Unit
Description
DL
DL
P tot txi ic F ortho BTS P tot txi ic
None

DL
1 F ortho BTS P b txi ic
282
AT271_TRG_E6
Forsk 2009
DL
I extra ic
DL
P tot txj ic


txj j i
Ptot txj icadj

DL
DL
I inter carrier ic
txj
j
------------------------------------------------
RF ic ic adj
DL
DL
I tot ic
DL
N tot ic
UL intra
I tot
DL
DL
Total effective interference at terminal

on carrier ic (after unscrambling)
DL
Term

carrier ic

ic

ic



None
I intra ic + I extra ic + I inter carrier ic

I tot ic + N 0
Pb
UL
txi ic
UL extra
I tot
txi
term
txi
ic
UL
UL
P b ic + P c ic
term
txj j i
Pb
UL
UL
I inter carrier txi
UL
ic + P c ic
ic
UL
ic adj + P c ic adj
term
txj
j
-----------------------------------------------------------------------------
RF ic ic adj
UL
I tot txi ic
UL extra
I tot
UL intra
Tx
UL
UL
N tot txi ic
UL
tx
I tot txi ic + N 0
UL
UL
txi ic
I tot txi ic
----------------------------UL
N tot txi ic
UL
txi ic
I tot txi ic
------------------------------------------------------------------------------------------UL intra
Tx
I tot
txi ic 1 F MUD term
None
UL
txi ic
1
-----------------------------UL
F txi ic
None

carrier ic
P tx txi ic
-------------------------------- 100
P max txi ic
None
Percentage of max transmitter power

used.
None
Downlink load factor on carrier ic
None
Downlink reuse factor on a carrier ic
UL
%Power
DL
txi ic
Simulation result available per cell

DL
I extra ic
DL
+ I inter carrier ic L T
--------------------------------------------------------------------------------------- + 1 F ortho BTS
P tx txi ic
-----------------------------------------------------------------------------------------------------------------------------------------1
------------- + 1 F ortho BTS
tch
DL
CI req
DL
txi ic
with
DL
CI req
SCH DL
FCH DL
Q req
Q req
= ------------------------ + -----------------------SCH DL
FCH DL
Gp
Gp
DL
I tot ic
Simulation result available per mobile: -------------------DL
N tot ic
DL
DL
I tot ic
------------------------------DL
I intra txi ic
txi ic
NR
DL
txi ic
10 log 1 X
DL
txi ic
dB
NR
UL
txi ic
10 log 1 X
UL
txi ic
dB
7.2.2
Parameters Used for CDMA2000 1xEV-DO Modelling
7.2.2.1
Inputs
etc.)
Forsk 2009
AT271_TRG_E6
283
Name
Value
Unit
Description
F ortho
None
F MUD
Tx
None
MUD factor
ic
Global parameter
None
Carrier number
req
min
req
req
None
Active set upper threshold

(used to determine the best server in
the active set)
min
min
None
Active set lower threshold

(used to determine other members of
the active set)
req
Min. Ec/I0 - Cell parameter
None

to be the best server in the active set
min
T_Drop - Cell parameter
None

not to be rejected from the active set
req
Delta Min. Ec/I0 - Mobility parameter
None

required from the cell to be the best
server in the active set
Q pilot
min
Delta T_Drop - Mobility parameter
None

required from the cell not to be
rejected from the active set
E
-----c-
N t min
Mobility parameter for 1xEV-DO Rev. 0 users

Parameter read in the 1xEV-DO Rev. A Radio Bearer Selection
(Uplink) table for 1xEV-DO Rev. A users
None
Minimum pilot quality level on uplink
n SF
1xEV-DO Rev. A Radio Bearer Selection (Uplink) table
None
Number of subframes associated to

uplink 1xEV-DO Rev. A bearer
R RLC peak
Uplink 1xEV-DO Rev. A Radio Bearer table
None
Uplink RLC peak rate provided by the

1xEV-DO Rev. A bearer
E
-----c-
N t min
Mobility parameter for 1xEV-DO Rev. 0 users

Parameter read in the 1xEV-DO Rev. A Radio Bearer Selection
(Downlink) table for 1xEV-DO Rev. A users
None
Minimum pilot quality level required to

obtain a data rate on downlink
n TS
1xEV-DO Rev. A Radio Bearer Selection (Downlink) table
None
Number of timeslots associated to

downlink 1xEV-DO Rev. A bearer
Downlink 1xEV-DO Rev. A Radio Bearer Table
None
Downlink RLC peak rate provided by

the 1xEV-DO Rev. A bearer
Q pilot
Q pilot
Q pilot txi ic
Q pilot txi ic
Q pilot
UL
UL
DL
DL
R RLC peak
EVDO CE
NI
Site parameter
None
Number of EVDO channel elements

available for a site on uplink and
downlink
EVDO CE
NI
Simulation result
None
Total number of EVDO channel

elements of a site consumed by users
on uplink and downlink
None

TCH on uplink
N max
TCH C E UL
MacIndexes
txi ic
None
Maximum number of MAC indexes

available per cell (59)
MacIndexes
txi ic
Simulation result
None
Number of MAC indexes used by the

cell
Simulation constraint (cell parameter)
None
Maximum number of EVDO users that

can be connected to the cell
Simulation result
None
Number of EVDO users connected to

the cell
NF term
Terminal parameter
None
NF Tx

None
1.38 10-23
J/K
Boltzman constant
293
Ambient temperature
1.23 MHz
Hz
Spreading Bandwidth
RF ic ic adj
Network parameter
None

N max
N
EVDO
n max txi ic
n
284
EVDO
txi ic
AT271_TRG_E6
Forsk 2009

UL
Tx
NF Tx K T W
Term
NF Term K T W
Rc
bps
Chip rate
f rake efficiency
Equipment parameter
None
Simulation result
bps
Uplink data rate
R TCP ACK
Simulation result
bps
Uplink data rate due to TCP

aknowledgements
R BCMCS
Cell parameter
bps
Downlink data rate for Broadcast/

Multicast services
DL
Simulation result
bps
Downlink maximum data rate supplied

to the terminal
DL
Simulation result
bps
Downlink average cell data rate
X max
N0
N0
UL
UL
UL
R max
R avg
DL
DL
R application
SF rate R max R
bps
Downlink user application throughput
SF Rate
Service parameter
Scaling factor
Service parameter
kbps
Offset
Gp
W
---------UL
R
None

FCH
G idle power
Cell parameter
None
Idle power gain
G MU
Cell parameter
None
Multi user gain
P max txi ic
Cell parameter
Max cell power
P tx txi ic b pilot
P max txi ic
Pilot burst transmitted by the

transmitter on carrier ic.
Traffic burst transmitted by the

transmitter on carrier ic.
UL
P tx txi ic b traffic
P max txi ic if users to support

P max txi ic G idle power if no user to support
ER DRC
Cell parameter
Error rate on the DRC channel
TS BCMCS
Cell parameter
Pourcentage of EVDO timeslots

dedicated to Broadcast/Multicast
services
TS EVDO CCH
Cell parameter
Pourcentage of EVDO timeslots

dedicated to control channels
P term ic
Simulation result
Terminal power transmitted on carrier

ic
P term
min
Terminal parameter
P term
max
Terminal parameter
BTS
BTS parameter

transmitted
term
Terminal parameter

transmitted
Clutter parameter

finger
G Tx
Antenna parameter
None
Forsk 2009
AT271_TRG_E6
285

G Term
Terminal parameter
None
Terminal gain
L Tx

None
Transmitter lossa
L body
Service parameter
None
Body loss
L Term
Terminal parameter
None
Terminal loss
L indoor
Clutter parameter
L path
None
Path loss
G ACK
Terminal parameter
None
Acknowledgement Channel gain
G RRI
Terminal parameter (for 1xEV-DO Rev A terminals only)
None
Reverse Rate Indicator Channel gain
G DRC
Terminal parameter
None
Data Rate Control Channel gain
G Auxiliary pilot
Terminal parameter (for 1xEV-DO Rev A terminals only)
None
Auxiliary Pilot Channel gain
G TCH
Terminal parameter
None
Traffic data Channel gain
M Shadowing model

standard deviation
None

M Shadowing Ec Io
Result calculated from cell edge coverage probability and Ec/I0

standard deviation
None
Ec/I0 Shadowing margin

DL
npaths
DL
G macro diversity
M Shadowing Eb Nt
Indoor loss
None
n=2 or 3
standard deviation
UL
UL
npaths
UL
G macro diversity
UL
M Shadowing Eb Nt
n=2 or 3
E Shadowing
UL
None
None

pilot signals at the mobile b.
in soft handoffc.
None

None
P tx txi ic b pilot
-------------------------------------------LT
Pilot bust received at terminal from a

P tx txi ic b traffic
----------------------------------------------LT
Traffic burst received at terminal from

P term
-------------LT
Bit received power at transmitter on

carrier ic
Cell parameter
dB
Cell uplink noise rise threshold
Cell parameter
dB
Cell uplink noise rise upgrading/

downgrading delta
Simulation result
In prediction studiesd
For Ec/I0 and Ec/Nt calculations
--------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
-----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
LT
In simulations
-------------------------------------------------------------------------------------------------------------------------------G Tx G term
DL
P tot txi ic b pilot

DL
P tot txi ic b traffic

UL
P b ic
UL
NR threshold txi ic
UL
NR threshold txi ic
a.
L Tx = L total UL on uplink and L Tx = L total DL on downlink.
b.
npaths
density function (n paths) in case of downlink Ec/I0 modelling.

c.
npaths
M Shadowing Eb Nt
UL
286
AT271_TRG_E6
Forsk 2009

d.
UL
the shadowing model ( M Shadowing Ec Io ) while extra-cell interference level is not. Therefore, M Shadowing Ec Io is set to
1 in downlink extra-cell interference calculation.
7.2.2.2
Ec/I0 and Ec/Nt Calculations

Ec
Ec
Ec
This table details ------ txi ic b pilot , ------ txi ic b pilot and ------ txi ic b traffic calculations.
I0
Nt
Nt
Name
Value
Unit
Description
txi ic
DL
I intra
b
or
b
pilot
traffic

terminal on carrier ic (only one mobile
is served at a time)
Downlink extra-cell interference based

on pilot at terminal on carrier ic
P tot txj ic b pilot
P tot txj ic b traffic

on traffic at terminal on carrier ic
Ptot txj icadj bpilot
Downlink inter-carrier interference

based on pilot at terminal on carrier ic

based on traffic at terminal on carrier
ic
Total noise based on pilot received at

Total noise based on traffic received

I extra ic b pilot + N 0
Total noise based on pilot received at

DL

None
Pilot quality level at terminal on carrier

ic
None
Pilot quality level at terminal on carrier

ic
None
Traffic quality level at terminal on

carrier ic
Unit
Description

ic

ic
DL
I extra ic b pilot
DL
txj j i
DL
I extra ic b traffic
DL
txj j i
DL
DL
I inter carrier ic b pilot
txj
j
-----------------------------------------------------------------
RF ic ic adj
Ptot txj icadj btraffic

DL
DL
I inter carrier ic b traffic
txj
j
--------------------------------------------------------------------
RF ic ic adj
DL
I 0 ic b pilot
DL
I 0 ic b traffic
DL
N tot ic b pilot
DL
N tot ic b traffic
Q pilot txi ic
Ec
------ txi ic b pilot
I0
DL
DL
DL
term
P tot txi ic b pilot + I extra ic b pilot + I inter carrier ic b pilot + N 0

DL
DL
DL
term
P tot txi ic b traffic + I extra ic b traffic + I inter carrier ic b traffic + N 0

DL
term
term
I extra ic b traffic + N 0
DL
BTS P tot txi ic b pilot

--------------------------------------------------------------------------DL
I 0 ic b pilot
DL
Ec
------ txi ic b pilot
Nt
BTS P tot txi ic b pilot

-----------------------------------------------------------------------------------------------------------------------DL
DL
N tot ic b pilot + 1 BTS P tot txi ic b pilot
Ec
------ txi ic b traffic
Nt
BTS P tot txi ic b traffic

------------------------------------------------------------------------------------------------------------------------------DL
DL
N tot ic b traffic + 1 BTS P tot txi ic b traffic
7.2.2.3
DL
This table details calculations of uplink quality ( Q
Name
UL intra
I tot
Forsk 2009
Eb
or ------- ).
Nt UL
Value
Pb
UL
txi ic
UL extra
I tot
txi
UL
ic
ic
term
txi
UL
P b ic
term
txj j i
AT271_TRG_E6
287
Pb
UL
UL
I inter carrier txi
ic
ic adj
term
txj j
--------------------------------------



RF ic ic adj
UL
I tot txi ic
UL extra
I tot
UL intra
Tx
UL
UL
N tot txi ic
UL
tx
I tot ic + N 0

UL
UL
term P b ic
--------------------------------------------------------------------------------------------------------------- G UL
p
UL
Tx
UL
Eb
txi ic ------
N t UL
None
UL
term P b ic
- G UL
Total noise: --------------------------------------p
UL
N tot txi ic
No HO: Q
Softer HO:
UL
f rake efficiency
UL
UL
Q tch tx k ic
Soft, Softer/Soft HO (No MRC):

UL
UL

tx k ActiveSet
UL
ica
txi ic
tx k ActiveSet
samesite
Q total ic
Quality level at transmitter on carrier
Softer/Soft HO (MRC):

and taking into account increase of the
None quality due to macro-diversity (macrodiversity gain).
Max UL
UL
UL
f rake efficiency Q tch tx k ic Q tch tx l ic
tx ,tx ActiveSet
k l
tx samesite
tx k
k
UL
In simulations, G macro diversity = 1 .
tx othersite
l
UL
G macro diversity
UL
Q total ic
--------------------------------------------------------UL
Q BestServer ic
UL
G SHO
None
Uplink soft handover gain on carrier ic
None
Eb/Nt target on uplink

Eb/Nt target at transmitter on carrier ic
For 1xEV-DO Rev 0 terminal

UL
E
UL
c
------
G p 1 + G ACK + G DRC + G TCH
N t min
UL
Q req
For 1xEV-DO Rev A terminalb

When the acknoledgement signal is considered
UL
E
-----c-
G p 1 + G ACK + G RRI + G DRC + G TCH + G Auxiliary Pilot
N t min
UL
When the acknoledgement signal is not considered

UL
E
UL
c
------
G p 1 + G RRI + G DRC + G TCH + G Auxiliary Pilot
N t min
UL
req
P term ic
Q req
----------------------- P term
UL
Q total ic
a.
b.
In simulations, the uplink Eb/Nt target is calculated whithout considering the aknoledgement signal.
7.2.2.4
Simulation Results
This table contains some simulation results provided in the Cells and Mobiles tabs of the simulation property dialog.
Name
Value
I intra txi ic b traffic
1 F ortho BTS P tot txi ic b traffic = 0
DL
288
DL
AT271_TRG_E6
Unit
Description

terminal on carrier ic (only one mobile
is served at a time)
Forsk 2009
DL
I extra ic b traffic
DL

on traffic at terminal on carrier ic

based on traffic at terminal on carrier
ic
DL
Total effective interference based on

traffic at terminal on carrier ic (after
unscrambling)
term


ic

ic


P tot txj ic b traffic
txj j i

DL
DL
I inter carrier ic
b traffic
txj
j
--------------------------------------------------------------------
RF ic ic adj
DL
DL
I tot ic b traffic
DL
DL
N tot ic b traffic
UL intra
I tot
txi
UL extra
I tot
DL
I intra ic b traffic + I extra ic b traffic + I inter carrier ic b traffic

I tot ic b traffic + N 0
Pb
UL
ic
txi ic
UL
P b ic
term
txj j i
Pb
UL
UL
I inter carrier txi
ic
term
txi
ic
ic adj
term
txj j
--------------------------------------
RF ic ic adj
UL
UL extra
I tot txi ic
I tot
UL intra
Tx
UL
N tot txi ic
I tot txi ic + N 0

N mobiles txi ic
Simulation result
None
Number of mobiles connected to

transmitter txi on carrier ic
None
Cell downlink load factor on carrier ic
None
UL
UL
tx
DL
DL
txi ic
I tot ic b traffic
--------------------------------------DL
N tot ic b traffic
UL
txi ic
I tot txi ic
----------------------------UL
N tot txi ic
UL
txi ic
I tot txi ic
------------------------------------------------------------------------------------------UL intra
Tx
I tot
txi ic 1 F MUD term
None
UL
txi ic
1
-----------------------------UL
F txi ic
None

carrier ic
UL
UL
NR
DL
txi ic
10 log 1 X
DL
txi ic
dB
NR
UL
txi ic
10 log 1 X
UL
txi ic
dB
7.3
Active Set Management

Active set (AS) management is detailed hereafter. The active set may consist of one or more cells; depending on whether
the service supports soft handoff and on the terminal active set size. The quality of the pilot (EcI0) is what determines
whether or not a cell can belong to the active set. Cells entering the mobiles active set must fulfill the following conditions:
1.
The best server (first cell entering active set)

The pilot quality from the best server cell must exceed an upper threshold equal to the sum of the Min. Ec/I0
defined in the properties of the best serer cell and the Delta Min. Ec/I0 defined in the properties of the mobility
type. The upper threshold is set for the carrier as defined in the cell properties and can also take into account the
user mobility type if the Delta minimum Ec/I0 defined in the mobility type is different from 0. Best server cell is the
cell with the highest pilot quality.
2.
Forsk 2009
Other cells of active set

- They must use the same carrier as the best server cell,
- The pilot quality from other candidate cells must exceed a lower threshold. The lower threshold depends both
on the type of carrier and the mobility type. It is equal to the sum of T_Drop defined in the properties of the
best server and the Delta T_Drop defined in the properties of the mobility type.
AT271_TRG_E6
289

-
7.4
If you have selected to restrict the active set to neighbours, the cell must be a neighbour of the best server
(the restricted to neighbours option is selected in the equipment properties).
Simulations
The simulation process is divided into two steps:
1.
Obtaining a realistic user distribution
Atoll generates a user distribution using a Monte-Carlo algorithm, which requires traffic maps and data as input. The resulting user distribution complies with the traffic database and maps provided to the algorithm.
Each user is assigned a service, a mobility type, and an activity status by random trial, according to a probability law that
uses the traffic database.
The user activity status is an important output of the random trial and has direct consequences on the next step of the
simulation and on the network interferences. A user may be either active or inactive. Both active and inactive users
consume radio resources and create interference.
Additionally, each 1xEV-DO Rev. 0 user is assigned a transition flag ("True" or "False") for each possible data rate transition (from 9.6 to 19.2 kbps, 19.2 to 38.4 kbps, 38.4 to 76.8 kbps, and 76.8 to 153.6 kbps for data rate upgrading and from
153.6 to 76.8 kbps, 76.8 to 38.4 kbps, 38.4 to 19.2 kbps, and 19.2 to 9.6 kbps for data rate downgrading). These transition
flags are based on the data rate downgrading and upgrading probabilities. If a transition flag is "True," the user data rate
can be downgraded or upgraded if necessary.
Then, Atoll randomly assigns a shadowing error to each user using the probability distribution that describes the shadowing effect.
Finally, another random trial determines user positions in their respective traffic zone and whether they are indoors or
outdoors (according to the clutter weighting and the indoor ratio per clutter class defined for the traffic maps).
2.
Modelling the network regulation mechanism
This algorithm depends on the network. Atoll uses a power control algorithm in case of cdmaOne/CDMA2000 1xRTT
networks and a different algorithm, which mixes data rate control on downlink and power control on uplink, for CDMA2000
1xEV-DO networks.
7.4.1
7.4.1.1
Number of Users, User Activity Status and User Data Rate

During the simulation, a first random trial is performed to determine the number of users and their activity status. The determination of the number of users and the activity status allocation depend on the type of traffic cartography used.
Note:
Atoll follows a Poisson distribution to determine the number of total connected users before
each simulation. In order to make Atoll use a constant number of total connected users, the
following lines must be added to the Atoll.ini file:
[CDMA]
RandomTotalUsers=0
7.4.1.1.1
Simulations Based on Raster Traffic and Vector Traffic Maps

a list of user profiles with an associated mobility type and a given density (number of subscribers with the same profile per
km).
Vector traffic maps are directly based on user profiles. Each polygon and line of the map is assigned a density of subscribers with given user profile and mobility type. If the map is composed of points, each point is assigned a number of subscribers with given user profile and mobility type.
The user profile models the behaviour of the different subscriber categories. Each user profile contains a list of services
and their associated parameters describing how these services are accessed by the subscriber.
From environment (or polygon) surface (S) and user profile density (D), a number of subscribers (X) per user profile is
inferred.
X = SD
Notes:
290
In case of vector traffic map composed of lines, the number of subscribers (X) per user
profile is calculated from the line length (L) and the user profile density (D) (nb of
subscribers per km) as follows: X = L D
The number of subscribers (X) is an input when the vector traffic map is composed of
points.
AT271_TRG_E6
Forsk 2009

For each behaviour described in a user profile, according to the service, frequency use and exchange volume, Atoll calculates the probability for the user being connected in uplink and in downlink at an instant t.
Calculation of the service usage duration per hour ( p 0 : probability of being connected):
N call d
p 0 = --------------------3600
where N call is the number of calls per hour and d is the average call duration (in second).
Then, Atoll deduces the total number of users trying to access a certain service.
Calculation of the total number of users trying to access the service j ( n j ):
nj = X p0
The next step determines the activity status of each user. This one depends on the type of service (Voice, 1xRTT data,
1xEV-DO data).
IS-95 cdmaOne and CDMA2000 1xRTT Services

Activity status of voice and data service users is determined as follows.
Users are always active on FCH in both directions, uplink and downlink. Therefore, we have:
Probability of being active on UL: p UL = 0
Probability of being active on DL: p DL = 0
Probability of being active both on UL and DL: p UL + DL = 1
Probability of being inactive: p inactive = 0
Thus, for voice and data services, we have:
Number of inactive users: n j inactive = n j p inactive = 0
Number of users active on UL: n j UL = n j p UL = 0
Number of users active on DL: n j DL = n j p DL = 0
Number of users active on UL and DL both: n j UL + DL = n j p UL + DL = n j
n j = n j UL + n j DL + n j UL + DL + n j inactive = n j UL + DL
Voice Service Users
Voice service users are active on uplink and downlink. However, the FCH can have inactivity periods on both links. This
UL
DL
is modelled by the FCH activity factor, AFFCH and AF FCH . Therefore, all voice service users try to access the service
UL
UL
DL
DL
with the following FCH rates, R FCH AF FCH on uplink and R FCH AF FCH on downlink.
UL
DL
R FCH and R FCH are respectively the uplink and downlink FCH nominal rates.
Data Service Users
Data service users are active on uplink and downlink. FCH is always allocated but can have inactivity periods on both links;
UL
DL
this is modelled by the FCH activity factor, AFFCH and AF FCH . SCH may be allocated with four possible rates (2x, 4x,
8x and 16xFCH nominal rate).
Therefore, data service users can access the service with different rates. Possible rates are detailed in the table below:
Allocated rates
SCH rate factor r k

Only FCH is used
UL
AF FCH
On DL
DL
R FCH
DL
AF FCH
UL
UL
R FCH AF FCH + 2
UL
UL
R FCH AF FCH + 4
UL
UL
R FCH AF FCH + 8
2x
R FCH AF FCH + 2
4x
R FCH AF FCH + 4
8x
R FCH AF FCH + 8
16x
R FCH AF FCH + 16
Both FCH and SCH are used
UL
On UL
UL
R FCH
UL
UL
DL
DL
DL
DL
DL
DL
DL
DL
R FCH AF FCH + 16
DL
R FCH and R FCH are respectively the uplink and downlink FCH nominal rates.
Forsk 2009
AT271_TRG_E6
291

Then, Atoll determines the distribution of users between the different possible rates.
UL
In case of a data service, j, several data rate probabilities, P k
DL
and P k , can be assigned to different rate factors, r k , for
SCH channel.
Note:
For non-data services, these probabilities are 0.
For data service users, a random trial compliant with data rate probabilities is performed for each link in order to determine
the rate for each user.
On uplink, we have:
rk
UL
UL
For each SCH rate factor, r k , the number of users n j with the data rate R FCH AF FCH + r k is calculated as follows,
rk
UL
nj = Pr
nj
FCH
Therefore, the number of users n j

FCH
nj
= nj
UL
UL
with the data rate, R FCH AF FCH , is:
rk
nj
rk
On downlink, we have:
rk
DL
DL
For each SCH rate factor, r k , the number of users, n j with the data rate, R FCH AF FCH + r k , is calculated as follows,
rk
DL
nj = Pr
nj
FCH

FCH
nj
= nj
DL
DL
with the data rate, R FCH AF FCH , is:
rk
nj
rk
CDMA2000 1xEV-DO Services

As power control is performed on uplink only, 1xEV-DO data service users will be considered either active on uplink or
inactive. 1xEV-DO data Rev. 0 service users can access the service with uplink rates of 9.6, 19.2, 38.4, 76.8 and 153.6
kbps. 1xEV-DO data Rev. A service users can access the service with uplink rates of 4.8, 9.6, 19.2, 38.4, 76.8, 115.2,
153.6, 230.4, 307.2, 460.8, 614.4, 921.6, 1,228.8 and 1,848.2 kbps.
UL
UL
For each service, j, several data rate probabilities, P k , can be assigned to different rates R k . The number of users
active on uplink ( n j UL ) and the number of inactive users ( n j inactive ) are calculated as follows:
Probability of being active on UL: p UL =
Pk
UL
UL
Rk
UL
Rk
Pk
UL
UL
Rk
UL
Rk

Probability of being active on UL and DL both: p UL + DL = 0
Therefore, we have:
Number of users active on UL: n j UL = n j p UL
Number of inactive users: n j inactive = n j p inactive
Number of users active on DL: n j DL = n j p DL = 0
Number of users active on UL and DL both: n j UL + DL = n j p UL + DL = 0
n j = n j UL + n j DL + n j UL + DL + n j inactive = n j UL + n j inactive
UL
Then, Atoll determines the distribution of users between the different possible rates, R k . The number of users with the
UL
UL
data rate R k , n j R k , is calculated as follows:

UL
UL
nj Rk = Pk nj
Inactive users have a requested data rate equal to 0.
292
AT271_TRG_E6
Forsk 2009

Notes:
The user distribution per service is an average distribution and the service of each user is
randomly drawn in each simulation. Therefore, if you compute several simulations at once,
the average number of users per service will correspond to the calculated distribution. But if
you check each simulation, the user distribution between services is different in each of
them.
It is the same for the SCH rate distribution between 1xRTT data service users and the
traffic data rate distribution between 1xEV-DO data service users.
In calculations detailed above, we assume that the sum of data rate probabilities is less
than or equal to 1. If the sum of data rate probabilities exceeds 1, Atoll considers
normalised data rate probabilities values, P r

k
Pr , instead of specified data rate

k
rk
probabilities P r .
k
7.4.1.1.2
Simulations Based on Traffic Maps per Service and per Transmitter

IS-95 cdmaOne and CDMA2000 1xRTT Services
Voice Service Users
This calculation can be divided into several steps. First of all, Atoll determines the number of users active on UL and DL
using the service j in the Txi cell.
For each transmitter, Txi, and each service j:
-
Either Atoll deduces the number of users active on UL and DL using the service j in the Txi cell, if you have
selected the throughput map option.
UL
DL
Rt
Rt
N UL = --------- and N DL = --------UL
DL
Rj
Rj
Where,
UL
Rt
is the number of kbits per second transmitted on UL in the Txi cell to provide the service j to the users
(user-defined value in the traffic map properties)

DL
Rt
is the number of kbits per second transmitted on DL in the Txi cell to provide the service j to the users
(user-defined value in the traffic map properties).

UL
Rj
DL
and R j
correspond to uplink and downlink rates of a user. FCH is always allocated to active users but
UL
can have inactivity periods on both links. Therefore, we have R j
UL
UL
UL
= R FCH AF FCH (where R FCH is the serv-
UL
ice FCH nominal rate on UL and AF FCH corresponds to the FCH activity factor on UL) and
DL
Rj
DL
DL
DL
DL
= R FCH AFFCH (where R FCH is the service FCH nominal rate on DL and AF FCH corresponds to the
FCH activity factor on DL).

-
Or Atoll directly uses the defined NUL and NDL values (number of users active on UL and DL using the service
j in the Txi cell), if you have selected the user map option.
Users active on UL and DL both are included in the NUL and NDL values. Therefore, it is necessary to accurately determine
the number of users active on UL ( n j UL ), on DL ( n j DL ) and on UL and DL ( n j UL + DL ) both.
Users are always active on FCH for both links. Therefore, we have following activity probabilities.
Let us consider n j active , the total number of active users using the service j in the Txi cell
We have,
p UL + p UL + DL n j active = N UL
p DL + p UL + DL n j active = N DL
Thus,
Forsk 2009
AT271_TRG_E6
293

N UL p UL + DL
N DL p UL + DL
n j UL + DL = min ------------------------------------------------------------------ ------------------------------------------------------------------
p UL 1 p DL + p UL + DL p DL 1 p UL + p UL + DL
n j UL = N UL n j UL + DL
n j DL = N DL n j UL + DL
and, n j active = n j UL + n j DL + n j UL + DL
The number of inactive users trying to access the voice service j, n j inactive , is,
n j active
n j inactive = ------------------------------ p inactive = 0
1 p inactive
Therefore, the total number of users trying to access the voice service j, n j , is,
n j = n j active
UL
UL
DL
DL
Voice service users try to access the service with the FCH rates, R FCH AF FCH on uplink and R FCH AF FCH on downlink.
All user characteristics determined, a second random trial is performed to obtain their geographical positions.
Data Service Users
FCH is always allocated to active users but can have inactivity periods on both links. This is modelled by the FCH activity
UL
DL
factors, AFFCH and AF FCH . SCH may be allocated with four possible rates (2x, 4x, 8x, 16xFCH nominal rate). Several
UL
DL
data rate probabilities, P k
and P k , can be assigned to different rates factor, r k , for SCH channel.
Note:
For non-data services, these probabilities are 0.
Atoll determines the number of users active on UL and DL using the service j in the Txi cell. For each transmitter, Txi, and
each service j:
-
Either Atoll deduces the number of users active on UL and DL using the service j in the Txi cell, if you have
selected the throughput map option.
UL
DL
Rt
Rt
N UL = --------- and N DL = --------UL
DL
Rj
Rj
Where,
UL
Rt
is the number of kbits per second transmitted on UL in the Txi cell to provide the service j (user-defined
value in the traffic map properties).

DL
Rt
is the number of kbits per second transmitted on DL in the Txi cell to provide the service j (user-defined
value in the traffic map properties).

UL
and R j
UL
Rj
Rj
DL
correspond to uplink and downlink rates of a user.
rk + AFFCH RFCH Pr
rk + AFFCH RFCH Pr
UL
UL
UL
rk
DL
Rj
DL
DL
rk
UL
DL
+ 1
Pr
UL
+ 1
Pr
rk
DL
rk
UL
UL
DL
DL
R FCH AF FCH
R FCH AF FCH
DL
R FCH and R FCH are the uplink and downlink FCH nominal rates respectively.
Note:

k
Pr ,
k
instead of specified data rate
rk
probabilities P r .
k
294
Or Atoll directly uses the defined values for NUL and NDL (number of users active on UL and DL using the
service j in the Txi cell), if you have selected the user map option.
AT271_TRG_E6
Forsk 2009

Users active on UL and DL both are included in the NUL and NDL values. Therefore, it is necessary to accurately determine
the number of users active on UL ( n j UL ), on DL ( n j DL ) and on UL and DL ( n j UL + DL ) both.
Users are always active on FCH for both links. Therefore, we have following activity probabilities.
Let us consider n j active , the total number of active users using the service j in the Txi cell
We have,
p UL + p UL + DL n j active = N UL
p DL + p UL + DL n j active = N DL
Thus,
N UL p UL + DL
N DL p UL + DL
n j UL + DL = min ------------------------------------------------------------------ ------------------------------------------------------------------
p UL 1 p DL + p UL + DL p DL 1 p UL + p UL + DL
n j UL = N UL n j UL + DL
n j DL = N DL n j UL + DL
and, n j active = n j UL + n j DL + n j UL + DL
The number of inactive users trying to access the service j, n j inactive , is,
n j active
n j inactive = ------------------------------ p inactive = 0
1 p inactive
Therefore, the total number of users trying to access the service j, n j , is,
n j = n j active
As explained above, data service users can access the service with different rates. Possible rates are detailed in the table
below:
Allocated rates
SCH rate factor r k

Only FCH is used
On UL
UL
On DL
UL
R FCH AF FCH
2x
R FCH AF FCH + 2
4x
R FCH AF FCH + 4
8x
R FCH AF FCH + 8
16x
R FCH AF FCH + 16
Both FCH and SCH are used
DL
DL
R FCH AF FCH
UL
UL
R FCH AF FCH + 2
UL
UL
R FCH AF FCH + 4
UL
UL
R FCH AF FCH + 8
UL
UL
DL
DL
DL
DL
DL
DL
DL
DL
R FCH AF FCH + 16
Atoll determines the distribution of users with the different possible rates. A random trial compliant with data rate probabilities is performed for each link in order to determine the data rate of each user.
On uplink, we have,
rk
UL
UL
For each SCH rate factor, r k , the number of users n j with the data rate R FCH AF FCH + r k is calculated as follows,
rk
UL
nj = Pr
nj
FCH

FCH
nj
= nj
UL
UL
with the data rate, R FCH AF FCH , is,
rk
nj
rk
On downlink, we have,
rk
DL
DL
For each SCH rate factor, r k , the number of users, n j with the data rate, R FCH AF FCH + r k , is calculated as follows,
Forsk 2009
AT271_TRG_E6
295

k
DL
nj = Pk nj
FCH

FCH
nj
= nj
DL
DL
with the data rate, R FCH AF FCH , is,
rk
nj
rk
CDMA2000 1xEV-DO Services

As power control is performed on uplink only, 1xEV-DO data service users will be considered either active on uplink or
inactive. 1xEV-DO data Rev. 0 service users can access the service with uplink rates of 9.6, 19.2, 38.4, 76.8 and 153.6
kbps. 1xEV-DO data Rev. A service users can access the service with uplink rates of 4.8, 9.6, 19.2, 38.4, 76.8, 115.2,
153.6, 230.4, 307.2, 460.8, 614.4, 921.6, 1,228.8 and 1,848.2 kbps.
UL
UL
For each service, j, several data rate probabilities, P k , can be assigned to different uplink rates R k . The number of
users active on uplink ( n j UL ) and the number of inactive users ( n j inactive ) are calculated into several steps. First
of all, Atoll determines the number of users active on UL using the service j in the Txi cell.
For each transmitter, Txi, and each service j:
-
Either Atoll deduces the number of users active on UL using the service j in the Txi cell, if you have selected
the throughput map option.
UL
Rt
N UL = --------UL
Rj
Where:
UL
Rt
is the number of kbits per second transmitted on UL in the Txi cell to provide the service j (user-defined value in the
traffic map properties).

UL
corresponds to the uplink data rate for a user.
UL
Rj
Rj
Pk
UL
UL
Rk
Note:

k
Pr , instead of specified data rate

k
rk
probabilities P r .
k
Or Atoll directly uses the defined values of NUL (number of users active on UL using the service j in the Txi
cell), if you have selected the user map option.
Activity probabilities are:

Probability of being active on UL: p UL =
Pk
UL
UL
Rk
UL
Rk
Pk
UL
UL
Rk
UL
Rk

Therefore, we have:
Number of users active on UL: n j UL = N UL p UL
Number of inactive users: n j inactive = N UL p inactive
Number of users active on DL: n j DL = 0
Number of users active on UL and DL both: n j UL + DL = 0
n j = n j UL + n j inactive
296
AT271_TRG_E6
Forsk 2009

Then, Atoll determines the distribution of users with the different possible rates. The number of users with the data rate
UL
UL
R k , n j R k , is calculated as follows:
UL
UL
nj Rk = Pk nj
Inactive users have a requested data rate equal to 0.
Note:
7.4.1.2
The user distribution per service is an average distribution and the service of each user is
randomly drawn In each simulation. Therefore, if you compute several simulations at once,
the average number of users per service will correspond to the calculated distribution. But if
you check each simulation, the user distribution between services is different in each of
them.
It is the same for the SCH rate distribution between 1xRTT data service users and the
traffic data rate distribution between 1xEV-DO data service users.
Transition Flags for 1xEV-DO Rev.0 User Data Rates

For 1xEV-DO Rev. 0 services supporting data rate downgrading, you can define the probability of the service being
UL
UL
UL
UL
UL
upgraded ( P Upg k R k ) or downgraded ( P Downg k R k ) on the uplink (reverse link) for each data rate ( R k ). The
probabilities are taken into account in order to determine if a user with a certain data rate can be upgraded or downgraded.
User data rate downgrading and upgrading occur during congestion control when the cell is over- or underloaded.
The following table shows the data rate changes that are possible when a data rate is upgraded or downgraded. The probabilities are defined with a number from 1 to 255 for each data rate.
Possible Data Rate Changes

During Upgrading
From
To
Possible Data Rate Changes

During Downgrading
From
To
9.6 kbps
19.2 kbps
153.6 kbps
76.8 kbps
19.2 kbps
38.4 kbps
76.8 kbps
38.4 kbps
38.4 kbps
76.8 kbps
38.4 kbps
19.2 kbps
76.8 kbps
153.6 kbps
19.2 kbps
9.6 kbps
During the generation of the user distribution, each 1xEV-DO Rev. 0 user is assigned a random number between 1 and
255 for each possible data rate transition. When this number is lower or equal to the value of the probability, the transition
flag for this data rate transition is set to "True" meaning that this data rate transition can be performed if necessary.
UL
The number of 1xEV-DO Rev. 0 users with a certain data rate that can be downgraded ( n j R k Downg ) and upgraded
UL
( n j R k Upg ) are calculated as follows:

UL
UL
UL
P Upg k R k n j R k
UL
n j R k Upg = ------------------------------------------------------------255
And
UL
UL
UL
P Downg k R k n j R k
UL
n j R k Downg = -------------------------------------------------------------------255
Note:
7.4.1.3
The number of users with a certain data rate that can be downgraded or upgraded is an
average. Therefore, if you compute several simulations at once, the average number of
users with a certain data rate that can be downgraded or upgraded will correspond to the
calculated value. But if you check each simulation, this number is different in each of them.
User Geographical Position

Once all the user characteristics determined, another random trial is performed to obtain their geographical positions and
whether they are indoors or outdoors according to the percentage of indoor users per clutter class defined for the traffic
maps.
7.4.2
Network Regulation Mechanism
7.4.2.1
IS-95 cdmaOne Power Control Simulation Algorithm

The IS-95 cdmaOne network automatically regulates itself by using uplink and downlink power control in order to minimize
interference and maximize capacity. Atoll simulates the power control mechanism for each user distribution. Simulations
use an iterative algorithm, where in each iteration, all the mobiles selected during the user distribution generation (1st step)
try to connect to the networks active transmitters with a calculation area. The process is repeated for each iteration and
Forsk 2009
AT271_TRG_E6
297

ends once the network is balanced, i.e. when the convergence criteria (on UL and DL) are satisfied. Algorithm steps are
detailed below.
Figure 7.1: IS-95 cdmaOne Power Control Algorithm
7.4.2.1.1
The
total
power
transmitted
by
the
base
station
txi
on
carrier
ic
P Tx txi ic
is
initialised
to
P pilot txi ic + P sync txi ic + P paging txi ic .

UL intra
Uplink received powers by the base station txi on carrier ic, I tot
UL extra
txi ic , I tot
UL
txi ic and I inter carrier txi ic , are
initialised to 0 W (no connected mobile).

UL
I tot txi ic
UL
X k txi ic = ----------------------------- = 0
UL
N tot txi ic
7.4.2.1.2
Presentation of the Algorithm

UL
The algorithm is detailed for any iteration k. Xk is the value of the variable X at the iteration k. In the algorithm, all Q req
DL
For each mobile Mb
Determination of Mbs Best Server (SBS(Mb))

For each station txi containing Mb in its calculation area,
Determination of BestCarrier k txi M b .
If a given carrier is specified for the service requested by Mb and if it is used by txi
BestCarrier k txi M b is the carrier specified for the service.
Else the carrier selection mode defined for txi is considered.
UL
I tot txi ic
UL
UL
- + X
X k txi ic = ----------------------------UL
N tot txi ic
EndFor
UL

298
AT271_TRG_E6
Forsk 2009

UL
UL

Calculation of Q pilot txi M b BestCarrier = -----------------------------------------------------------------------------------------------DL
k
I 0 BestCarrier k txi M b
req
If Q pilot txi M b BestCarrier Q pilot then txi is rejected by Mb

k
max

k
Admission control (If simulation respects a load factor constraint and Mb was not connected in previous iteration).
UL
If X K txi BestCarrier k txi M b X
UL
max
, then txi is rejected by Mb
Else
max

k
tx BS M b = txi
Endif
EndFor
If no txBS has been selected, Mb cannot get a connection to the network
In the following lines, we will consider ic = BestCarrier k tx BS M b M b
Determination of the Active Set

For each station txi containing Mb in its calculation area, using ic, and if neighbours are used, neighbour of txBS(Mi)
BTS P c txi M b ic
Calculation of Q pilot txi M b ic = ----------------------------------------------------------------DL
k
I 0 ic
pilot
If Q pilot txi M b ic Q min then txi is rejected by Mb

k
Else txi is included in the Mb active set

Rejection of txi if the Mb active set is full
k
EndFor

req
Calculation of the required power for Mb, P term M b ic k

For each cell (txi,ic) present in the Mb active set
Calculation of quality level on Mb traffic channel at (txi,ic), with the minimum power allowed on traffic channel for the Mb
service
FCH r eq
FCH UL
Pb
M b ic k 1
P term
txi M b ic = ----------------------------------------------------L T txi M b
FCH U L
term P b
txi M b ic
UL
FCH UL
Q FCH txi M b ic k = ------------------------------------------------------------------------------------------------------------------------------------------------ Gp
Service
UL
FCH U L
txi M b ic
N tot txi ic 1 F MUD term P b
If the user selects the option Total noise
FCH U L
term P b
txi M b ic
UL
UL
- G FCH
Q FCH txi M b ic k = ----------------------------------------------------------------------- Service
p
UL
N tot txi ic
Forsk 2009
AT271_TRG_E6
299

End For
If (Mb is not in handoff)
UL
UL
Q FCH M b = Q FCH txi M b ic k

k
Else if (Mb is in softer handoff)

UL
UL
Q FCH M b = f rake efficiency

k
UL
Q FCH txi M b ic k
txi ActiveSet
Else if (Mb is in soft or softer/soft without MRC)

UL
UL
Q FCH M b =
UL
Max Q FCH txi M b ic k G macro diversity 2 links
txi ActiveSet
Else if (Mb is in soft/soft)

UL
UL
Q FCH M b =
UL
Max Q FCH txi M b ic k G macro diversity 3 links
txi ActiveSet
Else if (Mb is in softer/soft with MRC)
UL
UL
UL
UL
Q FCH M b = Max f rake efficiency
Q FCH txi M b ic Q FCH
k
other site
txi ActiveSet
(same site)
UL
txi M b ic G macro diversity 2 links
EndIf
FCH r eq
P term
UL
Q req Service M b Mobility M b FCH

FCH r eq
M b ic k = ---------------------------------------------------------------------------------------------------------- P term
M b ic k 1
UL
Q FCH M b
k
req
FCH r eq
P term M b ic k = Max P term

FCH r eq
If P term
min
M b ic k P term txi M b
max
M b ic k P term M b then Mb cannot select any station and its active set is cleared

Calculation of quality level on (txi,ic) FCH at Mb, with the minimum power allowed on FCH for the Mb service
FCH D L
Pb
min
P FCH Service M b
txi M b ic = -----------------------------------------------------L T txi M b
FCH D L
BTS P b
txi M b ic
DL
FCH DL
Q FCH txi M b ic k = -------------------------------------------------------------------------------------------------------------------------------------- Gp
Service M b
DL
FCH D L
txi M b ic
N tot ic 1 F ortho BTS P b
FCH D L
BTS P b
txi M b ic
DL
DL
- G FCH
Q FCH txi M b ic k = ----------------------------------------------------------------------- Service M b
p
DL
N tot ic
End For
DL
DL

k
DL
Q FCH txi M b ic k
txi ActiveSet FCH
Do
Calculation of the required power for DL traffic channel between (txi,ic) and Mb:
DL
Q req Service M b Mobility M b FCH

req
- P min
P FCH txi M b ic k = -----------------------------------------------------------------------------------------------------FCH Service M b
DL
Q FCH M b
k
req
max
If P FCH txi M b ic k P FCH Service M b then txi ic is excluded from Mb active set
DL
Recalculation of a decreased Q req
300
AT271_TRG_E6
Forsk 2009

EndFor
DL
DL

k
DL
Q FCH txi M b ic k
txi ActiveSet FCH

DL
DL
While Q FCH M b Q req Service M b Mobility M b FCH and Mb active set is not empty
k
Uplink and Downlink Interference Updates

Update of interference on active mobiles only (old contributions of mobiles and stations are replaced by the new ones).
For each cell (Sj,ic)
UL
Update of N tot S j ic
EndFor
For each mobile Mi
DL
Update of N tot ic
EndFor
Control of Radio Resource Limits (Walsh Codes, Cell Power and Site Channel Elements)
For each cell (txi,ic) on a site Nl
P tx txi ic k
DL
While ------------------------------ %Power max
P max
req
Rejection of mobile with highest P FCH txi M b ic k for the lowest service priority
EndWhile
EndFor
For each site Nl
The list of rejected mobiles for the site Nl is L rejected N l
If the equipment installed on Nl supports power pooling between transmitters
Activation of power pooling between transmitters for each cell (txi,ic) containing rejected users
Control of the available power for the other cells (txj,ic) of the site where power pooling between transmitters is not activated
If
DL
%Power max P max P tx txj ic k 0
txj ic
txj N l
Then, the power unused by the cells (txj,ic) of the site can be allocated to cells (txi,ic)
Sort of all the rejected mobiles by priority in a descending order and by simulation rank in a descending order
For the first mobile Mb of the list ( M b L rejected N l )
req
DL
If P tx txi ic k + P FCH txi M b ic k %Power max P max + M Pooling txi ic

Mb is reconnected
EndIf
EndFor
EndIf
EndFor
While N
Codes
Codes
txi ic k N max txi ic
Rejection of last admitted mobile

EndFor
For each site (Node B) Nl
While N
CE DL
CE DL
N I k N max
NI k
req
Rejection of mobile with highest P tch txi M b k for the lowest service priority
Forsk 2009
AT271_TRG_E6
301

While N
CE UL
CE UL
N I k N max
NI k
req
Rejection of mobile with highest P term M b ic k for the lowest service priority
EndFor

For each cell (txi,ic) with X
UL
UL
txi ic X max
Rejection of a mobile with the lowest service priority

EndFor
7.4.2.1.3
UL
UL
txi ic X max exists
Convergence Criterion
The convergence criteria are evaluated at each iteration, and can be written as follow:
DL
DL

P tx ic k P tx ic k 1
N user ic k N user ic k 1
100
DL = max int ma x ----------------------------------------------------- 100 int ma x -----------------------------------------------------------------DL
P tx ic k
Stations
Stations
N
ic
user
UL
UL
UL
UL

I tot ic k I tot ic k 1
UL = max int ma x ---------------------------------------------------- 100
- 100 int ma x -----------------------------------------------------------------UL
UL
Stations
Stations
I ic
N
ic
tot
user

1st case: Between two successive iterations, UL and DL are lower ( ) than their respective thresholds (defined
when creating a simulation).
has been achieved.
2nd case: After 30 iterations, UL or/and DL are still higher than their respective thresholds and from the 30th iteration,
UL or/and DL do not decrease during the next 15 successive iterations.
1. After the 30th iteration, UL and/or DL equal 100 and do not decrease during the next 15 successive iterations: Atoll
stops the algorithm at the 46th iteration. Convergence has not been achieved.
under the thresholds) and then do not change during the next 15 successive iterations: Atoll stops the algorithm at the 56th
iteration without achieving convergence.
If UL and/or DL are still strictly higher than their respective thresholds, the simulation has not converged (specific divergence symbol).
If UL and DL are lower than their respective thresholds, the simulation has converged.
7.4.2.2
CDMA2000 1xRTT Power Control Simulation Algorithm

CDMA2000 1xRTT network automatically regulates itself using traffic driven uplink and downlink power control on the
fundamental and supplemental channels (FCH and SCH respectively) in order to minimize interference and maximize
capacity. Atoll simulates this network regulation mechanism with an iterative algorithm and calculates, for each user distribution, network parameters such as base station power, mobile terminal power, active set and handoff status for each
terminal.
The power control simulation is based on an iterative algorithm, where in each iteration, all the mobiles selected during
the user distribution generation (1st step) try to connect to network active transmitters with a calculation area. The process
is repeated from iteration to iteration until convergence is achieved. The algorithm steps are detailed below.
302
AT271_TRG_E6
Forsk 2009
Figure 7.2: CDMA2000 1xRTT Power Control Algorithm
7.4.2.2.1
Total power on carrier ic, P Tx ic , of base station Sj is initialised to P pilot ic + P sync ic + P paging ic .
UL intra
Uplink received powers on carrier ic, I tot
UL extra
ic , I tot
UL
ic and I inter carrier ic , at base station Sj are initialised to 0 W
(no connected mobile).

UL
I tot S j ic
UL
- = 0
X k S j ic = ---------------------------UL
N tot S j ic
7.4.2.2.2

UL
The algorithm is detailed for any iteration k. Xk is the value of the variable X at the iteration k. In the algorithm, all Q req
DL
For each mobile Mi
Determination of Mis Best Server (SBS(Mi))

For each station Sj containing Mi in its calculation area,
Determination of BestCarrier k S j M i .
If a given carrier is specified for the service requested by Mi and if it is used by Sj
BestCarrier k S j M i is the carrier specified for the service.
Else the carrier selection mode defined for Sj is considered.
For each carrier ic used by Sj, we calculate current loading factor:
UL
I tot S j ic
UL
UL
X k S j ic = ---------------------------- + X
UL
N tot S j ic
EndFor
UL
BestCarrier k S j M i is the carrier with the lowest X k S j ic

BestCarrier k S j M i is the carrier with the lowest P tx S j ic k
Forsk 2009
AT271_TRG_E6
303

BestCarrier k S j M i is randomly selected
UL
UL
BestCarrier k S j M i is the first carrier so that X k S j ic X max

BTS P c M i S j BestCarrier
Calculation of Q pilot M i S j BestCarrier = -------------------------------------------------------------------------------------------DL
k
I 0 BestCarrier k S j M i
Rejection of station Sj if the pilot is not received
pilot
If Q pilot M i S j BestCarrier Q req then Sj is rejected by Mi

k
max
If Q pilot M i S j BestCarrier Q pilot M i

k
Admission control (If simulation respects a load factor constraint and Mi was not connected in previous iteration).
UL
UL
If X k S j BestCarrier k S j M i X max , then Sj is rejected by Mi

Else
max
Q pilot M i = Q pilot M i S j BestCarrier

k
S BS M i = S j
Endif
EndFor
If no SBS has been selected, Mi cannot get a connection to the network
In the following lines, we will consider ic = BestCarrier k S BS M i M i

For each station Sj containing Mi in its calculation area, using ic, and if neighbours are used, neighbour of SBS(Mi)
BTS P c M i S j
Calculation of Q pilot M i S j ic = -----------------------------------------------------DL
k
I 0 ic
pilot
If Q pilot M i S j ic Q min then Sj is rejected by Mi

k
Else Sj is included in the Mi active set

Rejection of Sj if the Mi active set is full
k
EndFor

req
Calculation of the required power for Mi, P term M i ic k

For each cell (Sj,ic) present in the Mi active set
Calculation of quality level on Mi traffic channel at (Sj,ic), with the minimum power allowed on traffic channel for the Mi
service
FCH r eq
FCH U L
Pb
SCH r eq
M i ic k 1
M i ic k 1
P term
P term
SCH U L
and P b
M i S j ic = --------------------------------------------------- M i S j ic = ---------------------------------------------------L T M i S j
L T M i S j
FCH U L
term P b
M i S j ic
UL
FCH UL
Q FCH M i S j ic k = ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Gp
Service
UL
FCH U L
SCH U L
M i S j ic + P b
M i S j ic
N tot ic 1 F MUD term P b
SCH U L
term P b
M i S j ic
UL
SCH UL
Q SCH M i S j ic k = ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Gp
Service
UL
FCH U L
SCH U L
N tot ic 1 F MUD term P b
M i S j ic + P b
M i S j ic
304
AT271_TRG_E6
Forsk 2009

FCH U L
term P b
M i S j ic
UL
UL
- G FCH
Q FCH M i S j ic k = -------------------------------------------------------------------- Service
p
UL
N tot ic
SCH U L
term P b
M i S j ic
UL
UL
- G SCH
Q SCH M i S j ic k = -------------------------------------------------------------------- Service
p
UL
N tot ic
End For
If (Mi is not in handoff)
UL
UL
UL
UL
Q FCH M i = Q FCH M i S j ic k and Q SCH M i = Q SCH M i S j ic k

k

UL
UL
Q FCH M i = f rake efficiency

k
UL
Q FCH M i S j ic k
S j ActiveSet
UL
UL
Q SCH M i = f rake efficiency

k
UL
Q SCH M i S j ic k
S j ActiveSet
Else if (Mi is in soft or softer/soft without MRC)

UL
UL
UL
Q FCH M i = G macro diversity 2 links Max Q FCH M i S j ic k

k
S j ActiveSet
UL
UL
UL
Q SCH M i = G macro diversity 2 links Max Q SCH M i S j ic k

k
S j ActiveSet
Else if (Mi is in soft/soft)

UL
UL
UL
Q FCH M i = G macro diversity 3 links Max Q FCH M i S j ic k

k
S j ActiveSet
UL
UL
UL
Q SCH M i = G macro diversity 3 links Max Q SCH M i S j ic k

k
S j ActiveSet
Else if (Mi is in softer/soft with MRC)
UL
Q FCH M i
k
UL
UL
UL
= Max f rake efficiency
Q FCH ic Q FCH
other site
i AS ActiveSet
UL
ic G macro diversity 2 links
UL
UL
UL
Q SCH ic Q SCH
other site
i AS ActiveSet
UL
ic G macro diversity 2 links
(same site)
UL
Q SCH M i
k
(same site)
EndIf
FCH r eq
P term
UL
Q req Service M i Mobility M i FCH

FCH r eq
M i ic k = ------------------------------------------------------------------------------------------------------- P term
M i ic k 1
UL
Q FCH M i
k
SCH r eq
P term
UL
Q req Service M i Mobility M i SCH_rate_multiple SCH

r eq
- P SCH
M i ic k = ------------------------------------------------------------------------------------------------------------------------------------------------------------ M i ic k 1
term
UL
Q SCH M i
k
req
P term M i
ic k =
req
FCH r eq
P term
M i
ic k +
SCH r eq
P term
M i
ic k
min
If P term M i ic k P term M i then

FCH r eq
P term
SCH r eq
P term
min
P term M i S j
r eq
- P FCH
M i ic k = -------------------------------- M i ic k
term
req
P term M i k
min
P term M i S j
r eq
- P SCH
M i ic k = -------------------------------- M i ic k
term
req
P term M i k
EndIf
FCH r eq
If P term
Forsk 2009
max
M i ic k P term M i then Mi cannot select any station and its active set is cleared
AT271_TRG_E6
305

req
max
If P term M i ic k P term M i and Mi uses SCH then:

Downgrading the service SCH rate:
req
max
UL
UL
While P term M i ic k P term M i and R SCH Service M i R FCH Service M i 2

UL
R SCH Service M i
UL
R SCH Service M i -----------------------------------------------------2
SCH r eq
SCH r eq
P term
UL
UL
P term
M i ic k
Q req Service M i Mobility M i R SCH Service M i SCH
M i ic k = ---------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------UL
UL
2
Q Service M Mobility M R
Service M 2
req
req
FCH r eq
P term M i ic k = P term
SCH r eq
M i ic k + P term
SCH
SCH
M i ic k
EndWhile
req
max
If P term M i ic k P term M i then Mi will not use SCH

Endif
Endif
If the required number of channel elements exceeds the available quantity in the site of Sj (Best server of Mi) and Mi uses
SCH then:
Downgrading the service SCH rate:
While N
CE U L
CE U L
M i N max
UL
UL
S j and R SCH Service M i R FCH Service M i 2
UL
R SCH Service M i
UL
R SCH Service M i -----------------------------------------------------2
CE U L
N SCH M i k
CE U L
N SCH M i k = --------------------------------2
SCH r eq
SCH r eq
P term
SCH UL
UL
P term
M i ic k
Service M i Mobility M i R SCH Service M i
Q req
M i ic k = ---------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------------------------------------------SCH UL
UL
2
Q
Service M Mobility M R
Service M 2
req
req
FCH r eq
P term M i ic k = P term
N
CE U L
SCH r eq
M i ic k + P term
CE U L
CE U L
M i k = N SCH M i k + N FCH
SCH
M i ic k
Mi k
EndWhile
Endif

If Mi uses an SCH on the downlink
For each cell (Sj,ic) in Mi FCH active set
Calculation of quality level on (Sj,ic) FCH at Mi, with the minimum power allowed on FCH for the Mi service
FCH D L
Pb
min
P FCH Service M i
M i S j ic = ----------------------------------------------------L T M i S j
FCH D L
BTS P b
M i S j
DL
DL
- G FCH
Q FCH M i S j ic k = ---------------------------------------------------------------------------------------------------------------------- Service M i
p
DL
DL
N tot ic 1 F ortho BTS P b M i S j ic
FCH D L
BTS P b
M i S j
DL
Q FCH M i S j ic k = ------------------------------------------------------------DL
N tot ic
If cell (Sj,ic) in Mi SCH active set
Calculation of quality level on (Sj,ic) SCH at Mi, with the minimum power allowed on SCH for the Mi service
SCH D L
Pb
306
min
P SCH Service M i
M i S j ic = ----------------------------------------------------L T M i S j
AT271_TRG_E6
Forsk 2009

SCH D L
BTS P b
M i S j
DL
DL
- G SCH
Q SCH M i S j ic k = ---------------------------------------------------------------------------------------------------------------------- Service M i
p
DL
DL
N tot ic 1 F ortho BTS P b M i S j ic
SCH D L
BTS P b
M i S j
DL
Q SCH M i S j ic k = ------------------------------------------------------------DL
N tot ic
EndIf
End For
Recombination of the first f active set links (f is the number of fingers of the Mi terminal): only quality levels from the
first f cells (Sf,ic) of active set are recombined.
DL
DL

k
Q FCH M i S j ic k
Q SCH M i S j ic k
DL
S f ActiveSet FCH
DL
DL

k
DL
S f ActiveSet SCH
Do
For each cell (Sj,ic) in Mi FCH active set
Calculation of the required power for DL traffic channel between (Sj,ic) and Mi:
DL
DL
Q req Service M i Mobility M i R FCH Service M i FCH

req
- P min
P FCH M i S j ic k = ----------------------------------------------------------------------------------------------------------------------------------------------------------------FCH Service M i
DL
Q FCH M i
k
req
max
If P FCH M i S j ic k P FCH Service M i then S j ic is excluded from Mi active set

DL
Recalculation of a decreased Q req

If cell (Sj,ic) in Mi SCH active set
Calculation of the required power for DL traffic channel between (Sj,ic) and Mi:
DL
DL

req
- P min
P SCH M i S j ic k = ----------------------------------------------------------------------------------------------------------------------------------------------------------------SCH Service M i
DL
Q SCH M i
k
Downgrading the service SCH rate (only for (Sj,ic) best server cell of Mi):
req
max
DL
While P SCH M i S j ic k P SCH Service M i R SCH Service M i

req
DL
DL
Or P tx S j ic k + P tch M i S j ic k P max S j ic and R SCH Service M i R FCH Service M i 2

DL
R SCH Service M i
DL
R SCH Service M i = ----------------------------------------------------2
req
DL
DL
P SCH M i S j ic k
req
P SCH M i S j ic k = -------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------DL
DL
2
Service M 2
req
req
req
SCH
SCH
req
P tch M i S j ic k = P SCH M i S j ic k + P FCH M i S j ic k

EndWhile
req
max
req
If P SCH M i S j ic k P SCH Service M i or P tx S j ic k + P tch M i S j ic k P max S j ic then Mi will not use SCH
Endif
While N
CE D L
CE D L
M i N max
DL
DL
S j and R SCH Service M i R FCH Service M i 2
DL
R SCH Service M i
DL
R SCH Service M i = ----------------------------------------------------2
CE D L
N SCH M i k
CE D L
N SCH M i k = --------------------------------2
req
DL
DL
P SCH M i S j ic k
req
P SCH M i S j ic k = -------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------DL
DL
2
Service M 2
req
Forsk 2009
AT271_TRG_E6
SCH
SCH
307

req
req
req

N
CE D L
CE D L
CE D L
M i k = N SCH M i k + N FCH
Mi k
EndWhile
If N
CE D L
CE D L
M i N max
S j then Mi will not use SCH
Endif
While N
Codes
Codes
DL
DL
M i N max S j ic and R SCH Service M i R FCH Service M i 2

DL
R SCH Service M i
DL
R SCH Service M i = ----------------------------------------------------2
Codes
N SCH M i k
Codes
N SCH M i k = -------------------------------2
req
DL
DL
P SCH M i S j ic k
req
P SCH M i S j ic k = -------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------DL
DL
2
Service M 2
req
req
req
SCH
SCH
req

N
Codes
Codes
Codes
M i k = N FCH M i k + N SCH M i k
EndWhile
If N
Codes
Codes
M i N max S j then Mi will not use SCH
Endif
Endif
EndFor
Recombination of the first f active set links (f is the number of fingers of the Mi terminal): only quality levels from the
first f cells (Sf,ic) of active set are recombined.
DL
DL

k
Q FCH M i S f ic k
Q SCH M i S f ic k
DL
S f ActiveSet FCH
DL
DL

k
DL
S f ActiveSet SCH
DL
DL
While Q k M i Q req Service M i Mobility Mi and Mi FCH active set is not empty
DL
DL
And Q k M i Q req Service M i Mobility Mi (if SCH active set is not empty)
Endif
Uplink and Downlink Interference Updates

Update of interference on active mobiles only (old contributions of mobiles and stations are replaced by the new ones)
UL
EndFor
For each mobile Mi
DL
Update of N tot ic
EndFor
Control of Radio Resource Limits (Walsh Codes, Cell Power and Site Channel Elements)
For each cell (Sj,ic) on a site Nl
P tx S j ic k
DL
While ---------------------------- %Power max
P max
req
Rejection of mobile with highest P tch S j M b ic k for the lowest service priority
308
AT271_TRG_E6
Forsk 2009

EndWhile
EndFor
For each site Nl
The list of rejected mobiles for the site Nl is L rejected N l
If the equipment installed on Nl supports power pooling between transmitters
Activation of power pooling between transmitters for each cell (Sj,ic) containing rejected users
Control of the available power for the other cells (Si,ic) of the site where power pooling between transmitters is not activated
If
DL
%Power max P max P tx S i ic k 0
S i ic
Si Nl
Then, the power unused by the cells (Si,ic) of the site can be allocated to cells (Sj,ic)
Sort of all the rejected mobiles by priority in a descending order and by simulation rank in a descending order
For the first mobile Mb of the list ( M b L rejected N l )
req
DL
If P tx S j ic k + P tch S j M b ic k %Power max P max + M Pooling S j ic

Mb is reconnected
EndIf
EndFor
EndIf
EndFor
While N
Codes
Codes
S j ic k N max S j ic
Rejection of last admitted mobile

EndFor
While N
CE DL
CE DL
N I k N max
NI
req
Rejection of mobile with highest P tch M i S j k for the lowest service priority
While N
CE UL
CE UL
N I k N max
NI
req
Rejection of mobile with highest P term M i ic k for the lowest service priority
EndFor

For each cell (Sj,ic) with X
UL
UL
S j ic X max

EndFor
7.4.2.2.3
UL
UL
S j ic X max exists
The convergence criteria are evaluated at each iteration, and can be written as follow:
DL
DL

P tx ic k P tx ic k 1
100
DL = max int ma x ----------------------------------------------------- 100 int ma x -----------------------------------------------------------------DL
P tx ic k
Stations
Stations
N
ic
user
UL
UL
UL
UL

UL = max int ma x ---------------------------------------------------- 100
- 100 int ma x -----------------------------------------------------------------UL
UL
Stations
Stations
I ic
N
ic
tot
user
Forsk 2009
AT271_TRG_E6
309

1st case: Between two successive iterations, UL and DL are lower ( ) than their respective thresholds (defined
when creating a simulation).
has been achieved.
2nd case: After 30 iterations, UL or/and DL are still higher than their respective thresholds and from the 30th iteration,
UL or/and DL do not decrease during the next 15 successive iterations.
1. After the 30th iteration, UL and/or DL equal 100 and do not decrease during the next 15 successive iterations: Atoll
stops the algorithm at the 46th iteration. Convergence has not been achieved.
under the thresholds) and then do not change during the next 15 successive iterations: Atoll stops the algorithm at the 56th
iteration without achieving convergence.
If UL and/or DL are still strictly higher than their respective thresholds, the simulation has not converged (specific divergence symbol).
If UL and DL are lower than their respective thresholds, the simulation has converged.
7.4.2.3
CDMA2000 1xEV-DO Power/Data Rate Control Simulation Algorithm

In a CDMA2000 1xEV-DO system, the UL supports power control but the DL does not. When a DL connection is established, the transmitter transmits at full power (Pmax). Instead of power control, there is a data rate control based on the C/
I ratio computed at the mobile. For each distribution of users, Atoll simulates the power control mechanism for the UL and
the data rate control for the DL.
The simulation uses an iterative algorithm, where in each iteration, all the 1xEV-DO data service users selected during the
user distribution generation (1st step) try to connect to network active transmitters with a calculation area. The process is
repeated from iteration to iteration until convergence is achieved. The algorithm steps are detailed below.
Figure 7.3: CDMA2000 1xEVDO Power Control Algorithm
7.4.2.3.1
UL intra
Uplink received powers on carrier ic, I tot
UL extra
ic , I tot
UL
ic and I inter carrier ic , at base station Sj are initialised to 0 W
(no connected mobile).
310
AT271_TRG_E6
Forsk 2009

UL
I tot S j ic
UL
X k S j ic = ---------------------------- = 0
UL
N tot S j ic
7.4.2.3.2

E c UL
The algorithm is detailed for any iteration k. Xk is the value of the variable X at the iteration k. In the algorithm, ------
is
N t min
the minimum pilot quality level on uplink.
In case of 1xEV-DO Rev. 0 users, this threshold depends on the user mobility and is defined in the Mobility parameters
table.
Ec
For 1xEV-DO Rev. A users, the value of ------
depends on the user requested data rate. This data rate can be obtained
N t min
UL
by using a certain uplink 1xEV-DO Rev. A radio bearer ( Index UL RevA Bearer ) in a certain number of subframes ( n SF ).
E
-----c-
is the value defined in the 1xEV-DO Rev. A Radio Bearer Selection (Uplink) table for the combination (radio
N t min
UL
bearer Index, mobility and number of subframe) providing the user requested data rate. Two values are available for this
parameter, one when the service uplink mode is "Low Latency" and another one for high capacity services.
All variables are described in Definitions and formulas part (see "Definitions and Formulas" on page 276).
For each mobile Mi

For each station Sj containing Mi in its calculation area,
Determination of BestCarrier k S j M i .
If a given carrier is specified for the service requested by Mi and if it is used by Sj
BestCarrier k S j M i is the carrier specified for the service.
Else the carrier selection mode defined for Sj is considered.
For each carrier ic used by Sj, we calculate current loading factor:
UL
I tot S j ic
UL
UL
X k S j ic = ---------------------------- + X
UL
N tot S j ic
EndFor
UL
BestCarrier k S j M i is the carrier with the lowest X k S j ic

BestCarrier k S j M i is the carrier with the lowest P tx S j ic k
BestCarrier k S j M i is randomly selected
UL
UL
BestCarrier k S j M i is the first carrier so that X k S j ic X max

DL
BTS P tot M i S j BestCarrier b pilot

Calculation of Q pilot M i S j BestCarrier = ---------------------------------------------------------------------------------------------------------------DL
k
I 0 BestCarrier k S j M i b pilot
req
If Q pilot M i S j BestCarrier Q pilot then Sj is rejected by Mi

k
max
If Q pilot M i S j BestCarrier Q pilot M i

k
Admission control (If simulation respects a load factor constraint and Mi was not connected in previous iteration).
UL
UL
If X k S j BestCarrier k S j M i X max , then Sj is rejected by Mi

Else
max
Q pilot M i = Q pilot M i S j BestCarrier

k
Forsk 2009
AT271_TRG_E6
311

S BS M i = S j
Endif
EndFor
If no SBS has been selected, Mi cannot get a connection to the network
In the following lines, we will consider ic = BestCarrier k S BS M i M i

For each station Sj containing Mi in its calculation area, using ic, and if neighbours are used, neighbour of SBS(Mi)
DL
BTS P tot M i S j ic b pilot

Calculation of Q pilot M i S j ic = ---------------------------------------------------------------------------------DL
k
I 0 ic b pilot
min
If Q pilot M i S j ic Q pilot then Sj is rejected by Mi

k
Else Sj is included in the Mi active set

Rejection of Sj if the Mi active set is full
k
EndFor

req
Calculation of the required power for Mi, P term M i ic k

For each cell (Sj,ic) present in the Mi active set
Calculation of quality level on Mi traffic channel at (Sj,ic), with the minimum power allowed on traffic channel for the Mi
service
req
P term M i ic k 1
UL
P b M i S j ic = -----------------------------------------L T M i S j
UL
UL
term P b M i S j ic
UL
M i S j ic k = ---------------------------------------------------------------------------------------------------------------------- G p Service
UL
Tx
UL
N tot ic 1 F MUD term P b M i S j ic

UL
UL
term P b M i S j ic
UL
M i S j ic k = -------------------------------------------------------- G p Service
UL
N tot ic
End For
If (Mi is not in handoff)
UL
Q total M i = Q
UL
M i S j ic

UL
UL
Q total M i = f rake efficiency

k
UL
M i S j ic k
S j ActiveSet
Else if (Mi is in soft or softer/soft without MRC)

UL
Q total M i =
k
Max Q
UL
I AS ActiveSet
UL
M i S j ic k G macro diversity 2 links
Else if (Mi is in soft/soft)

UL
Q total M i =
k
Max Q
UL
I AS ActiveSet
UL
M i S j ic k G macro diversity 3 links
Else if (Mi is in softer/soft with MRC)
UL
Q total M i
k
UL
UL
UL
Q M i S j ic k Q M i S j ic k
othersite
i AS ActiveSet
(same site)
312
AT271_TRG_E6
G UL
macro diversity 2 links
Forsk 2009

EndIf
UL
Q req Service M i Mobility M i

req
req
P term M i ic k = ---------------------------------------------------------------------------------------- P term M i ic k 1
UL
Q total M i
k
If the service of Mi uses Transmission Control Protocol (TCP)

For the best server cell (Sk,ic) of Mi
Calculation of the Mi downlink application throughput
DL
Calculation of N tot ic b traffic

DL
DL
j
- + N0
Ptot Sj i c btraffic + txj-------------------------------------------------------------------RF ic ic adj
DL
N tot ic b traffic =
term
j j k
DL
Calculation of the maximum data rate supplied to Mi, R max M i S k

Calculation of pilot quality level at Mi
DL
Ec
P tot M i S k ic b pilot
------ M i S k ic b pilot = -----------------------------------------------------DL
Nt
N tot ic b pilot
If Mi is an 1xEV-DO Rev. 0 service user, determination of the maximum data rate from the graph (Max rate=f(C/I)) specified
for the mobility type of Mi
Ec
DL
R max M i S k = f ------ M i S k ic b pilot
Nt
If Mi is an 1xEV-DO Rev. A service user, selection of the downlink 1xEV-DO Rev. A radio bearer ( Index DL R evA Bearer ):
DL
E
E
Index DL R evA Bearer where -----c- M i S k ic b pilot -----c- Index DL R evA Bearer
Nt
min
Nt
DL
R RLC Peak Index DL R evA Bearer

DL
Determination of the maximum data rate: R max M i S k = -------------------------------------------------------------------------------------------n TS Index DL R evA Bearer
DL
DL
R application M i S k = R max M i S k SF rate Service Mi R Service Mi

UL
Determination of the uplink data rate due to TCP acknowledgements, R TCP ACK M i S k from the graph (UL Thr due to
TCP=f(DL Thr) specified for the service of Mi
UL
DL
R TCP ACK M i S k = f R application M i S k

UL
UL
UL
Determination of the nearest lower and higher supported rates ( R low and R high ) for R TCP ACK M i S k
UL
UL
UL
UL
UL
UL
For R low and R high , calculation of CI req R low and CI req R high
Ec
UL
CI req = ------
1 + G DRC + G TCH for DO Rev.0 terminals
N t min
UL
And
E c UL
UL
CI req = ------
1 + G DRC + G TCH + G RRI + G Auxiliary pilot for DO Rev.A terminals
N t min
EndFor
UL
UL
UL
UL
UL
UL
UL
UL
Linear interpolation of CI req R TCP ACK between CI req R low and CI req R high
UL
UL
CI req = CI req R
UL
+ CI req R TCP ACK
W
UL
UL
Q req = CI req ------------------------------------------------UL
UL
R + R TCP ACK
EndIf
req
req
min
P term M i ic k = Max P term M i ic k P term M i S j

req
max
If P term M i ic k P term M i then:

Downgrading the traffic data channel data rate:
Forsk 2009
AT271_TRG_E6
313

req
max
While P term M i ic k P term M i

And R
UL
Service M i 9.6kbps for DO-Rev.0 users or R
UL
Service M i 4.8kbps for DO-Rev.A users
req
UL
P term M i ic k
req
- R UL Service M i ( R low Service M i is the nearest lower supported data
P term M i ic k = -------------------------------------------------UL
R low Service M i
rate)
R
UL
UL
Service M i = R low Service M i
EndWhile
req
max
If P term M i ic k P term M i then Mi is rejected

Endif
Endif
Uplink Interference Updates

Update of interference on active mobiles only (old contributions of mobiles and stations are replaced by the new ones)
UL
EndFor
Control of Radio Resource Limits (Number of EVDO users, MAC Indices and Site Channel Elements)
While n
EVDO
EVDO
S j ic n max S j ic
Rejection of the last admitted mobile

EndFor
While N
MacIndexes
MacIndexes
S j ic N max
S j ic

EndFor
While N
EVDO CE
EVDO CE
N I k N max
NI

EndFor

For each cell (Sj,ic) with NR
While NR
UL
UL
UL
UL
S j ic NR threshold S j ic + NR threshold S j ic
UL
UL
S j ic NR threshold S j ic + NR threshold S j ic and there is at least one mobile that can be downgraded
Downgrading the traffic data channel rate for all 1xEV-DO Rev. 0 mobiles for which the data rate transition flag is set to
"True".
UL
Endwhile
While NR
UL
UL
UL
UL
S j ic NR threshold S j ic NR threshold S j ic
UL
UL
S j ic NR threshold S j ic NR threshold S j ic and there is at least one mobile that can be upgraded
Upgrading the traffic data channel rate for all 1xEV-DO Rev. 0 mobiles for which the data rate transition flag is set to "True".
UL
Endwhile
314
UL
UL
UL
S j ic NR threshold S j ic + NR threshold S j ic
AT271_TRG_E6
Forsk 2009

While NR
UL
UL
UL
S j ic NR threshold S j ic + NR threshold S j ic and there is at least one mobile to be downgraded
Downgrading the traffic data channel rate for the 1xEV-DO Rev. A mobile with the highest data rate
UL
Endwhile
While NR
UL
UL
UL
UL
S j ic NR threshold S j ic NR threshold S j ic
UL
UL
S j ic NR threshold S j ic NR threshold S j ic and there is at least one mobile to be upgraded
Upgrading the traffic data channel rate for the 1xEV-DO Rev. A mobile with the lowest data rate (only 1xEV-DO Rev. A
mobiles which have not been downgraded can be upgraded. In addition, the upgraded data rate cannot exceed the initial
user data rate drawn by the Monte-Carlo algorithm. This means that only mobiles downgraded during the uplink power
control step can be upgraded)
UL
Endwhile
For each cell (Sj,ic) with X
UL
UL
S j ic X max

EndFor
UL
UL
S j ic X max exists
Downlink Data Rate Control

For each mobile Mi connected to a cell (Sk,ic)
DL
Calculation of N tot ic b traffic

For each cell (Sj,ic) ( k j )
Determination of the number of mobiles connected to the cell (Sj,ic), N mobiles S j ic
If N mobiles S j ic = 0 then, P tx S j ic b traffic = G idle power P max S j ic
Else P tx S j ic b traffic = P max S j ic
EndFor
DL
N tot ic b traffic =
Ptot Sj ic btraffic + N0
DL
term
j j k
EndFor
DL
Calculation of the maximum data rate supplied to Mi, R max

For the best server cell (Sk,ic) of Mi
Calculation of pilot quality level at Mi
DL
Ec
P tot M i S k ic b pilot
------ M i S k ic b pilot = -----------------------------------------------------DL
Nt
N tot ic b pilot
If Mi is an 1xEV-DO Rev. 0 service user, determination of the maximum data rate from the graph (Max rate=f(C/I)) specified
for the mobility type of Mi
Ec
DL
R max M i S k = f ------ M i S k ic b pilot
Nt
If Mi is an 1xEV-DO Rev. A service user, selection of the downlink 1xEV-DO Rev. A radio bearer ( Index DL R evA Bearer ):
DL
Ec
Ec
Index DL R evA Bearer for which ------ M i S k ic b pilot ------ Index DL R evA Bearer
Nt
min
Nt
DL
R RLC Peak Index DL R evA Bearer

DL
Determination of the maximum data rate: R max M i S k = -------------------------------------------------------------------------------------------n TS
EndFor
DL
Calculation of the average cell data rate, R av
Forsk 2009
AT271_TRG_E6
315

DL
DL
If N mobiles S j ic = 1 , then R av S j ic = R max M i S j ic

Else if N mobiles S j ic 1 , determination of the multi-user gain G MU from the graph (MUG table=f(nb users)) specified
for (Sj,ic)
DL
R av S j
ic =
M N
DL
R max M i S j ic
S ic
i
mobiles j
- 1 ER DRC S j ic
G MU N mobiles S j ic --------------------------------------------------------------------N mobiles S j ic
N mobiles
1 TS BCMCS S j ic TS EVDO CCH S j ic + R BCMCS S j ic TS BCMCS S j ic

EndIf
EndFor
7.4.2.3.3
The algorithm convergence is studied on uplink only. The uplink convergence criterion is evaluated at each iteration, and
can be written as follow:
UL
UL
UL
UL

UL = max int ma x ---------------------------------------------------- 100
- 100 int ma x -----------------------------------------------------------------UL
UL
Stations
Stations
I ic
N
ic
tot
user

1st case: Between two successive iterations, UL is lower ( ) than the threshold (defined when creating a simulation).
Example: Let us assume that the maximum number of iterations is 100, UL convergence threshold is set to 5. If UL 5
between the 4th and the 5th iteration, Atoll stops the algorithm after the 5th iteration. Convergence has been achieved.
2nd case: After 30 iterations, UL is still higher than the threshold and from the 30th iteration, UL does not decrease
during the next 15 successive iterations.
Examples: Let us assume that the maximum number of iterations is 100, UL convergence threshold is set to 5.
1. After the 30th iteration, UL equals 100 and do not decrease during the next 15 successive iterations: Atoll stops the
algorithm at the 46th iteration. Convergence has not been achieved.
2. After the 30th iteration, UL equals 80, it starts decreasing slowly until the 40th iteration (without going under the threshold) and then does not change during the next 15 successive iterations: Atoll stops the algorithm at the 56th iteration without achieving convergence.
If UL is still strictly higher than the threshold, the simulation has not converged (specific divergence symbol).
If UL is lower than the threshold, the simulation has converged.
7.4.3
Appendices
7.4.3.1
Admission Control
During admission control, Atoll calculates the uplink load factor of a considered cell assuming the mobile concerned is
connected with it. Here, activity status assigned to users is not taken into account. So even if the mobile is not active on
UL, it can be rejected due to cell load saturation. To calculate the cell UL load factor, either Atoll takes into account the
mobile power determined during power control if mobile was connected in previous iteration, or it estimates a load rise due
to the mobile and adds it to the current load. The load rise ( X
X
UL
UL
) is calculated as follows:
1
= --------------------------------------W
1 + ----------------------------UL
UL
Q req R
In case of IS-95 cdmaOne and CDMA2000 1xRTT networks, we have:

UL
UL
UL
Q req = Q req FCH + Q req SCH and R
316
UL
UL
UL
= R FCH + R SCH
AT271_TRG_E6
Forsk 2009
7.4.3.2
Resources Management
7.4.3.2.1
Walsh Code Management

Walsh codes are managed on the downlink during a simulation in case of IS-95 cdmaOne and CDMA2000 1xRTT
networks. Atoll performs Walsh code allocation during the radio resource control step.
Walsh codes form a binary tree with codes of a longer length generated from codes of a shorter length. Length-k Walsh
codes are generated from length-k/2 Walsh codes. Therefore, if a channel needs 1 length-k/2 Walsh code, it is equivalent
to using 2 length-k Walsh codes, or 4 length-2k Walsh codes and so on.
Figure 7.4: Walsh Code Tree Indices (Not Walsh Code Numbers)
128 128-bit-length Walsh codes per cell are available in CDMA2000 and IS-95 cdmaOne documents.
During the resource control, Atoll determines the number of 128-bit-length Walsh codes that will be consumed by each
cell. Therefore, it allocates :
A code with the longest length (i.e. a 128 bit-length code) per common channel for each cell,
Two 128 bit-length codes per cell-receiver link for FCH in RC1, RC2, RC3 or RC5 and only one for FCH in RC4.
The number of 128 bit-length codes to be allocated per cell-receiver link for SCH (in case SCH is supported by the
Walsh codes
user radio configuration), N 128 bits

Walsh codes
N 128 bits
, is determined as follows:
DL
= Frate SCH 2 for RC1, RC2, RC3 and RC5,
And
Walsh codes
N 128 bits
DL
= Frate SCH for RC4.
Where
DL
Frate SCH is the SCH rate factor.

The Walsh code allocation follows the Buddy algorithm, which guarantees that:
7.4.3.2.2
If a k-length Walsh code is used, all of its children with lengths 2k, 4k, , cannot be used as they are not orthogonal.
If a k-length Walsh code is used, all of its ancestors with lengths k/2, k/4, , cannot be used as they are not orthogonal.
Notes:
The Walsh code allocation follows the mobile connection order (mobile order in the Mobiles
tab).
The Walsh code and channel element management is dealt with differently in case of
softer handoff. Atoll allocates Walsh codes for each transmitter-receiver link while it
assigns channel elements globally to a site.
Channel Element Management

Channel elements are controlled in the simulation.
IS95 cdmaOne and CDMA2000 1xRTT networks

Atoll checks the availability of this resource on uplink and downlink.
Overhead CE UL
FCH CE UL
CE UL
channel elements for control channels (Pilot channel),

UL
1 + Frate SCH per cell-receiver link, for TCH (TCH correspond to Traffic channels i.e. FCH and
SCH).
Forsk 2009
AT271_TRG_E6
317

N
CE UL
NI =
CE UL
CE UL
N I , is:
j NI
On downlink, Atoll consumes N

Overhead CE DL
N
channel),
FCH CE DL
CE DL
channel elements for control channels (Pilot channel, Synchronisation channel, Paging
DL
1 + Frate SCH per cell-receiver link, for TCH (TCH correspond to Traffic channels i.e. FCH and
SCH).
Therefore, the number of channel elements required on downlink at the site level, N
N
CE DL
NI =
CE DL
CE DL
N I , is:
j NI
Note:
In case of softer handover (the mobile has several links with co-site cells), Atoll allocates
channel elements for the best serving cell-mobile link only.
CDMA2000 1xEV-DO networks

CE UL
2 channel elements for control channels (Pilot channel, Data Rate Control channel, etc ). This value is fixed and
hard-coded.
TCH CE UL
per cell-receiver link, for TCH (Traffic channels).
N
CE UL
NI =
CE UL
CE UL
N I , is:
j NI
On downlink, only one user can be served by a cell at a time, so this resource is not limited.
7.4.3.3
Downlink Load Factor Calculation

Atoll calculates the downlink load factor for each cell (available in the Cells tab of any given simulation results) and each
connected mobile (available in the Mobiles tab of any given simulation results).
7.4.3.3.1
Downlink Load Factor per Cell

The downlink load factor is calculated for each CDMA2000 1xRTT and IS-95 cdmaOne cell.
Approach for downlink load factor evaluation is highly inspired by the downlink load factor defined in the book WCDMA
for UMTS by Harry Holma and Antti Toskala.
DL FCH
DL SCH
Q req
Q req
Let CI req = ------------------------ + ------------------------ be the required quality.
DL FCH
DL SCH
Gp
Gp
FCH
SCH
So, we have CI req = CI req + CI req
In case of soft handoff, required quality is limited to the effective contribution of the transmitter.
DL
ortho
P tx ic = P pilot ic + P sync ic + P paging ic + P SCH ic + P FCH ic = P CCH ic +
Ptch ic
tch
where
ortho
P CCH ic = P pilot ic + P sync ic + P paging ic
Ptch ic = PSCH ic + PFCH ic

tch
At mobile level, we have a required power, Ptch:

term
P tch ic = CI req I extra ic + I intra ic + I inter carrier ic + N 0
318
AT271_TRG_E6
LT
Forsk 2009

DL
P tx ic P tch ic
term
P tch ic = CI req I extra ic + I inter carrier ic + 1 F ortho BTS ----------------------------------------------- + N 0 L T
LT
DL
term
I extra ic + I inter carrier ic L T + 1 F ortho BTS P tx ic + N 0

LT
P tch ic = ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------1
------------- + 1 F ortho BTS
CI req
where
DL
I intra ic is the total power received at receiver from the cell to which it is connected.
DL
I extra ic is the total power received at receiver from other cells.

I inter carrier ic is the inter-carrier interference received at receiver.
DL
ortho
P tx ic = P CCH ic +
DL
term
I extra ic + I inter carrier ic L T + 1 F ortho BTS P tx ic + N 0

LT
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------1
------------+
tch
ortho
BTS
CI req
We have:
I extra ic + I inter carrier ic L T
DL
term
LT
--------------------------------------------------------------------------------------- + 1 F ortho BTS P tx ic + N 0
DL
P tx ic
DL
ortho
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------P tx ic = P CCH ic +
1
------------- + 1 F ortho BTS
tch
CI req
- + 1 F ortho BTS
--------------------------------------------------------------------------------------DL
P ic
DL
tx
tch
DL
P tx ic --------------------------------------------------------------------------------------------------------------------------------------------------------- P tx ic
1
------------- + 1 F ortho BTS
CI req
ortho
= P CCH ic +
term
N0
LT
--------------------------------------------------------------------
1
tch ------------- + 1 F ortho BTS

CI req
term
N0
LT
-------------------------------------------------------------------1
tch ------------- + 1 F ortho BTS
CI req
DL
P tx ic = ------------------------------------------------------------------------------------------------------------I extra ic L T
------------------------------------ + 1 F ortho BTS
DL
P tx ic
-------------------------------------------------------------------------------------1
1 + 1 F
------------ortho BTS
tch
CI req
ortho
P CCH ic +
Therefore, the downlink load factor can be expressed as:
DL

---------------------------------------------------------------------------------------+ 1 F ortho BTS
DL
P tx ic
-----------------------------------------------------------------------------------------------------------------------------------------=
1
------------- + 1 F ortho BTS
tch
CI req
The downlink load factor represents the signal degradation in relative to the reference interference (thermal noise).
7.4.3.3.2
Downlink Load Factor per Mobile

Atoll evaluates the downlink load factor for any connected mobile (CDMA2000 1xRTT 1xEV-DO or IS-95 cdmaOne user)
as follows,
X
DL
DL
I tot ic
= -------------------DL
N tot ic
7.5
IS-95 cdmaOne and CDMA2000 Prediction Studies
7.5.1
Point Analysis: The AS Analysis Tab

Let us imagine a probe receiver with associated terminal, mobility and service with certain UL and DL rates. This receiver
does not create any interference. The analysis is based on the uplink load percentage and the downlink total power of
Forsk 2009
AT271_TRG_E6
319

cells. These parameters can be either outputs of a given simulation, average values calculated from a group of simulations,
or user-defined cell inputs.
7.5.1.1
Bar Graph and Pilot Sub-Menu

Atoll provides the same outputs in the bar graph and pilot sub-menu whichever the studied network, IS-95 cdmaOne,
CDMA2000 1xRTT or 1xEV-DO.
Atoll proceeds as in power control simulation. It determines the best carrier of each transmitter i (its best cell) that contains
the receiver in its calculation area. The best carrier selection depends on the option chosen in Equipment (UL minimum
noise, DL minimum power, random, sequential). Then, Atoll calculates pilot quality at the receiver from these transmitters
on their best carriers and defines the best server (on its best carrier).
1.
Ec/I0 (or Q pilot ic ) evaluation
We assume that ic is the best carrier of a transmitter i containing the receiver in its calculation radius.
For, IS-95 cdmaOne and CDMA2000 1xRTT users we have,
BTS P c i ic
term
DL
DL
DL
DL
Q pilot i ic = ------------------------------------------------- with I 0 ic = P tot i ic + I extra ic + I inter carrier ic + N 0
DL
I 0 ic
For CDMA2000 1xEV-DO users, we have,
DL
BTS P tot i ic b pilot

Q pilot i ic = --------------------------------------------------------------------DL
I 0 ic b pilot
DL
DL
DL
DL
term
With I 0 ic b pilot = P tot i ic b pilot + I extra ic b pilot + I inter carrier ic b pilot + N 0
The calculation of Q pilot i ic can be divided into 6 steps explained in the table below.
IS-95 cdmaOne and CDMA2000 1xRTT users
CDMA2000 1xEV-DO users

DL
P tot i ic b pilot calculation for each cell (i,ic)

P c i ic calculation for each cell (i,ic)
DL
1st step
P c i ic is the pilot power from a transmitter i on the carrier ic at the
P tot i ic b pilot is the pilot burst from the transmitter i on the carrier
ic at the receiver.
receiver.
P tx i ic b pilot
DL
P tot i ic b pilot = ------------------------------------LT
P pilot i ic
P c i ic = --------------------------LT
and
P tx i ic b pilot = P max i ic

L T is the total loss between the transmitter i and the receiver: --------------------------------------------------------------------------------------------------------------------------------------------------I
G Tx G term
DL
DL
I extra ic and I inter carrier ic calculation

We have,
DL
I extra ic =
Ptot j ic
DL
DL
DL
2nd step
For each transmitter of the network, P tot j ic is the total power

received at the receiver from the transmitter j on the best carrier ic of
the transmitter i.
P tx j ic
DL
P tot j ic = --------------------LT
We have,
DL
I extra ic b pilot =
Ptot j ic bpilot
DL
j j i
and
P tx j ic is the total power transmitted by the transmitter j on the best

carrier of the transmitter i.
DL
I extra ic b pilot and I inter carrier ic b pilot calculation
j j i
Ptot j icadj bpilot

DL
DL
I inter carrier ic
Finally, we have,
j j
b pilot = ------------------------------------------------------RF ic ic adj
3rd step
DL
P tot j ic adj
DL
j j
I inter carrier ic = --------------------------------------RF ic ic adj
320
term
N0
calculation
term
N0
= F term K T W
AT271_TRG_E6
Forsk 2009
4th step
DL
I 0 ic and Q pilot i ic evaluation based on formulas defined above

DL
G macro diversity calculation

DL
The macro-diversity gain, G macro diversity , models the decrease in shadowing margin due to the fact there are several pilot signals at the
5th step
mobile.
DL
npaths

npaths
M Shadowing Ec Io is the shadowing margin for the mobile receiving n pilot signals (not necessarily from transmitters belonging to the mobile
active set).
Note: This parameter is determined from the fixed cell edge coverage probability and the model standard deviation. When the model standard
deviation is set to 0, the macro-diversity gain equals 0.
Determination of active set
Atoll takes the transmitter i with the highest Q pilot i ic and calculates the best pilot quality received with a fixed cell edge coverage
Resulting
probability Q pilot
Resulting
Q pilot
ic
Resulting
Q pilot
ic .
DL
G macro diversity
req
Q pilot
max Q pilot i ic
Resulting
means that the pilot quality at the receiver exceeds Q pilot
ic x% of times (x is the fixed cell edge coverage
6th step
probability). The cell with the highest Q pilot i ic enters the active set as best server ( Q pilot BS ic ) and the best carrier (icBS) of the best
server
BS will be the one used by other transmitters of active set (when active set size is greater than 1). Pilot is available.
Resulting
If Q pilot
req
ic Q pilot , no cell (i,ic) can enter the active set. Pilot is unavailable.
Then, pilot qualities at the receiver from transmitters i (other than the best server) on the best carrier of the best server, icBS, are recalculated
to determine the entire receiver active set (when active set is greater than 1). Same formulas and calculation method are used to update
DL
I 0 ic BS and determine Q pilot i ic BS .

Other cells (i,icBS) in active set must fulfill the following criteria:
pilot
Q pilot i ic BS Q min
i ic BS neighbour list BS ic BS (optional)

2.
Number of cells in active set
This is a user-defined input in the terminal properties. It corresponds to the active set size.
3.
Number of fingers
The number of fingers, f, of the rake receiver. This parameter is defined in the terminal properties. It is relevant in
CDMA2000 1xRTT only4. This is the maximum number of active set links that the terminal (rake) can combine.
4.
Thermal noise
This parameter is calculated as described above (3rd step).

5.
Io (Best server)
Io (Best server) is the total noise received at the receiver on icBS.

6.
Downlink macro-diversity gain
This parameter is calculated as described above (5th step).
7.5.1.2
Downlink Sub-Menu
Outputs calculated by Atoll depend on the studied network (IS-95 cdmaOne, CDMA2000 1xRTT or CDMA2000 1xEV-DO).
7.5.1.2.1
IS-95 cdmaOne
Let m denote the number of cells in the receiver active set and f be the number of fingers defined for the terminal. We
assume that f is less than or equal to m5.
Among the m cells of the receiver active set, only the first f cells will be considered in order to determine the traffic channel
availability on downlink. Each of these cells is noted (k,icBS).
Atoll calculates the traffic channel quality from each cell (k,icBS). No power control is performed as in simulations. Here,
Atoll determines the downlink traffic channel quality at the receiver for the maximum traffic channel power per transmitter
4.
In IS-95 cdmaOne, the number of fingers is the same than the active set size and CDMA2000 1xEV-DO systems
do not support soft handover on downlink.
5.
In IS-95 cdmaOne, the number of fingers is normally the same than the active set size.
Forsk 2009
AT271_TRG_E6
321

allowed on the fundamental channel (FCH). Then, after combination, the total downlink traffic channel quality is evaluated
and compared with the specified target quality.
1.
Eb/Nt target
The Eb/Nt target parameter is available in the Services table. This is the user-defined downlink traffic data quality target
DL
on FCH ( Q req FCH ) for a given service and a terminal.

2.
Required transmitter power on FCH

req
The calculation of the required transmitter power on FCH ( P FCH ) may be divided into three steps.
1st step: Eb/Nt max for the first f (number of fingers) cells of active set
DL
Let us assume the following notation: Eb/Nt max corresponds to Q max .

DL
BTS P b max k ic BS
DL
- G DL
Q max k ic BS = ------------------------------------------------------------p
DL
N tot ic BS
max
P FCH
DL
DL
DL
DL
DL
term
- and N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
With P b max k ic BS = ------------LT
k
Where
max
P FCH is the maximum power allowed on FCH. This parameter is user-defined in the Services table.
L T is the total loss between the transmitter i and the receiver.
k

DL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
DL
With
DL
I intra ic BS = 1 BTS F ortho P DL k ic

tot
BS
DL
I extra ic BS =
Ptot j icBS
DL
j j k
DL
For each transmitter in the network, P tot ic BS is the total power received at the receiver from this transmitter on icBS.
DL
I inter carrier ic BS is the inter-carrier interference at the receiver on the best carrier of the best server.
Ptot j icadj
DL
DL
I inter carrier ic BS
txj j
= ------------------------------------------RF ic BS ic adj

DL
Q MAX is the traffic channel quality at the receiver on icBS after combining the signal from each cell (k,icBS).
DL
DL
DL
DL
Qmax k icBS
DL
Where
DL
f rake efficiency is the downlink rake efficiency factor defined in the Terminal properties.
req
3rd step: P FCH calculation
322
AT271_TRG_E6
Forsk 2009

DL
Q req FCH
req
- P max
P FCH = ------------------------------FCH
DL
Q MAX ic BS
3.
Eb/Nt max for the first f (number of fingers) cells of active set
DL
Let us assume the following notation: Eb/Nt max corresponds to Q max .

DL
BTS P b max k ic BS
DL
- G DL
Q max k ic BS = ------------------------------------------------------------p
DL
N tot ic BS
max
P FCH
term
DL
DL
DL
DL
DL
- and N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
With P b max k ic BS = ------------LT
k
Where
max
P FCH is the maximum power allowed on FCH. This parameter is user-defined in the Services table.
k

DL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
DL
With
max
req
DL
P FCH P FCH
I intra ic BS = 1 BTS F ortho P DL k ic 1
tot
BS
BTS max ( ----------------------------------,0)
LT
k
And
DL
I extra ic BS =
Ptot j icBS
DL
j j k
DL
For each transmitter in the network, P tot ic BS is the total power received at the receiver from the transmitter on icBS.
DL
Ptot j icadj
DL
DL
txj j

4.
Eb/Nt max
DL
Q MAX is the traffic channel quality at the receiver on icBS after combining the signal from each cell (k,icBS).
DL
DL
DL
DL
Qmax k icBS
DL
Where
DL
f rake efficiency is the downlink rake efficiency factor defined in the Terminal properties.
DL
DL
Therefore, the service on the downlink traffic channel is available if Q MAX ic BS Q req FCH .
5.
Effective Eb/Nt
DL
Q eff is the downlink effective traffic channel quality at the receiver on icBS.
DL
DL
DL
Q eff = min Q MAX Q req FCH
Forsk 2009
AT271_TRG_E6
323

6.
Downlink soft handover gain

DL
G SHO corresponds to the DL soft handover gain.

DL
Q MAX ic BS
DL
G SHO = -------------------------------------------------------DL
max Q max k ic BS
k
max
k
7.5.1.2.2
DL
Q max k
DL
ic BS corresponds to the highest Q max k ic BS value.
CDMA2000 1xRTT
Let mFCH and mSCH respectively denote the number of cells in the receiver active set for the fundamental channel (FCH)
and the supplemental channel (SCH) and f be the number of rake fingers defined for the terminal. We assume that f is less
than or equal to mFCH and mSCH.
Among the mFCH cells of the receiver active set, only the first f cells will be considered in order to determine the FCH availability on downlink. In the same way, only the first f cells among the mSCH cells of the receiver active set will be considered
in order to determine the SCH availability on downlink. Each of these cells is noted (k,icBS).
Atoll calculates the traffic channel quality on FCH from each cell (k,icBS). No power control is performed as in simulations.
Here, Atoll determines the downlink traffic channel quality on FCH at the receiver for the maximum traffic channel power
per transmitter allowed on FCH. Then, after combination, the total downlink traffic channel quality on FCH is evaluated and
compared with the specified target quality.
Atoll calculates the traffic channel quality on SCH from each cell (k,icBS). No power control is performed as in simulations.
Here, Atoll determines the downlink traffic channel quality on SCH at the receiver for the maximum traffic channel power
per transmitter allowed on SCH. This value depends on the downlink data rate specified in the analysis. Then, after combination, the total downlink traffic channel quality on SCH is evaluated and compared with the specified target quality.
1.
Eb/Nt target on FCH and Eb/Nt target on SCH

DL
Eb/Nt target on FCH ( Q req FCH ) is the downlink traffic data quality target on the fundamental channel (FCH). This value
is user-defined for a given service and terminal.
DL
Eb/Nt target on SCH ( Q req SCH ) is the downlink traffic data quality target on the supplemental channel (SCH). This value
is specified for a given service, terminal and SCH rate.
2.
Required transmitter powers on FCH and SCH

req
req
The calculation of the required transmitter powers on FCH and SCH ( P FCH and P SCH ) may be divided into three steps.
1st step: Eb/Nt max for the first f (number of fingers) cells of active set
DL
Let us assume the following notations: Eb/Nt max on FCH and SCH respectively correspond to Q max FCH and
DL
Q max SCH .
DL FCH
BTS P b max k ic BS
DL
FCH
- G DL
Q max k ic BS FCH = ----------------------------------------------------------------p
DL
N tot ic BS
And
DL SCH
BTS P b max k ic BS
DL
SCH
- G DL
Q max k ic BS SCH = -----------------------------------------------------------------p
DL
N tot ic BS
DL FCH
With P b
max
max
P FCH
P SCH
DL SCH
k ic BS = ------------- , P b max k ic BS = ------------LT
LT
k
DL
DL
DL
DL
term
And N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0

Where
max
P FCH is the maximum power allowed on FCH. This parameter is user-defined in the Services table for a certain terminal.
max
P SCH is the maximum power allowed on SCH for the specified downlink data rate. This parameter is user-defined in the
Services table for a certain terminal and SCH rate.
k

DL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
324
AT271_TRG_E6
Forsk 2009

DL
With
DL
I intra ic BS = 1 BTS F ortho P DL k ic

tot
BS
And
DL
I extra ic BS =
Ptot j icBS
DL
j j k
DL
For each transmitter in the network, P tot ic BS is the total power received at the receiver from this transmitter on icBS.
DL
Ptot j icadj
DL
DL
txj j

2nd step: Calculation of the total traffic channel quality on FCH and SCH
DL
Q MAX FCH is the traffic channel quality on FCH at the receiver on icBS after combining the signal from each cell (k,icBS).
DL
DL
Q MAX ic BS FCH = Q max k ic BS FCH

DL
DL
Q MAX ic BS FCH = f rake efficiency
Qmax k icBS FCH

DL
Where
DL
DL
Q MAX SCH is the traffic channel quality on SCH at the receiver on icBS after combining the signal from each cell (k,icBS).
DL
DL
Q MAX ic BS SCH = Q max k ic BS SCH

DL
DL
Q MAX ic BS SCH = f rake efficiency
Qmax k icBS SCH

DL
req
req
3rd step: P FCH and P SCH calculation

DL
Q req FCH
req
- P max
P FCH = ---------------------------------------------FCH
DL
Q MAX ic BS FCH
DL
Q req SCH
req
- P max
P SCH = ---------------------------------------------SCH
DL
Q MAX ic BS SCH
3.
Eb/Nt max on FCH for the first f (number of fingers) cells of active set
DL
Let us assume the following notation: Eb/Nt max on FCH corresponds to Q max FCH .
DL FCH
BTS P b max k ic BS
DL
FCH
- G DL
Q max k ic BS FCH = ----------------------------------------------------------------p
DL
N tot ic BS
DL FCH
With P b max
max
P FCH
DL
DL
DL
DL
term
k ic BS = ------------- and N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
LT
k
Where
Forsk 2009
AT271_TRG_E6
325

max
P FCH is the maximum power allowed on FCH. This parameter is user-defined in the Services table for a certain terminal.
k

DL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
DL
With
max
req
DL
P FCH P FCH
I intra ic BS = 1 BTS F ortho P DL k ic 1
tot
BS
BTS max (----------------------------------,0)
LT
k
And
DL
I extra ic BS =
Ptot j icBS
DL
j j k
DL
DL
Ptot j icadj
DL
DL
txj j

4.
Eb/Nt max on SCH for the first f (number of fingers) cells of active set
DL
Let us assume the following notation: Eb/Nt max on SCH corresponds to Q max SCH .
DL SCH
BTS P b max k ic BS
DL
SCH
- G DL
Q max k ic BS SCH = -----------------------------------------------------------------p
DL
N tot ic BS
DL SCH
With P b max
max
P SCH
DL
DL
DL
DL
term
k ic BS = ------------- and N tot ic BS = I intra ic BS + I extra ic BS + I inter carrier ic BS + N 0
LT
k
Where
max
P SCH is the maximum power allowed on SCH for the specified downlink data rate. This parameter is user-defined in the
Services table for a certain terminal and SCH rate.
k

DL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
DL
With
max
req
DL
P SCH P SCH
I intra ic BS = 1 BTS F ortho P DL
tot k ic BS 1 BTS max (----------------------------------,0)
LT
k
And
DL
I extra ic BS =
Ptot j icBS
DL
j j k
DL
DL
326
AT271_TRG_E6
Forsk 2009
Ptot j icadj
DL
DL
txj j

5.
Eb/Nt max on FCH and Eb/Nt max on SCH
DL
Q MAX FCH
is the traffic channel quality on FCH at the receiver on icBS after combining the signal from each cell (k,icBS).

DL
DL
Q MAX ic BS FCH = Q max k ic BS FCH

DL
DL
Qmax k icBS FCH

DL
Where
DL
DL
Q MAX SCH is the traffic channel quality on SCH at the receiver on icBS after combining the signal from each cell (k,icBS).
DL
DL
Q MAX ic BS SCH = Q max k ic BS SCH

DL
DL
Q MAX ic BS SCH = f rake efficiency
Qmax k icBS SCH

DL
Therefore, the service on the downlink traffic channel is available if

DL
Q MAX ic BS SCH
6.
DL
Q req SCH
DL
DL
Q MAX ic BS FCH Q req FCH
and
Effective Eb/Nt on FCH and Eb/Nt on SCH
DL
Q eff FCH
DL
and Q eff SCH are respectively effective traffic channel qualities at the receiver on icBS supplied on FCH and
SCH.
DL
DL
DL
DL
DL
Q eff FCH = min Q MAX FCH Q req FCH

And
DL
Q eff SCH = min Q MAX SCH Q req SCH

7.
Downlink soft handover gain on FCH and downlink soft handover gain on SCH
DL
DL
G SHO FCH and G SHO SCH respectively correspond to DL soft handover gains on FCH and SCH.
DL
Q MAX ic BS FCH
DL
G SHO FCH = ----------------------------------------------------------------------DL
max Q max k ic BS FCH
k
And
DL
Q MAX ic BS SCH
DL
G SHO SCH = ----------------------------------------------------------------------DL
max Q max k ic BS SCH
k
max
k
7.5.1.2.3
DL
Q max k
DL
ic BS corresponds to the highest Q max k icBS value.
CDMA2000 1xEV-DO
Atoll calculates the effective pilot quality level at the receiver and compares this value with the required quality level.
1.
Required rate
DL
The required rate, R req , is the downlink data rate selected for the analysis.
2.
Forsk 2009
Required C/I
AT271_TRG_E6
327

C
For 1xEV-DO Rev. 0 users, the required C/I ( ---I-
req
) is determined from the graph Max Rate=f(C/I) defined for the mobil-
ity type selected in the analysis. It corresponds to the value read in the graph Max Rate=f(C/I) (Rev0) for the specified
DL
required rate, R req .

DL
For 1xEV-DO Rev. A users, the required data rate ( R req ) is obtained by using a certain downlink transmission format (i.e.
a 1xEV-DO Rev. A radio bearer ( Index DL R evA Bearer ) with a certain number of timeslots ( n TS )). It is calculated as
follows:
DL
R RLC peak Index DL RevA Bearer

DL
R req = --------------------------------------------------------------------------------------------n TS
C
----
is the value defined in the 1xEV-DO Rev. A Radio Bearer Selection (Downlink) table for this downlink transmission
I req
format (radio bearer Index, mobility and number of timeslots). It corresponds to the C/I required to obtain the defined
DL
required rate, R req .

3.
Effective C/I
Ec
Let ------ ic BS b pilot be the effective C/I at the receiver on icBS.
Nt
For the best cell (BS,icBS) of the receiver active set, we have:
pilot
Ec
Q resulting ic BS
------ ic BS b pilot = -------------------------------------------------pilot
Nt
Q resulting ic BS
Where
pilot
DL
Q resulting ic BS = G macro diversity Q pilot ic BS

BS
4.
Effective data rate
For 1xEV-DO Rev. 0 users, the effective data rate, R

for the mobility type selected in the analysis. R
DL
DL
, is determined from the graph Max Rate=f(C/I) (Rev0) defined
is the value read in the graph Max Rate=f(C/I) (Rev0) for the calculated
effective C/I, E c
.
------ ic BS b pilot
Nt
For 1xEV-DO Rev. A users, the effective data rate ( R
DL
) provided on downlink depends on the downlink transmission
format, i.e the radio bearer index ( Index DL R evA Bearer ) with the number of timeslots ( n TS ). For the defined mobility
Ec
C
type, Atoll selects the downlink transmission format where ------ ic BS b pilot ----
.
I req Then, it determines the downlink
Nt
effective data rate as follows:
R
DL
DL
R RLC peak Index DL R evA Bearer

= -------------------------------------------------------------------------------------------.
n TS
The traffic data channel on downlink is available if R
7.5.1.3
DL
DL
R req .
Uplink Sub-Menu
Outputs calculated by Atoll depend on the studied network (IS-95 cdmaOne, CDMA2000 1xRTT or CDMA2000 1xEV-DO).
7.5.1.3.1
IS-95 cdmaOne
For each cell (i,icBS) in the receiver active set, Atoll calculates the uplink traffic channel quality on FCH from the receiver.
No power control is performed as in simulations. Here, Atoll determines the uplink traffic channel quality at the cell for the
maximum terminal power allowed. Then, the total uplink traffic channel quality is evaluated with respect to the receiver
handover status. From this value, Atoll calculates the required terminal power and compares it to the maximum terminal
power allowed.
1.
Max terminal power

max
The Max terminal power parameter ( P term ) is user-defined for each terminal. It corresponds to the maximum terminal
power allowed.
2.
Required terminal power

req
The required terminal power ( P term ) calculation may be divided into three steps:
328
AT271_TRG_E6
Forsk 2009

UL
1st step: Evaluation of the uplink traffic channel quality, Q max i ic BS , for each cell of active set
For each cell (i,icBS), we have:
UL
term P b max i ic BS
UL
- G UL
Q max i ic BS = ---------------------------------------------------------p
UL
N tot i ic BS
max
P term
UL
With P b max i ic BS = ------------LT
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
N tot i ic BS is the total noise at the transmitter on the best carrier of the best server. This value is deduced from the cell
uplink load factor X
UL
i ic BS .
tx
N0
UL
N tot i ic BS = --------------------------------------UL
1 X i ic BS
tx

UL
2nd step: Calculation of the total traffic channel quality at the transmitter on icBS ( Q MAX ) based on the receiver handover
status.
If there is no handover, we have:
UL
UL
Q MAX ic BS = Q max i ic BS
For soft handover, we have:
UL
UL
UL
Q MAX ic BS = G macro diversity 2 links max Q max i ic BS

i
UL
UL
UL
max Q max i ic BS corresponds to the highest Q max i ic BS value.

i
For soft-soft handover, we have:

UL
UL
UL

i
UL
For softer and softer-softer handovers, we have:
UL
UL
Qmax i icBS
UL
For softer-soft handover, there are two possibilities. If the MRC option is selected (option available in Global parameters),
we have:
UL
UL
UL
i on the same site
UL
UL
Q max i ic BS Q max
i on the other site
i ic BS
otherwise,
UL
UL
UL

i
req
3rd step: Calculation of P term

UL
Q req FCH
req
- P max
P term = ------------------------------term
UL
Q MAX ic BS
Where
Forsk 2009
AT271_TRG_E6
329

UL
Q req FCH is the user-defined uplink data traffic quality target on FCH for a given service and a terminal. This parameter
is available in the Services table.
req
max
Therefore, the service on the uplink data traffic channel is available if P term P term .
3.
Eb/Nt max on FCH

UL
UL
- G UL
Q max i ic BS = ---------------------------------------------------------p
UL
N tot i ic BS
max
P term
UL
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
UL
i ic BS .
tx
max
req
N0
P FCH P FCH
UL
- + 1 term max (--------------------------------N tot i ic BS = ---------------------------------------,0)
UL
LT
1 X i ic BS
i
tx

UL
Q MAX ic BS is the traffic channel quality on FCH at the transmitter on icBS after signal combination of all the transmitters
of the active set.
If there is no handover, we have:
UL
UL
UL
UL
UL

i
UL
UL
UL
max Q max i ic BS corresponds to the highest Q max i icBS value.

i

UL
UL
UL

i
UL
UL
UL
Qmax i icBS
UL
we have:
UL
UL
UL
i on the same site
UL
UL
Q max i ic BS Q max
i on the other site
i ic BS
otherwise,
UL
UL
UL

i
4.
330
Effective Eb/Nt
AT271_TRG_E6
Forsk 2009

UL
Q eff is the uplink effective traffic channel quality at the receiver on icBS.
UL
UL
UL
Q eff = min Q MAX Q req FCH

5.
Uplink soft handover gain
UL
G SHO
corresponds to the UL soft handover gain.

UL
Q MAX ic BS
UL
G SHO = ----------------------------------------------------UL
max Q max i ic BS
i
UL
UL

i
7.5.1.3.2
CDMA2000 1xRTT
For each cell (i,icBS) in the receiver active set, Atoll calculates the uplink traffic channel quality on FCH and SCH from the
receiver. No power control is performed as in simulations. Here, Atoll determines the uplink traffic channel quality on FCH
at the cell for the maximum terminal power allowed on FCH. In the same way, it evaluates the uplink traffic channel quality
on SCH at the cell for the maximum terminal power allowed on SCH. Then, total uplink traffic channel qualities on FCH
and SCH are evaluated with respect to the receiver handover status. From these values, Atoll deduces required terminal
powers on FCH and SCH, calculates the total terminal power required and compares this value with the maximum terminal
power allowed.
1.
Max terminal power on FCH and SCH

max
power allowed. On uplink, the terminal power is shared between pilot, FCH and SCH channels. So, we may write:
max
max
max
max
P term = P term pilot + P term FCH + P term SCH

We have:
max
max
P term pilot = p P term

Where p is the percentage of the terminal power dedicated to pilot. This parameter is user-defined in the terminal properties.
And
UL
UL
UL
max
Q req FCH R FCH AF FCH
P term FCH
- ------------------------------------------------------------------ = -------------------------UL
UL
max
R SCH
Q req SCH
P term SCH
Therefore,
max
1 p P term
max
P term FCH = --------------------------------------------------------------------------------UL
UL
Q req SCH R SCH
1 + -----------------------------------------------------------------------UL
UL
UL
And
max
1 p P term
max
P term SCH = --------------------------------------------------------------------------------UL
UL
UL
1 + -----------------------------------------------------------------------UL
UL
Q req SCH R SCH
2.
Required terminal power on FCH and SCH

req
req
The required terminal powers on FCH and SCH, respectively P term FCH and P term SCH , are calculated as follows:
UL
1st step: Evaluation of uplink traffic channel qualities on FCH and SCH, Q max ic BS
i
UL
FCH
and Q max ic BS
i
SCH
, for
each cell of active set.

UL FCH
UL
FCH
- G UL
Q max i ic BS FCH = ---------------------------------------------------------------p
UL
N tot i ic BS
And
UL SCH
UL
UL SCH
Q max i ic BS SCH = ---------------------------------------------------------------- Gp
UL
N tot i ic BS
Forsk 2009
AT271_TRG_E6
331

UL FCH
With P b max
max
max
P term FCH
P term SCH
UL SCH
i ic BS = ---------------------------- and P b max i ic BS = ---------------------------LT
LT
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
UL
i ic BS .
tx
N0
UL
N tot i ic BS = --------------------------------------UL
1 X i ic BS
tx

UL
2nd step: Calculation of FCH and SCH total traffic channel qualities at the transmitter on icBS, Q MAX FCH and
UL
Q max SCH , based on the receiver handover status.

If there is no handoff, we have:
UL
UL
UL
UL
Q MAX ic BS FCH = Q max i ic BS FCH and Q MAX ic BS SCH = Q max i ic BS SCH

UL
UL
UL
Q MAX ic BS FCH = G macro diversity 2 links max Q max i ic BS FCH

i
And
UL
UL
UL
Q MAX ic BS SCH = G macro diversity 2 links max Q max i ic BS SCH

i
UL
UL
UL

i

UL
UL
UL

i
And
UL
UL
UL

i
UL
G macro diversity 3 links is the uplink macro-diversity gain.This parameter is determined from the fixed cell edge coverage
UL
UL
Qmax i icBS FCH

UL
UL
UL
And Q MAX ic BS SCH = f rake efficiency
Qmax i icBS SCH

UL
we have:
UL
UL
UL
Q MAX ic BS FCH = G macro diversity 2 links max f rake efficiency
Q max i ic BS FCH Q max
Q max i ic BS SCH Q max
UL
i on the same site
UL
i ic BS
i on the other site
FCH
And
UL
UL
UL
Q MAX ic BS SCH = G macro diversity 2 links max f rake efficiency
UL
i on the same site
UL
i ic BS
i on the other site
SCH
otherwise,
332
AT271_TRG_E6
Forsk 2009

UL
UL
UL

i
And
UL
UL
UL

i
req
req
3rd step: Calculation of P term FCH and P term SCH

UL
UL
Q req FCH
Q req SCH
req
req
- P max
- P max
P term FCH = ---------------------------------------------term FCH and P term SCH = ---------------------------------------------term SCH
UL
UL
Q MAX ic BS FCH
Q MAX ic BS SCH
Where
UL
Q req FCH is the user-defined uplink data traffic quality target on FCH for a given service and a terminal. This parameter
is available in the Services table.
UL
Q req SCH is the user-defined uplink data traffic quality target on SCH for a given service, terminal and SCH rate. This
parameter is available in the Services table.
req
Then, from the required terminal power on FCH and SCH, Atoll determines the total terminal power required ( P term ).
req
req
req
req
P term = P term FCH + P term SCH + P term pilot

req
req
As P term pilot = p P term , we have:

req
req
P term FCH + P term SCH

req
P term = ---------------------------------------------------------------1p
req
max
Therefore, the service on the uplink data traffic channel is available if P term P term .
3.
Eb/Nt max on FCH for each cell in active set

UL FCH
UL
FCH
- G UL
Q max i ic BS FCH = ---------------------------------------------------------------p
UL
N tot i ic BS
UL FCH
With P b max
max
P term FCH
i ic BS = ---------------------------LT
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
UL
i ic BS .
tx
max
req
N0
P FCH P FCH
UL
N tot i ic BS = ---------------------------------------,0)
- + 1 term max (--------------------------------UL
LT
1 X i ic BS
i
tx

4.
Eb/Nt max on SCH for each cell in active set

UL SCH
UL
UL SCH
Q max i ic BS SCH = ---------------------------------------------------------------- Gp
UL
N tot i ic BS
UL SCH
With P b max
max
P term SCH
i ic BS = ---------------------------LT
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
Forsk 2009
AT271_TRG_E6
333

UL
UL
i ic BS .
tx
max
req
N0
P SCH P SCH
UL
- + 1 term max (---------------------------------N tot i ic BS = --------------------------------------,0)
UL
LT
1 X i ic BS
i
tx

5.
Eb/Nt max on FCH and SCH

UL
UL
Q MAX ic BS FCH and Q MAX ic BS SCH are respectively the traffic channel qualities on FCH and SCH at the transmitter
on icBS after signal combination of all the transmitters of the active set.
UL
UL
UL
UL
Q MAX ic BS FCH = Q max i ic BS FCH and Q MAX ic BS SCH = Q max i ic BS SCH

UL
UL
UL

i
And
UL
UL
UL

i
UL
UL
UL

i

UL
UL
UL

i
And
UL
UL
UL

i
UL
UL
UL
Qmax i icBS FCH

UL
UL
UL
And Q MAX ic BS SCH = f rake efficiency
Qmax i icBS SCH

UL
we have:
UL
UL
UL
Q MAX ic BS FCH = G macro diversity 2 links max f rake efficiency
Q max i ic BS FCH Q max
Q max i ic BS SCH Q max
UL
i on the same site
UL
i ic BS
i on the other site
FCH
And
UL
UL
UL
Q MAX ic BS SCH = G macro diversity 2 links max f rake efficiency
UL
i on the same site
UL
i ic BS
i on the other site
SCH
otherwise,
UL
UL
UL

i
And
UL
UL
UL

i
6.
334
Effective Eb/Nt on FCH and SCH
AT271_TRG_E6
Forsk 2009

UL
Q eff FCH is the uplink effective traffic channel quality on FCH at the receiver on icBS.
UL
Q eff SCH is the uplink effective traffic channel quality on SCH at the receiver on icBS.
UL
UL
UL
UL
UL
UL
Q eff FCH = min Q MAX FCH Q req FCH and Q eff SCH = min Q MAX SCH Q req SCH
7.
Uplink soft handover gain FCH and SCH

UL
G SHO FCH corresponds to the UL soft handover gain on FCH.

UL
G SHO SCH corresponds to the UL soft handover gain on SCH.

UL
UL
Q MAX ic BS FCH
Q MAX ic BS SCH
UL
UL
G SHO FCH = -------------------------------------------------------------------- and G SHO SCH = -------------------------------------------------------------------UL
UL
max Q max i ic BS FCH
max Q max i ic BS SCH
I
max
I
7.5.1.3.3
UL
Q max i
ic BS corresponds to the highest
UL
Q max i
ic BS value.
CDMA2000 1xEV-DO
For each cell (l,icBS) in the receiver active set, Atoll calculates the uplink quality level from the receiver. No power control
is performed as in simulations. Here, Atoll determines the uplink quality level at the cell for the maximum terminal power
allowed. Then, the total uplink quality level is evaluated with respect to the receiver handover status. From this value, Atoll
calculates the required terminal power and compares it with the maximum terminal power allowed.
1.
Max terminal power

max
power allowed.
2.
Required terminal power with ACK

req
The required terminal power ( P term ) calculation may be divided into four steps:
UL
1st step: Evaluation of the uplink quality, Q max i ic BS , for each cell of active set
UL
UL
- G UL
Q max i ic BS = ---------------------------------------------------------p
UL
N tot i ic BS
max
P term
UL
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
UL
i ic BS .
tx
N0
UL
N tot i ic BS = --------------------------------------UL
1 X i ic BS
tx

UL
2nd step: Calculation of the total quality at the transmitter on icBS ( Q MAX ) based on the receiver handover status.
UL
UL
UL
UL
UL

i
UL
G macro diversity 2 links is the uplink macro-diversity gain.This parameter is determined from the fixed cell edge coverage
Forsk 2009
AT271_TRG_E6
335

UL
UL

i

UL
UL
UL

i
UL
UL
UL
Qmax i icBS
UL
we have:
UL
UL
UL
Qmax i icBS Qmax

UL
UL
i on the same site
i on the other site
i ic BS
otherwise,
UL
UL
UL

i
UL
3rd step: Evaluation of the required quality level on uplink, Q req

In case of a 1xEV-DO Rev. 0 capable terminal, we have:
E c UL
UL
UL
Q req = ------
N t min
Where
E
-----c-
is the minimum pilot quality level on uplink. This parameter is available in the Mobility types table.
N t min
UL
G ACK , G DRC and G TCH are respectively acknowledgement, data rate control and traffic data gains relative to the pilot.
They are defined in the terminal properties (1xEV-DO Rev. 0 tab).
In case of a 1xEV-DO Rev. A capable terminal, we have:
E c UL
UL
UL
Q req = ------
G p 1 + G ACK + G DRC + G TCH + G RRI + G Auxiliary pilot
N t min
Where
UL
E
-----c-
is the minimum pilot quality level required on uplink to obtain the defined data rate, R req . The required data rate,
N t min
UL
UL
R req (i.e. the uplink data rate selected for the analysis) is obtained by using a certain uplink transmission format (i.e. 1xEVDO Rev. A radio bearer ( Index UL RevA Bearer ) with a certain number of subframes ( n SF )) and calculated as follows:
UL
R RLC peak Index UL RevA Bearer

UL
R req = --------------------------------------------------------------------------------------------n SF
E
-----c-
is the value defined in the 1xEV-DO Rev. A Radio Bearer Selection (Uplink) table for this uplink transmission
N t min
UL
format (radio bearer Index, mobility and number of subframe). Two values are available for this parameter, one when the
service uplink mode is "Low Latency" and another one for high capacity services.
G ACK , G DRC , G TCH , G RRI and G Auxiliary pilot are respectively acknowledgement, data rate control, traffic data channel, reverse rate indicator and auxiliary pilot channel gains relative to the pilot. They are defined in the terminal properties
(1xEV-DO Rev. A tab). Two values of G TCH are available, one when the service uplink mode is "Low Latency" and another
one for high capacity services.
req
4th step: Calculation of P term

UL
Q req
req
- P max
P term = ------------------------------term
UL
Q MAX ic BS
req
max
Therefore, the service on the uplink traffic data channel is available if P term P term .
336
AT271_TRG_E6
Forsk 2009

3.
Required terminal power without ACK
Atoll also calculates the required terminal power without taking into account the ACK channel contribution. Calculations
are quite similar to those detailed in the previous paragraph, only the evaluation of the required quality on uplink is different.
E c UL
UL
UL
Q req withoutACK = ------
G p 1 + G DRC + G TCH for 1xEV-DO Rev. 0 capable terminals
N t min
And
Ec
UL
UL
Q req withoutACK = ------
G p 1 + G DRC + G TCH + G RRI + G Auxiliary pilot for 1xEV-DO Rev. A capable terminals
N t min
UL
And then,
UL
Q req withoutACK
req
max
P term withoutACK = ------------------------------------------ P term
UL
Q MAX ic BS
4.
UL SHO gain
UL
1st step: Evaluation of the uplink quality, Q max i ic BS , for each cell of active set.
UL
UL
- G UL
Q max i ic BS = ---------------------------------------------------------p
UL
N tot i ic BS
max
P term
UL
i

i

UL
L T = -----------------------------------------------------------------------------------------------------------------------------------------------------------G Tx G term
UL
UL
i ic BS .
tx
max
req
N0
P term P term
UL
- + 1 term max (--------------------------------N tot i ic BS = ---------------------------------------,0)
UL
LT
1 X i ic BS
i
tx

UL
2nd step: Calculation of the total quality at the transmitter on icBS ( Q MAX ) based on the receiver handover status.
UL
Q MAX ic BS is the traffic channel quality at the transmitter on icBS after signal combination of all the transmitters of the
active set.
UL
UL
UL
UL
UL

i
UL
UL
UL

i

UL
UL
UL

i
UL
Forsk 2009
AT271_TRG_E6
337

UL
UL
Qmax i icBS
UL
we have:
UL
UL
UL
Qmax i icBS Qmax

UL
UL
i on the same site
i on the other site
i ic BS
otherwise,
UL
UL
UL

i
3rd step: Calculation of the UL SHO gain

UL
G SHO corresponds to the uplink soft handover gain.

UL
Q MAX ic BS
UL
G SHO = ----------------------------------------------------UL
max Q max i ic BS
i
7.5.2
Coverage Studies
Let us assume each pixel of the map corresponds to a probe receiver with associated terminal, mobility and service. This
receiver may be using a specific carrier or all of them. Moreover, it does not create any interference. Coverage studies are
based on the uplink load percentage and the downlink total power of cells. These parameters can either be outputs of a
simulation, average values calculated from a group of simulations or user-defined cell inputs.
7.5.2.1
Pilot Reception Analysis

For further details on calculation formulas, see "Definitions and Formulas" on page 276 For further details on calculations,
see "Bar Graph and Pilot Sub-Menu" on page 320
7.5.2.1.1
1st Case: Analysis Based on all Carriers

Atoll proceeds as in point analysis. It determines the best carrier of each transmitter i containing the receiver in its calculation area. The best carrier selection depends on the option chosen in Equipment (UL minimum noise, DL minimum
power, random, sequential) and is based on the UL load percentage and the downlink total power of cells (simulation
results or cell properties). Atoll calculates the pilot quality at the receiver from these transmitters on their best carrier and
determines the best serving transmitter BS on its best carrier icBS ( Q pilot ic BS ). Then, it deduces the best pilot quality
BS
received with a fixed cell edge coverage probability,
Resulting
Q pilot
ic BS
7.5.2.1.2
2nd Case: Analysis Based on a Specific Carrier

The carrier that can be used by transmitters is fixed. In this case, for each transmitter i containing the receiver in its calculation area that may use the specified carrier (carrier specified in Cell Properties), Atoll calculates pilot quality at the
receiver on this carrier icgiven. Then, it determines the best serving transmitter BS using the carrier icgiven
( Q pilot ic given ) and deduces the best pilot quality received with a fixed cell edge coverage probability,
BS
Resulting
Q pilot
ic given
1.
Single colour
Resulting
req
ic Q pilot . Coverage consists of a single layer with a unique colour.
ic = ic BS or ic given
2.

Resulting
req
ic Q pilot ( ic = ic BS or ic given ). Coverage consists of several layers with asso-
ciated colours. There is a layer per transmitter with no intersection between layers. Layer colour is the colour assigned to
the best serving transmitter BS.
3.
Colour per mobility
In this case, the receiver is not completely defined and no mobility assigned. Coverage consists of several layers with a
layer per user-defined mobility type defined in the Mobility Types sub-folder. For each layer, area is covered if
338
AT271_TRG_E6
Forsk 2009

Resulting
Q pilot
req
ic Q pilot ( ic = ic BS or ic given ). Each layer is assigned a colour and displayed with intersections between
layers.
4.
and the All option in the Condition tab of prediction properties).
Coverage consists of several layers with a layer per user-defined probability level defined in the Display tab (Prediction
Resulting
req
ic Q pilot ( ic = ic BS or ic given ) in the required number of
simulations. Each layer is assigned a colour and displayed with intersections between layers.
5.
Resulting
tab (Prediction properties). For each layer, area is covered if Q pilot
req
ic p Q pilot ( ic = ic BS or ic given ). Each layer

6.
Colour per quality level (Ec/I0)
Resulting
ic Q pilot threshold ( ic = ic BS or ic given ). Each layer is

7.
Colour per quality margin (Ec/I0 margin)
Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction propResulting
erties). For each layer, area is covered if Q pilot
req
ic Q pilot Q pilot m arg in ( ic = ic BS or ic given ). Each layer is
7.5.2.2
Downlink Service Area Analysis

The downlink service area analysis depends on the studied network (IS-95 cdmaOne, CDMA2000 1xRTT or CDMA2000
1xEV-DO). Several display options are available when calculating this study, some of which are dedicated to IS-95
cdmaOne and CDMA2000 1xRTT networks while others are relevant when analysing CDMA2000 1xEV-DO systems only.
7.5.2.2.1
IS-95 cdmaOne and CDMA2000 1xRTT

As in point analysis, Atoll calculates downlink quality on FCH at the receiver for each cell (k,ic) (with ic=icBS or icgiven)
(these cells are the first f cells in the receivers active set and f is the number of fingers defined for the terminal). No power
control is performed as in simulations. Here, Atoll determines the downlink quality on FCH at the receiver for a maximum
traffic channel power per transmitter allowed on the fundamental channel (FCH). Then, the total downlink quality on FCH
DL
( Q MAX ic FCH ) is evaluated after recombination.

Note:
Best server and active set determination is performed as in point prediction.
Atoll displays total traffic channel quality at the receiver on the carrier ic ( ic BS or ic given ).
For further details on formulas, see "Definitions and Formulas" on page 276. For further details on calculation, see "Downlink Sub-Menu" on page 321.
You may choose following display options:
1.
Single colour
DL
DL
DL
Atoll displays a coverage with a unique colour if Q MAX ic FCH Q req FCH . Q req FCH is the downlink traffic data quality target on the fundamental channel (FCH). This parameter is user-defined for a given service and a terminal in the Services sub-folder.
2.

DL
DL
Atoll displays a coverage if Q MAX ic FCH Q req FCH . Coverage consists of several layers with associated colours.
There is a layer per transmitter with no intersection between layers. Layer colour is the colour assigned to best serving
transmitter.
3.
Colour per mobility
layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if
DL
DL
Q MAX ic FCH Q req FCH . Each layer is assigned a colour and displayed with intersections between layers.
4.
Colour per service
DL
DL
per user-defined service defined in Services sub-folder. For each layer, area is covered if Q MAX ic FCH Q req FCH .
Forsk 2009
AT271_TRG_E6
339

5.
and the All option in the Condition tab of prediction properties).
Coverage consists of several layers with a layer per user-defined probability level defined in the Display tab (Prediction
DL
DL
properties). For each layer, area is covered if Q MAX ic FCH Q req FCH in the required number of simulations. Each
layer is assigned a colour and displayed with intersections between layers.
6.
DL
DL
tab (Prediction properties). For each layer, area is covered if Q MAX ic p FCH Q req FCH . Each layer is assigned a
colour and displayed with intersections between layers.
7.
Colour per maximum quality level (max Eb/Nt)
DL
properties). For each layer, area is covered if Q MAX ic FCH Threshold . Each layer is assigned a colour and displayed
8.
DL
properties). For each layer, area is covered if Q eff ic FCH Threshold . Each layer is assigned a colour and displayed
9.
Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction propDL
DL
erties). For each layer, area is covered if Q MAX ic FCH Q req FCH M arg in . Each layer is assigned a colour and
10.

req
Atoll calculates the downlink power required on FCH, P FCH ic , as follows:

DL
Q req FCH
req
- P max
P FCH ic = -------------------------FCH
DL
Q MAX ic
max
Where P FCH is a user-defined input for a given service and terminal. It corresponds to the maximum traffic data power
allowed on FCH for a transmitter.
Coverage consists of several layers with a layer per user-defined required power threshold defined in the Display tab
req
(Prediction properties). For each layer, area is covered if P FCH ic Threshold . Each layer is assigned a colour and
11.
Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction propmax
req
erties). For each layer, area is covered if P FCH P FCH ic M arg in . Each layer is assigned a colour and displayed with
12.
Colour per data rate
This display option is relevant for CDMA2000 1xRTT data services only. For each possible data rate, R
DL
DL
DL
DL
DL
DL
DL
DL
DL
DL
DL
( R FCH AF FCH , R FCH AF FCH + 2 , R FCH AF FCH + 4 , R FCH AF FCH + 8 , R FCH AF FCH + 16 ), Atoll calculates traffic channel quality at the receiver for each cell (k,ic) (with ic=icBS or icgiven). Downlink traffic channel quality at the
receiver is evaluated from a maximum traffic channel power per transmitter allowed for the corresponding data rate. Then,
DL
the total downlink traffic channel quality ( Q MAX ic R
DL
) is calculated after recombination.
Coverage consists of several layers with a layer per possible data rate, R
DL
Q MAX ic
DL
Q req R
DL
DL
DL
DL
Q req R
DL
. For each layer, area is covered if
. Each layer is assigned a colour and displayed with intersections between layers.
is the downlink traffic data quality target for the data rate, R
DL
. This parameter is user-defined for a given
service, terminal and data rate in the Services sub-folder.
7.5.2.2.2
CDMA2000 1xEV-DO
Ec
As in point analysis, Atoll calculates the effective pilot quality level at the receiver from the best server cell, ------ ic b pilot .
Nt
Best server and active set determination is performed as in point prediction (AS analysis).Then, from this value, it determines the effective downlink data rate received, R
340
DL
AT271_TRG_E6
Forsk 2009

For further details on formulas, see "Definitions and Formulas" on page 276. For further details on calculations, see
"Downlink Sub-Menu" on page 321.
1xEV-DO Rev. 0 Users

For 1xEV-DO Rev. 0 users (users with EV-DO Rev. 0-capable terminals and EV-DO Rev. 0 services), the effective data
rate ( R
DL
) provided on downlink is determined from the graph Max Rate=f(C/I) (Rev0) defined for the mobility type
selected in the Condition tab (Prediction properties). R
DL
is the value read in the graph Max Rate=f(C/I) (Rev0) for the
calculated effective pilot quality level, E c

.
------ ic BS b pilot
Nt
1xEV-DO Rev. A Users

For 1xEV-DO Rev. A users (users with EV-DO Rev. A-capable terminals and EV-DO Rev. A services), the effective data
rate ( R
DL
) provided on downlink depends on the downlink transmission format, i.e the radio bearer index
( Index DL R evA Bearer ) with the number of timeslots ( n TS ). Atoll selects the downlink transmission format where
Ec
C
------ ic BS b pilot ----
.
I req Then, it determines the downlink effective data rate as follows:
Nt
R
DL
DL

= ---------------------------------------------------------------------------------------------.
n TS
The effective data rate corresponds to the guaranted data rate after a certain number of retransmissions (i.e. the number
of timeslots, n TS ).
When HARQ (Hybrid Automatic Repeat Request) is used, the required average number of retransmissions is smaller and
DL
the data rate is an average data rate ( R av ) calculated as follows:

DL

DL
R av = ----------------------------------------------------------------------------------------------DL
n Rtx (Index DL RevA Bearer,n TS) av
DL
The average number of retransmissions ( n Rtx av ) is determined from early termination probabilities defined for the
selected downlink transmission format. The Early Termination Probability graph shows the probability of early termination
DL
( p ) as a function of the number of retransmissions ( n Rtx ). Atoll calculates the average number of retransmissions
DL
( n Rtx av ) as follows:
n Rtx max
DL
DL
DL
DL
DL
p n Rtx p n Rtx 1 n Rtx
=1
Rtx
n Rtx av = ------------------------------------------------------------------------------------------------DL
p n Rtx max
Display Options
You may choose following display options:
1.
Colour per C/I
Coverage consists of several layers with a layer per quality threshold defined in the Display tab (Prediction properties).
Ec
For each layer, area is covered if ------ ic b pilot Threshold . Each layer is assigned a colour and displayed with intersecNt
tions between layers.
2.
Colour per rate
Coverage consists of several layers with a layer per possible data rate ( R
rate, R
3.
DL
DL
). For each layer, area is covered if the data
, can be obtained. Each layer is assigned a colour and displayed with intersections between layers.
Colour per average rate
This display option is available for 1xEV-DO Rev. A users only. It enables you to view the obtained downlink data rate
DL
when HARQ is used. Coverage consists of several layers with a layer per possible average data rate ( R av ). For each
DL
layer, area is covered if the average data rate, R av , can be obtained. Each layer is assigned a colour and displayed with
Forsk 2009
AT271_TRG_E6
341
7.5.2.3
Uplink Service Area Analysis

The results displayed when calculating the uplink service area analysis depend on the studied network (IS-95 cdmaOne,
CDMA2000 1xRTT or CDMA2000 1xEV-DO).
7.5.2.3.1
IS-95 cdmaOne and CDMA2000 1xRTT

As in point analysis, Atoll calculates uplink quality on FCH from receiver for each cell (l,ic) (with ic=icBS or icgiven) in receiver
active set. No power control simulation is performed. Atoll determines uplink quality on FCH at the transmitter for the maxiUL
mum terminal power. Then, the total uplink traffic channel quality ( Q MAX ic FCH ) is evaluated with respect to the
receiver handover status.
Note:
Atoll displays uplink quality on FCH at transmitters in active set on the carrier ic ( ic BS or ic given ) received from the
receiver.
For further details on formulas, see "Definitions and Formulas" on page 276. For further details on calculations, see "Uplink
Sub-Menu" on page 328.
1.
Single colour
UL
UL
UL
Atoll displays a coverage if Q MAX ic FCH Q req FCH . Coverage colour is unique. Q req FCH is the uplink data traffic
quality target on the fundamental channel (FCH). This parameter is user-defined for a given service and a terminal in the
Services sub-folder.
2.

UL
UL
Atoll displays a coverage if Q MAX ic FCH Q req FCH . Coverage consists of several layers with associated colours.
There is a layer per transmitter with no intersection between layers. Layer colour is the colour assigned to best server
transmitter.
3.
Colour per mobility
layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if
UL
UL
Q MAX ic FCH Q req FCH . Each layer is assigned a colour and displayed with intersections between layers.
4.
Colour per service
UL
UL
per user-defined service defined in Services sub-folder. For each layer, area is covered if Q MAX ic FCH Q req FCH .
5.
and the All option in the Condition tab of prediction properties). Coverage consists of several layers with a layer per userdefined probability level defined in the Display tab (Prediction properties). For each layer, area is covered if
UL
UL
Q MAX ic FCH Q req FCH in the required number of simulations. Each layer is assigned a colour and displayed with
6.
UL
UL
tab (Prediction properties). For each layer, area is covered if Q MAX ic p FCH Q req FCH . Each layer is assigned a
7.
UL
properties). For each layer, area is covered if Q MAX ic FCH Threshold . Each layer is assigned a colour and displayed
8.
UL
properties). For each layer, area is covered if Q effective ic FCH Threshold . Each layer is assigned a colour and
9.
UL
erties). For each layer, area is covered if Q MAX ic FCH Q req FCH M arg in . Each layer is assigned a colour and
342
AT271_TRG_E6
Forsk 2009

10.

FCH re q
The required terminal power, P term
, is calculated as described in the Point analysis AS analysis tab Uplink sub-
menu part. Coverage consists of several layers with a layer per user-defined power threshold defined in the Display tab
FCH re q
(Prediction properties). For each layer, area is covered if P term
ic Threshold . Each layer is assigned a colour and

11.
Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction properties). For each layer, area is covered if P max P FCH re q ic M arg in . Each layer is assigned a colour and displayed
term
term
12.
Colour per rate
This display option is relevant for CDMA2000 1xRTT data services only. For each possible data rate, R
UL
UL
UL
UL
UL
UL
UL
UL
UL
UL
UL
( R FCH AF FCH , R FCH AF FCH + 2 , R FCH AF FCH + 4 , R FCH AF FCH + 8 , R FCH AF FCH + 16 ), Atoll calcuUL
lates the total uplink traffic channel quality ( Q MAX ic R

data rate, R
UL
UL
). Coverage consists of several layers with a layer per possible
UL
Q MAX ic
. For each layer, area is covered if
displayed with intersections between layers. Q req R
UL
UL
UL
UL
Q req R
UL
. Each layer is assigned a colour and
is the uplink traffic data quality target for the data rate, R
UL
. This
parameter is user-defined for the service, a given terminal and data rate in the service properties.
7.5.2.3.2
CDMA2000 1xEV-DO
As in point analysis, Atoll calculates uplink quality from receiver for each cell (l,ic) (with ic=icBS or icgiven) in receiver active
set. No power control simulation is performed. For 1xEV-DO Rev. 0 users, Atoll determines uplink quality at the transmitter
for the maximum terminal power allowed and an uplink data channel rate of 9.6 kbps. For 1xEV-DO Rev. A users, Atoll
determines uplink quality at the transmitter for the maximum terminal power allowed and an uplink data channel rate of
UL
4.8 kbps. Then, the total uplink quality ( Q MAX ic ) is evaluated with respect to the receiver handover status.
Note:
Atoll displays uplink quality at transmitters in active set on the carrier ic ( ic BS or ic given ) received from the receiver.
For further details on formulas, see "Definitions and Formulas" on page 276. For further details on calculations, see "Uplink
Sub-Menu" on page 328.
1.
Single colour
UL
UL
UL
Atoll displays a coverage if Q MAX ic Q req . Coverage colour is unique. For 1xEV-DO Rev. 0 users, Q req is the quality
UL
required on uplink for a 9.6 kbps data channel rate. For 1xEV-DO Rev. A users, Q req is the quality required on uplink for
a 4.8 kbps data channel rate. This parameter is calculated from the minimum uplink pilot quality and gains on the different
uplink channels.
We have:
E c UL
UL
UL
Q req = ------
G p 1 + G ACK + G DRC + G TCH for 1xEV-DO Rev 0 terminals,
N t min
And
Ec
UL
UL
Q req = ------
G p 1 + G ACK + G RRI + G DRC + G TCH + G Auxiliary Pilot for 1xEV-DO Rev A terminals.
N t min
UL
2.

UL
UL
Atoll displays a coverage if Q MAX ic Q req . Coverage consists of several layers with associated colours. There is a layer
per transmitter with no intersection between layers. Layer colour is the colour assigned to best server transmitter.
3.
Colour per mobility
UL
UL
layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if Q MAX ic Q req . Each
layer is assigned a colour and displayed with intersections between layers.
4.
Colour per service
UL
UL
per user-defined service defined in Services sub-folder. For each layer, area is covered if Q MAX ic Q req . Each layer is
5.
Forsk 2009

AT271_TRG_E6
343

and the All option in the Condition tab of prediction properties). Coverage consists of several layers with a layer per userdefined probability level defined in the Display tab (Prediction properties). For each layer, area is covered if
UL
UL
Q MAX ic Q req in the required number of simulations. Each layer is assigned a colour and displayed with intersections
between layers.
6.
UL
UL
tab (Prediction properties). For each layer, area is covered if Q MAX ic p Q req . Each layer is assigned a colour and
7.

UL
Here, Atoll calculates the total uplink traffic channel quality ( Q MAX ic TCH ).
UL
Q req TCH
UL
- P max
Q MAX ic TCH = -------------------------term
req
P term
With
Ec
UL
UL
Q req TCH = ------
G p G TCH
N t min
UL
UL
properties). For each layer, area is covered if Q MAX ic TCH Threshold . Each layer is assigned a colour and displayed
8.
UL
properties). For each layer, area is covered if Q effective ic TCH Threshold . Each layer is assigned a colour and
UL
UL
UL
Q effective ic TCH = min Q MAX ic TCH Q req TCH

9.
UL
erties). For each layer, area is covered if Q MAX ic Q req M arg in . Each layer is assigned a colour and displayed with
10.

TCH re q
The required terminal power on traffic data channel, P term
, is calculated as described in the Point analysis AS anal-
ysis tab Uplink sub-menu part.

TCH re q
P term
req
P term
= ------------------------------------------------------------------- G TCH for 1xEV-DO Rev 0 terminals,
1 + G ACK + G DRC + G TCH
And
TCH re q
P term
req
P term
= ------------------------------------------------------------------------------------------------------------------------------------- G TCH for 1xEV-DO Rev A terminals.
1 + G ACK + G RRI + G DRC + G TCH + G Auxiliary Pilot
Coverage consists of several layers with a layer per user-defined power threshold defined in the Display tab (Prediction
properties). For each layer, area is covered if P TCH re q ic Threshold . Each layer is assigned a colour and displayed
term
11.
Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction properties). For each layer, area is covered if P max P req ic M arg in . Each layer is assigned a colour and displayed with
term
term
12.
Colour per rate
For each possible data rate, R
UL
UL
, Atoll calculates the total uplink quality ( Q MAX ic R

UL
UL
). Coverage consists of several
layers with a layer per possible data rate. For each layer, area is covered if Q MAX ic R
UL
UL
Q req R
UL
. Each layer is

UL
Q req R
344
UL
is the uplink quality required to obtain the data rate, R
UL
1xEV-DO Rev. 0 users
AT271_TRG_E6
Forsk 2009

The possible data rates on uplink, R
UL
, are: 9.6, 19.2, 38.4, 76.8 and 153.6 kbps
Ec
UL
UL
Q req = ------
N t min
UL
Where
E
-----c-
is the minimum pilot quality level on uplink. This parameter is available in the Mobility types table.
N t min
UL
G ACK , G DRC and G TCH are respectively acknowledgement, data rate control and traffic data gains relative to the pilot.
They are defined in the terminal properties (1xEV-DO Rev. 0 tab).
1xEV-DO Rev. A users
The data rate, R
UL
is obtained when a certain uplink transmission format (i.e. 1xEV-DO Rev. A radio bearer
( Index UL RevA Bearer ) with a certain number of subframes ( n SF )) is used. It is calculated as follows:
UL
R RLC peak Index UL RevA Bearer

UL
R req = --------------------------------------------------------------------------------------------n SF
E c UL
UL
UL
Q req = ------
G p 1 + G ACK + G DRC + G TCH + G RRI + G Auxiliary pilot
N t min
Where
UL
E
-----c-
is the minimum pilot quality level required on uplink to obtain the data rate, R . The value is defined in the 1xEV N t min
UL
DO Rev. A Radio Bearer Selection (Uplink) table for the uplink transmission format (radio bearer Index, mobility and
number of subframe). Two values are available for this parameter, one when the service uplink mode is "Low Latency"
and another one for high capacity services.
G ACK , G DRC , G TCH , G RRI and G Auxiliary pilot are respectively acknowledgement, data rate control, traffic data channel, reverse rate indicator and auxiliary pilot channel gains relative to the pilot. They are defined in the terminal properties
(1xEV-DO Rev. A tab). Two values of G TCH are available, one when the service uplink mode is "Low Latency" and another
one for high capacity services.
13.
Colour per average rate
This display option is available for 1xEV-DO Rev. A users only. When HARQ (Hybrid Automatic Repeat Request) is used,
UL
the required average number of retransmissions is smaller and the data rate is an average data rate ( R av ) calculated as
follows:
UL
R RLC peak Index UL R evA Bearer

UL
R av = ----------------------------------------------------------------------------------------------UL
n Rtx Index UL R evA Bearer n SF av
UL
The average number of retransmissions ( n Rtx av ) is determined from early termination probabilities defined for the
selected uplink transmission format (i.e. the radio bearer index ( Index UL R evA Bearer ) with the number of subframes
( n SF )). The Early Termination Probability graph shows the probability of early termination ( p ) as a function of the number
UL
UL
of retransmissions ( n Rtx ). Atoll calculates the average number of retransmissions ( n Rtx av ) as follows:
n Rtx max
UL
UL
UL
UL
UL
p n Rtx p n Rtx 1 n Rtx
=1
Rtx
n Rtx av = ------------------------------------------------------------------------------------------------UL
p n Rtx max
UL
UL
UL
For each possible average data rate, R av , Atoll calculates the total uplink quality ( Q MAX ic R av ). Coverage consists of
UL
UL
UL
UL
several layers with a layer per possible average data rate. For each layer, area is covered if Q MAX ic R av Q req R av .
UL
UL
UL
Q req R av is the uplink quality required to obtain the average data rate, R av .
7.5.2.4
Downlink Total Noise Analysis

Atoll determines downlink total noise generated by cells.
For IS-95 cdmaOne and CDMA2000 1xRTT systems, we have:
Forsk 2009
AT271_TRG_E6
345
Ptot icadj
DL
DL
N tot ic =
j
- + N0
Ptot ic + txj-------------------------------------RF ic ic adj
DL
term
txj j
For CDMA2000 1xEV-DO systems, we have:
Ptot icadj bpilot

DL
DL
N tot ic =
j
+ N0
Ptot ic bpilot + txj------------------------------------------------------RF ic ic adj
DL
term
txj j
term
N0
DL
-
Downlink noise rise, NR DL ic , is calculated from the downlink total noise, N tot , as: NR DL ic = 10 log ------------ N DL
tot
7.5.2.4.1
Analysis on all Carriers

If all the carriers are selected, Atoll determines DL total noise for all the carriers. Then, allows the user to choose different
displays.
1.
Colour per minimum noise level
erties). For each layer, area is covered if min NR tot ic Threshold . Each layer is assigned a colour and displayed
ic

2.
Colour per maximum noise level
erties). For each layer, area is covered if max NR tot ic Threshold . Each layer is assigned a colour and displayed with
ic

3.
Colour per average noise level
erties). For each layer, area is covered if average NR tot ic Threshold . Each layer is assigned a colour and displayed
ic

4.
Colour per minimum noise rise
Atoll displays bins where min NR DL ic Threshold . Coverage consists of several areas with an area per user-defined
ic
5.
Colour per maximum noise rise
Atoll displays bins where max NR DL ic Threshold . Coverage consists of several areas with an area per user-defined
ic
6.
Colour per average noise rise
Atoll displays bins where average NR DL ic Threshold . Coverage consists of several areas with an area per useric
defined noise rise threshold defined in the Display tab. Each area is assigned a colour with intersections between areas.
7.5.2.4.2
Analysis on a Specific Carrier

When only one carrier is analysed, Atoll determines DL total noise or DL noise rise on this carrier. In this case, the
displayed coverage is the same for any selected display per noise level (average, minimum or maximum) or any display
per noise rise (average, minimum or maximum).
1.
Colour per noise level
erties). For each layer, area is covered if N tot ic Threshold . Each layer is assigned a colour and displayed with intersections between layers.
2.
Colour per noise rise
Atoll displays bins where NR DL ic Threshold . Coverage consists of several areas with an area per user-defined noise
rise threshold defined in the Display tab. Each area is assigned a colour with intersections between areas.
7.6

Atoll permits the automatic allocation of intra-technology neighbours in the current network. Two allocation algorithms are
available, one dedicated to intra-carrier neighbours and the other for inter-carrier neighbours.
346
AT271_TRG_E6
Forsk 2009

The intra-technology neighbour allocation algorithms take into account all the cells of TBC transmitters. It means that all
the cells of TBC transmitters of your .atl document are potential neighbours.
The cells to be allocated will be called TBA cells. They must fulfill the following conditions:
They are active,


Note:
7.6.1
Global Allocation for all Transmitters

We assume a reference cell A and a candidate neighbour, cell B. When automatic allocation starts, Atoll checks following
conditions:
1.
The distance between both cells must be less than the user-definable maximum inter-site distance. If the distance
is discarded.
2.
Force co-site cells as neighbours: This option enables you to force cells located on the reference cell site in the candidate
neighbour list. This constraints can be weighted among the others and ranks the neighbours through the importance field
(see after).
Force adjacent cells as neighbours (only for intra-carrier neighbours): This option enables you to force cells geographically
adjacent to the reference cell in the candidate neighbour list.This constraints can be weighted among the others and ranks
the neighbours through the importance field (see after).
Force neighbour symmetry: This option enables user to force the reciprocity of a neighbourhood link. Therefore, if the reference cell is a candidate neighbour of another cell, this one will be considered as candidate neighbour of the reference cell.
force/forbid a cell to be candidate neighbour of the reference cell.
Note:
Adjacence criterion: Let CellB be a candidate neighbour cell of CellA. CellB is considered
adjacent to CellA if there exists at least one pixel in the CellA Best Server coverage area
where CellB is Best Server (if several cells have the same best server value) or CellB is the
second best server that enters the Active Set (respecting the T_Drop of the allocation).
the least, depending on the above criterion. Adjacence is relative to the number of pixels satisfying the criterion.
There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability:
3.
Intra-carrier neighbours: intra-carrier handover is a soft handover.
The reference cell A and the candidate cell B are located inside a continuous layer of cells with carrier c1 (c1 is the selected
carrier on which you run the allocation).
SA is the area where the cell A is the best serving cell. It means that the cell A is the first one in the active set.
-
The pilot quality from A exceeds Min. Ec/I0.
The pilot quality from A is the best.
SB is the area where the cell B can enter the active set.
The pilot quality from B is greater than T_Drop.
Inter-carrier neighbours: inter-frequency handover is a hard handover. It is needed in a multi-carrier (1xRTT and
1xEV-DO carriers) CDMA network:
-
To balance loading between carriers and layers (1st case),
To make a coverage reason handover from micro cell frequency to macro cells (2nd case).
1st case: the reference cell A is located inside a continuous layer of cells with carrier c1 (c1 is the selected carrier on which
you run the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.
Forsk 2009
AT271_TRG_E6
347

SA is the area where:
-
The pilot signal from A is not the highest one. It is strictly lower than the best pilot signal received and higher
than the best pilot signal minus the margin.
SB is the area where:

-
The pilot signal from B is the highest one.
Figure 7.5: Overlapping Zones - 1st Case

2nd case: the reference cell A is located on the border of a layer with carrier c1 (c1 is the selected carrier on which you run
the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.
SA is the area where:
-
The pilot signal from A is the highest one
The pilot signal from A is lower than the minimum pilot signal level plus the margin.
SB is the area where:

-
The pilot signal from B is the highest one.
Figure 7.6: Overlapping Zones - 2nd Case

Note:
Two ways enable you to determine the I0 value:

1 - A reduction factor (% of maximum powers contributing to I0) may be applied to cell
maximum powers (defined in Cell properties) to customize their contribution to I0. Thus, I0
represents the sum of effective powers received from the other cells. The entered
percentage is a kind of downlink load factor estimation.
If the % of maximum powers contributing to I0 is too low, i.e. if % P max P pilot , Atoll
takes into account the pilot powers to evaluate the I0 value.
348
AT271_TRG_E6
Forsk 2009

2 - Atoll takes into account load parameters defined per cell (such as the total downlink
power used). I0 represents the sum of total transmitted powers.
SA SB
SA
after).
4.
are 15 candidate neighbours and the maximum number of neighbours to be allocated to the reference cell is 8. Among
these 15 candidate neighbours, only 8 (having the highest importances) will be allocated to the reference cell. Note that
specific maximum numbers of neighbours (maximum number of intra-carrier neighbours, maximum number of inter-carrier
neighbours) can be defined at the cell level (property dialog or cell table). If defined there, this value is taken into account
instead of the default one available in the dialog.
to 100%.
Neighbourhood cause
When
Importance
value
Existing neighbour

Existing
importance
Exceptional pair
100 %
Co-site transmitter
(IF) function
Only if the Force adjacent cells as neighbours option is

selected
(IF) function

coverage conditions
(IF) function
relationship
(IF) function
(IF) Importance Function evaluation. The importance is evaluated through a function (IF), taking into account 3 factors:
The co-site factor (C) which is a Boolean factor,

The adjacency factor (A) which deals with the percentage of adjacency,
Factor
Min
importance
Default value
Max
importance
Default value

Min(O)
1%
Max(O)
30%
Min(A)
30%
Max(A)
60%
Co-site factor (C)
Min(C)
60%
Max(C)
100%
Neighbourhood cause
(IF) function

no
1% + 29%(O)
no
yes
30% + 30%{30%(O) + 70%(A)}
yes
yes
60% + 40%{30%(O) + 70%(A)}
Co-site
Adjacent
no
Where Delta(x)= Max(x) - Min(x)

Notes:
Forsk 2009
AT271_TRG_E6
349
factors.
accepted for co-site, adjacency and coverage reasons, Atoll displays the percentage of area meeting the coverage conditions and the corresponding surface area (km2), the percentage of area meeting the adjacency conditions and the corresponding surface area (km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.
Notes:
No simulation or prediction study is needed to perform an automatic neighbour allocation.

When starting an automatic neighbour allocation, Atoll automatically calculates the path
loss matrices if not found.
Even if no specific terminal, mobility or service is selected in the automatic allocation, it is

interesting to know that the algorithm works such as finding the maximum number of
neighbours by selection the multi-service traffic data as follows:
Service: selection of the one with the lowest body loss.
Mobility: no impact on the allocation, no specific selection.
Terminal: selection of the one with the greatest (Gain - Loss) value, and, if equal, the one
with the lowest noise figure.
The neighbour lists may be optionally used in the power control simulations to determine
the mobile's active set.
The percentage of area is calculated with the resolution specified in the properties dialog of
the predictions folder (default resolution parameter).

Thus, if the cell B is a neighbour of the cell A while the cell A is not a neighbour of the cell
B, two cases are possible:
1st case: There is space in the cell B neighbour list: the cell A will be added to the list. It will
be the last one.
2nd case: The cell B neighbour list is full: Atoll will not include cell A in the list and will
cancel the link by deleting cell B from the cell A neighbour list.
7.6.2

TBA cells,
Neighbours of TBA cells marked as exceptional pair, adjacent and symmetric,
Neighbours of TBA cells that satisfy coverage conditions.
Automatic neighbour allocation parameters are described in "Global Allocation for all Transmitters" on page 347.
7.7
PN Offset Allocation
PN offset is used to identify a cell. It is a time offset used by a cell to shift a Pseudo Noise sequence. Mobile processes
the strongest received PN sequence and reads its phase that identifies the cell.
By default, there are 512 PN Offsets. PN Offsets are numbered (0...511).
350
AT271_TRG_E6
Forsk 2009

The cells to which Atoll allocates PN Offsets are referred to as the TBA cells (cells to be allocated). TBA cells fulfil following
conditions:
-
They are active,

They belong to the folder on which allocation has been executed. This folder can be either the Transmitters
folder or a group of transmitters or a single transmitter.
Note:
7.7.1
7.7.1.1
Options and Constraints

The PN Offset allocation algorithm can take into account following constraints and options:
1.
PILOT_INC parameter,
Atoll uses this parameter to determine the pool of possible PN offsets (512 divided by PILOT_INC value). The first
PN offset is PILOT_INC and other ones are multiples of this value.
For example: When PILOT_INC is set to 4, the pool of possible PN offsets consists of PN offsets from 4 to 508
with a separation interval of 4 (i.e. [4,8,12,16,...508]).
2.
Neighbourhood between cells,
You may consider:
Notes:
3.
In the context of the PN Offset allocation, the term "neighbours" refers to intra-carrier
neighbours.
different PN Offsets to the CDMA neighbours of a GSM transmitter. In order to consider
inter-technology neighbour relations in the PN Offset allocation, you must make the
Transmitters folder of the GSM .atl document accessible in the CDMA .atl document. For
information on making links between GSM and CDMA .atl documents, see the User
Manual.
Cells fulfilling a criterion on Ec/I0 (option Additional Ec/I0 conditions),
For a reference cell A, Atoll considers all the cells B that can enter the active set on the area where the reference cell
is the best server (area where (Ec/I0)A exceeds Min. Ec/I0 and is the highest one and (Ec/I0)B exceeds T_Drop).
Note:
4.
Atoll takes the total downlink power used by the cell into account in order to evaluate Io. Io
equals the sum of total transmitted powers. In case this parameter is not specified in the
cell properties, Atoll uses 50% of the maximum power.
Co-PN Reuse distance,
Reuse distance is a constraint on the allocation of PN offsets. A PN offset cannot be reused at a site that is not at least as
far away as the reuse distance from the site allocated with the particular PN offset.
Note:
5.
6.
7.
PN offset reuse distance can be defined at cell level. If this value is not defined, then Atoll
will use the default reuse distance defined in the PN offset Automatic Allocation dialog.
PN-cluster size. Within the context of PN offset allocation, the term "PN-cluster" refers to a sub-group of PN
offsets.
Exceptional pairs,
Domains of PN Offsets,
Note:
Forsk 2009
When no domain is assigned to cells, Atoll considers the PILOT_INC parameter only to
determine available PN offsets (e.g., If PILOT_INC is set to 4, all PN offsets from 4 to 508
with a separation interval of 4 can be allocated).
AT271_TRG_E6
351

8.
The carrier on which the allocation is run: It can be a given carrier or all of them. In this case, either Atoll
independently plans PN Offsets for the different carriers, or it allocates the same PN Offset to each carrier of a
transmitter if the option "Allocate carriers identically" is selected.
The possibility to use a maximum of PN offsets (option "Use a Maximum of PN Offsets"): Atoll will try to spread
the PN offset spectrum the most.
The "Delete All Codes" option: When selecting this option, Atoll deletes all the current PN Offsets and carries out
a new PN Offset allocation. If not selected, the existing PN Offsets are kept.
9.
10.
In addition, it depends on the selected allocation strategy. Allocation strategies can be:
PN offset per cell: The purpose of this strategy is to reduce the spectrum of allocated PN offsets the maximum
possible. Atoll will allocate the first possible PN offsets in the domain.
Adjacent PN-Clusters per site: This strategy consists of allocating one cluster of adjacent PN offsets to each site,
then, one PN offset of the cluster to each cell of each transmitter according to its azimuth. When all the clusters
have been allocated and there are still sites remaining to be allocated, Atoll reuses the clusters at another site.
Distributed PN-clusters per site: This strategy consists of allocating one cluster of PN offsets to each site in the
network, then, one PN offset of the cluster to each cell of each transmitter according to its azimuth. With this
strategy, the cluster is made of PN offsets separated as much as possible. When all the clusters have been allocated and there are still sites remaining to be allocated, Atoll reuses the clusters at another site.
In the Results table, Atoll only displays PN offsets allocated to TBA cells.
7.7.1.2
Allocation Process
and "First Order"),
The cells that fulfil Ec/I0 condition (option Additional Ec/I0 conditions),
One additional constraint is considered when:
and the cost of the PN Offset plan. During the allocation, Atoll tries to assign different PN Offsets to the TBA cell and its
near cells. If it respects all the constraints, the cost of the PN Offset plan is 0. When a cell has too many constraints and
there are not anymore PN Offsets available, Atoll breaks the constraint with the lowest cost so as to generate the PN Offset
plan with the lowest cost. For information on the cost generated by each constraint, see "Cell Priority" on page 353.
7.7.1.2.1

Strategy: PN offset per cell

Atoll processes TBA cells according to their priority. It allocates PN Offsets starting with the highest priority cell and its near
cells, and continuing with the lowest priority cells not allocated yet and their near cells. For information on calculating cell
priority, see "Cell Priority" on page 353.
Strategy: Adjacent PN-Clusters per site

Atoll assigns a PN-cluster of adjacent PN offsets to each site, starting with the highest priority site and its near sites, and
continuing with the lowest priority sites not allocated yet and their near sites. When all the clusters have been allocated
and there are still sites remaining to be allocated, Atoll reuses the clusters at another site. When the Co-PN Reuse
Distance option is selected, the algorithm reuses the clusters as soon as the Co-PN reuse distance is exceeded. Otherwise, when the option is not selected, the algorithm tries to assign reused clusters as spaced out as possible.
Then, Atoll allocates a PN offset from the cluster to each cell of each transmitter located on the sites according to the transmitter azimuth. It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not allocated
yet and their near cells.
Strategy: Distributed PN-Clusters per site

Atoll assigns one cluster to each site, starting with the highest priority site and its near sites, and continuing with the lowest
priority sites not allocated yet and their near sites. When all the clusters have been allocated and there are still sites
remaining to be allocated, Atoll reuses the clusters at another site. When the Co-PN Reuse Distance option is selected,
352
AT271_TRG_E6
Forsk 2009

the algorithm reuses the clusters as soon as the Co-PN reuse distance is exceeded. Otherwise, when the option is not
selected, the algorithm tries to assign reused clusters as spaced out as possible.
Then, Atoll assigns a PN offset from the cluster to each cell of each transmitter located on the sites according to the transmitter azimuth. It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not allocated
yet and their near cells.
7.7.1.2.2
In case you have a multi-carrier network and you run the PN Offset allocation on all the carriers, the allocation process
depends on wether the option "Allocate Carriers Identically" is selected or not.
When the option is not selected, algorithm works for each strategy, as explained above. On the other hand, when the
option is selected, allocation order changes. It is no longer based on the cell priority but depends on the transmitter priority.
All transmitters which have constraints with the studied transmitter will be referred to as near transmitters.
In case of a "Per cell" strategy (PN offset per cell), Atoll starts PN offset allocation with the highest priority transmitter and
its near transmitters and continues with the lowest priority transmitters not allocated yet and their near transmitters. The
same PN offset is assigned to each cell of the transmitter.
In case of a "Per site" strategy (Adjacent and Distributed PN-clusters per site strategies), Atoll assigns a cluster to each
site and then, allocates a PN offset to each transmitter. It starts with the highest priority transmitter and its near transmitters
and continues with the lowest priority transmitters not allocated yet and their near transmitters. The same PN offset is
assigned to each cell of the transmitter.
For information on calculating cell priority, see "Cell Priority" on page 353. For information on calculating transmitter priority, see "Transmitter Priority" on page 355.
Note:
7.7.1.2.3
alphanumeric order.
Difference between Adjacent and Distributed PN-Clusters

The following example explains the difference between "Adjacent PN-clusters" and "Distributed PN-clusters". The
PILOT_INC has been set to 4 and the PN-cluster size to 3. There are:
128 PN offsets that can be allocated: they are all PN offsets from 4 to 508 with a separation interval of 4.
Each PN-cluster consists of three PN offsets. So, there are 42 PN-clusters available.
If you select "Adjacent PN-cluster per site" as allocation strategy, Atoll will consider PN-clusters consisted of adjacent PN
offsets (e.g. {4,8,12}, {16,20,24}, ...,{496,500,504}).
If you select "Distributed PN-cluster per site" as allocation strategy, Atoll will consider PN-clusters consisted of PN offsets
separated as much as possible (e.g. {4,172,340}, {8,176,344}, ...,{168,336,504}).
7.7.1.3
7.7.1.3.1
Cell Priority
PN Offset allocation algorithm in Atoll allots priorities to cells before performing the actual allocation. Priorities assigned to
cells depend upon how much constrained each cell is and the cost defined for each constraint. A cell without any constraint
has a default cost, C , equal to 0. The higher the cost on a cell, the higher the priority it has for the PN Offset allocation
process.
There are five criteria employed to determine the cell priority:
PN Offset Domain Criterion
The cost due to the domain constraint, C i Dom , depends on the number of PN Offsets available for the allocation. The
When no domain is assigned to cells, 512 PN Offsets are available and we have:
C i Dom = 0
When domains of PN Offsets are assigned to cells, each unavailable PN Offset generates a cost. The higher the number
of codes available in the domain, the less will be the cost due to this criterion. The cost is given as:
C i Dom = 512 Number of PN Offsets in the domain
Distance Criterion
C i Dist =
Cj Dist i
j
Forsk 2009
AT271_TRG_E6
353

Where
C i EP =
cEP i j
j
Where

Ci N =
j

C j N1 i = I j c N1
Where
Because two first order neighbours must not have the same PN Offset, Atoll considers the cost created by two first order
neighbours to be each other.
C j N1 i + C j N1 i
C j j N1 i = ---------------------------------------------------------2
Where
Because two second order neighbours must not have the same PN Offset, Atoll considers the cost created by two second
C k N2 i + C k N2 i
C k k N2 i = -----------------------------------------------------------2
C l N3 i = Max j
c N3
Where
354
AT271_TRG_E6
Forsk 2009

Because two third order neighbours must not have the same PN Offset, Atoll considers the cost created by two third order
neighbours to be each other.
C l N3 i + C l N3 i
C l l N3 i = ---------------------------------------------------------2
Note:
.
And
C i N 2G =
cN
2G
j Tx 2G
Where
cN
2G
With
C i U = C i Dist + C i EP + C i N + C i N 2G
7.7.1.3.2
In case you have a multi-carrier network and you run PN Offset allocation on "all" the carriers with the option "allocate
carriers identically", algorithm in atoll allots priorities to transmitters. Priorities assigned to transmitters depend on how
much constrained each transmitter is and the cost defined for each constraint. The higher the cost on a transmitter, the
higher the priority it has for the PN Offset allocation process.
as:
With C Tx U =
Max C U and C Dom = 512 Number of PN offsets in the domain

i
Tx
i Tx
7.7.1.3.3
Site Priority
In case of "Per Site" allocation strategies (Adjacent PN-clusters per site and Distributed PN-clusters per site), algorithm in
Atoll allots priorities to sites. Priorities assigned to sites depend on how much constrained each site is. The higher the
constraint on a site, the higher the priority it has for the PN Offset allocation process.
Let us consider a site S with three transmitters; each of them has two cells using carriers 0 and 1. The site constraint is
given as:
With, C S U =
CTx U , and CS Dom = 512 Number of PN offsets in the domain

Tx
Here, the domain considered for the site is the intersection of domains available for transmitters of the site.
Forsk 2009
AT271_TRG_E6
355
7.7.2
Allocation Examples
In order to understand the differences between the different allocation strategies and the behaviour of the algorithm when
using a maximum of PN offsets or not, let us consider the following sample scenario:
Figure 7.8: PN Offset Allocation

Let Site0, Site1, Site2 and Site3 be four sites with 3 cells using carrier 0 whom PN Offsets have to be allocated. The
PILOT_INC parameter has been set to 4 and the PN Cluster Size is 3. Therefore, all PN offsets from 4 to 508 with a separation interval of 4 can be allocated. The reuse distance is supposed to be lower than the inter-site distance. Only co-site
neighbours exist and all of them have the same importance.
The following section lists the results of each combination of options with explanation where necessary.
7.7.2.1
Strategy: PN Offset per Cell

Since the restrictions of neighbourhood only apply to co-sites with the same importance and inter-site distances are greater
than reuse distances, every cell has the same priority. Then, the PN offset allocation to cells is performed in an alphanumeric order.
356
Without Use a Maximum of PN Offsets
With Use a Maximum of PN Offsets
Atoll allocates the first three PN offsets in the domain (4, 8

and 12) to the Site0s cells. Under given constraints of
neighbourhood and reuse distance, same PN offsets can
be allocated to each sites cells.
Atoll allocates the first three PN offsets in the domain (4, 8

and 12) to the Site0s cells. As it is allowed to use a maximum of PN offsets, Atoll allocates different PN offsets to
each sites cells so that there is least repetition.
AT271_TRG_E6
Forsk 2009
7.7.2.2
Strategy: Adjacent PN-Clusters Per Site

Atoll allocates a PN cluster of adjacent PN offsets to Site0 As it is possible to use a maximum of PN offsets, Atoll alloand then, one PN offset of the PN cluster to each cell. cates different PN clusters of adjacent PN offsets to sites so
Under given constraints of neighbourhood and reuse that there is least repetition of PN offsets.
distance, the same PN cluster can be allocated to each site
and same PN offsets to each sites cells.
7.7.2.3
Strategy: Distributed PN-Clusters Per Site

Atoll allocates a PN cluster of distributed PN offsets to Site0 As it is possible to use a maximum of PN offsets, Atoll alloand then, one PN offset of the PN cluster to each cell. cates different PN clusters of distributed PN offsets to sites
Under given constraints of neighbourhood and reuse so that there is least repetition of PN offsets.
distance, the same PN cluster can be allocated to each site
and same PN offsets to each sites cells.
7.8
Automatic GSM-CDMA Neighbour Allocation
7.8.1
Overview
It is possible to automatically calculate and allocate neighbours between GSM/TDMA and CDMA (i.e. IS-95 cdmaOne or
CDMA2000) networks. In Atoll, it is called inter-technology neighbour allocation.
Forsk 2009
When the CDMA coverage is not continuous. In this case, the CDMA coverage is extended by CDMA-GSM
handover into the GSM network,
And in order to balance traffic and service distribution between both networks.
AT271_TRG_E6
357

Note that the automatic inter-technology neighbour allocation algorithm takes into account both cases.
An .atl document containing the GSM/TDMA network, GSM.atl, and another one containing the CDMA (i.e. IS-95
cdmaOne or CDMA2000) network, CDMA.atl,
An existing link on the Transmitters folder of GSM.atl into CDMA.atl.
transmitters of GSM.atl are potential neighbours. The cells to be allocated will be called TBA cells which, being cells of
CDMA.atl, fulfill following conditions:
They are active,

Only CDMA TBA cells may be assigned neighbours.
7.8.2

The inter-transmitter distance,

The maximum number of neighbours fixed,
Allocation options,
The selected allocation strategy,
We assume we have a CDMA reference cell, A, and a GSM candidate neighbour, transmitter B.
7.8.2.1

1.
The distance between the CDMA reference cell and the GSM neighbour must be less than the user-definable
maximum inter-site distance. If the distance between the CDMA reference cell and the GSM neighbour is greater

Note:
2.
Transmitter azimuths are taken into account to evaluate the inter-transmitter distance. For
further information on inter-transmitter distance calculation, see "Calculation of InterTransmitter Distance" on page 270.
the reference CDMA cell in the candidate neighbour list. This option is automatically selected.
force/forbid a GSM transmitter to be candidate neighbour of the reference CDMA cell.
3.
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

distance conditions
358
AT271_TRG_E6
Forsk 2009

be marked as exceptional pair or co-site. If the neighbour is not forced but satisfies distance conditions, Atoll displays the
7.8.2.2

1.
The distance between the CDMA reference cell and the GSM neighbour must be less than the user-definable
maximum inter-site distance. If the distance between the CDMA reference cell and the GSM neighbour is greater
Note:
2.
Here, real inter-transmitter distance is considered.
the reference CDMA cell in the candidate neighbour list. This option is automatically selected.
force/forbid a GSM transmitter to be candidate neighbour of the reference CDMA cell.
3.
There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability where:

-
1st case: SA is the area where the cell A is the best serving cell of the CDMA network.
- The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0) and is the highest one.
In this case, the Ec/I0 margin must be equal to 0dB and the max Ec/I0 option disabled.
2nd case: SA represents the area where the pilot quality from the cell A strats decreasing but the cell A is still
the best serving cell of the CDMA network.
The Ec/I0 margin must be equal to 0dB, the max Ec/I0 option selected and a maximum Ec/I0 user-defined.
-
The pilot quality from A exceeds the minimum Ec/I0 but is lower than the maximum Ec/I0.
The pilot quality from A is the highest one.
3rd case: SA represents the area where the cell A is not the best serving cell but can enter the active set.
Here, the Ec/I0 margin has to be different from 0dB and the max Ec/I0 option disabled.
-
The pilot quality from A is within a margin from the best Ec/I0, where the best Ec/I0 exceeds the minimum
Ec/I0.
4th case: SA represents the area where:

- The pilot quality from A is within a margin from the best Ec/I0 (where the best Ec/I0 exceeds the minimum
Ec/I0) and lower than the maximum Ec/I0.
In this case, the margin must be different from 0dB, the max Ec/I0 option selected and a maximum Ec/I0
user-defined.

-
1st case: SB is the area where the cell B is the best serving cell of the GSM network.
In this case, the margin must be set to 0dB.
-
is the highest one.
2nd case: The margin is different from 0dB and SB is the area where:
SA SB
SA
area.
Forsk 2009
AT271_TRG_E6
359
Guidelines for the automatic allocation

When the automatic allocation is based on coverage overlapping, we recommend you to perform two successive automatic allocations:
- A first allocation in order to find handovers due to non-continuous CDMA coverage. In this case, you have to select the
max Ec/I0 option and define a high enough value.
- A second allocation in order to complete the previous list with handovers motivated for reasons of traffic and service
distribution. Here, the max Ec/I0 option must be disabled.
4.
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

coverage conditions
be marked as exceptional pair or co-site. If the neighbour is not forced but fulfils coverage conditions, Atoll displays the
Notes:
7.8.2.3

not found.
dialog of the predictions folder (default resolution parameter).


A new TBA cell i is created in CDMA.atl. Therefore, if you start a new allocation without selecting the Delete existing neighbours option, Atoll determines the neighbour list of the cell i,
If you change some allocation criteria (e.g. increase the maximum number of neighbours or create a new GSM TBC transmitter) and start a new allocation without selecting the Delete existing neighbours option, it examines the neighbour list of
TBA cells and checks allocation criteria if there is space in their neighbour lists. A new GSM TBC transmitter can enter the
TBA cell neighbour list if allocation criteria are satisfied. It will be the first one in the neighbour list.
7.9
Automatic cdmaOne-CDMA2000 Neighbour

Allocation
7.9.1
Overview
It is possible to automatically calculate and allocate neighbours between cdmaOne and CDMA2000 networks. In Atoll, it
is called inter-technology neighbour allocation.
360
AT271_TRG_E6
Forsk 2009

In order to be able to use the external neighbour allocation algorithm, you must have:
An .atl document containing the CDMA2000 network, CDMA2000.atl, and another one containing the cdmaOne
network, CDMA.atl,
An existing link of the Transmitters folder of CDMA.atl into CDMA2000.atl or vice-versa.
The external neighbour allocation algorithm takes into account all the CDMA TBC cells. It means that all the TBC cells of
CDMA.atl are potential neighbours. The cells to be allocated will be called TBA cells which, being cells of CDMA2000.atl,
fulfill following conditions:
They are active,

Only CDMA2000 TBA cells may be assigned neighbours.
7.9.2

We assume we have a CDMA2000 reference cell, A, and a CDMA candidate neighbour, cell B.
1.
The distance between the CDMA2000 reference cell and the CDMA neighbour must be less than the userdefinable maximum inter-site distance. If the distance between the CDMA2000 reference cell and the CDMA
neighbour is greater than this value, then the candidate neighbour is discarded.

Note:
2.
information on inter-transmitter distance calculation, please refer to paragraph "Calculation
of Inter-Transmitter Distance" on page 270.
Force co-site cells as neighbours: It enables you to automatically include CDMA cells located on the same site as the reference CDMA2000 cell in the candidate neighbour list. This option is automatically selected.
force/forbid a CDMA cell to be candidate neighbour of the reference CDMA2000 cell.
3.
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

distance conditions
be marked as exceptional pair or co-site. If the neighbour is not forced but fulfils distance conditions, Atoll displays the
Notes:
Forsk 2009

AT271_TRG_E6
361
7.9.2.1

A new TBA cell i is created in CDMA2000.atl. Therefore, if you start a new allocation without selecting the Delete existing
neighbours option, Atoll determines the neighbour list of the transmitter i,
If you change some allocation criteria (e.g. increase the maximum number of neighbours or create a new CDMA TBC cell)
and start a new allocation without selecting the Delete existing neighbours option, it examines the neighbour list of TBA
cells and checks allocation criteria if there is space in their neighbour lists. A new CDMA TBC cell can enter the TBA cell
neighbour list if allocation criteria are satisfied. It will be the first one in the neighbour list.
362
AT271_TRG_E6
Forsk 2009
Chapter 8
TD-SCDMA Networks
This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations
and prediction studies available in TD-SCDMA projects.
Atoll
364
AT271_TRG_E6
Forsk 2009
Chapter 8: TD-SCDMA Networks
TD-SCDMA Networks
This chapter describes in detail the algorithms, calculation parameters, and processes of the coverage predictions and the
simulations available in TD-SCDMA documents. The first part of this chapter lists all the input and output parameters in
the TD-SCDMA documents, their significance, location in the Atoll GUI, and their usage. Detailed explanation of the basic
coverage predictions, which do not require simulation results, is provided in the second part. The third part describes the
traffic scenario generation and Montel Carlo simulation algorithms including smart antenna modelling and dynamic channel allocation. The next sections are dedicated to TD-SCDMA coverage predictions which can be based on results
obtained from simulations. The last three sections describe in detail the allocation of frequencies, i.e., master and slave
carriers, the allocation of neigbours, and the allocation of scrambling codes.
8.1

The tables in the following subsections list the input and output parameters and formulas used in simulations and other
computations.
8.1.1
Inputs
This table lists the inputs to computations, coverage predictions, and simulations.
Name
Value
Unit
Description
R Ch
Global parameter
Mcps
Chip rate (or Spreading rate) (1.28)
Spread
Global parameter
None
Minimum spreading factor (1)
F Max
Spread
Global parameter
None
Maximum spreading factor (16)
Proc
Global parameter
None
P-CCPCH processing gain (13.8 dB)
N TS
SF
Global parameter
None
Number of timeslots per subframe (7)
SF
Global parameter
ms
Subframe duration (5)
Frame
Global parameter
ms
Frame duration (10)
N Ch TS
GP
Global parameter
None
Number of guard period chips per

timeslot (16)
N Ch TS
Data
Global parameter
None
Number of data chips per timeslot

(704)
N Ch TS
Midamble
Global parameter
None
Number of midamble chips per

timeslot (144)
N Ch PTS
GP
Global parameter
None
Number of guard period chips per pilot

timeslot (96)
N Ch DwPTS
GP
Global parameter
None

DwPTS timeslot (32)
SYNC_DL
Global parameter
None
Number of SYNC_DL chips per

DwPTS timeslot (64)
None
Total number of chips per DwPTS

timeslot (96)
F Min
G P CCPCH
D
D
N Ch DwPTS
Total
N Ch DwPTS
Global parameter
Total
N Ch DwPTS
GP
SYNC_DL
= N Ch DwPTS + N Ch DwPTS
N Ch UpPTS
GP
Global parameter
None

UpPTS timeslot (32)
SYNC_UL
Global parameter
None
Number of SYNC_UL chips per

UpPTS timeslot (128)
None
Total number of chips per UpPTS

timeslot (160)
N Ch UpPTS
Total
N Ch UpPTS
Global parameter
Total
N Ch UpPTS
GP
SYNC_UL
= N Ch UpPTS + N Ch UpPTS
Calculated global parameter

Data
N Ch TS
W = -----------------SF
D
bps
Chip rate (140800 bps)
F Avg
MHz
Average frequency range of the

frequency band (2010)
Forsk 2009
AT271_TRG_E6
365

BW
MHz
Channel bandwidth of the carriers of a

frequency band (1.6)
F IRF
Cell parameter
None
Interference reduction factor
None
Joint Detection (JD) factor
(user-defined or calculated from transmitter equipment characteristics)
None
BTS Noise Figure
(user-defined or calculated from transmitter equipment characteristics)
None
Transmitter loss
L Tx = L Total UL on uplink
TX
BTS parameter
None

transmitted
P TCH
Max
Cell parameter
Maximum cell traffic timeslot power
P P CCPCH
Cell parameter
P-CCPCH power on TS0
P DwPCH
Cell parameter
DwPCH power on DwPTS
P OCCH TS0
Cell parameter
Other common channel power on TS0
TComp P CCPCH
Cell parameter
None
P-CCPCH RSCP comparative

threshold for baton handover
P Max
Cell parameter
None
Maximum difference between two

transmitted powers
Req
Cell parameter
None
Required resource units in uplink
RU DL
Req
Cell parameter
None
Required resource units in downlink
P HS PDSCH
Available
Cell parameter
HS-PDSCH power available per

downlink timeslot
P HR
Cell parameter
None
Power headroom
P HS SCCH
Cell parameter
HS-SCCH power per downlink

timeslot
N HS SCCH
Cell parameter
None
Number of HS-SCCH channels
N HS SICH
Cell parameter
None
Number of HS-SICH channels
Cell parameter
None
Maximum number of HSDPA users
Min
Cell parameter
None
Minimum number of HS-PDSCH

codes
Max
Cell parameter
None
Maximum number of HS-PDSCH

codes
Max
Cell parameter
None
Maximum number of intra-technology

neighbours
N Intra Neigh
Max
Cell parameter
None
Maximum number of inter-technology

neighbours
P OCCH
Timeslot parameter
Other common channel powera
DL
Timeslot parameter
Downlink traffic power
%P DL
Timeslot parameter
(Simulation constraint)
None
Maximum percentage of downlink

used power
X UL
Timeslot parameter
(Simulation result)
None
Uplink load factor
Timeslot parameter
(Simulation constraint)
None
P HS PDSCH
Timeslot parameter
HS-PDSCH power available
Min
Timeslot parameter
None
Minimum number of HS-PDSCH

codes
Max
Timeslot parameter
None
Maximum number of HS-PDSCH

codes
TX
F JD
NF
TX
TX
RU UL
Max
N HSDPA
N HS-PDSCH Codes
N HS-PDSCH Codes
N Intra Neigh
P TCH
Max
Max
X UL
Available
N HS-PDSCH Codes
N HS-PDSCH Codes
366
AT271_TRG_E6
L Tx = L Total DL on downlink
Forsk 2009

RU Overhead
Timeslot parameter
Overhead resource units
Body
Service parameter
None
Body loss
f DL
Act
Service parameter
None
Downlink activity factor for circuitswitched services and the A-DPCH

activity factor for HSDPA services
f UL
Act
Service parameter
None
Uplink activity factor for circuitswitched services and the A-DPCH

activity factor for HSDPA services
f DL
Eff
Service parameter
None
Downlink efficiency factor for circuitswitched services
f UL
Eff
Service parameter
None
Uplink efficiency factor for circuitswitched services
F Scaling
Service parameter
None
Application througput scaling factor
O TP
Service parameter
kbps
Application throughput offset
UL
Service parameter (packet session modelling)
None
Average number of packet calls on the

uplink during a session
DL
None
Average number of packet calls on the

downlink during a session
UL
ms
Average time between two packet

calls on the uplink
T PacketCall
DL
ms
Average time between two packet

calls on the downlink
UL
KBytes
Minimum packet call size on the uplink
DL
KBytes
Minimum packet call size on the

downlink
UL
KBytes
Maximum packet call size on the

uplink
S Max PacketCall
DL
KBytes
Maximum packet call size on the

downlink
T Packet
UL
ms
Average time between two packets on

the uplink
DL
ms
Average time between two packets on

the downlink
UL
Bytes
Packet size on uplink
DL
Bytes
Packet size on downlink
Nom
R99 bearer parameter
kbps
Downlink nominal bit rate
Nom
kbps
Uplink nominal bit rate
W
-)
R99 bearer parameter (Can be calculated as ------------Nom
R DL
None
Downlink processing gain
W
-)
R99 bearer parameter (Can be calculated as ------------Nom
R UL
None
Uplink processing gain
Min
Allowed minimum downlink traffic

channel power
Max
Allowed maximum downlink traffic

channel power
N DL
TS
None
Number of downlink timelots
TS
None
Number of uplink timelots
E b Req
C Req
R99 bearer parameter per mobility ( ------
or ----
)
N t TCH UL
I TCH UL
None
Eb/Nt or C/I target on uplink
E b Req
C Req
R99 bearer parameter per mobility ( ------
or ----
)
N t TCH DL
I TCH DL
None
Eb/Nt or C/I target on downlink
N PacketCall
N PacketCall
T PacketCall
S Min PacketCall
S Min PacketCall
S Max PacketCall
T Packet
S Packet
S Packet
R DL
R UL
Proc
G DL
Proc
G UL
P TCH DL
P TCH DL
N UL
Req
Q TCH UL
Req
Q TCH DL
Forsk 2009
AT271_TRG_E6
367

Req
R99 bearer parameter per mobility
Target RSCP on uplink TCH
Req
Target RSCP on downlink TCH
G DL
Div
None
Downlink diversity gain
Div
None
Uplink diversity gain
Term
Terminal parameter
Maximum terminal power
P Min
Term
Terminal parameter
Minimum terminal power
P UpPCH
Terminal parameter
UpPCH power
Term
Terminal parameter
None
Term
Terminal parameter
None
Joint Detection (JD) factor
Term
Terminal parameter
None

transmitted
Term
Terminal parameter
None
Terminal gain
Term
Terminal parameter
None
Terminal loss
TAdd P CCPCH
Mobility parameter
Required RSCP T_Add for P-CCPCH
TDrop P CCPCH
Mobility parameter
Required RSCP T_Drop for P-CCPCH
Req
Mobility parameter
Required RSCP threshold for DwPCH
Req
Mobility parameter
Required RSCP threshold for UpPCH
E b Req
C Req
Mobility parameter ( ------
or ----
)
N t P CCPCH
I P CCPCH
None
Required quality threshold for PCCPCH
E c Req
)
N t HS SCCH
None
Required quality threshold for HSSCCH
E c P CCPCH
)
N t HS SICH
None
Required quality threshold for PCCPCH
C Req
Mobility parameter ( ----
)
I DwPCH
None
Required quality threshold for DwPCH
Clutter class parameter
None
Model standard deviation
None
P-CCPCH Eb/Nt or C/I standard

deviation
CI
None
Downlink Eb/Nt or C/I standard

deviation
CI
None
Uplink Eb/Nt or C/I standard deviation
L Indoor
None
Indoor loss
Ortho
None
Downlink orthogonality factor
F UL
Ortho
None
Uplink orthogonality factor
Spread
Spreading angle
1.38 x 10-23
J/K
Boltzman constant
293
Ambient temperature
TX
N0
NF TX K T BW
Term
NF Term K T BW
TX
Antenna parameter
None
RSCP TCH UL
RSCP TCH DL
G UL
P Max
NF
F JD
G
L
RSCP DwPCH
RSCP UpPCH
Req
Q P CCPCH
Req
Q HS SCCH
Req
Q HS SICH
Req
Q DwPCH
Model
Eb/Nt
P CCPCH or
CI
P CCPCH
Eb/Nt
or DL
Eb/Nt
or UL
DL
UL
F DL
N0
368
AT271_TRG_E6
Forsk 2009

L Path
None
Path loss

standard deviation
None
Model shadowing margin used in

coverage predictions
P CCPCH
Result calculated from cell edge coverage probability and P-CCPCH

Eb/Nt standard deviation
None
P-CCPCH Eb/Nt shadowing margin

used in coverage predictions
Eb Nt DL

standard deviation
None
DL Eb/Nt shadowing margin used in

Eb Nt UL

standard deviation
None
UL Eb/Nt shadowing margin used in

Model
M Shadowing
M Shadowing
M Shadowing
M Shadowing
For RSCP calculation

Model
LT
TX
Term
Body
Transmitter-terminal total loss in

Model
M Shadowing
L Path L L
L
L Indoor
= -----------------------------------------------------------------------------------------------------------------------------------TX
Term
G G
In UL, only carrier power is attenuated

Eb Nt UL
For P-CCPCH Eb/Nt calculation

TX
Term
Body
by M Shadowing .
P CCPCH
L Path L L
L
L Indoor M Shadowing
= -----------------------------------------------------------------------------------------------------------------------------------TX
Term
G G
Eb Nt P
LT
LT
None

Eb Nt DL
LT
TX
Term
L Path L L
L
= -----------------------------------------------------------------------------------------------------------------------------------TX
Term
G G
Eb Nt UL
a.
TX
Term
P CCPCH
extra-cell interference is not.

Eb Nt DL

LT
Eb Nt DL
M Shadowing or M Shadowing while
Eb Nt DL
Body
In DL, carrier power and intra-cell

interference are attenuated by
Therefore, M Shadowing or
Eb Nt UL
Body
L Path L L
L
= -----------------------------------------------------------------------------------------------------------------------------------TX
Term
G G
P CCPCH
M Shadowing are set to 1 in DL extracell interference calculation.
For the calculation of interference, P OCCH also includes the MBMS SCCPCH channel power when the optional MBMS
feature is activated. You must modify the data structure for activating the optional MBMS feature. For more information, see
the Administrator Manual.
8.1.2
P-CCPCH Eb/Nt and C/I Calculation
Name
TX i ic
E
-----b-
N t P CCPCH
Value
TX i ic
RSCP P CCPCH
-----------------------------------------------------------TX i ic
N Tot DL
TX i
TX i ic
TX i ic
TX i ic
Term
I Intra DL + I Extra DL + I IC DL ic jc + N 0
TX i ic
RSCP P CCPCH
TX i ic
I Intra DL
With
TX i
TX i
TX i ic
I IC DL ic jc
8.1.3
Name
Forsk 2009
TX i
TX i ic
+ RSCP OCCH TS0

Ortho
1 F DL
= 0
1
I Extra DL
Description
None
P-CCPCH Eb/Nt for the cell TX i ic
None
P-CCPCH C/I for the cell TX i ic
Downlink total noise for the cell

TX i ic
Downlink intra-cell interference for the

cell TX i ic
Downlink extra-cell interference for the

cell TX i ic
Inter-carrier interference
Unit
Description
TX i ic
TX i
RSCP P CCPCH
-----------------------------------------------------------TX i ic
N Tot DL
TX i ic
C
----
I P CCPCH
N Tot DL
Proc
G P CCPCH
Unit
Term
1 F JD
TX i
and
Without Useful Signal

Total Noise
TX j ic
TX j ic
TX j jc
TX j jc
RSCPP CCPCH + RSCPOCCH TS0

ji

TX
j
----------------------------------------------------------------------------------------------------------F IRF ic jc
DwPCH C/I Calculation

Value
AT271_TRG_E6
369
TX i ic
N Tot DL
TX i ic
TX i
RSCP DwPCH
---------------------------------------------------TX i ic
N Tot DL
TX i ic
C
----
I DwPCH
TX i ic
TX i ic
Term
TX i ic
RSCP DwPCH
TX i ic
I Intra DL
With
TX i
TX i
Ortho
1 F DL
= 0
1
DwPCH C/I for the cell TX i ic

TX i ic
Intra-cell interference for the cell

TX i ic
Extra-cell interference for the cell

TX i ic
TX i
Term
1 F JD
and

Total Noise
TX j ic
RSCPDwPCH
TX i ic
I Extra DL
None
ji
TX j jc
RSCPDwPCH
I IC DL ic jc
8.1.4
TX
j
----------------------------------------------F IRF ic jc
DL TCH Eb/Nt and C/I Calculation

When the optional MBMS feature is activated, the MBMS Eb/Nt is also calculated in the same manner. You must modify
the data structure for activating the optional MBMS feature. For more information, see the Administrator Manual.
Name
TX i ic
E
-----b-
N t TCH DL
Value
TX i ic
RSCP TCH DL
------------------------------------------------------TX i ic
N Tot DL
TX i
TX i ic
TX i ic
I Intra DL
Div
TX i ic
Description
None
Downlink TCH Eb/Nt for the cell

TX i ic
None
Downlink TCH C/I for the cell TX i ic

TX i ic
Downlink intra-cell interference for the

cell TX i ic
Downlink extra-cell interference for the

cell TX i ic
Unit
Description
None
Uplink TCH Eb/Nt for the cell TX i ic
None
Uplink TCH C/I for the cell TX i ic
Uplink required power for the terminal
TX i ic
TX i
RSCP TCH DL
------------------------------------------------------- G Div
DL
TX i ic
N Tot DL
TX i ic
C
----
I TCH DL
N Tot DL
Proc
G DL G DL
Unit
TX i ic
Term
TX i
TX i
Ortho
Term
1 F DL 1 F JD + 1
TX i ic
TX i ic
RSCP TCH DL + RSCP OCCH

TX i ic
I Extra DL
I IC DL ic jc
8.1.5
TX j ic
TX j jc
TX j jc
ji
RSCPTCH DL + RSCPOCCH
TX
j
-----------------------------------------------------------------------------------------F IRF ic jc
UL TCH Eb/Nt and C/I Calculation
Name
TX i ic
E
-----b-
N t TCH UL
TX i ic
C
----
I TCH UL
Value
TX i ic
Term
RSCP TCH UL
----------------------------------------------------------TX i ic
N Tot UL
Term
Term
P Req
Proc
Div
G UL G UL
TX i ic
RSCP TCH UL
----------------------------------------------------------- G Div
UL
TX i ic
N Tot UL
Req
370
TX j ic
Req
Q TCH UL
Q TCH UL
Term
- or P Term
P Max -------------------------------Max -----------------------------TX i ic
TX i ic
E
b
C
------
----
I TCH UL
N t TCH UL
AT271_TRG_E6
Forsk 2009
8.1.6
Interference Calculation
Name
Value
TX j jc
I C2C TX i TX j
TX j ic
TX j ic
Unit
Description
Cell to cell interference
UpPCH interference
TX j jc
RSCPTCH DL + RSCPOCCH +-----------------------------------------------------------------------------------------F IRF ic jc

TX j
TX j
8.1.7
TX i ic
X TS1 UL
------------------------------------TX i ic
1 X TS1 UL
TX i
TX i ic
I TS1 UL
N0
HSDPA Dynamic Power Calculations
Name
Value
Unit
Description
TX i ic
P HS SCCH
TX i ic
TX i
TX i ic
E
-----c-
N Tot DL RSCP HS SCCH
N t HS SCCH
------------------------------------------------------------------------------------------------------------------------------------------------- L Model
T
TX i
HS-SCCH power
HS-PDSCH power
HS-SICH power
TX i ic
TX i ic
P HS PDSCH
TX i ic
TX i ic
TX i ic
P Max DL Eff P R99 DL P HR
TX i ic
P HS SCCH
TX i ic
Mi
P HS SICH
8.1.8
TX i ic
Mi
Mi
E
-----c-
N Tot UL RSCP HS SICH
N t HS SICH
------------------------------------------------------------------------------------------------------------------------------------------- L Model
T
Mi
Smart Antenna Modelling
Name
Value
Unit
Description
E SA
Smart antenna model parameter
None
Number of smart antenna elements
Calculation parameter
Degrees
Angle of arrival for the useful signal
Degrees
Angle at which the smart antenna

effect is calculated
--- , where is the wavelength of the signal

2
Distance between two adjacent

antenna elements
gn
None
Gain of a single element
None
Steering vector for the direction of
None
Complex smart antenna weight
None
Array correlation matrix for a given

user direction
None
Smart antenna gain in any direction
None
Average array correlation matrix
1 e
2
j ------- d sin
2
j ------- 2d sin
... e
wn
e
2
j ------- nd sin
j n sin
with d = --2
H
R = S S
H
SA
T
2
j ------- E SA 1 d sin
gn S R S = gn S S S S
10 Log G
SA
in dB
R Avg
j pj Rj
j=1
RN S
None
Vector of ESA complex weights for the

beam former
E SA
---------------------------------H
1
S RN S
None
Smart antenna gain in the uplink
Forsk 2009
AT271_TRG_E6
371

J
RN
Rn + RI =
2
n
I+
pj Sj Sj
None
Total noise correlation matrix
None
Thermal noise correlation matrix
None
Interference correlation matrix
Total noise power, including thermal

noise and interference from all uplink
interferers
Total power received from the served

user
None
Signal quality in the uplink
None
Uplink smart antenna beam forming

gain in the direction of the served user
Average of the inverse noise

correlation matrices
j=1
2
Rn
n I
J
pj Sj Sj
RI
j=1
P N
S RN S
p S RN S
2
H
2
1
p S RN S
H
P
1
------- = --------------------------------------------------------------= p S RN S
H
2
1
P
N
S RN S
SA
Q UL
S I S = E SA
SA
10 Log E SA in dB
K
1
----
K
1
RN
Avg
RN
1
k
k=1
8.2
Signal Level Based Calculations

Two types of signal level based calculations are available in Atoll:
1. Point Analysis: Real-time calculations for profile and reception analysis using the mouse to move a probe mobile
on the map.
2. RSCP Based Coverage Predictions: Calculation of RSCP related parameters on each pixel and colouring
according to the selected display.
8.2.1
Point Analysis
For the selected transmitted TXi and carrier (ic), you can study three parameters in point analysis Profile tab:
Study criteria
Formulas
Signal level ( RSCP ) in dBm
RSCP
TX i ic
= EIRP
TX i ic
Model
L Path M Shadowing L Indoor

TX i
Path loss ( L Path ) in dB
L Path = L Model + L Ant
Total losses ( L T ) in dB
L T = L Path + L
TX i
Model
+ L Indoor + M Shadowing G
TX i
Where,
RSCP is the received signal code power for the P-CCPCH.
EIRP is the effective isotropic radiated power of the transmitter. EIRP
TX i ic
TX i ic
= P P CCPCH + G
TX i
TX i
ic is a carrier number
L Model is the loss on the transmitter-receiver path (path loss) calculated by the propagation model
TX i
L Ant is the transmitter antenna attenuation (from antenna patterns)

Model
M Shadowing is the shadowing margin. This parameter is taken into account when the option Shadowing taken into
account is selected
L Indoor are the indoor losses. These losses are defined for each clutter class. They are taken into account when the option
"Indoor coverage" is selected
G
372
TX i
is the transmitter antenna gain
AT271_TRG_E6
Forsk 2009

L
TX i
is the transmitter loss ( L
TX i
= L Total DL )
Notes:
8.2.1.1
It is possible to analyse all the carriers. In this case, Atoll takes the highest P-CCPCH
power of cells to calculate the signal level received from a transmitter.
Profile Tab
TX i ic
Atoll displays either the signal level received from the selected transmitter on a carrier ( RSCP P CCPCH ), or the highest
signal level received from the selected transmitter on all the carriers.
Notes:
For a selected transmitter, it is also possible to study the path loss, L Path , or the total
losses, L T . Path loss and total losses are the same on any carrier.
8.2.1.2
Reception Tab
Analysis provided in the Reception tab is based on path loss matrices. You can study reception from TBC transmitters for
which path loss matrices have been calculated on their calculation areas.
TX i ic
For each transmitter, Atoll displays either the signal level received on a carrier, ( RSCP P CCPCH ), or the highest signal
level received on all the carriers.
Received signal level bar graphs are displayed in a decreasing signal level order. The number of bars in the graph depends
on the signal level received from the best server. Only bars for transmitters whose signal level is within a 30 dB margin
from the best server signal are displayed.
Note:
8.2.2
RSCP Based Coverage Predictions

For each TBC transmitter, TXi, Atoll determines the value of the selected parameter on each studied pixel inside the TXi
calculation area. Each pixel within the TXi calculation area is considered a probe receiver.
The study conditions to determine the service area of each TBC transmitter
The display settings to for colouring the covered pixels
Atoll uses the parameters entered in the Condition tab of the coverage study properties dialogue to determine pixels
covered by the each transmitter. Coverage prediction display resolution is independent of the path loss matrix and
geographic data resolutions, and can be different for each coverage prediction. Coverage predictions are calculated using
bilinear interpolation of multi-resolution path loss matrices (similar to the evaluation of site altitudes).
8.2.2.1
The RSCP from a transmitter TXi and a selected carrier (ic) is given by:
RSCP
TX i ic
= EIRP
TX i ic
Model
L Path M Shadowing L Body L Indoor + G
Term
Term
Where,
RSCP is the received signal code power. RSCP can be calculated for P-CCPCH, DwPCH, or the downlink TCH.
TX i ic
TX i ic
EIRP is the effective isotropic radiated power of the transmitter. EIRP P CCPCH = P P CCPCH + G
TX i ic
EIRP DwPCH
TX i ic
P DwPCH
+G
TX i
TX i
, or
TX i ic
EIRP DL TCH
TX i ic
P DL TCH
+G
TX i
TX i
TX i
TX i
ic is a carrier number
TX i

TX i
Forsk 2009
AT271_TRG_E6
373

Model
account is selected
L
Term
is the terminal loss
L Body is the body loss defined in the service

G
G
L
Term
TX i
TX i
is the receiver total gain

TX i
= L Total DL )
8.2.2.2
P-CCPCH RSCP Coverage Prediction
8.2.2.2.1
Coverage Condition
This coverage prediction calculates and displays the Received Signal Code Power (RSCP) for the P-CCPCH. The coverage prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0. The best
server for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest PCCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage
prediction is calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any
pixels covered by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the RSCP
considering:
the preferred carrier of the selected service, or

the carrier with the highest P-CCPCH power, if no preferred carrier is defined for the service, or
the master carrier in case of N-frequency mode compatible transmitters.
You can select the display colours according to the RSCP, or on any best server parameter.
TX i ic
The pixels in the TX i ic coverage area where RSCP P CCPCH TAdd P CCPCH Mobility are covered and coloured
according to the selected display parameter.
8.2.2.2.2
Coverage Display
It is possible to colour the transmitter service areas using a unique colour per transmitter, or colour the pixels in the coverage areas by any transmitter attribute or other criteria such as:
Best Signal Level (dBm)

TX i ic
Atoll calculates the best RSCP P CCPCH received from each transmitter TX i ic on each pixel. Where other
service areas overlap the studied one, Atoll chooses the highest RSCP. A pixel of a service area is coloured if
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility . The pixel colour depends on the RSCP level. Coverage consists
of several independent layers whose visibility in the workspace can be managed. There are as many layers as
defined thresholds. Each layer corresponds to an area where the RSCP from the best server exceeds a defined
minimum threshold.
RSCP Margin (dB)

Coverage consists of several layers with a layer per user-defined RSCP margin defined in the Display tab (PredicTX i ic
RSCP
tion properties). For each layer, area is covered if RSCP P CCPCH TAddP CCPCH Mobility M P CCPCH .

TX i ic
On each pixel of each transmitter service area, the coverage corresponds to the pixels where the RSCP P CCPCH
from the transmitter exceeds TAddP CCPCH defined in the mobility selected in the Conditions tab, with different
cell edge coverage probabilities. There is one coverage area per transmitter in the explorer.
8.2.2.3
Best Server P-CCPCH Coverage Prediction

This coverage prediction calculates and displays the best server RSCP for the P-CCPCH. The coverage prediction is
calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0. The best server for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power, or from
the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage prediction is calculated
for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any pixels covered by this
transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the RSCP considering:
374
AT271_TRG_E6
Forsk 2009
The pixels in the

TX i ic
RSCP P CCPCH =
8.2.2.4
TX i ic
coverage area where
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility
and where
Best RSCP TXj jc

P CCPCH will be covered and coloured according to the transmitter colour.
j = All
P-CCPCH Pollution Coverage Prediction

This coverage prediction calculates and displays the number of P-CCPCH polluters. Atoll calculates the Received Signal
Code Power (RSCP) for the P-CCPCH for each pixel in the TX i ic coverage area where
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility and determines the polluting transmitters according to:
TX i ic
TX j jc
RSCP P CCPCH Best RSCP P CCPCH M
ji
Where M is the specified pollution margin.

The coverage prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0.
The best server for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the
coverage prediction is calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there will
not be any pixels covered by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the
RSCP considering:

Atoll determines the number of transmitters covering each pixel and colours the pixel according to the number of polluting
transmitters. Coverage consists of several independent layers whose visibility in the workspace can be managed. There
are as many layers as defined thresholds. Each layer corresponds to an area where the number of servers exceeds ( )
a defined minimum threshold.
8.2.2.5
DwPCH RSCP Coverage Prediction
8.2.2.5.1
Coverage Condition
This coverage prediction calculates and displays the Received Signal Code Power (RSCP) for the DwPCH. The coverage
prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for DwPTS. The best
server for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest PCCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage
prediction is calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any
pixels covered by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the RSCP
considering:

TX i ic
Req
The pixels in the TX i ic coverage area where RSCP DwPCH RSCP DwPCH Mobility are covered and coloured
according to the selected display parameter.
8.2.2.5.2
Coverage Display
DwPCH RSCP (dBm)

TX i ic
Atoll calculates the best RSCP DwPCH received from each transmitter TX i ic on each pixel.. Where other service areas overlap the studied one, Atoll chooses the highest RSCP. A pixel of a service area is coloured if
TX i ic
Req
RSCP DwPCH RSCP DwPCH Mobility . The pixel colour depends on the RSCP level. Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as
defined thresholds. Each layer corresponds to an area where the RSCP from the best server exceeds a defined
minimum threshold.
RSCP Margin (dB)

Coverage consists of several layers with a layer per user-defined RSCP margin defined in the Display tab (PredicTX i ic
Req
RSCP
tion properties). For each layer, area is covered if RSCP DwPCH RSCP DwPCH Mobility M DwPCH . Each layer
Forsk 2009
AT271_TRG_E6
375
Cell edge coverage probability (%)

TX i ic
On each pixel of each transmitter service area, the coverage corresponds to the pixels where the RSCP DwPCH
TX i ic
from the transmitter TX i ic exceeds RSCP DwPCH defined in the mobility selected in the Conditions tab, with
different cell edge coverage probabilities. There is one coverage area per transmitter in the explorer.
8.2.2.6
UpPCH RSCP Coverage Prediction
8.2.2.6.1
Coverage Condition
This coverage prediction calculates and displays the Received Signal Code Power (RSCP) for the UpPCH in the uplink.
The coverage prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for UpPTS.
RSCP considering:

Atoll uses the UpPCH power of the selected terminal to calculate the RSCP from each pixel of each transmitters best
server coverage area.
Term
Req
The pixels where RSCP UpPCH RSCP UpPCH Mobility are covered and coloured according to the selected display
parameter.
8.2.2.6.2
Coverage Display
UpPCH RSCP (dBm)

Term
Atoll calculates the best RSCP UpPCH received from each pixel of each transmitter service area at the transmitter.
Where other service areas overlap the studied one, Atoll chooses the highest RSCP. A pixel of a service area is
Term
Req
coloured if RSCP UpPCH RSCP UpPCH Mobility . The pixel colour depends on the RSCP level. Coverage
consists of several independent layers whose visibility in the workspace can be managed. There are as many
layers as defined thresholds. Each layer corresponds to an area where the RSCP at the best server exceeds a
RSCP Margin (dB)

Coverage consists of several layers with a layer per user-defined RSCP margin defined in the Display tab (PredicTerm
Req
RSCP
tion properties). For each layer, area is covered if RSCP UpPCH RSCP UpPCH Mobility M UpPCH . Each layer

On each pixel of each transmitter service area, the coverage corresponds to the pixels from which the
Term
Term
RSCP UpPCH at the transmitter exceeds RSCP UpPCH defined in the mobility selected in the Conditions tab, with
different cell edge coverage probabilities. There is one coverage area per transmitter in the explorer.
8.2.2.7
Baton Handover Coverage Prediction
8.2.2.7.1
Coverage Condition
This coverage prediction determines the pixels which receive RSCP from cells other than the best server high enough to
perform baton handovers. Received Signal Code Power (RSCP) is calculated for the P-CCPCH. The coverage prediction
is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0. The best server for the
coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power,
or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage prediction is
calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any pixels covered
by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the RSCP considering:
The

pixels
are
TX i ic
RSCP P CCPCH
376
covered
and
coloured
TAdd P CCPCH Mobility and
according
to
TX j jc
RSCP P CCPCH
AT271_TRG_E6
the
selected
display
TDrop P CCPCH Mobility
parameters,
where
TX j jc
TComp P CCPCH
Forsk 2009
8.2.2.7.2
Coverage Display
It is possible to display the potential handover areas or the number of transmitters covering each pixel.
Handover Areas
Atoll displays the pixels where there are transmitters other than the best server that satisfy the above criteria.
Coverage consists of a single layer with a defined colour whose visibility in the workspace can be managed.
Number of Potential Servers

Atoll determines the number of transmitters covering each pixel and colours the pixel according to the number of
transmitters. Coverage consists of several independent layers whose visibility in the workspace can be managed.
There are as many layers as defined thresholds. Each layer corresponds to an area where the number of servers
exceeds ( ) a defined minimum threshold.
8.2.2.8
Scrambling Code Interference Analysis

This coverage prediction calculates and displays the pixels covered by two cells using the same scrambling code. Atoll
calculates the Received Signal Code Power (RSCP) for the P-CCPCH for each pixel in the TX i ic coverage area where
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility and determines the interfering transmitters according to:
TX i ic
TX j jc
RSCP P CCPCH Best RSCP P CCPCH M
ji
Where M is the specified pollution margin.

The coverage prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0.
RSCP considering:

Atoll determines whether the cells of two transmitters covering a pixel have the same scrambling code. If the pixel is interfered, Atoll colours it according to the colour assigned to the scrambling code in the display parameters. Coverage consists
of several independent layers whose visibility in the workspace can be managed. There are as many layers as scrambling
codes. Each layer corresponds to the area where the corresponding scrambling code has interference. A layer corresponding to areas where more than one scrambling code interferes is also available.
8.3
Monte Carlo Simulations

The simulation process is divided into two steps.
Generating a realistic user distribution as explained in "Generating a Realistic User Distribution" on page 377.
Atoll generates user distributions as part of the Monte Carlo algorithm based on traffic data. The resulting user
distribution complies with the traffic database and maps selected when creating simulations.
8.3.1
Dynamic channel allocation and power control as explained under "Power Control Simulation" on page 383.

During each simulation, Atoll performs two random trials. The first random trial generates the number of users and their
activity status as explained in the following sections depending on the type of traffic input.
"Simulations Based on Raster and Vector Traffic Maps" on page 378.

"Simulations Based on Traffic Map Based on Transmitters and Services" on page 381.
Notes:
Atoll follows a Poisson distribution to determine the total number of users attempting a
connection in each simulation. In order for Atoll to use a constant total number of users
attempting a connection, the following lines must be added to the Atoll.ini file:
[CDMA]
RandomTotalUsers=0
Each user is randomly assigned a service, a terminal, and a mobility type. The activity status is determined based on the
calculations of activity probabilities using the traffic inputs.
The user activity status influences the next step of the simulation, i.e., the interference in the network. Both active and
inactive users use radio resources and generate interference.
Forsk 2009
AT271_TRG_E6
377

Once all the user characteristics have been determined, a second random trial is performed to obtain their geographical
locations weighted according to the clutter classes, and whether they are indoor or outdoor according to the percentage
of indoor users per clutter class defined for the traffic maps.
Atoll also calculates the shadowing margin for each user based on the standard deviations defined for the clutter class of
each user.
In TD-SCDMA networks users accessing packet-switched services can transmit either on uplink or on downlink, but never
on both simultaneously. Users accessing circuit-switched services transmit on both uplink and downlink simultaneously.
Circuit-switched service users, mobiles connected in uplink and downlink both, are modelled in Atoll by two mobiles generated at the same location with one connected on the uplink and the other on the downlink. If one of these two mobiles is
rejected for some reason, the other is also rejected due to the same reason.
8.3.1.1
Simulations Based on Raster and Vector Traffic Maps

a list of user profiles with an associated mobility type and a given density, i.e., number of users of a user profile per km.
Vector traffic maps are based on user profiles. Each polygon or line of the map is assigned a density of users with a given
user profile and mobility type. If the map is composed of points, each point is assigned a number of users with given user
profile and mobility type.
The user profile models the behaviour of the different user categories. Each user profile contains a list of services and their
associated parameters describing how these services are accessed by the user.
The number of users of each user profile is calculated from the surface area (SEnv) of each environment class map (or
each polygon) and the user profile density (DUP).
N Users = S Env D UP
Notes:
In case of vector traffic map composed of lines, the number of users per user profile is
calculated from the line length (L) and the user profile density (DUP) (users per km):
The number of users is an input when the vector traffic map is composed of points.
N Users = L D UP
At any given instant, Atoll calculates the probability for a user being active in the uplink and in the downlink according to
the service usage characteristics described in the user profiles, i.e., the number of voice calls or data sessions, the average
duration of each voice call, or the volumes of the data exchanged in the uplink and the downlink in each data session.
8.3.1.1.1

User profile parameters for circuit switched services are:
The user terminal equipment used for the service (from the Terminals table),
The average number of calls per hour N Call ,
The average duration of a call (seconds) D Call .
N Call D Call
Calculation of the service usage duration per hour ( p 0 : probability of a connection): p 0 = -------------------------------3600
Calculation of the number of users trying to access the service i ( n i ): n i = N Users p 0
The activity status of each user depends on the activity periods during the connection, i.e., the uplink and downlink activity
UL
DL
factors defined for the circuit switched service i, f Act and f Act .
UL
DL
Probability of being inactive: p Inactive = 1 f Act 1 f Act

UL
UL
DL
DL
DL
UL
Probability of being active on UL: p Active = f Act 1 f Act

Probability of being active on DL: p Active = f Act 1 f Act
UL + DL
Probability of being active both on UL and DL: p Active
UL
DL
= f Act f Act

Number of inactive users: n i Inactive = n i p Inactive
UL
UL
Number of users active in the uplink: n i Active = n i p Active

DL
DL
Number of users active in the downlink: n i Active = n i p Active

UL + DL
UL + DL
Number of users active in the uplink and downlink both: n i Active = n i p Active
378
AT271_TRG_E6
Forsk 2009

Therefore, a connected user can be either active on both links, inactive on both links, active on UL only, or active on DL
only.
8.3.1.1.2

User profile parameters for packet switched services are:
The user terminal equipment used for the service (from the Terminals table),
The average number of packet sessions per hour N Sess ,
The volume (in kBytes) which is transferred on the downlink V
DL
and the uplink V
UL
during a session.
A packet session consists of several packet calls separated by a reading time. Each packet call is defined by its size and
may be divided in packets of fixed size (1500 Bytes) separated by an inter-packet arrival time.
Figure 8.1: Description of a Packet Session

Calculation of the average packet call size (kBytes):
UL
DL
UL
DL
V
V
S PacketCall = ------------------------------------------- and S PacketCall = ------------------------------------------UL
UL
DL
DL
N PacketCall f Eff
N PacketCall f Eff
UL
DL
In case of HSDPA services, f Eff and f Eff are the uplink and downlink A-DPCH activity factors, respectively.
Calculation of the average number of packets per packet call:
UL
DL
S PacketCall
S PacketCall
UL
- + 1 and N DL
- + 1
N Packet = Int -----------------------------------Packet = Int -----------------------------------UL
S Packet 1024
S DL
Packet 1024
Note:
1 kBytes = 1024 Bytes.
Calculation of the average duration of inactivity within a packet call (c):

UL
UL
DL
DL
N Packet 1 T Packet
N Packet 1 T Packet
UL
DL
D Inactivity PacketCall = --------------------------------------------------------------- and D Inactivity PacketCall = --------------------------------------------------------------1000
1000
Calculation of the average duration of inactivity in a session (s):
UL
UL
UL
DL
DL
DL
D Inactivity Session = N PacketCall D Inactivity PacketCall and

D Inactivity Session = N PacketCall D Inactivity PacketCall
Calculation of the average duration of activity in a session (s):
UL
UL
DL
DL
N Packet S Packet 8
UL
UL
- and
D Activity Session = N PacketCall -----------------------------------------------------UL
R Nom 1000
N Packet S Packet 8
DL
DL
D Activity Session = N PacketCall -----------------------------------------------------DL
R Nom 1000
Therefore, the average duration of a connection in the session s is:
UL
UL
UL
DL
DL
DL
D Connection = D Activity Session + D Inactivity Session and D Connection = D Activity Session + D Inactivity Session
Calculation of the service usage duration per hour (probability of a connection):
Forsk 2009
AT271_TRG_E6
379

N Sess
N Sess
UL
UL
DL
DL
p Connection = --------------- D Connection and p Connection = --------------- D Connection
3600
3600
Calculation of the probability of being connected:
UL
DL
p Connected = 1 1 p Connection 1 p Connection

Therefore, the number of users trying to access the service j is:
n j = N Users p Connected
As Figure 8.1: on page 379 shows, there can be three possible cases when a user is connected:
a. 1st case: At a given time, packets are downloaded and uploaded.
UL
DL
p Connection p Connection
UL + DL
The probability of being connected is: p Connected = ---------------------------------------------------------------p Connected
b. 2nd case: At a given time, packet are uploaded only.
UL
DL
UL
The probability of being connected is: p Connected = -----------------------------------------------------------------------------p Connected
c. 3rd case: At a given time, packet are downloaded only.
DL
UL
DL
The probability of being connected is: p Connected = -----------------------------------------------------------------------------p Connected
Calculation of the probability of being active:
To determine the activity status of each user, the activity periods during the connection are taken into account.
f
UL
UL
DL
D Activity Session
D Activity Session
DL
= -------------------------------------------------------------------------------------------------- and f
= -------------------------------------------------------------------------------------------------UL
UL
DL
DL
D Inactivity Session + D Activity Session
D Inactivity Session + D Activity Session
Therefore, we have:
a. 1st case: At a given time, packets are downloaded and uploaded.
UL
UL
1 f
DL
p Connected
DL
DL
1 f
UL
p Connected
DL
p Connected
The probability of the user being active on UL and inactive on DL: p1 Active = f
The probability of the user being active on DL and inactive on UL: p1 Active = f
UL + DL
The probability of the user being active on both UL and DL: p1 Active
= f
UL
The probability of the user being inactive on both UL and DL: p1 Inactive = 1 f
UL + DL
UL + DL
UL + DL
UL
1 f
DL
UL + DL
p Connected
b. 2nd case: At a given time, packet are uploaded only.

UL
The probability of the user being active on UL and inactive on DL: p2 Active = f
UL
UL
p Connected
UL
UL
p Connected
c. 3rd case: At a given time, packet are downloaded only.

DL
The probability of the user being active on DL and inactive on UL: p1 Active = f
DL
DL
p Connected
DL
DL
p Connected

Number of inactive users on UL and DL: n j Inactive = n j p1 Inactive + p2 Inactive + p3 Inactive
UL
UL
UL
DL
DL
DL
Number of users active on UL and inactive on DL: n j Active = n j p1 Active + p2 Active

Number of users active on DL and inactive on UL: n j Active = n j p1 Active + p3 Active
UL + DL
UL + DL
Number of users active on UL and DL: n j Active = n j p1 Active

Therefore, a connected user can be active on both links, inactive on both links, active on UL only, or active on DL
only.
380
AT271_TRG_E6
Forsk 2009

Notes:
8.3.1.2
The user distribution per service, and the activity status distribution between the users are
average distributions. The service and the activity status of each user are random in each
simulation. Therefore, if you compute several simulations at once, the average number of
users per service and average numbers of inactive, active on UL, active on DL, and active
on UL and DL users, will correspond to calculated distributions. But, if you compare each
simulation, you will observe that the user distribution between services as well as the
activity status distribution between users is different in each simulation.
Simulations Based on Traffic Map Based on Transmitters and Services

Traffic maps based on transmitters and services are also referred to as live traffic maps. Live traffic data from the OMC is
spread over the best server coverage areas of the transmitters included in the traffic map. Throughput demands per service, the numbers of active users per service, or Erlangs per service are assigned to the coverage areas of each transmitter.
For each transmitter TXi and each service S,
Traffic Map Based on Transmitters and Services (Throughputs)

Atoll calculates the number of active users of each service S on UL and DL in the coverage area of TXi as follows:
-
UL
DL
UL
RS
UL
DL
RS
RS
DL
= -------------= -------------and N
for R99 circuit and packet switched services
UL
DL
R Nom
R Nom
DL
RS
= ------------ for HSDPA service
DL
R Avg
DL
and R S
are the uplink and downlink rates for service S in the TXi cell from the traffic map.
Traffic Map Based on Transmitters and Services (# Active Users)

UL
Atoll directly uses the defined N

and N
coverage area using the service S.
DL
values, i.e., the number of active users on UL and DL in the TXi
Traffic Map Based on Transmitters and Services (Erlangs)

UL
DL

and N
values, i.e., the number of active and inactive users on UL and DL
in the TXi coverage area using the service S.
8.3.1.2.1

There can be two cases for circuit-switched service:
1. Number of active users on uplink and downlink (NUL and NDL),
2. Erlangs, i.e, a total number of users (n).
Active Users on Uplink and Downlink

N
UL
and N
DL
UL
DL
values include users active on uplink ( n i Active ), on downlink ( n i Active ) and on both links
UL + DL
( n i Active ). They are calculated as follows:

UL
DL

UL
UL
DL
DL
DL
UL

UL + DL
UL
DL
= f Act f Act
Calculation of the total number of active users, n i Active , attempting to access the circuit switched service
i:
We have: N
UL
UL
UL + DL
= p Active + p Active n i Active and N
DL
DL
UL + DL
= p Active + p Active n i Active
Therefore,
UL
UL + DL
DL
UL + DL
N p Active
N p Active
UL + DL
- -------------------------------------------
n i Active = Min ------------------------------------------UL
UL + DL
UL + DL
p Active + p Active p DL
Active + p Active
Forsk 2009
AT271_TRG_E6
381

UL
UL
n i Active
DL
DL
n i Active
UL
DL
n i Active = N
n i Active = N
UL + DL
UL + DL
And
UL + DL
n i Active = n i Active + n i Active + n i Active

Calculation of the number of inactive users, n i Inactive , for the circuit switched service i
The number of inactive users is calculated from the total number of active users as follows:
n i Active
n i Inactive = ------------------------------ p Inactive
1 p Inactive
Erlangs
n i is the total number of users trying to access the service i. This figure includes both active and inactive users.
They are determined as follows:
UL
DL

UL
UL
DL
DL
DL
UL

UL + DL
UL
DL
= f Act f Act

Number of inactive users: n i Inactive = n i p Inactive
UL
UL
DL
DL
Number of users active on UL: n i Active = n i p Active

Number of users active on DL: n i Active = n i p Active
UL + DL
UL + DL
Number of users active on UL and DL both: n i Active = n i p Active
Therefore, a connected user can have four different activity status: either active on both links, active on UL only,
active on DL only, or inactive.
8.3.1.2.2

Atoll considers all the users as active. Activity probabilities are not calculated.
nj is the total number of users in the TXi cell using the packet switched service j.
N
UL
and N
DL
UL
DL
UL + DL
values include users active on uplink ( n j Active ), on downlink ( n j Active ) and on both links ( n j Active ).

If N
UL
DL
UL + DL
n j Active = N
UL
UL
n j Active = 0
DL
n j Active = N
If N
UL
DL
UL
DL
DL
UL + DL
n j Active = N
DL
DL
n j Active = 0
UL
n j Active = N
UL
Therefore, we have:
382
AT271_TRG_E6
Forsk 2009

UL
DL
UL + DL
n j Active = n j Active + n j Active + n j Active

Note:
8.3.2
The activity status of users is based on an average distribution. The activity status of each
user is random in each simulation. Therefore, if you compute several simulations at once,
the average numbers of inactive, active on UL, active on DL, and active on UL and DL
users, will correspond to calculated distributions. But, if you compare each simulation, you
will observe that the activity status distribution between users is different in each
simulation.
Power Control Simulation

Based on CDMA air interface, a TD-SCDMA network automatically regulates itself by using uplink and downlink power
control in order to minimise interference and maximise capacity. For each user distribution, Atoll simulates these network
regulation mechanisms using an iterative algorithm and calculates network parameters such as traffic power per cell and
per timeslot, mobile terminal power, and handoff status for each terminal.
In each iteration, all the mobiles (R99 and HSDPA service users) selected during generation of the user distribution
attempt to connect to the network one by one. The process is repeated from iteration to iteration and ends when the
network is balanced, i.e., when the convergence criteria on uplink and downlink are satisfied.
The simulation algorithm also models the impact of smart antennas in the power control loop. The influence of smart antennas is taken into account in signal quality calculations. Smart antennas improve the signal quality of each served mobile,
decrease the required powers and the loads of all the surrounding cells. Interference on the downlink and the uplink is
calculated on a per user. Power control is simulated over a sub-frame, i.e., 7 timeslots.
For HSDPA users, uplink and downlink power control is performed on the associated A-DCH bearer before fast link adaptation on downlink. The steps of this algorithm are detailed below.
Figure 8.2: TD-SCDMA Power Control Algorithm
8.3.2.1
Algorithm Initialisation
At the start of each simulation, the system loads for each carrier and timeslot are reset to initial values:
Forsk 2009
Downlink traffic powers of cells P TCH DL are initialised to 0 Watts
AT271_TRG_E6
383

Uplink interference powers received on all the carriers and timeslots I Intra UL and I Extra UL are initialised to 0
Watts (i.e., no connected mobiles)

Term
Uplink required power for mobiles is set to P Min
8.3.2.2
R99 Part of the Algorithm

The algorithm is described for an iteration k. Here, Xk is the value of the variable X at the iteration k. In the algorithm, all
Req
Q UL
Req
and Q DL
thresholds depend on the user mobility, and are defined in the Service and Mobility parameter tables.
All the variables used in the description below are listed in "Definitions and Formulas" on page 365.
The following calculations are made for all R99 and HSDPA mobiles (Mi) using R99 bearers.
8.3.2.2.1

This step is performed for TS0 for each station TXi containing Mi in its calculation area.
The best server for Mi is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power,
or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the P-CCPCH RSCP is calculated for:
the preferred carrier of the service used by Mi, or

The RSCP from a transmitter TXi and a selected carrier ic is given by:
TX i ic
TX i ic
RSCP P CCPCH = P P CCPCH + G
TX i
TX i
Model
Mi
L Path M Shadowing L Body L Indoor + G
Mi
Mi
in dBm
Where,
TX i

TX i

Model
account is selected
L
Mi
is the los of the terminal used by Mi
Mi
L Body is the body loss defined in the service used by Mi

G
G
L
Mi
TX i
TX i
is the receiver gain of the terminal user by Mi

TX i
= L Total DL )
A cell TX i ic is considered the best server of a mobile Mi if it satisfies the following conditions:
TX i ic
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility and RSCP P CCPCH =
Best RSCP TXj jc

P CCPCH .
j = All
The best server is determined once for the whole simulation during the first iteration, i.e., k = 0, because the best server
does not change during the simulation and smart antennas do not influence this step.
Mi is considered unable to connect to the network if no best server has been selected. In this case, Mi is rejected for the
reason P-CCPCH RSCP < Min P-CCPCH RSCP. If Mi has no best server, it is not taken into account in the next steps.
8.3.2.2.2
Dynamic Channel Allocation

The dynamic channel allocation is performed once for the whole simulation during the first iteration, i.e., k = 0. The DCA
controls the mobile admission. Once a mobile has been admitted for a simulation, it remains admitted for the all the iterations unless there are other reasons to reject it (following steps).
The aim of Dynamic Channel Allocation (DCA) is to reduce interference in order to maximise the usage of the radio
resources. In other words, the DCA tries to find the "best carrier" and the "best timeslots", which when allocated to the
mobiles will optimise the load balance between carriers.
384
AT271_TRG_E6
Forsk 2009

If a preferred carrier is defined for the service requested by Mi and if it is available at TX i . BestCarrier TX i M i =
the
carrier preferred for the service. In the case of N-frequency compatible transmitters, Mi can be allocated timeslots over
more than one slave carrier.
Mi is considered unable to connect to the network if no carrier or not enough timeslots have been selected. In this case,
the mobile Mi will be rejected for the reason "RU Saturation". If the carrier and timeslot(s) selected by the DCA do not
satisfy the control of radio resource limits for DL power or UL load, then the mobile will be rejected for the reason "DL Load
Saturation" or "Admission Rejection" respectively.
There are four strategies for the DCA available in Atoll. These strategies are described below one by one.
1. Load
Carrier Selection by Load: The DCA determines the least loaded carrier with enough timeslots to accomodate
the service being used by each mobile Mi. The best carrier for a mobile is the one that is least loaded:
BestCarrier TX i M i = Carrier
Where, X
And, X
DCA
DCA
DCA
DCA
DCA
= X DL
Min X
DCA
TX i ic TS M i
= N Tot DL
if the mobile is connected in the downlink.
TX i ic TS M i
N Tot UL
- X DCA if the mobile is connected in the uplink.
= ----------------------------------------------------TX i ic TS M i
TX i
N Tot UL
+ N0
DCA
X UL
is the load increment given by:

Mi
TX i
Ortho
1 f UL 1 f JD
= -----------------------------------------------------------------------1
1 + ------------Req
Q UL
E
-----b-
N t UL
Proc
= --------------------- is the uplink required signal quality. The uplink processing gain G UL calcuProc
G UL
Req
Req
C
Req
Where Q UL = ----
I UL
lated from the service parameters, if no smart antenna is used by the transmitter in the uplink.
If a smart antenna is used by the transmitter in the uplink, the smart antenna gain is taken into account in calcuReq
lating Q UL .
Notes:
TX i ic TS M i
N Tot UL
N Tot DL
The carrier is the same in the uplink and in the downlink for mobiles accessing circuitswitched services.
TX i ic TS M i
is described in "Uplink Power Control" on page 386.

is described in "Downlink Power Control" on page 387.
Timeslot selection by Load: From the selected carrier, Atoll selects the timeslots which are the least loaded and
have enough resource units for the service being accessed by Mi.
2. Available RUs
Carrier selection by Available RUs: The DCA determines the carrier which has the highest number of available
resource units with enough timeslots to accomodate the service being used by each mobile Mi. The best carrier
for a mobile is the one that has the highest number of resource units:
Max RUs
Timeslot selection by Available RUs: From the selected carrier, Atoll selects the timeslots which have the highest numbers of available resource units.
3. Direction of Arrival
Carrier selection by Direction of Arrival: The DCA determines the direction of arrival of the signal from the
served user Mi and checks whether there is an interfering mobile in the same direction as Mi. Atoll searches for
interfering mobiles within the angle defined by the Angular Step. For example, if you enter an angular step of 15
degrees, Atoll searches for interfering mobiles within 15 degrees to the right and to the left of the served user, and
allocates a different carrier than the ones used by any interfering mobiles found. The best carrier for a mobile is
the one which is not interfered by another mobile in the direction of the mobile Mi.
DoA Mi DoA Mj
In other words, the direction of arrival for the served user Mi should not be the direction of arrival of an interfering
mobile.
Forsk 2009
AT271_TRG_E6
385

Timeslot selection by Direction of Arrival: From the selected carrier, Atoll selects the timeslots which are not
being used by any other mobile Mj located in the same direction as the served user Mi.
4. Sequential
Sequential carrier selection: The DCA allocates carriers to served users Mi in a sequential order.
Sequential timeslot selection: From the selected carrier, Atoll allocates timeslots to served users Mi in a sequential order.
At the end of the DCA, each admitted mobile has an associated carrier and timeslots. In case of N-frequency mode
compatible transmitters, an admitted mobile can have associated timeslots over more than one slave carrier.
8.3.2.2.3

For each mobile Mi, the uplink power control step calculates the uplink power required to satisfy the required quality level
on the traffic channel, which is defined for the service being accessed by Mi.
If the mobile Mi is connected (active or inactive) in the uplink and has a best server TX i ic assigned to it, Atoll calculates
the signal quality on the uplink timeslots allocated to Mi by the DCA:
TX i ic TS M i
E
-----b-
N t TCH UL
TX i ic TS M i
Mi
Mi
TX i ic TS Mi
TX i ic TS Mi
RSCP TCH UL
RSCP TCH UL
Div
C-
- G Proc
- G Div
= --------------------------------------------------------------= --------------------------------------------------------------UL G UL or --UL
TX i ic TS M i
TX i ic TS Mi
I TCH UL
N Tot UL
N Tot UL
Calculation of Uplink Total Noise ( N Tot UL ):

The uplink total noise is calculated for the uplink connection between each mobile Mi and its best server TX i ic .
TX i ic TS Mi
N Tot UL
TX i ic TS Mi
= I Tot UL
TX i
+ N0
Where
TX i ic TS Mi
I Tot UL
i
= RSCP TCH
UL TX i ic TS M i
Mi
Mj
RSCP TCH UL TX i ic TS M i
Mi
M j TX i ic TS Mi
Mj Mi
Mj
Mj
1 RSCP TCH UL TX i ic TS M i +
M j TX i ic TS Mi
Mj Mi
Mj
RSCP TCH UL TX i ic TS M i
M j TX i ic TS Mi
Mi
Mi
Ortho
1 F UL
TX i
1 F JD and = 0
1

Total Noise
The above formula gives the value of I Tot UL for the uplink connection between Mi and TX i ic , taking into account the
interference received from other mobiles, Mj, which are located in the Mi best server coverage area, as well as located in
the coverage areas of other cells. The mobile Mi is the focus, i.e., the mobile that is listened to by the transmitter TX i ic .
The four terms comprising I Tot UL are:
The useful signal for which the received mobile is the focus (Mi).
The intra-cell interference for which the best-server is the same for the received mobile Mj and the focus Mi,
TX i ic .
The intra-cell interference due to distortion in the terminal transmission.
The extra-cell interference for which the best-server for the received mobile Mj is not TX i ic .
Mi
The uplink received signal code power is:

TX i
Mi
Mi
Mi
RSCP TCH UL TX i ic
P Req TX i ic TS M i
k1
TS M i = ----------------------------------------------------------------------Model
LT
Model
Mi
L Path L L L Body L Indoor M Shadowing
= ------------------------------------------------------------------------------------------------------------------------------ and P Req TX i ic TS M i is the uplink required mobile
TX i
Mi
G G
power
calculated
for
the
timeslot
allocated
to
Mi .
If
Mi
is
an
HSDPA
user,
Model
LT
Mi
Mi
P Req TX i ic TS M i = 0.1 P Req TX i ic TS M i
386
AT271_TRG_E6
Forsk 2009

Model
In L T
, G
TX i
SA
= G UL and L
TX i
SA
= L UL are calculated according to the smart antenna modelling method used, for
Mi
P Req TX i ic TS M i , if a smart antenna is available in the uplink. Otherwise, G
TX i
and L
TX i
are read from the main
antenna model.
Notes:
Interference is updated only for active mobiles on the uplink for circuit- and packetswitched services. However, if these mobiles are rejected, they are considered in the
number of rejected mobiles.
M
i
Calculation of Uplink Required Power ( P Req
):
Then Atoll determines the required uplink power by:

E
-----b-
N t TCH UL
Mi
= P Req TX i ic TS M i
------------------------------------------TX i ic TS Mi
k1
E
-----b-
N t TCH UL
Req
Mi
Req
or
Mi
P Req TX i ic
TS M i
Mi
P Req TX i ic
Mi
Mi
C
----
I TCH UL
TS M i
----------------------------------------TX i ic TS M i
k1
C
----
I TCH UL
Mi
Mi
And if P Req TX i ic TS M i P Min then P Req TX i ic TS M i = P Min

Mi
Mi
If P Req TX i ic TS M i P Max
Mi
P Req TX i ic
Mi
then the mobile Mi is rejected for the reason "Pmob > PmobMax", and
TS M i is set to 0.
Mi
P Min and P Max are set in the properties of the terminal used by the mobile Mi.
Where
TX i ic TS M i
RSCP TCH UL
TX i
Mi
k1
= ----------------------------------------------------------------------Model
LT
Mi
Mi
Model
Mi
= ------------------------------------------------------------------------------------------------------------------------------ and P Req TX i ic TS M i
TX i
Mi
G G
power for iteration k - 1 transmitted on the timeslot allocated to Mi.
Model
LT
Model
In L T
, G
TX i
SA
= G UL and L
TX i
k1
is the uplink required mobile
SA
Mi
P Req TX i ic TS M i , if a smart antenna is available in the uplink. Otherwise, G
TX i
and L
TX i
antenna model.
Note:
8.3.2.2.4
The uplink required powers for mobiles inactive in the uplink accessing circuit- or packetswitched services are calculated for information only. However, if these mobiles are
rejected, they are considered in the number of rejected mobiles.

For each mobile Mi, the downlink power control step calculates the downlink power for the best server TX i ic required
to satisfy the required quality level on the traffic channel, which is defined for the service being accessed by Mi.
If the mobile Mi is connected (active or inactive) in the downlink and has a best server TX i ic assigned to it, Atoll calculates the signal quality on the uplink timeslots allocated to Mi by the DCA:
TX i ic TS M i
E
-----b-
N t TCH DL
TX i
TX i ic TS M i
TX i
TX i ic TS M i
TX i ic TS M i
RSCP TCH DL
RSCP TCH DL
Div
C-
- G Proc
- G Div
= -----------------------------------------------------------------= -----------------------------------------------------------------DL G DL or --DL
TX i ic TS M i
TX i ic TS M i
I TCH DL
N Tot DL
N Tot DL
Calculation of Downlink Total Noise ( N Tot DL ):

The downlink total noise is calculated for the downlink connection between each mobile Mi and its best server TX i ic .
TX i ic TS M i
N Tot DL
Forsk 2009
TX i ic TS M i
= I Tot DL
Mi
+ I IC DL ic jc + I MM M i M j + N 0
AT271_TRG_E6
387

Where
TX i ic TS Mi
TX i ic TS Mi
= RSCP Tot DL
I Tot DL
Mi
TX i
TX i ic TS Mi
Mj
RSCP Tot DL
TX i
M j TX i ic TS Mi
Mj Mi
TX i
TX i ic TS Mi
RSCP Tot DL
Mj +
M j TX i ic TS M i
Mj Mi
TX j ic TS M i
RSCP Tot DL
Mj
M j TX i ic TS M i
The four terms comprising I Tot DL are:
The useful signal for which the received mobile is the focus (Mi).
The intra-cell interference for which the best-server is the same for the received mobile Mj and the focus Mi,
TX i ic .
The intra-cell interference due to distortion in the transmitter.
The extra-cell interference for which the best-server for the received mobile Mj is not TX i ic .
TX j jc TS Mi
RSCP Tot DL
Mi
All TX j
I IC DL ic jc = --------------------------------------------------------------------------F IRF ic jc
TX i
TX i
Ortho
1 F DL
Mi
1 F JD and = 0
1

Total Noise
I IC DL ic jc is the inter-carrier interference from a carrier jc to another carrier ic on the downlink, which is reduced by
the interference reduction factor F IRF ic jc defined for the pair (ic, jc).
TX i ic TS Mi
RSCP Tot DL
With
TX i ic TS M i
= RSCP TCH DL
TX i ic TS M i
RSCP TCH DL
Model
LT
TX i
TX i ic TS Mi
+ RSCP OCCH
TX i ic TS M i
TX i ic TS M i
P TCH DL
TX i ic TS Mi
P OCCH
k1
= ---------------------------------------------- and RSCP OCCH
= ----------------------------------Model
Model
LT
LT
Mi
Mi
Model
TX i ic TS M i
= ------------------------------------------------------------------------------------------------------------------------------ and P TCH DL
TX i
Mi
G G
k1
is the downlink traffic power transmitTX i ic TS M i
ted on the timeslot allocated to Mi during the iteration k - 1. If Mi is an HSDPA user, P TCH DL
Model
In L T
, G
TX i
TX i ic TS M i
P TCH DL
SA
= G DL and L
TX i
TX i ic TS M i
= 0.1 P TCH DL
SA
= L DL are calculated according to the smart antenna modelling method used, for
TX i ic TS M i
only and not for P OCCH
, if a smart antenna is available in the downlink. Otherwise, G
TX i
and L
TX i
are read from the main antenna model.

Mj
RSCPTCH UL Mi
M
j
I MM M i M j = --------------------------------------------------------- is the interference from each mobile Mj transmitting in the uplink on the same timesF IRF ic jc
lots as those on which the mobile Mi is receiving in the downlink. Mj can interfere Mi directly if and only if:
Mi Mj
) is less than the Max Distance between interfering mobiles defined by

The distance between Mi and Mj ( d
the user when starting the simulation, and
The downlink timeslot of Mi (TSMi) is the same as the uplink timeslot of Mj, (TSMj).
The interference received from the mobile Mj at the mobile Mi is calculated using either the free-space propagation model
or the Xia model.
Mj
Mj
P TCH UL
RSCP TCH UL M i = -----------------------L MM
388
AT271_TRG_E6
Forsk 2009
32.4 + 20 Log F Avg + 20 Log d

L MM =
49 + 30 Log F Avg + 40 Log d
If d
If d
Mi Mj
Mi Mj
3m
with F Avg being the average frequency in MHz of
3m
the frequency band used by the best server of the mobile Mi, and d is the distance between the mobiles Mi and Mj in km.
TX ic TS M i
Calculation of Downlink Required Power ( P Reqi
):
Then Atoll determines the required downlink power by:

E
-----b-
N t TCH DL
------------------------------------------TX i ic TS M i
E
-----b-
N t TCH DL
Req
TX i ic TS Mi
P Req
k
TX i ic TS M i
P Req
k1
Req
or
TX i ic TS Mi
P Req
k
TX i ic TS M i
And if P Req
TX i ic TS M i
If P Req
TX i ic TS M i
P Req
k1
C
----
I TCH DL
----------------------------------------TX i ic TS M i
C
----
I TCH DL
TX i ic TS M i
Min
P TCH DL Service then P Req
Min
= P TCH DL Service
TX i ic TS M i
Max
P TCH DL Service then the mobile Mi is rejected for the reason "Ptch > PtchMax", and P Req
is set to 0.
Min
Max
P TCH DL Service and P TCH DL Service are set in the properties of the R99 bearer associated with the service used
by the mobile Mi.
TX i ic TS M i
Otherwise, the downlink traffic power is incremented P TCH DL
TX i ic TS M i
= P TCH DL
TX i ic TS M i
+ P Req
For each mobile, Atoll also calculates the downlink traffic power for the different values of the Angular Step Step .
TX i ic TS Mi
RSCP TCH DL
Where
TX i ic TS Mi
Step
= RSCP TCH DL
TX i ic TS M i
RSCP TCH DL
TX i
SA
G DL
---------------------------SA
L DL Step
TX i ic TS M i
P Req
k1
= ---------------------------------------------Model
LT
Mi
Mi
Model
TX i ic TS M i
= ------------------------------------------------------------------------------------------------------------------------------ and P Req
TX i
Mi
G G
tion k - 1 transmitted on the timeslot allocated to Mi.
Model
LT
Model
In L T
, G
TX i
TX i ic TS M i
P Req
SA
= G DL and L
TX i
k1
is the downlink traffic power for itera-
SA
, if a smart antenna is available in the downlink. Otherwise, G
TX i
and L
TX i
are read from the main antenna
model.
Note:
8.3.2.2.5
The downlink power for mobiles inactive in the downlink accessing circuit- or packetswitched services are calculated for information only.
Uplink Signals Update

This step uses the uplink terminal powers calculated for each timeslot allocated to the mobiles. The Dynamic Channel Allocation allocates timeslots and carriers to all the connected and active mobiles. The Dynamic Channel Allocation is
performed once only, during the first iteration, and the timeslot and carrier allocation remains the same for all the following
iterations of a simulation.
This step updates the received signals for all the mobiles Mi interfered in the uplink by the uplink connection between interfering mobiles Mj and their best servers TX j ic , only if TX j ic contain Mi in their coverage areas. TX i ic is the interfered receiver and Mi is the focus, i.e., the mobile that is listened to by the transmitter TX i ic .
For each mobile Mi interfered by Mj in the uplink by the connection between Mj and TX j ic , Atoll updates
TX i ic TS Mi
RSCP TCH UL
Forsk 2009
AT271_TRG_E6
389
8.3.2.2.6
Downlink Signals Update

For the first iteration, i.e., k = 0, the downlink traffic powers for all the downlink timeslots are set to 0 Watts. Therefore, for
the first iteration, this step is performed for any downlink timeslot for each mobile Mi that is connected and active.
However, for the following iterations, the downlink signals update step uses the actual downlink traffic powers calculated
for each timeslot and the actual timeslots allocated to the mobiles. The Dynamic Channel Allocation allocates timeslots
and carriers to all the connected and active mobiles. The Dynamic Channel Allocation is performed once only during the
first iteration and the timeslot and carrier allocation remains the same for all the following iterations of a simulation.
Therefore, this step is performed for any downlink timeslot for each mobile Mi that is connected and active for the first iteration, and this step is performed for all the downlink timeslots allocated to the mobile Mi on which it is connected and active,
for the following iterations since the DCA has been performed.
This step updates the received signals for all the mobiles in the TX i ic coverage area which are interfered in the downlink
by the connection between TX i ic and Mi.
TX i ic TS M i
For each mobile interfered by Mi, Atoll updates RSCP TCH DL
Where TX i ic is the transmitter considered and Mi is the focus, i.e., the mobile that is the target for TX i ic .
8.3.2.2.7
Control of Radio Resource Limits (Downlink Traffic Power and Uplink Load)
This step checks whether the downlink traffic powers of the downlink timeslots and the uplink loads of the uplink timeslots
of all the cells satisfy the conditions defined globally or per cell and timeslot.
Downlink Power Control:

Atoll verifies that the total R99 power transmitted by any cell on any timeslot does not exceed the effective maximum cell
power per timeslot. The effective maximum cell traffic power per timeslot is calculated as:
TX i ic TS M i
TX i ic TS M i
P Max DL Eff = P Max DL

TX i ic TS M i
Where P Max DL
%P Max DL
is the maximum cell power per timeslot defined per cell, and %P Max DL is the maximum allowed
downlink load either taken from the properties of each cell or from the simulation properties if a global value is defined.
For each transmitter TXi, carrier ic, and downlink timeslot TS M ,
i
TX i ic TS M i
P R99 DL
TX i ic TS M i
= P TCH DL
TX i ic TS M i
If P R99 DL
TX i ic TS M i
+ P OCCH
TX i ic TS M i
P Max DL Eff the mobile with the lowest service priority is rejected for the reason "DL Load Saturation".
Uplink Load Control:

Atoll verifies that the uplink load of any cell on any timeslot does not exceed the maximum uplink cell load allowed per
timeslot.
TX i ic TS M i
The maximum allowed uplink cell load, X Max UL
, is either taken from the properties of each cell or from the simulation
properties if a global value is defined.

For each transmitter TXi, carrier ic, and uplink timeslot TS M ,
i
TX i ic TS M i
If X UL
TX i ic TS M i
X Max UL
the mobile with the lowest service priority is rejected for the reason "UL Load Saturation".
The uplink load is given by:

TX i ic TS M i
X UL
TX i ic TS Mi
N Tot UL
= ----------------------------------------------------- if no smart antenna is used by the transmitter in the uplink.
TX i ic TS M i
TX i
N Tot UL
+ N0
If a smart antenna is used by the transmitter in the uplink, the smart antenna gain is taken into account in the calculation
of uplink load.
8.3.2.3
HSDPA Part of the Algorithm

The following calculations are made for all HSDPA mobiles (Mi).
8.3.2.3.1
HSDPA Power Allocation

TX i ic
The total transmitted power of the cell ( P Tot DL ) is the sum of the R99 transmitted power and the HSDPA powers.
TX i ic
TX i ic
TX i ic
P Tot DL = P R99 DL + P HR
390
TX i ic
TX i ic
+ P HS SCCH + P HS PDSCH
AT271_TRG_E6
Forsk 2009

The HSDPA powers, i.e., the HS-SCCH and HS-PDSCH powers are calculated as follows:
HS-SCCH Power:
HS-SCCH channels are transmitted on DL traffic timeslots. The maximum number of supported HS-SCCH channels is defined per cell. Power can be allocated to HS-SCCH statically or dynamically:
-
Static Allocation
The static HS-SCCH power is defined in the properties of the HSDPA cell.
Dynamic Allocation
HS-SCCH
power
TX i ic
TX i ic
is
calculated
TX i ic
Req
E
-----c-
= Q HS SCCH Mobility
N t HS SCCH
for
so
that
P HS SCCH P Available HS SCCH .

TX i ic
TX i ic
TX i ic
TX i ic
Where P Available HS SCCH = P Max DL Eff P R99 DL P HR

TX i ic
TX i ic
is the power available for HS-SCCH in the
TX i ic
cell TX i ic , and P R99 DL = P TCH DL + P OCCH .

The
effective
TX i ic
P Max DL Eff
maximum
TX i ic
P Max DL
cell
%P Max DL .
traffic
power
TX i ic
P Max DL
per
timeslot
is
calculated
as:
is the maximum power defined per cell, and %P Max DL
is the maximum allowed downlink load either taken from the properties of each cell or from the simulation properties if a global value is defined.
TX i ic
TX i ic
P HS SCCH
TX i ic
TX i
TX i ic
E
-----c-
N Tot DL RSCP HS SCCH
N t HS SCCH
- L Model
= -----------------------------------------------------------------------------------------------------------------------------------------------T
TX i
TX i ic
Where N Tot DL is the downlink total noise calculated in "Downlink Power Control" on page 387,
TX i
TX i
Ortho
1 F DL
Mi
1 F JD and = 0
1
TX i
Mi
Mi

Total Noise
Model
TX i ic
= ------------------------------------------------------------------------------------------------------------------------------ and P HS SCCH is the HS-SCCH power calculatTX i
Mi
G G
ed for the timeslots allocated to Mi.
Model
LT
Model
In L T
, G
TX i
SA
= G DL and L
TX i
SA
= L DL are calculated according to the smart antenna modelling method
TX i ic
used, for P HS SCCH , if a smart antenna is available in the downlink. Otherwise, G
TX i
and L
TX i
are read from
the main antenna model.
HS-PDSCH Power:
HS-PDSCH channels are transmitted on DL traffic timeslots. Power can be allocated to HS-PDSCH statically or
dynamically:
-
Static Allocation
The static HS-PDSCH power is defined in the properties of the HSDPA cell.
Dynamic Allocation
HS-PDSCH power is calculated as follows:
TX i ic
TX i ic
TX i ic
TX i ic
P HS PDSCH = P Max DL Eff P R99 DL P HR

TX i ic
TX i ic
TX i ic
P HS SCCH
TX i ic
Where P R99 DL = P TCH DL + P OCCH . The effective maximum cell traffic power per timeslot is calculated as:
TX i ic
TX i ic
TX i ic
P Max DL Eff = P Max DL %P Max DL . P Max DL is the maximum power defined per cell, and %P Max DL
is the maximum allowed downlink load either taken from the properties of each cell or from the simulation properties if a global value is defined.
The HS-SICH power is calculated as follows:
HS-SICH Power:
HS-SICH channels can be transmitted on any UL traffic timeslot. The maximum number of supported HS-SICH
channels is defined per cell. Power can be allocated to HS-SICH statically or dynamically:
-
Static Allocation
The static HS-SICH power is defined in the properties of the terminal used by the HSDPA mobile Mi.
Forsk 2009
AT271_TRG_E6
391

-
Dynamic Allocation
TX i ic
Ec
Mi
TX i ic
Req
= Q HS SICH Mobility so that P HS SICH P Max HS SICH
HS-SICH power is calculated for ------
N t HS SICH
Mi
Mi
and P HS SICH P Max HS SICH .

TX i ic
Mi
P HS SICH
Where
Mi
Mi
TX i ic
Mi
E
-----c-
N Tot UL RSCP HS SICH
N t HS SICH
- L Model
= -----------------------------------------------------------------------------------------------------------------------------------------T
Mi
TX i ic
N Tot UL
Mi
is the uplink total noise calculated in "Uplink Power Control" on page 386,
Ortho
1 F UL
TX i
1 F JD and = 0
1
TX i
Mi
Mi

Total Noise
Model
Mi
= ------------------------------------------------------------------------------------------------------------------------------ and P HS SICH is the HS-SICH power calculated
TX i
Mi
G G
for the timeslots allocated to Mi.
Model
LT
Model
In L T
, G
TX i
SA
= G UL and L
TX i
SA
= L UL are calculated according to the smart antenna modelling method
Mi
used, for P HS SICH , if a smart antenna is available in the uplink. Otherwise, G
TX i
and L
TX i
are read from
the main antenna model.

TX i ic
The total transmitted power of the cell ( P Tot DL ) is the sum of the R99 transmitted power and the HSDPA powers.
TX i ic
TX i ic
TX i ic
P Tot DL = P R99 DL + P HR
8.3.2.3.2
TX i ic
TX i ic
+ P HS SCCH + P HS PDSCH
Connection Status and Number of HSDPA Users

HSDPA users cannot receive HS-SCCH and HS-PDSCH powers simultaneously. HS-PDSCH arrives 3 timeslots after the
HS-SCCH. HS-SICH is 9 timeslots afte the HS-PDSCH. Atoll assumes that an active HSDPA user has the same probability of receiving HS-SCCH and HS-PDSCH, and transmitting HS-SICH because their occurence is equally likely. Therefore, each HSDPA user is assigned a sub-connection status randomly. The sub-connection status can be:
HS-SCCH: HSDPA mobile that is receiving HS-SCCH power

HS-PDSCH: HSDPA mobile that is receiving traffic power
HS-SICH: HSDPA mobile that is transmitting HS-SICH power
The number of active HSDPA users belonging to each sub-connection status is 1/3rd of the total number of active HSDPA
users.
n HS SCCH is the maximum number of HS-SCCH channels and n HS SICH is the maximum number of HS-SICH channels
that the cell can manage. Each HSDPA user consumes one HS-SCCH and HS-SICH channels. Therefore, at a given
instance, the number of connected HSDPA users cannot exceed the number of HS-SCCH and HS-SICH channels per
cell. The maximum number of HSDPA users ( n Max ) corresponds to the maximum number of HSDPA users that the cell
can support.
8.3.2.3.3
HSDPA Admission Control

HS-SCCH
HS-SCCH admission control is performed for active HSDPA users connected to A-DCH bearers on the downlink and
having an HS-SCCH sub-connection status. Each cell is able to manage a maximum number of HS-SCCH channels,
n HS SCCH . During the R99 part, the DCA provides a DL timeslot with one SF16 resource unit that has the downlink Ec/
Nt higher than the required quality. If no cell with such a resource unit is available, the user is rejected.
HS-SICH
HS-SICH admission control is performed for active HSDPA users connected to A-DCH bearers on the uplink and having
an HS-SICH sub-connection status. Each cell is able to manage a maximum number of HS-SICH channels, n HS SICH .
During the R99 part, the DCA provides an UL timeslot with one SF16 resource unit that has the uplink Ec/Nt higher than
the required quality. If no cell with such a resource unit is available, the user is rejected.
392
AT271_TRG_E6
Forsk 2009
HS-PDSCH
Scheduling is performed for active HSDPA users connected to A-DCH bearers on the downlink and having an HS-PDSCH
sub-connection status. The scheduling is performed as follows:
1. Each HS-PDSCH user is considered as the only served user. The scheduler allocates the best available HSDPA
bearer to each user. The best available HSDPA bearer is selected depending on the users Ec/Nt. If no bearer can
be allocated due to low Ec/Nt, the user is rejected for the reason "HSDPA Scheduler Saturation".
The required HS-PDSCH Ec/Nt value is read from receiver equipment properties. For each bearer, Atoll checks
that the Ec/Nt reaches the quality target. HS-PDSCH Ec/Nt is calculated by taking into account all intra and extra
cells interferences.
2. The scheduler sorts the HS-PDSCH users to whom bearers have been assigned in the order of decreasing RLC
peak rates. If two users have the same bearer, the user with the higher Ec/Nt has the higher rank.
3. The scheduler considers the group of HS-PDSCH users to whom bearers, HS-SCCH, and HS-SICH have been
assigned. The number of HS-PDSCH users cannot exceed the maximum number of HSDPA users ( n Max ) supported by the cell. If there are enough HSDPA power and resource units available in order to obtain a HSDPA
bearer, the users will be connected. Otherwise, they will be delayed and their connection status will be HSDPA
Delayed.
4. Other HS-PDSCH users will be rejected for the reason "HSDPA Scheduler Saturation".
For N-frequency mode compatible transmitters, the resource units available in the master and slave carriers can be
shared, i.e., a mobile can be connected to timeslots belonging more than one carrier.
8.3.2.3.4
HSDPA Dynamic Channel Allocation

For each mobile connected to the A-DPCH bearer:
1. Atoll selects the HSDPA bearers that match to the mobile terminal and UE category parameters.
2. For each bearer supported by a mobile:
a. The scheduler searches for the best collection of "n" ordered timeslots that can provide enough resource units
to support the service, and whose Ec/Nt is better than the minimum required and enough to reach the bearers
resource unit requirements. The best is determined by applying the R99 Dynamic Channel Allocation algorithm.
b. The scheduler calculates the HS-PDSCH Ec/Nt for each timeslot of the best collection. The Ec/Nt value associated with the mobile-bearer pair is the worst one of all selected timeslots.
c. If the scheduler is unable to find a satisfactory timeslot collection, the bearer is removed from the list of supported bearers.
3. The mobile is connected to the supported bearer having the highest RLC peak rate. If two bearers have the same
RLC peak rate, the best one is the one with the highest Ec/Nt.
8.3.2.3.5
Ressource Unit Saturation

For each time slot, a minimum and maximum number of resource units for HSDPA users are defined in the cell properties.
Atoll dynamically allocates the required number of codes respecting these limitations. The minimum number of HSDPA
codes is excluded from the set of codes available for R99 users. The scheduler checks if enough codes are available for
the selected HSDPA bearer (taking into account the maximum number of HSDPA codes). If not, the scheduler allocates
a lower HSDPA bearer which needs fewer codes. If there are no more resource units available for the lowest HSDPA
bearer, the user will be delayed or rejected.
8.3.2.4
Convergence Criteria
The convergence criteria are evaluated for each iteration and can be written as follows:
Max
TX i ic TS M i
DL = Int
P
100
All TX i Err
TX i ic TS M i
TX i ic TS M i
N Tot UL
Max N Tot UL
k
k 1 100
UL = Int
---------------------------------------------------------------------------------------------TX i ic TS M i
All TX i
N
Tot UL
k
TX i ic TS M i
Where, P Err
TX i ic TS M i
P Err
is given by:
TX i ic TS Mi
TX i ic TS M i
P Rec
P Rec
Step k
Step k 1
=
----------------------------------------------------------------------------------------------------------------------- with smart antennas.
TX
ic
TS
Mi
0 Step 360
i
P Rec
Max
Step k
TX i ic TS M i
P Err
TX i ic TS Mi
TX i ic TS M i
P Rec
P Rec
k
k 1 without smart antennas.
= -------------------------------------------------------------------------------------------TX i ic TS Mi
P Rec
k
Forsk 2009
AT271_TRG_E6
393

Atoll stops the simulations in the following cases:
Convergence: Between two successive iterations, DL and UL are less than or equal to their respective thresh-
olds (defined when creating a simulation).

Example: Let us assume that the maximum number of iterations is 100, and the UL and DL convergence thresholds are set to 5 %. If DL 5 and UL 5 between the 4th and the 5th iteration, Atoll stops the algorithm after
the 5th iteration. The simulation has converged.
Divergence: After 30 iterations, DL and/or UL are still higher than their respective thresholds and from the 30th
iteration, DL and/or UL do not decrease during the next 15 successive iterations.

Examples: Let us assume that the maximum number of iterations is 100, and the UL and DL convergence thresholds are set to 5 %.
a. After the 30th iteration, DL and/or DL equal 100 and do not decrease during the next 15 successive iterations. Atoll stops the algorithm at the 46th iteration. The simulation has not converged.
b. After the 30th iteration, DL and/or UL equal 80, they start decreasing slowly until the 40th iteration (without
going under the thresholds) and then, do not change during 15 successive iterations. Atoll stops the algorithm
at the 56th iteration without converging.
Last Iteration: If DL and/or UL are still much higher than their respective thresholds after the last iteration, the
simulation has not converged. If DL and UL are lower than their respective thresholds, the simulation has
reached convergence.
8.4
TD-SCDMA Prediction Studies

For each TBC transmitter, TXi, Atoll determines the value of the selected parameter on each studied pixel inside the TXi
calculation area. Each pixel within the TXi calculation area is considered a probe receiver.
The study conditions to determine the service area of each TBC transmitter
The display settings to for colouring the covered pixels
Atoll uses the parameters entered in the Condition tab of the coverage study properties dialogue to determine pixels
covered by the each transmitter. Coverage prediction display resolution is independent of the path loss matrix and
geographic data resolutions, and can be different for each coverage prediction. Coverage predictions are calculated using
bilinear interpolation of multi-resolution path loss matrices (similar to the evaluation of site altitudes).
8.4.1
P-CCPCH Reception Analysis (Eb/Nt) or (C/I)

Eb
C
These coverage predictions calculate and display the Eb/Nt or C/I on the P-CCPCH, ------
or ----
. The
N t P CCPCH
I P CCPCH
coverage predictions are calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS0. The
best servers for the coverage predictions are determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the
coverage predictions are calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there
will not be any pixels covered by this transmitter. If you perform these coverage predictions for all the carriers, Atoll calculates the Eb/Nt or C/I considering:

The
pixels
in
TX i ic
the
TX i ic
coverage
area
where
TX i ic
and
TX i ic
C
Req
Req
E
-----b-
Q P CCPCH or ----
Q P CCPCH are covered and coloured according to the selected display
N t P CCPCH
I P CCPCH
option.
TX i
TX i ic
TX
TX ic
i
i
TX i ic
E b TX i ic
RSCP P CCPCH
RSCP P CCPCH
C
- G Proc
----
Where ------
and
= ----------------------------------------------------------=
----------------------------------------------------------P
CCPCH
TX i ic
N t P CCPCH
I P CCPCH
TX i ic
N Tot DL
N Tot DL
TX i ic
TX i ic
P P CCPCH
RSCP P CCPCH = ---------------------------LT
The downlink total noise is calculated as follows:

TX i ic
TX i ic
TX i ic
Term
N Tot DL = I Intra DL + I Extra DL + I IC DL ic jc + N 0
394
AT271_TRG_E6
Forsk 2009

Where
TX i ic
TX i ic
I Intra DL = RSCP P CCPCH

With
TX i
TX i ic
TX i
Ortho
1 F DL
TX i
TX i ic
+ RSCP OCCH TS0

Term
1 F JD
TX j ic
TX i
and = 0
1

Total Noise
TX j ic
I Extra DL =
ji
TX j jc
TX j jc

TX
j
I IC DL ic jc = ----------------------------------------------------------------------------------------------------------F IRF ic jc
TX i ic
TX i ic
P OCCH TS0
RSCP OCCH TS0 = ------------------------------LT
TX i
Eb Nt
Term
L Path L L
L Body L Indoor M Shadowing
L T = ------------------------------------------------------------------------------------------------------------------------------------TX i
Term
G G
TX i
Term
and N 0
are defined in "Definitions and Formulas" on page 365.
Coverage Display
Eb/Nt or C/I (dB)

Atoll calculates the Eb/Nt or C/I on each pixel of the TX i ic best server coverage area. The pixel colour depends
on the Eb/Nt or C/I level. Coverage consists of several independent layers whose visibility in the workspace can
be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties). For each
TX ic
TX ic
i
i
Eb
C
layer, a pixel is covered if ------
Threshold or ----
Threshold . Each layer is assigned a
N t P CCPCH
I P CCPCH
Eb/Nt Margin or C/I Margin (dB)

Atoll calculates the Eb/Nt or C/I margin on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the Eb/Nt or C/I margin value. Coverage consists of several independent layers whose visibility in the
workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties).
For
each
layer,
pixel
is
covered
if
TX i ic
Req
Eb Nt
E
-----b-
Q P CCPCH M P CCPCH
N t P CCPCH
or
TX i ic
Req
CI
C
----
Q P CCPCH M P CCPCH . Each layer is assigned a colour and displayed with intersections
I P CCPCH
between layers.

Atoll calculates the cell edge coverage probability on each pixel of the TX i ic best server coverage area. The
pixel colour depends on the cell edge coverage probability value. Coverage consists of several independent layers
whose visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display
E b TX i ic
tab (Prediction properties). For each layer, a pixel is covered if ------
N t P CCPCH
TX i ic
C
----
I P CCPCH
Req
Q P CCPCH or
CECP
Req
Q P CCPCH . Each layer is assigned a colour and displayed with intersections between
CECP
layers.
8.4.2
DwPCH Reception Analysis (C/I)

C
This coverage prediction calculates and displays the C/I on the DwPCH, ----
. The coverage prediction is calculated
I DwPCH
for a given set of a terminal type, a mobility type, a service, a carrier, and for DwPTS. The best server for the coverage
Forsk 2009
AT271_TRG_E6
395

prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power, or from the
master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage prediction is calculated for
the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any pixels covered by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the C/I considering:

TX i ic
TX i ic
C
Req
The pixels in the TX i ic coverage area where RSCP P CCPCH TAdd P CCPCH Mobility and ----
Q DwPCH
I DwPCH
are covered and coloured according to the selected display option.

TX i ic
TX i
TX i ic
RSCP DwPCH
C
Where ----
= -------------------------------------------------- I DwPCH
TX i ic
N Tot DL
TX i ic
TX i ic
P DwPCH
RSCP DwPCH = -------------------LT
The downlink total noise is calculated as follows:

TX i ic
TX i ic
TX i ic
Term

Where
TX i ic
TX i ic
I Intra DL = RSCP DwPCH

With
TX i
TX i ic
I Extra DL =
TX i
Ortho
1 F DL
TX i
Term
1 F JD
and = 0
1

Total Noise
TX j ic
RSCPDwPCH
ji
TX j jc
RSCPDwPCH
TX
j
I IC DL ic jc = ----------------------------------------------F IRF ic jc
TX i
Eb Nt
Term
L Path L L
L T = ------------------------------------------------------------------------------------------------------------------------------------TX i
Term
G G
TX i
Term
and N 0
are defined in "Definitions and Formulas" on page 365.
Coverage Display
C/I (dB)
Atoll calculates the C/I on each pixel of the TX i ic best server coverage area. The pixel colour depends on the
C/I level. Coverage consists of several independent layers whose visibility in the workspace can be managed.
There are as many layers as thresholds defined in the Display tab (Prediction properties). For each layer, a pixel
TX i ic
C
is covered if ----
Threshold . Each layer is assigned a colour and displayed with intersections between
I DwPCH
layers.
C/I Margin (dB)

Atoll calculates the C/I margin on each pixel of the TX i ic best server coverage area. The pixel colour depends
on the C/I margin value. Coverage consists of several independent layers whose visibility in the workspace can
TX i ic
C
Req
CI
Q DwPCH M DwPCH . Each layer is assigned a colour and displayed with
layer, a pixel is covered if ----
I DwPCH
396
AT271_TRG_E6
Forsk 2009

TX i ic
C
tab (Prediction properties). For each layer, a pixel is covered if ----
I DwPCH
Req
Q DwPCH . Each layer is

CECP
8.4.3
Downlink TCH RSCP Coverage

This coverage prediction calculates and displays the RSCP for the downlink traffic channel, RSCP TCH DL . The coverage
prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for a downlink timeslot.
RSCP considering:
The

pixels
in
TX i ic
RSCP TCH DL
TX i ic
the
coverage
area
where
TX i ic
and
Req
RSCP TCH DL Service Mobility are covered and coloured according to the selected display option.
TX i ic
Where RSCP TCH DL is given by:

Max
TX i ic
P TCH DL Service
RSCP TCH DL = ---------------------------------------------------Model
LT
TX i
Term
Model
L Path L L
Max
= ------------------------------------------------------------------------------------------------------------------------------------- and P TCH DL Service is the maximum downlink traffic
TX i
Term
G G
power defined for the selected service.
Model
LT
Model
In L T
, G
TX i
SA
= G DL and L
TX i
SA
Max
P TCH DL Service , if a smart antenna is available in the downlink. Otherwise, G
TX i
and L
TX i
antenna model.
Coverage Display
DL TCH RSCP (dBm)

Atoll calculates the DL TCH RSCP on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the RSCP level. Coverage consists of several independent layers whose visibility in the workspace
can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties). For
TX i ic
each layer, a pixel is covered if RSCP TCH DL Threshold . Each layer is assigned a colour and displayed with
RSCP Margin (dB)

Atoll calculates the RSCP margin on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the RSCP margin value. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties).
TX i ic
Req
RSCP
For each layer, a pixel is covered if RSCP TCH DL RSCP TCH DL Service Mobility M TCH DL . Each layer

tab
(Prediction
properties).
For
each
layer,
a
pixel
is
covered
if
TX i ic
RSCP TCH DL
Req
CECP
RSCP TCH DL Service Mobility . Each layer is assigned a colour and displayed with
Forsk 2009
AT271_TRG_E6
397
8.4.4
Uplink TCH RSCP Coverage

This coverage prediction calculates and displays the RSCP for the uplink traffic channel, RSCP TCH UL . The coverage
prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for an uplink timeslot.
RSCP considering:
The

pixels
in
TX i ic
RSCP TCH UL
the
TX i ic
coverage
Req
RSCP TCH UL Service
area
where
TX i ic
and
Mobility are covered and coloured according to the selected display option.
TX i ic
Where RSCP TCH UL is given by:

Term
TX i ic
P Max
RSCP TCH UL = ---------------Model
LT
TX i
Term
Model
L Path L L
Term
= ------------------------------------------------------------------------------------------------------------------------------------- and P Max is the maximum uplink traffic power defined for
TX i
Term
G G
the selected terminal.
Model
LT
Model
In L T
, G
TX i
SA
= G UL and L
TX i
SA
Term
P Max , if a smart antenna is available in the uplink. Otherwise, G
TX i
and L
TX i
Coverage Display
UL TCH RSCP (dBm)

Atoll calculates the UL TCH RSCP on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the RSCP level. Coverage consists of several independent layers whose visibility in the workspace
can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties). For
TX i ic
each layer, a pixel is covered if RSCP TCH UL Threshold . Each layer is assigned a colour and displayed with
RSCP Margin (dB)

Atoll calculates the RSCP margin on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the RSCP margin value. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties).
TX i ic
Req
RSCP
For each layer, a pixel is covered if RSCP TCH UL RSCP TCH UL Service Mobility M TCH UL . Each layer

tab
(Prediction
properties).
For
each
layer,
a
pixel
is
covered
if
TX i ic
RSCP TCH UL
Req
CECP
RSCP TCH UL Service Mobility . Each layer is assigned a colour and displayed with
8.4.5
Downlink Total Noise

This coverage prediction calculates and displays the total noise on the downlink, N Tot DL . The coverage prediction is
calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for a downlink timeslot. The best server
for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH
power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage prediction is calculated for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any pixels
398
AT271_TRG_E6
Forsk 2009

covered by this transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the downlink noise
for all the carriers but keeps the worst case value, i.e., the most interfered carrier. You can choose to display the minimum,
the maximum, or the average total noise values from among the values calculated for all the carriers. Pixels are covered
and coloured according to the total downlink noise thresholds defined in the display options.
Total downlink noise is given by: N Tot DL =
Term
RSCP TCH DL + RSCP OCCH + N 0
All TX, c, and TS
P TCH DL
P OCCH
With RSCP TCH DL = ----------------------- and RSCP OCCH = ----------------Model
Model
LT
LT
TX i
Term
Model
L Path L L
= ------------------------------------------------------------------------------------------------------------------------------------- and P TCH DL and P TCH DL are respectively the downTX i
Term
G G
link traffic power and the other common control channel power for the selected timeslot.
Model
LT
Model
In L T
, G
TX i
SA
= G DL and L
TX i
SA
Max
TX i
and L
TX i
antenna model.
8.4.6
Downlink Service Area (Eb/Nt) or (C/I)

Eb
These coverage predictions calculate and display the Eb/Nt or C/I on the downlink traffic channel, ------
or
N t TCH DL
C
----
. The coverage predictions are calculated for a given set of a terminal type, a mobility type, a service, a carrier,
I TCH DL
and for a downlink timeslot. The best servers for the coverage predictions are determined according to the P-CCPCH
RSCP from the carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage predictions are calculated for the selected carrier. If the selected carrier does
not exist on a transmitter, there will not be any pixels covered by this transmitter. If you perform these coverage predictions
for all the carriers, Atoll calculates the Eb/Nt or C/I considering:
The

pixels
in
TX i ic
the
coverage
area
where
TX i ic
RSCP P CCPCH TAdd P CCPCH Mobility ,
TX i ic
TX i ic
E b TX i ic
C
Req
Req
Req
RSCP TCH DL RSCP TCH DL Service Mobility , and ------
Q TCH DL or ----
Q TCH DL are
N t TCH DL
I TCH DL
covered and coloured according to the selected display option.

TX i ic
TX i
TX i ic
TX i
TX i ic
E b TX i ic
RSCP TCH DL
RSCP TCH DL
Div
Div
C-
- G Proc
Where ------
= -----------------------------------------------------= ------------------------------------------------------ G DL
DL G DL and --TX i ic
TX i ic
N t TCH DL
I TCH DL
N Tot DL
N Tot DL
Max
TX i ic
P TCH DL Service
With RSCP TCH DL = --------------------------------------------------- Eb Nt DL
LT
TX i
Eb Nt DL
Term
L Path L L
Max
= ------------------------------------------------------------------------------------------------------------------------------------- and P TCH DL Service is the maximum downlink trafTX i
Term
G G
fic power defined for the selected service.
Eb Nt DL
LT
Eb Nt DL
In L T
, G
TX i
SA
= G DL and L
TX i
SA
Max
TX i
and L
TX i
antenna model.
TX i ic
TX i ic
TX i ic
Term

Where
TX i
TX i ic
TX i ic
TX i ic
TX i
Ortho
Term
I Intra DL = 1 F DL 1 F JD + 1 RSCP TCH DL + RSCP OCCH
TX i ic
TX i ic
P OCCH
With RSCP OCCH = ------------------------ Eb Nt DL
LT
Forsk 2009
AT271_TRG_E6
399

TX i ic
I Extra DL =
TX j ic
TX j ic
ji
TX j jc
TX j jc
TX
j
I IC DL ic jc = -----------------------------------------------------------------------------------------F IRF ic jc
Coverage Display
Max Eb/Nt or Max C/I (dB)

TX i ic
E b TX i ic
C
Threshold or ----
N t TCH DL
I TCH DL
Effective Eb/Nt or Effective C/I (dB)

Atoll calculates the effective Eb/Nt or C/I on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the effective Eb/Nt or C/I level. Coverage consists of several independent layers whose visibility in the
For
each
layer,
pixel
is
covered
E b TX i ic
Req
Min ------
Q TCH DL Threshold
N t TCH DL
if
or
TX i ic
C
Req
Min ----
Q
I TCH DL TCH DL
between layers.

For
each
layer,
pixel
is
covered
TX i ic
Req
Eb Nt
E
-----b-
Q TCH DL M TCH DL
N t TCH DL
if
or
TX i ic
Req
CI
C
----
Q TCH DL M TCH DL . Each layer is assigned a colour and displayed with intersections between
I TCH DL
layers.
Required Power (dBm)

Atoll calculates the downlink required power on each pixel of the TX i ic best server coverage area. The pixel
colour depends on the required power level. Coverage consists of several independent layers whose visibility in
the workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction
properties).
Req
P TCH DL =
For
each
Req
Q TCH DL
-------------------------------TX i ic
E b
layer,
Max
pixel
is
Req
covered
P TCH DL Service or P TCH DL =
----- N t TCH DL
Req
P TCH DL Threshold ,
if
Req
Q TCH DL
-----------------------------TX i ic
C
----
I TCH DL
where
Max
P TCH DL Service . Each layer
Required Power Margin (dB)

Atoll calculates the downlink required power margin on each pixel of the TX i ic best server coverage area. The
pixel colour depends on the required power margin value. Coverage consists of several independent layers whose
visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab
Req
Max
(Prediction properties). For each layer, a pixel is covered if P TCH DL P TCH DL Service M arg in , where
Req
Req
Q TCH DL
Q TCH DL
Req
Req
Max
- P Max
P TCH DL = ------------------------------- P TCH DL Service . Each layer
TCH DL Service or P TCH DL = -----------------------------TX i ic
TX i ic
E
C
----
-----b-
I TCH DL
N t TCH DL
400
AT271_TRG_E6
Forsk 2009

E b TX i ic
N t TCH DL
TX i ic
C
----
I TCH DL
8.4.7
Req
Q TCH DL
or
CECP
Req
Q TCH DL . Each layer is assigned a colour and displayed with intersections between layers.
CECP
Uplink Service Area (Eb/Nt) or (C/I)

Eb
These coverage predictions calculate and display the Eb/Nt or C/I on the uplink traffic channel, ------
or
N t TCH UL
C
----
. The coverage predictions are calculated for a given set of a terminal type, a mobility type, a service, a carrier,
I TCH UL
and for an uplink timeslot. The best servers for the coverage predictions are determined according to the P-CCPCH RSCP
from the carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible
transmitters. Afterwards, the coverage predictions are calculated for the selected carrier. If the selected carrier does not
exist on a transmitter, there will not be any pixels covered by this transmitter. If you perform these coverage predictions
for all the carriers, Atoll calculates the Eb/Nt or C/I considering:
The

pixels
in
TX i ic
the
coverage
area
where
TX i ic
TX ic
TX ic
i
i
TX i ic
Eb
C
Req
Req
Req
RSCP TCH UL RSCP TCH UL Service Mobility , and ------
Q TCH UL or ----
Q TCH UL are
N t TCH UL
I TCH UL

TX i ic
Term
Term
TX i ic
TX i ic
E b TX i ic
RSCP TCH UL
RSCP TCH UL
Div
C-
- G Proc
- G Div
Where ------
= ----------------------------------------------------------= ----------------------------------------------------------UL G UL and --UL
TX i ic
TX i ic
N t TCH UL
I TCH UL
N Tot UL
N Tot UL
Req
Req
Term
TX i ic
Q TCH UL
Q TCH UL
P Max
Term
Term
Term
- or P Term
With RSCP TCH UL = ------------------------- and P Req = P Max -------------------------------Req = P Max -----------------------------TX i ic
TX i ic
Eb Nt UL
E
C
-----b-
LT
----
I TCH UL
N t TCH UL
TX i
Eb Nt UL
Term
L Path L L
Term
= ------------------------------------------------------------------------------------------------------------------------------------- and P Max is the maximum power defined for the
TX i
Term
G G
selected terminal.
Eb Nt UL
LT
Eb Nt UL
In L T
, G
TX i
SA
= G UL and L
TX i
SA
Term
P Max , if a smart antenna is available in the uplink. Otherwise, G
TX i
and L
TX i
Coverage Display
Max Eb/Nt or Max C/I (dB)

TX i ic
E b TX i ic
C
Threshold or ----
N t TCH UL
I TCH UL
Effective Eb/Nt or Effective C/I (dB)

Atoll calculates the effective Eb/Nt or C/I on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the effective Eb/Nt or C/I level. Coverage consists of several independent layers whose visibility in the
workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction prop-
Forsk 2009
AT271_TRG_E6
401
erties).
For
each
layer,
pixel
is
covered
if
E b TX i ic
Req
Min ------
Q TCH UL Threshold
N t TCH UL
or
TX i ic
C
Req
Min ----
Q
I TCH UL TCH UL
between layers.

For
each
layer,
pixel
is
covered
if
TX i ic
Req
Eb Nt
E
-----b-
Q TCH UL M TCH UL
N t TCH UL
or
TX i ic
Req
CI
C
----
Q TCH UL M TCH UL . Each layer is assigned a colour and displayed with intersections between
I TCH UL
layers.
Required Power (dBm)

Atoll calculates the uplink required power on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the required power level. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties).
Req
Q TCH UL
Term
Term
- P Term
For each layer, a pixel is covered if P Req Threshold , where P Req = -------------------------------Max
TX i ic
E
-----b-
N t TCH UL
or
Req
Q TCH UL
Term
Term
P Req = ----------------------------- P Max . Each layer is assigned a colour and displayed with intersections between layers.
TX i ic
C
----
I TCH UL
Required Power Margin (dB)

Atoll calculates the uplink required power margin on each pixel of the TX i ic best server coverage area. The
pixel colour depends on the required power margin value. Coverage consists of several independent layers whose
visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab
(Prediction
Term
P Req =
properties).
Req
Q TCH UL
-------------------------------TX i ic
E b
For
each
Term
layer,
Term
P Max or P Req =
----- N t TCH UL
pixel
Req
Q TCH UL
-----------------------------TX i ic
C
----
I TCH UL
is
covered
if
Term
Term
P Req P Max M arg in ,
where
Term
P Max . Each layer is assigned a colour and displayed

TX i ic
Eb
N t TCH UL
TX i ic
C
----
I TCH UL
8.4.8
Req
Q TCH UL
or
CECP
Req
Q TCH UL . Each layer is assigned a colour and displayed with intersections between layers.
CECP
Effective Service Area (Eb/Nt) or (C/I)

These coverage predictions consist of pixels covered by the both the uplink and the downlink service areas. These coverEb
C
age predictions calculate the Eb/Nt or C/I on the downlink and uplink traffic channels, ------
or ----
and
N t TCH DL
I TCH DL
C
E
-----b-
or ----
, and display the pixels where both downlink and uplink Eb/Nt or C/I are above the required
N t TCH UL
I TCH UL
quality thresholds.
The coverage predictions are calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for all
the 6 timeslots. The best servers for the coverage predictions are determined according to the P-CCPCH RSCP from the
carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters.
Afterwards, the coverage predictions are calculated for the selected carrier. If the selected carrier does not exist on a trans-
402
AT271_TRG_E6
Forsk 2009

mitter, there will not be any pixels covered by this transmitter. If you perform these coverage predictions for all the carriers,
Atoll calculates the Eb/Nt or C/I considering:

The pixels in the TX i ic coverage area are covered and coloured according to the selected display option if all the following conditions are satisfied:
8.4.9
TX i ic
RSCP TCH DL RSCP TCH DL Service Mobility
RSCP TCH UL RSCP TCH UL Service Mobility
C
Req
Req
E
-----b-
Q TCH DL or ----
Q TCH DL for any of the 6 timeslots
N t TCH DL
I TCH DL
C
Req
Req
E
-----b-
Q TCH UL or ----
Q TCH UL for any of the 6 timeslots
N t TCH UL
I TCH UL
TX i ic
Req
TX i ic
Req
TX i ic
TX i ic
TX i ic
TX i ic
Cell to Cell Interference

This coverage prediction calculates and displays the interference received by cells receiving in uplink from other cells
which are transmitting in downlink. The timeslot configuration of each cell defines the direction of the link at any given
instance. During each subframe, the direction of the link changes twice (downlink to uplink, and then uplink to downlink).
These transitions are referred to as switching points.
The coverage prediction is calculated for a given set of a terminal type, a mobility type, a service, a carrier, and a timeslot.
RSCP considering:

The mobility, service, and terminal are used to calculate the best server coverage of the interfered cell.
Assuming that a transmitter TX j is interfering a studied transmitter TX i on a timeslot, on the same carrier ic or on another
carrier jc, the cell to cell interference is given by:
TX j jc
I C2C TX i TX j =
TX j ic
TX j ic
TX j jc
RSCPTCH DL + RSCPOCCH + -----------------------------------------------------------------------------------------F IRF ic jc

TX j
TX j
TX j ic
Where
TX j ic
P TCH DL
RSCP TCH DL = -------------------------------LT
TX j ic
RSCP TCH DL
TX j ic
TX j jc
and
TX j jc
P TCH DL
RSCP TCH DL = -------------------------------LT
TX j jc
TX j
using
smart
antenna,
and
TX j
P TCH DL G Ant
TX j jc
P TCH DL G Ant
- and RSCP TCH
- ------------ otherwise.
= ------------------------ ---------- DL = ----------------------TX j
TX j
LT
LT
L Ant
L Ant
TX j ic
TX j jc
TX j
TX j
TX j ic
TX j jc
P OCCH G Ant
P OCCH G Ant
- and RSCP OCCH
- otherwise.
RSCP OCCH = ------------------ ----------= ------------------ ----------TX j
TX j
LT
LT
L Ant
L Ant
ITU526 5
L T = L Path
ITU526 5
L Path
TX j
TX i
L TX L RX
is the path loss calculated using the ITU526-5 propagation model without antenna loss.
is the angle for the smart antenna pattern.

TX j
L Ant is the main antenna attenuation.

TX j
G Ant is the main antenna gain.

Atoll calculates the cell to cell interference on each pixel of the TX i ic best server coverage area. The pixel colour
depends on the cell to cell interference level. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties). For
Forsk 2009
AT271_TRG_E6
403

each layer, a pixel is covered if I C2C TX i TX j Threshold . Each layer is assigned a colour and displayed with intersections between layers.
8.4.10
UpPCH Interference
UpPCH is usually carried by the UpPTS timeslot. However, if the interference on UpPTS is high, from unsynchronised
DwPTS or TS0 timeslots of other cells, it is possible to shift the UpPCH to TS1. This is called UpPCH shifting. If some cells
in a network use UpPCH shifting, you can use this coverage prediction to study the interference on the shifted UpPCH of
these cells from other cells. The interference from other cells is in this case generated by the traffic on the TS1 of interfering
cells.
This coverage prediction calculates and displays the uplink interference on the TS1, I TS1 UL . The coverage prediction is
calculated for a given set of a terminal type, a mobility type, a service, a carrier, and for TS1. The best server for the coverage prediction is determined according to the P-CCPCH RSCP from the carrier with the highest P-CCPCH power, or from
the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage prediction is calculated
for the selected carrier. If the selected carrier does not exist on a transmitter, there will not be any pixels covered by this
transmitter. If you perform this coverage prediction for all the carriers, Atoll calculates the interference for all the carriers
but keeps the worst case value, i.e., the most interfered carrier. You can choose to display the minimum, the maximum,
or the average total noise. The coverage prediction is calculated using the main antenna.
TX i ic
TX i ic
Pixels in the TX i ic coverage area where RSCP P CCPCH TAdd P CCPCH Mobility and I TS1 UL Threshold are
The uplink interference on TS1 is calculated from the uplink load calculated in the simulations or manually defiend for the
TS1.
TX i ic
TX i
The uplink interference on TS1 is given by: I TS1 UL = N 0
8.4.11
TX i ic
X TS1 UL
------------------------------------TX i ic
1 X TS1 UL
HSDPA Coverage
This coverage prediction calculates and displays the RLC peak rate or the MAC rate per pixel covered by HSDPA cells.
The coverage prediction is calculated for a given set of an HSDPA terminal type, a mobility type, an HSDPA service, a
carrier, and for all downlink timeslots. The best server for the coverage prediction is determined according to the P-CCPCH
RSCP from the carrier with the highest P-CCPCH power, or from the master carrier in case of N-frequency mode compatible transmitters. Afterwards, the coverage predictions are calculated for the selected carrier. If the selected carrier does
not exist on a transmitter or if it does not support HSDPA, there will not be any pixels covered by this transmitter. If you
perform these coverage predictions for all the carriers, Atoll calculates the RLC or MAC rate considering:

The pixels in the TX i ic coverage area are covered and coloured if:
TX i ic
Req
E
------C-
Q HS PDSCH , and
N t HS PDSCH
E
------C-
is enough to select a bearer for the pixels.
N t HS PDSCH
TX i ic
TX i ic
For more information on HSDPA bearer selection, see "HSDPA Part of the Algorithm" on page 390.
Coverage Display
RLC Peak Rate

After selecting the bearer, Atoll reads the corresponding RLC peak rate. This is the highest rate that the bearer
can provide on each pixel. The pixel colour depends on the RLC peak rate. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as thresholds defined
in the Display tab (Prediction properties). Each layer is assigned a colour and displayed with intersections between
layers.
MAC Rate
MAC
Atoll displays the MAC rate ( R DL

MAC
R DL
404
) provided on each pixel. The MAC rate is calculated as follows:
= S Block 500
AT271_TRG_E6
Forsk 2009

Where, S Block is the transport block size (in kbits) of the selected HSDPA bearer; it is defined for each HSDPA
bearer in the related table. The value 500 corresponds to the number of blocks per second (there are 4 blocks per
2000
TTI and 2000 TTI in one second, i.e ------------- blocks per second).
4
The pixel colour depends on the MAC rate. Coverage consists of several independent layers whose visibility in the
workspace can be managed. There are as many layers as thresholds defined in the Display tab (Prediction properties). Each layer is assigned a colour and displayed with intersections between layers.
8.5
Smart Antenna Modelling

Atoll calculates the smart antenna gains and losses in the direction of a user during the simulations, and in the direction
of each pixel in coverage predictions. During simulations, Atoll determines the gains and losses using the smart antenna
models. In coverage predictions, Atoll determines the gains and losses from the angular distributions calculated during the
simulations for each timeslot and stored in the Cell Parameters per Timeslot table.
If a smart antenna model is only downlink or only uplink, the other direction uses the main antenna gain and losses for
calculations. Therefore,
If a smart antenna is available on the downlink and uplink:

TX
SA
TX
SA
TX
SA
TX
SA
G UL = G UL , L UL = L UL and G DL = G DL , L DL = L DL
If a smart antenna is available on the downlink only:

TX
SA
TX
SA
TX
TX
TX
G DL = G DL , L DL = L DL and G UL = G Ant , L UL = L
= L Total UL
TX
= L Total DL
If a smart antenna is available on the uplink only:

TX
SA
TX
SA
TX
TX
TX
G UL = G UL , L UL = L UL and G DL = G Ant , L DL = L
TX
If no smart antenna equipment is defined:

TX
TX
TX
TX
G DL = G UL = G Ant , L UL = L
TX
TX
= L Total UL , and L DL = L
8.5.1
Modelling in Simulations
8.5.1.1
Grid of Beams Modelling
TX
= L Total DL
A grid-of-beams smart antenna, called GOB, consists of more than one directional antenna pattern (beam) in different
directions. Each beam of a GOB has a different azimuth so that the GOB as a whole covers an entire sector. During the
simulations, Atoll determines the most suitable beam from the GOB for each user served by the smart antenna. The most
suitable beam (best beam) is the one which provides the highest gain towards the served user:
Beam Best = Beam
H
Max G Beam L Beam L Beam

V
Where G Beam , L Beam , and L Beam are the gains, horizontal, and vertical attenuations of the beams of the GOB. In words,
the best beam is the one among all the beams of a GOB that has the highest difference between gain, and horizontal and
SA
SA
SA
SA
vertical attenuations. The gains and losses of the GOB ( G DL , G UL , L DL , and L UL ) are determined from the selected
best beam.
The following example shows how Atoll calculates the GOB gains and losses.
Example:
Let us assume a GOB with 5 beams that have the same vertical patterns, and whose horizontal patterns are pointed
towards different directions as shown in the figure below:
Forsk 2009
AT271_TRG_E6
405
Figure 8.3: Grid Of Beams Modelling

Let us assume that all the beams and the main antenna have the same 18 dBi gain, and the vertical attenuation at the
user location is 15 dB, which is also the same for all the beams because we assume that the vertical patterns are the same.
If the user is located at = 70 azimuth, as shown in the figure below, Atoll determines the best beam, which has the
highest gain towards , as follows:
Beam
Gain (dBi)
Horizontal
Attenuation
(dB)
Vertical
Attenuation
(dB)
G Beam L Beam L Beam
Total Gain
(dB)
18
60
15
18 - 60 - 15
-57
30
18
60
15
18 - 60 - 15
-57
60
18
2.21
15
18 - 2.21- 15
0.79
-30
18
60
15
18 - 60 - 15
-57
-60
18
60
15
18 - 60 - 15
-57
Transmitter
Centre of the pixel where the served user is located
Angle between the user and the transmitter azimuth
Figure 8.4: GOB Modelling - Determination of the Best Beam

In our example, the total gain of the beam at 60 is the highest. Therefore this beam is selected as the best beam.
If this beam has been selected in the downlink,
SA
SA
G DL = 18 dB and L DL = L Beam + L Beam = 17.21 dB

If this beam has been selected in the uplink,
SA
SA
G UL = 18 dB and L UL = L Beam + L Beam = 17.21 dB
8.5.1.2
Adaptive Beam Modelling

An adaptive beam smart antenna is capable of steering a given antenna pattern towards the direction of the served signal.
In Atoll, this is modelled using a single antenna pattern, called a beam because of its highly directional shape. During the
simulations, this adaptive beam is oriented in the direction of each served user in order to model the effect of the smart
antenna.
406
AT271_TRG_E6
Forsk 2009

SA
SA
The adaptive beam gains ( G DL and G UL ) are the antenna gains defined for the beam, and the adaptive beam losses
SA
SA
( L DL and L UL ) are the horizontal and vertical pattern attenuations L Beam + L Beam towards the user direction.
The following example shows how Atoll calculates the adaptive beam gains and losses.
Example:
Let us assume an adaptive beam smart antenna selected for a transmitter along with a main antenna. Let us assume that
the adaptive beam and the main antenna have the same 18 dBi gain, and the vertical attenuation at the user location is
15 dB.
If the user is located at = 60 azimuth, as shown in the figure below:
Transmitter
Centre of the pixel where the served user is located
Angle between the user and the transmitter azimuth
Figure 8.5: Adaptive Beam Modelling - Determination of the Best Beam

If the adaptive beam smart antenna is selected in the downlink, the gain and losses of the adaptive beam at are:
SA
SA
G DL = 18 dB and L DL = L Beam + L Beam = 15 dB

If the adaptive beam smart antenna is selected in the uplink, the gain and losses of the adaptive beam at are:
SA
SA
G UL = 18 dB and L UL = L Beam + L Beam = 15 dB

H
In fact, as the ideal beam steering algorithm steers the beam towards the served user, L Beam = 0 . These values are used
in interference calculation to determine the downlink interfering signal due to transmission towards the served user, as well
as for calculating the uplink interfering signals received at transmitter when decoding signal received from the served user.
8.5.1.3
Statistical Modelling
A statistical modelling approach is also available in Atoll which can be used to model the effect of smart antennas through
C/I gains. You can create smart antenna equipment in Atoll based on the statistical approach by providing C/I gains and
their cumulative probabilities for different spreading angles, Spread .
You can assign a spreading angle to each clutter class in your document. Atoll reads the clutter class in which the served
user is located to determine the spreading angle. Different clutter types have different spreading effects on the propagation
of radio waves. Urban and dense urban clutter types introduce more multipath and spread the signal at a wider angle than
an open or rural clutte type.
Once you have assigned the spreading angles to clutter classes, you can enter the C/I gains and their cumulative probabilities for each spreading angle, in the smart antenna equipment based on the statistical model. For each smart antenna
equipment based on statistical modelling, you can set a probability threshold, TProb
SA
To find the smart antenna gain, Atoll determines the clutter class of the served user, it reads the spreading angle from the
clutter class properties, it reads the probability threshold from the smart antenna properties, and reads the smart antenna
C/I gain defined for the Probability = 1 TProb
SA
corresponding to the spreading angle.
The following example shows how Atoll calculates the statistical C/I gains and losses.
Example:
Let us assume that the served user is located at a an urban clutter class with Spread = 10 . The smart antenna equipSA
SA
ment has TProb

for Prob = 20 % . If a gain for the exact
= 80 % . Atoll will read the smart antenna C/I gain G
probability value of 20% is not defined, Atoll linearly interpolates the gain value from the two surrounding values.
If G
SA
Prob = 19%
= 4.6298 dB and G
SA
Prob = 20.4%
= 4.7196 dB , then G
SA
Prob = 20%
= 4.6941 dB
The smart antenna gains are the same for uplink and downlink. Their are no losses for this type of smart antenna equipment. Negative values of C/I gains are considered as losses.
Forsk 2009
AT271_TRG_E6
407
8.5.1.4
Optimum Beamformer Model

The optimum beamformer performs simple beam forming in the downlink to steer the main beam towards the served user,
and uses the Minimum Mean Square Error algorithm in the uplink in order to cancel interference.
Adaptive antenna systems use more than one antenna element, along with smart signal processing, to locate and track
various types of signals, to dynamically minimize interference, and maximize useful signal reception. The signal processor
dynamically applies weights to each element of the adaptive antenna system to create array patterns in real-time.
Figure 8.6: Linear Adaptive Array System

The antenna patterns created for downlink transmission have a main beam pointed in the direction of the useful signal. In
the uplink, in addition to the main beam pointed in the direction of the useful signal, there can also be one or more nulls in
the direction of the interfering signals. If the adaptive antenna system is using E SA antenna elements, it is possible to
create E SA 1 nulls to cancel E SA 1 interfering signals. In a mobile environment where the interference is not stationary, antenna patterns are adjusted so that the nulls remain in the direction of the moving interference. A system using
adaptive antennas adjusts the weights on each antenna element to achieve such a pattern.
The optimum beamformer supports linear adaptive array systems, such as the one shown in Figure 8.6: on page 408.
In the following explanations, we assume:
8.5.1.4.1
There are a total of E SA elements in the adaptive antenna system.
is the angle of arrival for the useful signal.
is the angle at which we want to calculate the smart antenna gain.

d is the distance between two adjacent antenna elements.
Downlink
Figure 8.7: Downlink Beamforming
408
AT271_TRG_E6
Forsk 2009

The downlink algorithm models a conventional beam former. The smart antenna processor applies complex weights, w n ,
to each antenna element in order to form a beam towards the served user. The magnitude of these complex weights is set
to 1. The beamforming is performed using only the phase of the complex weights. The steering vector, S , representing
the complex weights for forming a beam towards the served user, i.e., at the angle of arrival , is given by:
S = 1 e
2
j ------- d sin
2
j ------- 2d sin
... e
T
2
j ------- E SA 1 d sin
Where the notation T represents the transpose of a matrix.

Therefore, the complex weight at any nth antenna element can be given by:
wn = e
2
j ------- nd sin
j n sin
In Atoll, d = --- , therefore, w n = e

.
2
The smart antenna gain in any direction can be given by:

G
SA
= gn S R S
Where the notation H represents the Hilbert transform, which is the complex conjugate transpose of a matrix, g n is the
gain of the nth antenna element in the direction , and R is the array correlation matrix for a given user direction , given
by:
H
R = S S
For the direction of the served user, i.e., , the smart antenna gain is calculated as follows:
G
SA
= g n S R S = g n S S S S = g n E SA
The smart antenna gain includes the gain of the beamforming as well as the gain of power combination.
The smart antenna gain in dB is G
SA
= 10 Log G
SA
The smart antenna is able to form the beam only in the horizontal plane, therefore, the vertical pattern is assumed to
remain the same.
Power Combination Gain

Cell transmission power is fed to each antenna element of the smart antenna system. Since each element transmits the
same input power, this results in a gain due to power combination, i.e., the powers fed to each antenna element are
combined for transmission.
Additional Processing During Monte Carlo Simulations

During Monte Carlo simulations, as Atoll calculates the smart antenna gains for each victim mobile in a cells coverage
area, it averages the array correlation matrix R over all the iterations in order to generate the angular distribution of the
downlink traffic power.
The average array correlation matrix is given by:
J
R Avg =
j pj Rj
j=1
Where R Avg is the average downlink array correlation matrix, J is the number of served mobiles during the simulation, j
is the probability of presence of the mobile j, p j is the EIRP transmitted towards the mobile j, and R j is the array correlation
matrix for the mobile j.
8.5.1.4.2
Uplink
The uplink models the adaptive Minimum Mean Square Error algorithm which optimizes the useful signal as well as
cancels the interference from the most interfering E SA 1 interfering mobiles. The optimal beam forming method used in
Atoll overcomes the limitations of a null steering beam former and maximizes the signal quality.
A simple null steering beam former requires the knowledge of the directions of interference sources, and the estimated
weights do not maximize the signal quality. Whereas, an optimal beam former does not require knowledge of directions
and power levels of interference to maximize the output. It only requires the direction of the useful signal in order to calculate the optimum signal quality.
Forsk 2009
AT271_TRG_E6
409
Figure 8.8: Uplink Adaptive Algorithm

represent the vector of E complex weights for the beam former. w
is given by:
Let w
SA
= R1 S
w
Where S is the steering vector in the direction of the served user, . , which is a constant value for a given useful
signal, represents the smart antenna gain in the uplink given by the equation:
E SA
= ---------------------------------H
1
S RN S
1
R N is the inverse of the total noise correlation matrix. The total noise correlation matrix is the sum of the thermal noise
correlation matrix R n , and the interference correlation matrix R I , given by:
J
2
RN = Rn + RI = n I +
pj Sj Sj
j=1
J
pj Sj Sj
Where R n = n I and R I =
j=1
2
n is the thermal noise power. I is the identity matrix. p j is the power received by one element of the smart antenna from
the jth interfering mobile. S j is the steering vector in the direction of the jth interfering mobile, . J is the total number of
interfering mobiles.
The total noise power, including thermal noise and interference from all uplink interferers, received by a cell is given by:
= 2 S H R1 S
P
N
The total power received from the served user is given by:
= p 2 S H R1 S 2
P
Where p is the power received by one element of the smart antenna from the served user.
The uplink signal quality is calculated by:
2
H
2
1
p S RN S
H
P
SA
1
Q UL = ------- = --------------------------------------------------------------= p S RN S
H
2
1
P
N
S RN S
From the above equation, we can determine the uplink smart antenna beam forming gain in the direction of the served
user. RSCP TCH UL can be calculated from the above equation by considering the interference and noise to be null, i.e.,
1
R N = I . This gives:
H
RSCP TCH UL = p S I S = p E SA
410
AT271_TRG_E6
Forsk 2009

From the above equation, the uplink smart antenna beam forming gain equals the number of smart antenna elements, i.e.,
G
SA
= E SA .

1
The inverse noise correlation matrix R N for each iteration k includes the effect of the matrix calculated for the previous
iteration. Hence, Atoll is able to calculate an average of the smart antenna interference-cancellation effect. The result is
the angular distribution of the uplink load, which is calculated from the inverse of the noise correlation matrix obtained at
the end of the last iteration of a Monte Carlo simulation. This angular distribution of the uplink load can be stored in the
Cell Parameters per Timeslot table. The average of the inverse noise correlation matrices is calculated as follows:
K
1
RN
Avg
1
= ----
K
RN
1
k
k=1
1
Where R N
Avg
is the average of the inverse noise correlation matrices of all the iterations from k = 1 to K, and R N
is
the inverse noise correlation matrix of the kth iteration. The uplink load is calculated from the average inverse noise correlation matrix.
8.5.1.5
3rd Party Smart Antenna Modelling

3rd party smart antenna models can be used in Atoll to determine the gains and losses during the simulations for a given
user distribution generated. The smart antenna gains and losses are used during the simulations and the results are stored
in the Cell Parameters per Timeslot table, which can be used in coverage predictions.
8.5.2
Construction of the Geographic Distributions

During simulations, Atoll uses the smart antenna model selected for each transmitter to calculate the smart antenna gains
and losses. These values are calculated and stored for each user generated for the simulations. Therefore, these values
are calculated and are available for the given locations of the users, i.e., points, only. Atoll uses the Angular Step value
that you set when creating and running simulations to construct the geographic distributions of these results.
Once Atoll has calculated the downlink traffic power and the uplink load using the smart antenna gains and losses determined as explained in the previous section, at the location of a given user, it calculates the same for points located at the
angle equal to that of the Angular Step of the simulations.
At the end of the simulations, Atoll has a number of points, Angular Step apart, available with the values of these results.
The geographic distribution of these results, i.e., downlink traffic power and uplink loads, is constructed by connecting the
resulting value points.
The following example explains how the geographic distribution of downlink traffic power is created. The geographic distribution of uplink loads is constructed in the same manner.
Example:
Let us assume a smart antenna equipment using adaptive beam modelling. The angular step defined for the simulations
is Step = 30 . Therefore, the results are calculated for each point located at regular steps of 30 , i.e., 12 points. The
downlink traffic power at the served user (W) with the adaptive beam pointing in the users direction is P W . The downlink
traffic powers, using the same adaptive beam pointed towards the served user, at the 12 other points are also determined.
Forsk 2009
AT271_TRG_E6
411
Figure 8.9: Construction of the Geographic Distribution of Downlink Traffic Power

The resulting geographic distribution is formed by linearly joining the obtained results.
Figure 8.10: Geographic Distribution of Downlink Traffic Power

The accuracy of the geographic distribution depends upon the value of the angular step. A radiation pattern created at a
1 step will be much more accurate than one created at 45 , for example. But, the latter will be computed 45 times faster
than the first. The value of the Angular Step should be the best possible compromise between calculation speed and accuracy.
8.5.3
Modelling in Coverage Predictions

The results of Monte Carlo simulations, including the smart antenna results, can be stored in the Cells and in the Cell
Parameters per Timeslot tables, and can be used to carry out coverage predictions. The main results of Monte Carlo simulations used in coverage predictions are:
If a smart antenna is used in both uplink and downlink:

Geographic distribution of UL load X
UL
DL
and DL traffic power P Traffic
If a smart antenna is used in downlink only:

DL
Geographic distribution of DL traffic power P Traffic
Without smart antenna:

UL load X
UL
DL
and DL traffic power P Traffic
The uplink load and the downlink traffic power at a given pixel are determined by calculating the angle of that pixel with
respect to the transmitter azimuth, and reading the uplink load and downlink traffic power from the geographic distribution
results. If an exact value for the angle is not available, the load and power are determined using linear interpolation for the
given angle between two available values.
412
AT271_TRG_E6
Forsk 2009

For example, the figure below shows the distribution of downlink traffic power and uplink traffic load results from a simuDL 315
lation. For a pixel located at = 315 , the downlink traffic power P Traffic
from these results. In this example,
DL 315
P Traffic
30 dBm , and X
UL 315
and the uplink load X
UL 315
are read
= 2.75 % .
For each pixel, Atoll determines the downlink traffic powers and the uplink loads from all the transmitters.
Figure 8.11: Geographic Distribution of downlink traffic power and uplink load
8.6
N-Frequency Mode and Carrier-Type Allocation

Transmitters that support N-frequency mode are multi carrier transmitters with a master and one or more slave carriers.
You can assign master and slave carriers to transmitters manually, or use the automatic frequency allocation in Atoll to
assign carrier types automatically.
8.6.1
Automatic Carrier-Type Allocation

For each transmitter, Atoll determines a list of "near" transmitters. For any transmitter TXi, its "near" transmitters would be
the ones that are geographically near the transmitter TXi, according to a sort function. Among the "near" transmitters, the
transmitters that are directed towards the transmitter TXi are considered before the transmitters that are directed in the
same direction as TXi. This is because transmitters directed towards each other interfere more compared to transmitters
pointing in the opposite direction.
Next, Atoll assigns one master carrier to each transmitter TXi, such that the master carrier of TXi is different from the
master carrier of TXj, where TXj belongs to the list of "near" transmitters. The master carrier is one of the cells defined in
the transmitter. All the other cells of the transmitter are assigned the carrier-type "slave".
For transmitters that support the N-frequency mode and have master carriers properly assigned, Atoll performs the neighbour and scrambling code allocation for the master carrier only.
8.7
Neighbour Allocation
Atoll permits the automatic allocation of intra-technology neighbours in a TD-SCDMA document. Two allocation algorithms
are available, one dedicated to intra-carrier neighbours and the other for inter-carrier neighbours. The intra-technology
neighbour allocation algorithms take into account all the cells of TBC transmitters. It means that all the cells of TBC transmitters of your .atl document are potential neighbours.
The cells to be allocated will be called TBA cells. They must fulfil following conditions:
They are active

They satisfy the filter criteria applied to the Transmitters folder
They are located inside the focus zone

Note:
Forsk 2009
AT271_TRG_E6
413
8.7.1

We assume a reference cell A and a candidate neighbour, cell B. When automatic allocation starts, Atoll checks following
conditions:
1. The distance between both cells must be less than the user-defined maximum inter-site distance. If the distance
between the reference cell and the candidate neighbour is greater than this value, the candidate neighbour is
discarded.
2. The calculation options:
Carriers: This option enables you to select the carrier(s) on which you want to run the allocation. You may choose
one or more carriers. Atoll will allocate neighbours to cells using the selected carriers.
Force co-site cells as neighbours: This option enables you to force cells located on the same site as reference cell
in the candidate neighbour list. This constraint can be weighted among the others and ranks the neighbours
through the importance field.
Force adjacent cells as neighbours (only for intra-carrier neighbours): This option enables you to force cells
geographically adjacent to the reference cell in the candidate neighbour list. This constraint can be weighted
among the others and ranks the neighbours through the importance field.
Note:
Adjacency criterion: Geographically adjacent cells are determined on the basis of their
best server coverages in TD-SCDMA projects. Let CellA be a candidate neighbour cell of
CellB. CellA is considered adjacent to CellB if there exists at least one pixel in the CellB
best server coverage area (and P-CCPCH RSCP of CellB > P-CCPCH RSCP T_Add)
where CellA is best server (of several cells have the same best server value) or CellA is the
second best server that enters the handover set (i.e., P-CCPCH RSCP of CellA > PCCPCH RSCP T_Drop and P-CCPCH RSCP of CellA > P-CCPCH RSCP of CellB T_Comp.)
When this option is selected, adjacent cells are sorted and listed from the most adjacent to
the least, depending on the above criterion. Adjacency is relative to the number of pixels
Force neighbour symmetry: This option enables you to force the reciprocity of a neighbourhood link. Therefore, if
the reference cell is a candidate neighbour of another cell, the later will be considered as candidate neighbour of
the reference cell.
Force exceptional pairs: This option enables you to force/forbid some neighbourhood relationships. Therefore, you
may force/forbid a cell to be candidate neighbour of the reference cell.
Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current
neighbours and carries out a new neighbour allocation. If not selected, the existing neighbours are kept.
3. There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability:
-
Intra-carrier neighbours: intra-carrier handover is a baton handover.

The reference cell A and the candidate cell B are located inside a continuous layer of cells with carrier c1 (c1
is the selected carrier on which you run the allocation).
SA is the area where the cell A is the best serving cell.
-
The P-CCPCH RSCP from the cell A is greater than the P-CCPCH RSCP T_Add.
The P-CCPCH RSCP from the cell A is greater than the P-CCPCH RSCP from all other cells.
SB is the area where the cell B can enter the handover set.
414
AT271_TRG_E6
Forsk 2009

-
The P-CCPCH RSCP from the cell B is greater than the P-CCPCH RSCP T_Drop.
The P-CCPCH RSCP from the cell B is greater than the P-CCPCH RSCP from the cell A minus the PCCPCH RSCP T_Comp.
Figure 8.12: Intra-carrier Neighbours

-
Inter-carrier neighbours: Inter-carrier handover is a hard handover. It is needed in a multi-carrier CDMA network:
- To balance and share the load between carriers and layers.
- To make a coverage reason handover from micro cells to macro cells.
SA is the area where the received signal level from the transmitter A is greater than a minimum signal level.
SA is the coverage area of reference cells transmitter A restricted between an upper and a lower limit. The
upper limit represents the start of the handover area (best server area of A plus the handover margin named
"handover start"), and the lower limit represents the end of the handover area (best server area of A plus the
margin called "handover end").
SB is the coverage area where the candidate cells transmitter B is the best server.
Figure 8.13: Overlapping Coverages
Forsk 2009
AT271_TRG_E6
415

SA SB
Atoll calculates the percentage of covered area ( ---------------------- 100 ), which it compares with the % minimum covered
SA
The coverage condition can be weighted among the others and ranks the neighbours through the importance field.
4. The importance values are used by the allocation algorithm to rank the neighbours according to the allocation
maximum number of neighbours to be allocated to each cell is exceeded. If there are 15 candidate neighbours and the
maximum number of neighbours to be allocated to the reference cell is 8, among these 15 candidate neighbours, only 8
that have the most importance will be allocated to the reference cell. You can set different maximum numbers of neighbours for intra-carrier and inter-carrier neighbours for each cell. If you have defined these values for each cell, Atoll uses
these limits instead of the value set in the allocation dialog.
As indicated in the table below, the neighbour importance depends on the neighbourhood cause; this value varies between
from 0 and 100%.
Neighbourhood cause
When
Importance
value
Existing neighbour
If the Delete existing neighbours option is not selected
Existing
importance
Exceptional pair
If the Force exceptional pairs option is selected
100 %
Co-site transmitter
If the Force co-site cells as neighbours option is selected
IF
If the Force adjacent cells as neighbours option is selected
IF

coverage conditions
If the % minimum covered area is exceeded
IF
relationship
If the Force neighbour symmetry option is selected
IF
Except the case of forced neighbours (importance = 100%), priority assigned to each neighbourhood cause is determined
using the Importance Function (IF). The IF considers three factors for calculating the importance:
The co-site factor (C) which is a Boolean

The adjacency factor (A) which deals with the percentage of adjacency
The IF is user-definable using the Min importance and Max importance fields.
Factor
Min importance
Default value
Max importance
Default value
Min(O)
1%
Max(O)
30%
Min(A)
30%
Max(A)
60%
Co-site factor (C)
Min(C)
60%
Max(C)
100%
The IF evaluates importance as follows:
Neighbourhood
cause
IF
Resulting IF using the

default values from the
table above
Co-site
Adjacent
No
No
Min O + O O
1% + 29% O
No
Yes
Min A + A Max O O + 100% Max O A
30% + 30% 30% O + 70% A
Yes
Yes
Min C + C Max O O + 100% Max O A
60% + 40% 30% O + 70% A
Where X = Max X Min X

Notes:
fields, neighbours will be ranked in the following order:
i.
Co-site neighbours
ii. Adjacent neighbours

iii. Neighbours based on coverage overlapping
416
If the ranges of the importance factors overlap, the neighbours may not be ranked
according to the neighbourhood cause.
The ranking between neighbours from the same category depends on the factors (A) and
(O).
AT271_TRG_E6
Forsk 2009
The default value of Min(O) = 1% ensures that neighbours selected for symmetry will have
an importance greater than 0%. With a value of Min(O) = 0%, neighbours selected for
symmetry will have an importance field greater than 0% only if there is some coverage
overlapping.
In the Results part, Atoll provides the list of neighbours, the number of neighbours, and the maximum number of neighbours allowed for each cell. In addition, it indicates the importance (in %) of each neighbour and the allocation reason, i.e.,
a neighbour may be marked as exceptional pair, co-site, adjacent, coverage, or symmetric. For neighbours accepted for
co-site, adjacency, and coverage reasons, Atoll displays the percentage of area that satisfies the coverage conditions and
the corresponding surface area (km2), the percentage of area that satisfies the adjacency conditions and the corresponding surface area (km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.
Notes:
You donot require simulations or coverage predictions for an automatic neighbour

allocation. For automatic neighbour allocation, Atoll automatically calculates the missing
path loss matrices.
Although no specific terminal, mobility, or service is selected for automatic neighbour

allocation, the algorithm tries to find the maximum number of neighbours by selecting:
-
The service with the lowest body loss

The terminal with the highest difference between Gain and Losses. If this is the same
for all terminals, Atoll uses the terminal with the lowest noise figure.
Mobility does not impact the allocation
The percentage of area is calculated using the resolution specified in the Properties dialog
of the Predictions folder (Default Resolution).

relationship already exists and the Delete existing neighbours option is not selected. In this
case, Atoll displays a warning message in the Event Viewer indicating that the constraint
on the forbidden neighbour will be ignored by the algorithm because the neighbour already
exists.
Symmetric neighbour relations are only added to the neighbour lists if the neighbour lists
are not already full. Thus, if the cell B is a neighbour of the cell A, but cell A is not a
neighbour of the cell B, there can be two possibilities:
i.
There is space in the cell B neighbour list: cell A will be added to the list. It will be the
last one.
ii. The cell B neighbour list is full: Atoll will not include cell A in the list and will remove the
symmetric relation by deleting cell B from the cell A neighbour list.
8.7.2
If you select Force exceptional pairs and Force symmetry options, Atoll considers the
constraints between exceptional pairs in both directions so as to respect the symmetric
relation. On the other hand, if a neighbour relation is forced in one direction and forbidden
in the other, symmetry cannot be respected. In this case, Atoll displays a warning message
in the Event Viewer.
In the results, Atoll displays only the cells for which it finds new neighbours. Therefore, if a

TBA cells
Neighbours of TBA cells marked as exceptional pair, adjacent, or symmetric
Neighbours of TBA cells that satisfy coverage conditions
8.8
Scrambling Code Allocation

Downlink scrambling codes enable mobile to distinguish one cell from another. In TD-SCDMA, there are 128 scrambling
codes (or P-CCPCH midamble codes) distributed in 32 clusters of 4 codes each. A different DL synchronisation code, or
SYNC_DL code, is assigned to each cluster. Scrambling codes are numbered from 0 to 127, and SYNC_DL codes from
0 to 31.
Depending on the options you select for automatic allocation of scrambling and SYNC_DL codes, Atoll takes into account
either all the cells of TBC transmitters, or only cells of active and filtered transmitters located inside the computation zone.
Atoll calculates a scrambling code and a SYNC_DL code to all these cells. But, it allocates scrambling codes and
SYNC_DL codes only to TBA cells (cells to be allocated). TBA cells are the cells that fulfill the following conditions:
Forsk 2009
They are active

They satisfy the filter criteria applied to the Transmitters folder
AT271_TRG_E6
417

Furthermore, if there are transmitters that support the N-frequency mode among the TBC transmitters of your network, the
scrambling code allocation also considers the master and slave carrier allocations.
Note:
8.8.1
8.8.1.1
Allocation Constraints and Options

The scrambling code and SYNC_DL code allocation algorithm can take into account following constraints:
1. Neighbour relations between cells
You may consider:
-
Notes:
In the context of the scrambling code allocation, the term "neighbours" refers to intra-carrier
neighbours.
different scrambling codes to the TD-SCDMA neighbours of a GSM transmitter. In order to
consider inter-technology neighbour relations in the scrambling code allocation, you must
make the Transmitters folder of the GSM .atl document accessible in the TD-SCDMA .atl
document. For information on making links between GSM and TD-SCDMA .atl documents,
see the User Manual.
2. The scrambling code reuse distance

Reuse Distance: It is a constraint on the allocation of scrambling codes. The same scarmbling code or SYNC_DL
code cannot be allocated to two sites that are not farther apart than the reuse distance. Scrambling code reuse
distance can be defined for each cell in the cell properties. If this value is not defined, Atoll uses the default reuse
distance defined in the Automatic Scrambling Code and SYNC_DL code Allocation dialog. The reuse distance
constraint is used for clustered and distributed per cell allocation strategies.
3. The carrier for which you want to perform the automatic allocation
Carrier: You can select "All" or a specific carrier. If you select "All", Atoll allocates the same scrambling code to
each carrier of a transmitter.
4. The number of scrambling codes per SYNC_DL code
Each SYNC_DL code corresponds to a group of scrambling codes as defined in 3GPP specifications. 3GPP specifications define 32 SYNC_DL codes with 4 corresponding scrambling codes each (SYNC_DL codes are
numbered from 0 to 31). However, it is possible to define a different value (e.g. if you set the number of scrambling
codes per SYNC_DL codes to 2, scrambling codes will be distributed among 64 SYNC_DL codes).
5. Atoll can use a maximum of codes
Use a Maximum of Codes: If you choose to use a maximum of codes, Atoll will try to spread the allocated spectrum
of scrambling codes as much as possible.
6. Existing allocation
Delete All Codes: If you select this option, Atoll will delete any existing scrambling code allocation and perform a
fresh allocation. Otherwise, Atoll keeps the existing allocation.
8.8.1.2
Allocation Strategies
You can choose from the following four allocation strategies:
418
Clustered: The purpose of this strategy is to choose for a group of mutually constrained cells, scrambling codes
among a minimum number of clusters. In this case, Atoll will preferentially allocate all the codes within the same
cluster.
Distributed per Cell: This strategy consists in using as many clusters as possible. Atoll will preferentially allocate
codes from different clusters.
One SYNC_DL code per site: This strategy allocates one cluster, i.e., one SYNC_DL code, per site, then one
scrambling code from the cluster to each cell of the site. When all the clusters have been allocated but there are
still sites remaining, Atoll reuses the clusters as far as possible at another site.
AT271_TRG_E6
Forsk 2009

Notes:
Same carriers must be assigned different scrambling codes. Different carriers of the same
site can be assigned the same scrambling code. Therefore, cells of a transmitter (i.e.
different carriers) are assigned the same scrambling code, if the scrambling code domains
associated with the carriers have a common cluster or enough codes in one cluster.
Distributed per site: This strategy allocates a group of adjacent clusters, i.e., consecutive SYNC_DL codes, to
each site, then one cluster, or SYNC_DL code, to each transmitter on the site according to its azimuth, and finally
one scrambling code from each cluster to each cell of each transmitter. The number of adjacent clusters, or consecutive SYNC_DL codes, depends on the number of transmitters per site. When all the sites have been allocated
adjacent clusters, and there are still sites remaining to be allocated, Atoll reuses the adjacent clusters as far as
possible at another site.
In the Results table, Atoll only displays scrambling codes and SYNC_DL codes allocated to TBA cells.
8.8.1.3
Allocation Process
and "First Order"),
Additional constraints are considered when:
The neighbour cells cannot share the same cluster (for the "Distributed per site" allocation strategy only).
and the cost of the scrambling code plan. During the allocation, Atoll tries to assign different scrambling codes to the TBA
cell and its near cells. If it respects all the constraints, the cost of the scrambling code plan is 0. When a cell has too many
constraints and there are not anymore scrambling codes available, Atoll breaks the constraint with the lowest cost so as
to generate the scrambling code plan with the lowest cost. For information on the cost generated by each constraint, see
8.8.1.3.1

Strategies: Clustered and Distributed per Cell

Atoll processes TBA cells according to their priority. It allocates scrambling codes starting with the highest priority cell and
its near cells, and continuing with the lowest priority cells not allocated yet and their near cells. For information on calculating cell priority, see "Cell Priority" on page 420.
Strategy: One SYNC_DL Code per Site

Atoll assigns a cluster, i.e., a SYNC_DL code, to each site, starting with the highest priority site and its near sites, and
continuing with the lowest priority sites not allocated yet and their near sites. When all the clusters have been allocated
but there are still sites remaining, Atoll reuses the clusters at the other sites. When the Reuse Distance option is selected,
the algorithm reuses the clusters as soon as the reuse distance is exceeded. Otherwise, when the option is not selected,
the algorithm tries to assign reused clusters as spaced out as possible.
Then, Atoll allocates a scrambling code from the cluster to each cell located on the sites (codes belong to the assigned
clusters). It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not allocated yet
and their near cells.

Atoll assigns a group of adjacent clusters, i.e., SYNC_DL codes, to each site, starting with the highest priority site and its
near sites, and continuing with the lowest priority sites not allocated yet and their near sites. When all the sites have been
allocated adjacent clusters, and there are still sites remaining to be allocated, Atoll reuses the adjacent clusters at other
sites. When the Reuse Distance option is selected, the algorithm reuses the clusters as soon as the reuse distance is
exceeded. Otherwise, when the option is not selected, the algorithm tries to assign reused clusters as spaced out as possible.
Then, Atoll assigns each cluster of the group to each transmitter of the site according to the transmitter azimuth and
selected neighbourhood constraints (options "Neighbours in Other Clusters" and "Secondary Neighbours in Other Clus-
Forsk 2009
AT271_TRG_E6
419

ters"). Then, Atoll allocates a scrambling code to each cell located on the transmitters (codes belong to the assigned clusters). It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not allocated yet and
their near cells.
For information on calculating site priority, see "Site Priority" on page 423. or information on calculating cell priority, see
Determination of Groups of Adjacent Clusters

In order to determine the groups of adjacent clusters to be used, Atoll:
Defines theoretical groups of adjacent clusters, independent of the defined domain, considering the 128 scrambling codes available and 4 codes per cluster.
Starts the distribution of clusters to groups from the cluster 0
Takes into account the maximum number of transmitters per site in order to determine the number of clusters in
each group
Determines the total number of groups
If the number of scrambling codes per cluster is set to 4 and the maximum number of transmitters per site in the network
is 3, the theoretical groups of adjacent clusters will be:
Group 1
Group 2
Group 3
Group 4
Cluster 0
Cluster 3
Cluster 6
Cluster 9
Cluster 1
Cluster 4
Cluster 7
Cluster 10
Cluster 2
Cluster 5
Cluster 8
Cluster 11
...
Group 11
Cluster 30
...
Cluster 31
If no domain is assigned to cells, Atoll can use all these groups for the allocation. On the other hand, if a domain is used,
Atoll compares adjacent clusters actually available in the assigned domain with the theoretical groups and only keeps adjacent clusters common with the theoretical groups.
If we have a domain comprising 12 clusters: clusters 1 to 8 and clusters 12 to 15. In this case, Atoll will use the following
groups of adjacent clusters:
Group 2 with cluster 3, 4 and 5

The clusters 1, 2 and 15 will not be used.

If a domain does not contain any adjacent clusters, Atoll displays a warning message in the Event Viewer.
8.8.1.3.2
In case you have a multi-carrier network and you run the scrambling code allocation on all the carriers, the allocation order
changes. It is no longer based on the cell priority but depends on the transmitter priority. All transmitters which have
constraints with the studied transmitter will be referred to as near transmitters.
In case of a "Per cell" strategy (Clustered and Distributed per cell), Atoll starts scrambling code allocation with the highest
priority transmitter and its near transmitters and continues with the lowest priority transmitters not allocated yet and their
near transmitters. The same scrambling code is assigned to each cell of the transmitter.
In case of the "One SYNC_DL code per site" strategy, Atoll assigns a cluster, i.e., a SYNC_DL code, to each site and then,
allocates a scrambling code to each transmitter. It starts with the highest priority transmitter and its near transmitters and
continues with the lowest priority transmitters not allocated yet and their near transmitters. The same scrambling code is
assigned to each cell of the transmitter.
In case of the "Distributed per site" strategy, Atoll assigns a group of adjacent clusters, i.e., SYNC_DL codes, to each site,
then a cluster to each transmitter and finally, allocates a scrambling code to each transmitter. It starts with the highest priority transmitter and its near transmitters and continues with the lowest priority transmitters not allocated yet and their near
transmitters. The same scrambling code is assigned to each cell of the transmitter.
For information on calculating transmitter priority, see "Transmitter Priority" on page 423.
Note:
alphanumeric order.
8.8.1.4
8.8.1.4.1
Cell Priority
Scrambling code allocation algorithm in Atoll allots priorities to cells before performing the actual allocation. Priorities
assigned to cells depend upon how much constrained each cell is and the cost defined for each constraint. A cell without
any constraint has a default cost, C , equal to 0. The higher the cost on a cell, the higher the priority it has for the scrambling
code allocation process.
There are six criteria employed to determine the cell priority:
420
Scrambling Code Domain Criterion
AT271_TRG_E6
Forsk 2009

The cost due to the domain constraint, C i Dom , depends on the number of scrambling codes available for the allocation.
The domain constraint is mandatory and cannot be broken.
When no domain is assigned to cells, 128 scrambling codes are available and we have:
C i Dom = 0
When domains of scrambling codes are assigned to cells, each unavailable scrambling code generates a cost. The higher
the number of codes available in the domain, the less will be the cost due to this criterion. The cost is given as:
C i Dom = 128 Number of scrambling codes in the domain
Distance Criterion
Cj Dist i
C i Dist =
Where
C i EP =
cEP i j
j
Where

Ci N =
j

C j N1 i = I j c N1
Where
Because two first order neighbours must not have the same scrambling code, Atoll considers the cost created by two first
C j N1 i + C j N1 i
C j j N1 i = ---------------------------------------------------------2
Forsk 2009
AT271_TRG_E6
421

Where
Because two second order neighbours must not have the same scrambling code, Atoll considers the cost created by two
second order neighbours to be each other.
C k N2 i + C k N2 i
C k k N2 i = -----------------------------------------------------------2
C l N3 i = Max j
c N3
Where
Because two third order neighbours must not have the same scrambling code, Atoll considers the cost created by two third
C l N3 i + C l N3 i
C l l N3 i = ---------------------------------------------------------2
Note:

And
C i N 2G =
cN
2G
j Tx 2G
Where
cN
2G
Cluster Criterion
When the "Distributed per Site" allocation strategy is used, you can consider additional constraints on allocated clusters
(one cell, its first order neighbours and its second order neighbours must be assigned scrambling codes from different clusters). In this case, the constraint level of any cell i depends on the number of first and second order neighbours, j and k.
The total cost due to the cluster constraint is given as:
C i Cluster =
Cj N1 i cCluster + Ck N2 i cCluster
j
Where
c Cluster is the cost of the cluster constraint. This value can be defined in the Constraint Cost dialogue.
With
C i U = C i Dist + C i EP + C i N + C i N 2G + C i Cluster
422
AT271_TRG_E6
Forsk 2009
8.8.1.4.2
In case you have a multi-carrier network and you run scrambling code allocation on "all" the carriers, Atoll allots priorities
to transmitters. Priorities assigned to transmitters depend on how much constrained each transmitter is and the cost
defined for each constraint. The higher the cost on a transmitter, the higher the priority it has for the scrambling code allocation process.
as:
With C Tx U =

i
Tx
i Tx
8.8.1.4.3
Site Priority
In case of "Per Site" allocation strategies (One SYNC_DL code per Site and Distributed per Site), Atoll allots priorities to
sites. Priorities assigned to sites depend on how much constrained each site is and the cost defined for each constraint.
The higher the cost on a site, the higher the priority it has for the scrambling code allocation process.
Let us consider a site S with three transmitters; each of them has two cells using carriers 0 and 1. The cost due to
constraints on the site is given as:
With C S U =

Tx
S
Tx S
Here, the domain considered for the site is the intersection of domains available for transmitters of the site. The domain
constraint is mandatory and cannot be broken.
8.8.2
IScrambling Code Allocation Example
8.8.2.1

In order to understand the differences between the different allocation strategies and the behaviour of algorithm when
using a maximum of codes or not, let us consider the following sample scenario:
Figure 8.15: Scrambling Code Allocation Example

Let Site0, Site1, Site2, and Site3 be four sites, with 3 transmitters each using carrier 0, to whom scrambling codes have
to be allocated out of 6 clusters of 4 scrambling codes. This implies that the domain of scrambling codes for the four sites
is from 0 to 23 (cluster 0 to cluster 5). The reuse distance is supposed to be less than the inter-site distance. Only co-site
neighbours exist.
The following section shows the results of each combination of options with explanations where necessary.
Forsk 2009
AT271_TRG_E6
423
8.8.2.1.1
Strategy: Clustered
Since the restrictions of neighbourhood only apply to co-sites and, in our case, the distances between sites are greater
than the reuse distance, every cell has the same priority. Allocation is performed in an alphanumeric order.
8.8.2.1.2
Without "Use a Maximum of Code"
With "Use a Maximum of Code"
Atoll starts allocating the codes from the start of cluster 0

at each site.
As it is possible to use a maximum of codes, Atoll starts

allocation at the start of a different cluster at each site.
When a cluster is reused, and there are non allocated
codes left in the cluster, Atoll first allocates those codes
before reusing the already used ones.
Strategy: Distributed per Cell

Since the restrictions of neighbourhood only apply to co-sites and, in our case, the distances between sites are greater
than the reuse distance, every cell has the same priority. Allocation is performed in an alphanumeric order.
424
Atoll allocates codes from different clusters to each cell of

the same site. Under given constraints of neighbourhood
and reuse distance, same codes can be allocated to each
sites cells.
Atoll allocates codes from different clusters to each sites

cells. As it is possible to use a maximum of codes, Atoll
allocates the codes so that there is least repetition of
codes.
AT271_TRG_E6
Forsk 2009
8.8.2.1.3
Strategy: One SYNC_DL Code per Site

Since the restrictions of neighbourhood only apply to co-sites, therefore, every site has the same priority. Cluster allocation
to sites is performed in an alphanumeric order.
8.8.2.1.4
In this strategy, a cluster of codes is limited to be used at

just one site at a time unless all codes and clusters have
been allocated and there are still sites remaining to be
allocated. In this case, Atoll reuses the cluster as far as
possible at another site.
When it is possible to use a maximum of codes, Atoll can

allocate different codes from a reused cluster at another
site.

Since the restrictions of neighbourhood only apply to co-sites, therefore, every site has the same priority. Cluster allocation
to sites is performed in an alphanumeric order.
8.8.2.2
A group of adjacent clusters is allocated to one site at a

time, unless all the codes and groups of adjacent clusters
have been allocated but there are still sites remaining to
be allocated. In this case (here only one group of adjacent
clusters 0, 1, and 2 is available), Atoll reuses the group as
far as possible at another site.
When it is possible to use a maximum of codes, Atoll can

allocate different codes from a reused groups of adjacent
clusters at another site.
Multi Carrier Network

If you have a multi carrier network, i.e., transmitters with more than one cells using different carriers, and you run scrambling code allocation on "all" the carriers, Atoll allocates the same scrambling code to each carrier of a transmitter.
Let Site0, Site1, Site2, and Site3 be four sites with 3 cells using carrier 0 and 3 cells using carrier 1. Scrambling codes
have to be allocated out of 6 clusters consisted of 4 scrambling codes. This implies that the domain of scrambling codes
for the four sites is from 0 to 23 (cluster 0 to cluster 5). The reuse distance is supposed to be less than the inter-site
Forsk 2009
AT271_TRG_E6
425

distance. Only co-site neighbours exist. Every site has the same priority and the cluster allocation to sites is performed in
an alphanumeric order.
Atoll allocates one cluster at each site and then, one code to each transmitter. Then, the same code is given to each cell
of the transmitter.
Figure 8.16: Scrambling Code Allocation to All Carriers
8.9
Automatic GSM/TD-SCDMA Neighbour Allocation

It is possible to automatically calculate and allocate neighbours between GSM and TD-SCDMA networks. In Atoll, it is
called inter-technology neighbour allocation.
When the TD-SCDMA coverage is not continuous. In this case, the TD-SCDMA coverage is extended by TDSCDMA to GSM handovers.
In order to balance traffic and service distribution between both networks.
Atolls automatic inter-technology neighbour allocation algorithm takes into account both cases.
An .atl document containing the GSM network, GSM.atl, and another one containing the TD-SCDMA network, TDSCDMA.atl,
An existing link on the Transmitters folder of GSM.atl into TD-SCDMA.atl.
transmitters of GSM.atl are potential neighbours. The TD-SCDMA cells, in TD-SCDMA.atl, to be allocated neighbours are
called TBA cells which fulfill following conditions:
They are active

They satisfy the filter criteria applied to Transmitters folder
or one of its subfolders.
Only TD-SCDMA TBA cells can be assigned neighbours.
8.9.1

The inter-transmitter distance

The maximum number of neighbours
Allocation options
The selected allocation strategy
We assume we have a TD-SCDMA reference cell, A, and a GSM candidate neighbour transmitter, B.
426
AT271_TRG_E6
Forsk 2009
8.9.1.1

1. The distance between the TD-SCDMA reference cell and the GSM neighbour must be less than the user-defined
maximum inter-site distance. If the distance between the TD-SCDMA reference cell and the GSM neighbour is
greater than this value, then the candidate neighbour is discarded.
Note:
further information on inter-transmitter distance calculation, please refer to "Calculation of
Inter-Transmitter Distance" on page 429.

Force co-site cells as neighbours: It enables you to automatically include GSM transmitters located on the same
site than the reference TD-SCDMA cell in the candidate neighbour list. This option is automatically selected.
may force/forbid a GSM transmitter to be candidate neighbour of the reference TD-SCDMA cell.
neighbours and carries out a new neighbour allocation. If not selected, existing neighbours are kept.
3. Atoll lists all candidate neighbours and sorts them by priority so as to eliminate some of them from the neighbour
list if the maximum number of neighbours to be allocated to each cell is exceeded. The candidate neighbour priority depends on the neighbourhood cause. Priority assigned to each neighbourhood cause is listed in the table
below (1 is a higher than 2, and so on).
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

distance conditions
In the Results part, Atoll provides the list of neighbours, the number of neighbours, and the maximum number of neighbours allowed for each cell. In addition, it indicates the reason of allocation for each neighbour. Therefore, a neighbour
may be marked as exceptional pair or co-site. If the neighbour is not forced but fulfills distance conditions, Atoll displays
the distance from the reference cell. Finally, if cells have previous allocations in the list, neighbours are marked as existing.
8.9.1.2

1. The distance between the TD-SCDMA reference cell and the GSM neighbour must be less than the user-defined
maximum inter-site distance. If the distance between the TD-SCDMA reference cell and the GSM neighbour is
greater than this value, then the candidate neighbour is discarded.
Note:
The inter-transmitter distance is not effected by the azimuths. Only the geographical intertransmitter distance is considered.

Force co-site cells as neighbours: It enables you to automatically include GSM transmitters located on the same
site than the reference TD-SCDMA cell in the candidate neighbour list. This option is automatically selected.
may force/forbid a GSM transmitter to be candidate neighbour of the reference TD-SCDMA cell.
neighbours and carries out a new neighbour allocation. If not selected, existing neighbours are kept.
3. There must be an overlapping zone ( S A S B ) with a given cell edge coverage probability.
Forsk 2009
AT271_TRG_E6
427

-
1st case: SA is the area where the cell A is the best serving cell of the TD-SCDMA network.
- The pilot signal received from A is greater than the minimum pilot signal level and is the highest one.
- The margin is set to 0 dB.
2nd case: The margin is different from 0 dB and SA is the area where:
- The pilot signal level received from A exceeds the user-defined minimum pilot signal level and is within a
margin from the highest signal level.

-
1st case: SB is the area where the cell B is the best serving transmitter of the GSM network.
In this case, the margin must be set to 0 dB.
-
is the highest one.
2nd case: The margin is different from 0 dB and SB is the area where:
SA SB
SA
area. If this percentage is less than the minimum, the candidate neighbour B is discarded.
area.
4. Atoll lists all candidate neighbours and sorts them by priority so as to eliminate some of them from the neighbour
list if the maximum number of neighbours to be allocated to each cell is exceeded. The candidate neighbour priority depends on the neighbourhood cause. Priority assigned to each neighbourhood cause is listed in the table
below (1 is a higher than 2, and so on).
Neighbourhood cause
When
Priority
Existing neighbour

Exceptional pair
Co-site cell

coverage conditions
Therefore, among 15 candidate neighbours, only 8 (those with the highest priority) will be allocated to the reference cell.
In the Results part, Atoll provides the list of neighbours, the number of neighbours, and the maximum number of neighbours allowed for each cell. In addition, it indicates the allocation cause for each neighbour. Therefore, a neighbour may
be marked as exceptional pair or co-site. If the neighbour is not forced but fulfills coverage conditions, Atoll displays the
Notes:
Coverage predictions are not required to perform an automatic neighbour allocation. When
starting an automatic neighbour allocation, Atoll automatically calculates the path loss
matrices, if not found.
dialog of the Predictions folder (Default Resolution).

relationship already exists and the Delete existing neighbours option is not selected when
you start a new allocation. In this case, Atoll displays a warning in the Event Viewer
indicating that the constraint on the forbidden neighbour will ignored by algorithm because
the neighbour already exists.
In Results, Atoll displays only the cells for which it finds new neighbours. Therefore, if a
TBA cell has already reached its maximum number of neighbours before starting a new
8.9.1.3
Appendices
8.9.1.3.1

not selected. If a new TBA cell i is created in TD-SCDMA.atl, you can run the automatic allocation with the Delete existing
neighbours option not selected, in order to allocate neighbours to the new cell i only.
If you change some allocation criteria (e.g., increase the maximum number of neighbours or create a new GSM TBC transmitter) and start a new allocation without selecting the Delete existing neighbours option, Atoll examines the neighbour list
428
AT271_TRG_E6
Forsk 2009

of the TBA cells and checks allocation criteria only if there is still space left in their neighbour lists. A new GSM TBC transmitter can enter the TBA cell neighbour list if allocation criteria are satisfied. It will be the first one in the neighbour list.
8.9.1.3.2
Calculation of Inter-Transmitter Distance

When allocation algorithm is based on distance, Atoll takes into account the real distance and azimuths of antennas in
order to calculate the effective inter-transmitter distance.
Dist CellA CellB = D 1 + x cos x cos
Where x = 0.5% so that the maximum variation in D does not to exceed 1%. D is stated in m.
Figure 8.17: Inter-Transmitter Distance Computation

The formula above implies that two cells facing each other have a smaller effective distance than the real physical
distance. It is this effective distance that is taken into account rather than the real distance.
This formula is not used when allocation algorithm is based on coverage overlapping. In this case, the actual inter-transmitter distance is considered.
Forsk 2009
AT271_TRG_E6
429
430
AT271_TRG_E6
Forsk 2009
Chapter 9
WiMAX BWA Networks
simulations and coverage predictions available in WiMAX BWA projects.
Atoll
432
AT271_TRG_E6
Forsk 2009
Chapter 9: WiMAX BWA Networks
WiMAX BWA Networks

This chapter describes all the calculations performed in Atoll WiMAX BWA documents, i.e, WiMAX 802.16d and WiMAX
802.16e. The titles of sections that present 16d- or 16e-specific algorithms include the name of the technology for you to
recognize them easily.
The first part of this chapter lists all the input parameters in the WiMAX BWA documents, their significance, location in the
Atoll GUI, and their usage. It also contains the lists of the formulas used during the calculations.
The second part describes all the calculation processes, i.e., signal level coverage predictions, point analysis calculations,
signal quality coverage predictions, calculations on subscriber lists, Monte Carlo simulations, and traffic C/(I+N)-based
coverage predictions. The calculation algorithms used by these calculation processes are available in the next part.
The third part describes all the calculation algorithms used in all the calculations. These algorithms include the calculation
of signal levels, noise, and interference for downlink and uplink considering the effects of smart antennas, power control,
subchannelisation, MIMO etc., and the radio resource management algorithms used by the different available schedulers.
The third part also describes Forsks optimum beamformer model in detail.
If you are new to WiMAX, you can also see "Glossary of WiMAX Terms" on page 514 for information on WiMAX terms and
concepts, especially in the context of their user in Atoll.
Important:
All the calculations are performed on TBC (to be calculated) transmitters. For the definition
of TBC transmitters please refer to "Path Loss Matrices" on page 74.
A cell refers to a transmitter-carrier (TX-c) pair. The cell being studied during a calculation
is referred to as TXi(ic) in this chapter.
All the calculation algorithms in this section are described for two types of cells.
-
All the calculation algorithms in this section are described for two types of receivers.
-
9.1
A studied cell (represented by the subscript "i") comprising the studied transmitter TXi and
its carrier ic. It is the cell which is currently the focus of the calculation. For example, a victim
cell when calculating the interference it is receiving from other cells.
Other cells (represented by the subscript "j") comprising the other transmitter TXj and its carrier jc. The other cells in the network can be interfering cells (downlink) or the serving cells
of interfering mobiles (uplink).
Mi: A pixel (coverage predictions), subscriber (calculations on subscriber lists), or mobile
(Monte Carlo simulations) covered/served by the studied cell TXi(ic).
Mj: A mobile (Monte Carlo simulations) covered/served by any other cell TXj(jc).

The tables in the following subsections list the input and output parameters, and formulas used in simulations and other
computations.
9.1.1
Input
This table lists the input to computations, coverage predictions, and simulations.
Name
Value
Unit
Description
D Frame
Global parameter
ms
Frame Duration
Choice List: 2, 2.5, 4, 5, 8, 10, 12.5, 20
ms
r CP
Global parameter
None
Cyclic Prefix Ratio

Choice List: 1/4, 1/8, 1/16, 1/32
O Fixed
DL
Global parameter
SD
Fixed time-domain overhead (DL)
UL
Global parameter
SD
Fixed time-domain overhead (UL)
DL
Global parameter
Variable time-domain overhead (DL)
UL
Global parameter
Variable time-domain overhead (UL)
TDD
Global parameter
Ratio of the DL subframe to the entire

frame (TDD only)
Global parameter
ms
Transmit Time Guard (TDD only)
O Fixed
O Variable
O Variable
r DL-Frame
TDD
D TTG
Forsk 2009
AT271_TRG_E6
433

D RTG
TDD
Global parameter
ms
Receive Time Guard (TDD only)
M PC
Global parameter
dB
Uplink power control margin
Global parameter (WiMAX 802.16d)

Permutation zone parameter (WiMAX 802.16e)
None
Number of subchannels per channel

in UL subframe
N SC DL
None
Number of subchannels per channel

in DL subframe
N SCa Total

Frame configuration parameter (WiMAX 802.16e)
None
Total number of subcarriers per

channel (FFT size)
N SCa Preamble
Frame configuration parameter (WiMAX 802.16e)
None
Number of subcarriers used by the

preamble

None
Number of used subcarriers per

channel

None
Number of subcarriers per channel

used for data transfer
N SCa DC
Hard-coded parameter ( N SCa DC = 1 )
None
Number of DC subcarriers per

channel
N SCa Pilot or
Calculation result ( N SCa Pilot = N SCa Used N SCa Data or
None
Number of pilot subcarriers per

channel
None
Number of guard subcarriers per

channel
PZ
N SC UL or N SC UL
PZ
N SCa Used or
PZ
N SCa Used
N SCa Data or
PZ
N SCa Data
PZ
N SCa Pilot
N SCa Guard or
PZ
N SCa Guard
PZ
N SCa Pilot
PZ
N SCa Used
PZ
N SCa Data
Calculation result
( N SCa Guard = N SCa Total N SCa Used N SCa DC or
PZ
N SCa Guard
= N SCa Total
PZ
N SCa Used
N SCa DC )
PZ UL
None
Uplink permutation zone
PZ DL
None
Downlink permutation zone
CNR Req PZ
dB
Required signal quality for accessing

a permutation zone
Speed Max PZ
Km/hr
Speed limit for mobiles trying to

access a permutation zone
d Max PZ
Maximum distance from the

transmitter covered by a permutation
zone
p PZ
None
Permutation zone priority
Segmented permutation zone parameter (WiMAX 802.16e)

f Segment
3 PSG + 2 SSG
f Segment = ---------------------------------------------------15
None
Ratio of channel bandwidth allocated

to each segment of a segmented
frame configuration
W Channel
MHz
Channel bandwidth
First
None
First channel number of the frequency

band
N Channel
Last
None
Last channel number of the frequency

band
F Start FB TDD
MHz
Start frequency of the TDD frequency

band
F Start FB FDD DL
MHz
DL Start frequency of the FDD

frequency band
F Start FB FDD UL
MHz
UL Start frequency of the FDD

frequency band
f Sampling
None
Sampling factor
f ACS FB
dB
Adjacent Channel Suppression Factor
Bearer parameter
None
Bearer index
Mod B
Bearer parameter
None
Modulation used by the bearer
CR B
Bearer parameter
None
Coding rate of the bearer
Bearer parameter
bits/
symbol
Bearer Efficiency
N Channel
434
AT271_TRG_E6
Forsk 2009

TB
Bearer parameter
dB
Bearer selection threshold
1.38 x 10-23
J/K
Boltzmanns constant
290
Ambient temperature

dB
Transmitter noise figure
N Ant TX
None
Number of antennas used for MIMO in

transmission
N Ant RX
None

reception
TX
Antenna parameter
dB
TX

dB
Transmitter loss
N Channel
Cell parameter
None
Cells channel number
P Preamble
Cell parameter
dBm
Preamble power
dB
Traffic power offset
dB
Pilot power offset
dB
Idle pilot power offset
nf
G
L
TX
Cell parameter P Traffic = P Preamble P Traffic in dB

P Traffic
Ratio
P Traffic = 10
P Traffic
-----------------------10
in %
Cell parameter P Pilot = P Preamble P Pilot in dB

P Pilot
Ratio
P Pilot
= 10
P Pilot
------------------10
in %
Cell parameter P Idle Pilot = P Preamble P Idle Pilot in dB

P Idle Pilot
Ratio
P Idle Pilot = 10
P Idle Pilot
--------------------------------10
in %
TL DL
Cell parameter
Downlink traffic load
TL UL
Cell parameter
Uplink traffic load
NR UL
Cell parameter
dB
Uplink noise rise
N Users Max
Cell parameter
None
Maximum number of users per cell
SU
Cell parameter
Segmentation usage ratio
AU
Cell parameter
AAS usage ratio
T AMS
Cell parameter
dB
Adaptive MIMO switch threshold
PI
Cell parameter
None
Preamble index
T Preamble
Cell parameter
dB
Preamble quality threshold
D Reuse
Cell parameter
Channel and preamble index reuse

distance
QoS
Service parameter
None
QoS class of the service
Service parameter
None
Service priority
B DL Highest
Service parameter
None
Highest bearer used by a service in

the downlink
B UL Highest
Service parameter
None
Highest bearer used by a service in

the uplink
f Act
UL
Service parameter
Uplink activity factor for voice services
f Act
DL
Service parameter
Downlink activity factor for voice

services
TPD Min UL
Service parameter
kbps
Minimum throughput demand in the

uplink
TPD Min DL
Service parameter
kbps
Minimum throughput demand in the

downlink
TPD Max UL
Service parameter
kbps
Maximum throughput demand in the

uplink
Forsk 2009
AT271_TRG_E6
435

TPD Max DL
Service parameter
kbps
Maximum throughput demand in the

downlink
TP Offset
Service parameter
kbps
Throughput offset
f TP Scaling
Service parameter
Scaling factor
L Body
Service parameter
dB
Body loss
P Min
Terminal parameter
dBm
P Max
Terminal parameter
dBm
nf
Terminal parameter
dB
Terminal noise figure
Terminal parameter
dB
Terminal antenna gain
Terminal parameter
dB
Terminal loss
N Ant TX
Terminal parameter
None

transmission
N Ant RX
Terminal parameter
None

reception
G STTD
UL
Clutter parameter
dB
Uplink STTD/MRC gain offset
G STTD
DL
Clutter parameter
dB
Downlink STTD/MRC gain offset
f SM Gain
Clutter parameter
None
Spatial Multiplexing gain factor
L Indoor
Clutter parameter
dB
Indoor loss
L Path
dB
Path loss
M Shadowing Model

standard deviation
dB

Only used in coverage predictions
M Shadowing C I
Result calculated from cell edge coverage probability and C/I standard
deviation
dB
C/I Shadowing margin

Only used in coverage predictions
BTS
BTS parameter
Rho factor
% of BTS signal correctly transmitted
n0
Calculation result ( 10 Log 10 K T 1000 = 174 dBm/Hz )
dBm/Hz
Power spectral density of thermal

noise
QoS
Scheduler parameter
QoS class bias factor
G SM
MIMO configuration parameter
None
Maximum spatial-multiplexing gain for

a given number of transmission and
reception antennas
G STTD
UL
dB
Uplink STTD/MRC gain for a given

number of transmission and reception
antennas
DL
dB
Downlink STTD/MRC gain for a given

number of transmission and reception
antennas
f Bias
Max
G STTD
9.1.2
Co- and Adjacent Channel Overlaps Calculation
Name
Value
TX i ic
TX i ic
TX i ic
F End
TX i ic TX j jc
W CCO
TX i ic
TX i ic
TX i ic
TX i ic TX j jc
436
TX i ic
TX i ic
F Start FB + N Channel W Channel + W Channel

TX j jc
TX i ic
Min FEnd F End
TX j jc
TX i ic
Max FStart F Start
Description
MHz
Start frequency for the channel

number assigned to a cell
MHz
End frequency for the channel number

assigned to a cell
MHz
Co-channel overlap bandwidth
None
Co-channel overlap ratio
MHz
Bandwidth of the lower-frequency

adjacent channel overlap
TX i ic TX j jc
W CCO
--------------------------------------TX i ic
W Channel
TX i ic TX j jc
r CCO
W ACO
TX i ic
F Start FB + N Channel W Channel
F Start
Unit
TX j jc
TX i ic
TX j jc
TX i ic
TX i ic
Min F End F Start Max F Start F Start W Channel
AT271_TRG_E6
Forsk 2009

TX i ic TX j jc
W ACO
L
--------------------------------------TX i ic
W Channel
TX i ic TX j jc
r ACO
L
TX i ic TX j jc
W ACO
TX j jc
TX i ic
Min F End F End
TX i ic
TX j jc
TX i ic
+ W Channel Max F Start F End
None
Lower-frequency adjacent channel

overlap ratio
MHz
Bandwidth of the higher-frequency

adjacent channel overlap
None
Higher-frequency adjacent channel

overlap ratio
None
Adjacent channel overlap ratio
None
FDD TDD overlap ratio
None
Total overlap ratio
TX i ic TX j jc
W ACO
H
--------------------------------------TX i ic
W Channel
TX i ic TX j jc
r ACO
H
TX i ic TX j jc
TX i ic TX j jc
r ACO
r ACO
TX i ic TX j jc
+ r ACO
TDD
TX i ic TX j jc
r FDD TDD
r DL Frame
-------------------------- if interferer uses a TDD frequency band and victim uses
100
an FDD frequency band
1 other wise
TX ic
i
f ACS FB
---------------------------
TX i ic TX j jc
TXi ic TXj jc TX i ic TX j jc
10
+ r ACO
10
r CCO
r FDD TDD
TX i ic
TX i ic TX j jc
rO
TX j jc
if W Channel W Channel
TX ic
i
f ACS FB
TX ic
--------------------------- TX ic TX jc W i
TXi ic TXj jc TXi ic TXj jc
10
i
j
Channel
+ r ACO
10
---------------------- r CCO
r FDD TDD
TX j jc
W Channel
TX i ic
TX j jc
9.1.3
Preamble Signal Quality Calculations
9.1.3.1
Preamble Signal Level Calculation
Name
TX i ic
C Preamble
Value
TX i ic
EIRP Preamble L Path M Shadowing Model L Indoor + G

L
Mi
Mi
Without smart antenna equipment: P Preamble + G
Forsk 2009
TX i
Received preamble signal level
dBm
Preamble EIRP of a cell
dB
Path loss
dB
Total losses
TX i
With smart antenna equipment:

TX i ic
P Preamble
+G
TX i
TX i
+ 10
TX i
Log E SA
TX i
L Model + L Ant
L Path + L
Mi
dBm
Mi
Mi
L Path
L Total
Description
L Ant L Body
TX i ic
TX i ic
EIRP Preamble
Unit
TX i
Mi
+ L Indoor + M Shadowing Model G
TX i
+L
Mi
Mi
+ L Ant + L Body
AT271_TRG_E6
437
9.1.3.2
Preamble Noise Calculation
Name
Value
Unit
Description
dBm
Preamble thermal noise for a cell
None
Preamble segmenting factor
dBm
Preamble noise for a cell
Unit
Description
dBm
Total interference generated by an

interfering cell
dB
Interference reduction factor due to

the co- and adjacent channel overlap
dB

segmentation (WiMAX 802.16e)
Unit
Description
dB
Preamble C/N for a cell
Unit
Description
dB
Preamble C/(I+N) for a cell
dBm
Preamble Total Noise (I+N) for a cell
TX i ic
TX ic
N SCa Preamble
i
n 0 + 10 Log 10 F Sampling --------------------------------------TX i ic
N SCa Total
TX i ic
With Segmentation (WiMAX 802.16e):
n 0 Preamble
TX i ic
N SCa Preamble
---------------------------------------TX i ic
N SCa Total
TX ic
i
n 0 + 10 Log 10 F Sampling
1
--3
Preamble
f Segment
TX i ic
TX i ic
n Preamble
9.1.3.3
n 0 Preamble + nf
Mi
Preamble Interference Calculation
Name
Value
TX j jc
TX j jc
I Preamble
TX i ic TX j jc
TX i ic TX j jc
+ f Seg
TX i ic TX j jc
10 Log r O
TX i ic TX j jc
TX i ic TX j jc
10 Log 10 p Collision
f Seg
9.1.3.4
TX i ic TX j jc
C Preamble + f O
fO
Preamble C/N Calculation
Name
Value
TX i ic
TX i ic
CNR Preamble
9.1.3.5
TX i ic
C Preamble n Preamble
Preamble C/(I+N) Calculation
Name
TX i ic
CINR Preamble
Value
TX jc
j
I Preamble
TX i ic
TX i ic
------------------------C Preamble 10 Log

10
n Preamble M Shadowing C I
10
All TXj jc
TX jc
j
I +
Preamble
f Segment
TX ic
i
I Preamble
n Preamble
---------------------------------------------------
10
10
10 Log
10
+ 10
All TX j jc
TX i ic
N Preamble
9.1.4
Traffic and Pilot Signal Quality Calculations
9.1.4.1
Traffic and Pilot Signal Level Calculation (DL)
Name
TX i ic
C Traffic
438
Value
TX i ic
EIRP Traffic L Path M Shadowing Model L Indoor + G

L
Mi
Mi
Mi
Unit
Description
dBm
Received traffic signal level
Mi
L Ant L Body
AT271_TRG_E6
Forsk 2009
TX i ic
C Pilot
TX i ic
EIRP Pilot
L
Mi
Mi
L Path M Shadowing Model L Indoor + G
dBm
Received pilot signal level
TX i
dBm
Traffic EIRP of a cell
TX i
dBm
Pilot EIRP of a cell
dBm
Traffic transmission power of a cell
dBm
Pilot transmission power of a cell
Unit
Description
dBm
Thermal noise for a cell
None
Segmenting factor
dBm
Downlink noise for a cell
Mi
L Ant L Body
TX i ic
TX i ic
EIRP Traffic
P Traffic + G
TX i ic
TX i ic
EIRP Pilot
P Pilot
TX i ic
+G
TX i
TX i
L
L
TX i ic
TX i ic
TX i ic
TX i ic
P Preamble P Traffic
P Traffic
TX i ic
P Preamble P Pilot
P Pilot
9.1.4.2
Mi
Noise Calculation (DL)
Name
Value
N SCa Used
TX i ic
WiMAX 802.16d: n 0 + 10 Log 10 F Sampling ------------------------------
N SCa Total
M
TX i ic
n 0 DL
PZ DL
N SCa Used
TXi ic
-
WiMAX 802.16e: n 0 + 10 Log 10 F Sampling ----------------------------TX i ic
N SCa Total
With Segmentation (WiMAX 802.16e):

M
PZ DL
N SCa Used
TX i ic
- f Segment
n 0 + 10 Log 10 F Sampling ----------------------------TX i ic
N SCa Total
3 PSG + 2 SSG
---------------------------------------------------15
f Segment
TX i ic
TX i ic
n DL
9.1.4.3
n 0 DL + nf
Mi
Traffic and Pilot Interference Calculation (DL)
Name
Value
TX jc
j
Unit
Description
dBm
Total interference generated by an

interfering cell
TX jc
j
I Idle
I Non AAS
------------------
-------------------------10
10
Monte Carlo Simulations: 10 Log 10
+ 10
without
TX jc
j
TX j jc
I Total
I AAS
----------------10
smart antennas, or 10 Log 10
with smart antennas
TX jc
j
TX jc
j
TX jc
j
I Idle
I AAS
I Non AAS
-----------------------------------
-------------------------10
10
10
+ 10
+ 10
Coverage Predictions: 10 Log 10
TX j jc
I Traffic
TX j jc
I Pilot
TX j jc
EIRP Traffic L Path L Indoor + G

TX j jc
EIRP Pilot
TX j jc
TX j jc
EIRP Traffic
P Traffic + G
TX j jc
TX j jc
EIRP Pilot
TX j jc
I Non AAS
Forsk 2009
L Path L Indoor + G
P Pilot
+G
TX j
TX j
Mi
Mi
Mi
dBm
Traffic interference power of an

interfering cell
Mi
Mi
dBm
Pilot interfering power of an interfering

cell
TX j
dBm
Traffic EIRP of an interfering cell
TX j
dBm
Pilot EIRP of an interfering cell
dBm
Interference from the loaded part of

the frame transmitted using the
transmitter antenna of an interfering
cell
Mi
L
L
Mi
Mi
L Ant L Body
L Ant L Body
TX j jc
TXj jc
10 Log TL DL
1 AU
TX j jc
TXj jc
I Pilot
I Traffic
TX j jc
TX jc
TX j jc
------------------ N j
N SCa Data
----------------10
SCa Used N SCa Data
- + 10 10 ----------------------------------------------------------------
---------------------------- 10
TX j jc
TX j jc
N SCa Used
N SCa Used
AT271_TRG_E6
439

TX j jc
TX j jc
I AAS
EIRP AAS
L Path L Indoor + G
TX j jc
TX j jc
EIRP AAS
TX j jc
I Idle Pilot
P Traffic + G
TX j jc
EIRP Idle Pilot
TX j
Mi
L Path L Indoor + G
TX j jc
TX j jc
EIRP Idle Pilot
P Idle Pilot + G
TX j
Mi
Mi
Mi
L Ant L Body
TX j
Mi
Mi
Mi
L Ant
Mi
L Body
TX j
dBm
Interference power of an interfering

cell transmitted using smart antenna
dBm
Traffic EIRP of an interfering cell using

smart antenna
dBm
Interference from empty part of the

frame transmitted using the
cell
dBm
Idle pilot EIRP of an interfering cell
dBm
Interference from the empty part of the

frame transmitted using the
cell
dB

dB

segmentation (WiMAX 802.16e)
Unit
Description
dB
Traffic C/N for a cell
dB
Pilot C/N for a cell
Unit
Description
dB
Traffic C/(I+N) for a cell
dB
Pilot C/(I+N) for a cell
TX jc
j
TX j jc
I Idle
I Idle Pilot
TX jc
TX j jc
- N j
TX j jc
-------------------------
10
SCa Used N SCa Data
-
10 Log 1 TL DL
10
---------------------------------------------------------------TX j jc
N SCa Used
TX i ic TX j jc
TX i ic TX j jc
10 Log r O
fO
TX i ic TX j jc
TX i ic TX j jc
10 Log 10 p Collision
f Seg
9.1.4.4
Traffic and Pilot C/N Calculation (DL)
Name
Value
TX i ic
TX i ic
C Traffic n DL
With MIMO (STTD/MRC):

TX i ic
CNR Traffic
TX i ic
CNR Traffic
TX i ic
TX i ic
CNR Preamble
With MIMO (AMS) if

TX i ic
CNR Traffic
DL
DL
= CNR Traffic + G STTD + G STTD
TX i ic
CNR Traffic
TX i ic
C Pilot
DL
TX i ic
T AMS
DL
+ G STTD + G STTD
TX i ic
n DL

TX i ic
CNR Pilot
TX i ic
CNR Pilot
TX i ic
= CNR Pilot
With MIMO (AMS) if

TX i ic
CNR Pilot
9.1.4.5
DL
TX i ic
CNR Preamble
TX i ic
CNR Pilot
DL
+ G STTD + G STTD
DL
TX i ic
T AMS
DL
+ G STTD + G STTD
Traffic and Pilot C/(I+N) Calculation (DL)
Name
Value
TX jc
TX i ic
j
I DL
n DL
TX i ic
-----------------------------------
C Traffic 10 Log
10 +
10 M Shadowing C I
10
10
All TX j jc
TX i ic
CINR Traffic
TX i ic
TX i ic
DL
DL
CINR Traffic = CINR Traffic + G STTD + G STTD

With MIMO (AMS) if
TX i ic
TX i ic
TX i ic
CNR Preamble T AMS :

TX i ic
DL
DL
TX i ic
C Pilot
TX i ic
CINR Pilot
TX jc
TX i ic
j
I DL
n DL
-----------------------------------
10 Log
10 +
10 M Shadowing C I
10
10
All TX j jc
TX i ic
CINR Pilot
TX i ic
= CINR Pilot
With MIMO (AMS) if

TX i ic
CINR Pilot
440
DL
DL
+ G STTD + G STTD
TX i ic
TX i ic

TX i ic
= CINR Pilot
DL
DL
+ G STTD + G STTD
AT271_TRG_E6
Forsk 2009

TX jc
j
TX i ic
dBm
Traffic Total Noise (I+N) for a cell
Unit
Description
dBm
Received uplink signal level
dBm
Uplink EIRP of a user equipment
Unit
Description
dBm
Thermal noise for a cell
dBm
Uplink noise for a cell
Unit
Description
dBm
Uplink interference received at a cell
dB

10 Log 10 TL UL
dB

the interfering mobiles uplink traffic
load
Mj
TX i ic
I UL
n UL
------------------------TX i ic
10
10
10 Log 10
10 + 10
n UL
All M j
All TX jc
dB
Uplink noise at a cell without smart

antenna
I UL + n I
----------------------------------2
n I
dB
Uplink noise at a cell with smart

antenna
M
TX ic
j
i
I UL
n UL
------------------------
10
10
10 Log 10
10 + 10
All Mj
All TX jc
dBm
Total Noise (I+N) for a cell
Unit
Description
I + N DL
9.1.4.6
TX ic
i
I DL
n DL
-------------------------------------
10
10
10 Log
10
+ 10
All
TX
jc
Traffic Signal Level Calculation (UL)
Name
Value
Mi
EIRP UL L Path M Shadowing Model L Indoor + G
Mi
C UL
TX i
Mi
L Ant
P
Mi
EIRP UL
With P
Mi
TX i
Mi
L Body
Mi
+G
Mi
Mi
Mi
= P Max without power control and P
Mi
Mi
= P Eff after power
control
9.1.4.7
Noise Calculation (UL)
Name
Value
TX i ic
N SCa Used
WiMAX 802.16d: n 0 + 10 Log 10 F Sampling ------------------------------
N SCa Total
TX i ic
n 0 UL
PZ UL
N SCa Used
TXi ic
-
WiMAX 802.16e: n 0 + 10 Log 10 F Sampling ----------------------------TX i ic
N SCa Total
TX i ic
TX i ic
n UL
9.1.4.8
n 0 UL + nf
TX i ic
Traffic Interference Calculation (UL)
Name
Mj
I UL
TX i ic TX j jc
fO
Mj
f TL UL
TX i ic
NR UL
Value
Mj
TX i ic TX j jc
C UL + f O
Mj
+ f TL UL
TX i ic TX j jc
10 Log r O
Mj
NR UL
TX i ic
I + N UL
9.1.4.9
Name
Forsk 2009
Traffic C/N Calculation (UL)

Value
AT271_TRG_E6
441

TX i ic
Mi
C UL n UL
Mi
Mi
UL
UL
With MIMO (STTD/MRC): CNR UL = CNR UL + G STTD + G STTD
Mi
CNR UL
TX i ic
Mi
TX i ic
dB
Uplink C/N at a cell
Unit
Description
dB
Uplink C/(I+N) at a cell
With MIMO (AMS) if

Mi
UL
UL
CNR UL = CNR UL + G STTD + G STTD
9.1.4.10
Traffic C/(I+N) Calculation (UL)
Name
Value
TX i ic
Mi
Without smart antenna: CNR UL NR UL

With smart antenna:
Mi
CINR UL
Mi
CNR UL
TX i ic
NR UL
Mi
CINR UL
M Shadowing C I
Mi
CINR UL
TX i ic
UL
UL
+ G STTD + G STTD
TX i ic
With MIMO (AMS) if

Mi
M Shadowing C I
Mi
UL
UL
CINR UL = CINR UL + G STTD + G STTD
9.1.5
Channel Throughput Calculation
9.1.5.1
Calculation of Cell Capacity (Total Resources)
Name
Value
Unit
Description
Hz
Sampling frequency
F Sampling 10
------------------------------------------TX i ic
N SCa Total
kHz
Inter-subcarrier distance
D Sym Useful
1
---------------------TX i ic
F
ms
Useful symbol duration
D CP
r CP
-------F
ms
Cyclic prefix duration
D Symbol
D Sym Useful + D CP
ms
Symbol duration
D Frame
Used
D Frame D TTG D RTG
ms
Used frame duration
TX i ic
N SD Used Frame
D Used
Frame
Floor ------------------- TXi ic
D Symbol
SD
Frame duration in terms of symbol

durations
SD
Downlink subframe duration in terms

of symbol durations
Symbols
Total downlink cell resources, i.e., the

number of symbols in the downlink
subframe
SD
Uplink subframe duration in terms of

symbol durations
TX i ic
TX i ic
F Sampling
TX i ic
TX i ic
TX i ic
TX i ic
TX i ic
TDD
TX i ic
N SD DL Subframe
W Channel 10
Floor f Sampling ---------------------------------------- 8000

8000
TDD
TX i ic
TDD
DL
RoundUp N SD Used Frame r DL Frame O Fixed

WiMAX 802.16d:
DL
TX i ic
R DL
TX i ic
N Sym DL Subframe
TX i ic
N SD UL Subframe
442
TX i ic
O Variable
TX i ic
Floor N SD DL Subframe N SCa Data 1 ----------------------
100
WiMAX 802.16e:
Mi
DL
PZDL
O Variable
TX i ic
Floor N SD DL Subframe N SCa Data 1 ----------------------
100
TX i ic
TDD
UL
RoundDown N SD Used Frame 1 r DL Frame O Fixed
AT271_TRG_E6
Forsk 2009

WiMAX 802.16d:
UL
TX i ic
R UL
TX i ic
O Variable
TXi ic
Floor N SD UL Subframe N SCa Data 1 ----------------------
100
WiMAX 802.16e:
TX i ic
N Sym UL Subframe
Symbols
Total uplink cell resources, i.e., the

number of symbols in the uplink
subframe
Unit
Description
Mi
B DL Highest Service
None
Bearer assigned to the pixel,

subscriber, or mobile Mi in the
downlink
Mi
B UL Highest Service
None
Bearer assigned to the pixel,

subscriber, or mobile Mi in the uplink
Unit
Description
kbps
Downlink peak MAC channel

throughput
kbps
Downlink effective MAC channel

throughput
kbps
Downlink application level channel

throughput
kbps
Uplink peak MAC channel throughput
kbps
Uplink effective MAC channel

throughput
kbps
Uplink application level channel

throughput
UL
i
PZUL
O Variable
TXi ic
Floor N SD UL Subframe N SCa Data 1 ----------------------
100
9.1.5.2
Bearer Determination
Name
Value
Min Best Bearers
Mi
B DL
Min Best Bearers
Mi
B UL
9.1.5.3
TB
TX ic
i
CINR Traffic
TX ic
i
TB
i
CINR UL
Name
Value
TX i ic
R DL
Mi
DL
Without segmentation: -----------------------------------D Frame
TX i ic
R DL
Mi
DL
- f Segment
With segmentation: -----------------------------------D Frame
Mi
CTP P DL
With MIMO (SM):
Mi
B DL
Max
Mi
B DL
1 + f SM Gain G SM 1
With MIMO (AMS):
B DL
TX i ic
Max
B DL
Mi
1 + f SM Gain G SM 1 if
Mi
TX i ic
CNR Preamble T AMS

Mi
CTP P DL 1 BLER B DL
CTP E DL
Mi
Mi
f TP Scaling
Mi
CTP E DL ----------------------------- TP Offset
100
Mi
CTP A DL
TX i ic
R UL
UL
Without segmentation: -----------------------------------D Frame
Mi
With MIMO (SM):
CTP P UL
B UL
Max
B UL
With MIMO (AMS):
Mi
B UL
Mi
CTP E UL
Mi
CTP A UL
9.1.6
Max
Mi
B UL
1 + f SM Gain G SM 1 if
Mi
TX i ic
TX i ic
CNR Preamble T AMS

Mi
CTP P UL 1 BLER B UL
Mi
Mi
f TP Scaling
Mi
CTP E UL ----------------------------- TP Offset
100
Scheduling and Radio Resource Management
Name
Value
Sel
TPD Min DL
-----------------------------Sel
Unit
Description
None
Resources allocated to a mobile to

satisfy its minimum throughput
demand in downlink
Sel
Mi
R Min DL
Forsk 2009
Mi
Mi
CTP P DL
AT271_TRG_E6
443

Sel
Mi
None

satisfy its minimum throughput
demand in uplink
None
Remaining downlink cell resources

after allocation for minimum
throughput demands
None
Remaining uplink cell resources after

allocation for minimum throughput
demands
kbps
Remaining throughput demand for a

mobile in downlink
kbps
Remaining throughput demand for a

mobile in uplink
None
Remaining resource demand for a

mobile in downlink
None
Remaining resource demand for a

mobile in uplink
None

satisfy its maximum throughput
demand in downlink
None

satisfy its maximum throughput
demand in uplink
Sel
Mi
RD Rem DL
Sel
None
Effective remaining downlink

resources in a cell
(Proportional Demand)
Sel
Mi
RD Rem UL
Sel
None
Effective remaining uplink resources

in a cell
(Proportional Demand)
None
QoS class bias

(Biased (QoS Class))
TPD Min UL
-----------------------------Sel
Sel
Mi
R Min UL
Mi
CTP P UL
Sel
100 %
TX i ic
R Rem DL
Mi
RMin DL
Sel
Mi
Sel
100 %
TX i ic
R Rem UL
Mi
RMin UL
Sel
Mi
Sel
Sel
Mi
Sel
Mi
TPD Rem DL
Mi
TPD Max DL TPD Min DL
Sel
Sel
Mi
Sel
Mi
TPD Rem UL
Mi
TPD Max UL TPD Min UL

Sel
Mi
TPD Rem DL
--------------------------------Sel
Sel
Mi
RD Rem DL
Mi
CTP P DL
Sel
Mi
TPD Rem UL
--------------------------------Sel
Sel
Mi
RD Rem UL
Mi
CTP P UL
TX i ic
Sel
R Rem DL
Mi
Proportional Fair: Min RD Rem DL -------------------------

N
Sel
Proportional Demand:
TX i ic
R Eff Rem DL
Sel
Mi
Mi
RD Rem DL
--------------------------------------Sel
Mi
RDRem DL
Sel
Mi
R Max DL
TX i ic
Sel
R QoS DL
Mi
Biased (QoS Class): Min RD Rem DL ------------------------

N QoS
Sel
Mi
TPD Rem DL
Max Aggregate Throughput: --------------------------------Sel
Mi
CTP P DL
TX i ic
Sel
Mi
R Rem UL
Proportional Fair: Min RD Rem UL -------------------------

N
Sel
Proportional Demand:
TX i ic
R Eff Rem UL
Sel
Mi
Mi
RD Rem UL
--------------------------------------Sel
Mi
RDRem UL
Sel
Mi
R Max UL
TX i ic
Sel
R QoS UL
Mi
Biased (QoS Class): Min RD Rem UL ------------------------

N QoS
Sel
Mi
TPD Rem UL
Max Aggregate Throughput: --------------------------------Sel
Mi
CTP P UL
TX ic
i
Min R Rem DL
TX i ic
R Eff Rem DL
TX ic
i
Min R Rem UL
TX i ic
R Eff Rem UL
QoS
444
Sel
Mi
Mi
Mi
Sel
Mi
Sel
Mi
f Bias
R Max rtPS
R Max nrtPS
R Max ErtPS
1 + ---------- = -----------------------------= ------------------------------ = -----------------------------Sel
Sel
Sel
100
Mi
Mi
Mi
R Max rtPS
R Max nrtPS
R Max BE
AT271_TRG_E6
Forsk 2009

r
1 QoS
N QoS ---

TX i ic
R Rem DL -----------------------------------------------------------r
1 QoS
N QoS ---

TX i ic
R QoS DL
None

throughput demands for a QoS class
None

throughput demands for a QoS class
None
Total resources assigned to a mobile

in downlink
(Downlink traffic load of the mobile)
None
Total resources assigned to a mobile

in uplink
(Uplink traffic load of the mobile)
Unit
Description
kbps
Downlink peak MAC user throughput
kbps
Downlink effective MAC user

throughput
kbps
Downlink application level user

throughput
kbps
Uplink peak MAC user throughput
kbps
Uplink effective MAC user throughput
kbps
Uplink application level user

throughput
All QoS
1 QoS
N QoS ---

TX i ic
R Rem UL -----------------------------------------------------------r
1 QoS
N QoS ---

TX i ic
R QoS UL
All QoS
Sel
Sel
Mi
Sel
Mi
TL DL
Sel
Sel
Mi
Sel
Mi
= R UL
9.1.6.1
Mi
R Min DL + R Max DL
Sel
Mi
TL UL
Sel
Mi
= R DL
Mi
R Min UL + R Max UL
User Throughput Calculation
Name
Value
Sel
Sel
Mi
Mi
Sel
Sel
Sel
Mi
Mi
UTP P DL 1 BLER B DL
Mi
UTP E DL
Sel
Sel
Sel
Mi
UTP A DL
Mi
UTP E DL
Sel
Mi
Sel
f TP Scaling
Mi
----------------------------- TP Offset
100
Sel
Mi
Sel
Mi
Mi
R UL CTP P UL
UTP P UL
Sel
Sel
Sel
Mi
Mi
UTP P UL 1 BLER B UL
Mi
UTP E UL
Sel
Sel
Sel
Mi
UTP A UL
9.1.7
Sel
Mi
R DL CTP P DL
UTP P DL
Mi
UTP E UL
Mi
Sel
f TP Scaling
Mi
----------------------------- TP Offset
100
Name
Value
Unit
Description
E SA
None
Number of smart antenna elements
Degrees
Angle of arrival for the useful signal
Degrees
Angle at which the smart antenna

effect is calculated
--- , where is the wavelength of the signal

2
Distance between two adjacent

antenna elements
gn
None
Gain of a single element
None
Steering vector for the direction of
None
Complex smart antenna weight
None
Array correlation matrix for a given

user direction
None
Smart antenna gain in any direction
1 e
2
j ------- d sin
wn
e
2
j ------- nd sin
j n sin
with d = --2
R = S S
H
Forsk 2009
... e
T
2
j ------- E SA 1 d sin
G SA
2
j ------- 2d sin
gn S R S = gn S S S S
10 Log G SA in dB
AT271_TRG_E6
445

J
j pj Rj
R Avg
None
Average array correlation matrix
j=1
RN S
None
Vector of ESA complex weights for the

beam former
E SA
---------------------------------H
1
S RN S
None
Smart antenna gain in the uplink
None
Total noise correlation matrix
None
Thermal noise correlation matrix
None
Interference correlation matrix
Total noise power, including thermal

noise and interference from all uplink
interferers
Total power received from the served

user
None
C/(I+N) in the uplink
None
Uplink smart antenna beam forming

gain in the direction of the served user
Average of the inverse noise

correlation matrices
Uplink interference
None
Angular distribution of uplink noise

rise
RN
2
n
Rn + RI =
pj Sj Sj
I+
j=1
2
Rn
n I
J
pj Sj Sj
RI
j=1
P N
S RN S
p S RN S
2
H
2
1
p S RN S
H
P
1
------- = --------------------------------------------------------------= p S RN S
H
2
1
P
N
S RN S
CINR UL
S I S = E SA
G SA
10 Log E SA in dB
K
1
----
K
1
RN
Avg
RN
1
k
k=1
E SA
2
------------------------------------------ n
H
1
S RN
S
I UL
Avg
I UL + n
----------------------------2
n
NR UL
9.2
Calculation Processes
The following sections describe the processes of different calculations performed in Atoll and their results.
9.2.1
Point Analysis
Point analysis profile tab and reception tab display the following calculation results based on the calculation algorithm
described in "Preamble Signal Level Calculation" on page 466.
TX i ic
Preamble signal level C Preamble
Path loss L Path
Total losses L Total

Note:
Mi
, G
Mi
Mi
Mi
, L Ant , and L Body are not considered in the calculations performed for the point
analysis tool.
446
AT271_TRG_E6
Forsk 2009
9.2.1.1
Profile Tab
TX i ic
Atoll displays the preamble signal level received from the studied cell, C Preamble , which you select on the map or in the
list of transmitters. You can also choose to display the path loss, L Path , or the total losses, L Total for a selected cell.
9.2.1.2
Reception Tab
Analysis provided in the Reception tab is based on path loss matrices. So you can display received preamble signal levels
from the cells for which path loss matrices have been calculated over their calculation areas. For each cell, Atoll displays
TX i ic
the received preamble signal level, C Preamble .

Reception level bar graphs show the preamble signal levels in a decreasing order. The maximum number of bars in the
graph depends on the preamble signal level received from the best server. The bar graph displays cells whose received
preamble signal level is within a 30 dB margin from the highest preamble signal level.
Note:
9.2.2
You can use a value other than 30 dB for the margin from the highest preamble signal
level, for example a smaller value for improving the calculation speed. For more
information on defining a different value for this margin, see the Administrator Manual.
Preamble Signal Level Coverage Predictions

The following coverage predictions are based on the received preamble signal levels:
Coverage by Transmitter
Coverage by Signal Level
Overlapping Zones
These coverage predictions can be performed at any stage of network planning because they do not depend on the traffic
input. Therefore, these calculations are of special interest before and during the deployment stage of the network to study
the coverage footprint of the system.
TX i ic
For these calculations, Atoll calculates the received preamble signal level ( C Preamble ) as explained in "Preamble Signal
Level Calculation" on page 466. Then, Atoll determines the selected display criterion on each pixel inside the cells calculation area. Each pixel within the calculation area of TXi(ic) is considered a non-interfering receiver.
Coverage prediction parameters to be set are:
The coverage prediction conditions to determine the coverage area of each studied cell, and
The display settings to colour the coverage areas.
Note:
Mi
, G
Mi
Mi
Mi
, L Ant , and L Body are not considered in the calculations performed for the
preamble signal level based coverage predictions.

The following sections describe the determination of coverage area of each cell ("Coverage Area Determination" on
page 447), and the display options ("Coverage Display" on page 448) of the coverage predictions.
9.2.2.1

Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine coverage
areas to display. There are three possibilities.
9.2.2.1.1
All Servers
The coverage area of each cell TXi(ic) corresponds to the pixels where.
TX i ic
TX i ic
MinimumThreshold C Preamble or L Total
9.2.2.1.2
TX i ic
or L Path
MaximumThreshold

TX i ic
TX i ic
TX i ic
or L Path
MaximumThreshold
AND
TX i ic
TX j jc
C Preamble Best C Preamble M
ji
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
Forsk 2009
AT271_TRG_E6
447

Notes:
If M = 0 dB, Atoll considers pixels where the received preamble signal level from TXi(ic) is
the highest.
either the highest or within a 2 dB margin from the highest.
If M = -2 dB, Atoll considers pixels where the received preamble signal level from TXi(ic) is
2 dB higher than the received preamble signal levels from the cells which are 2nd best
servers.
9.2.2.1.3

TX i ic
TX i ic
TX i ic
or L Path
MaximumThreshold
AND
TX ic
i
C Preamble 2
nd
TX jc
Best C j
Preamble M
ji
Where M is the specified margin (dB). The 2nd Best function considers the second highest value from a list of values.
Notes:
the second highest.
either the second highest or within a 2 dB margin from the second highest.
If M = -2 dB, Atoll considers pixels where the received preamble signal level from TXi(ic) is
2 dB higher than the received preamble signal levels from the cells which are 3rd best
servers.
9.2.2.2
Coverage Display
9.2.2.2.1
Coverage Resolution
The resolution of the coverage prediction does not depend on the resolutions of the path loss matrices or the geographic
data and can be defined separately for each coverage prediction. Coverage predictions are generated using a bilinear
interpolation method from multi-resolution path loss matrices (similar to the one used to calculate site altitudes).
9.2.2.2.2
Display Types
It is possible to display the coverage predictions with colours depending on any transmitter or cell attribute, and other criteria such as:
Signal Level (dBm, dBV, dBV/m)

Atoll calculates preamble signal levels received from cells on each pixel of the cells coverage areas. A pixel of a coverage
area is coloured if the preamble signal level exceeds ( ) the defined minimum thresholds (pixel colour depends on
received preamble signal level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as cell coverage areas. Each layer shows the different preamble signal levels
received in the cell coverage area.
Best Signal Level (dBm, dBV, dBV/m)

Atoll calculates preamble signal levels received from cells on each pixel of the cells coverage areas. Where other coverage areas overlap the studied one, Atoll chooses the highest value. A pixel of a coverage area is coloured if the preamble
signal level exceeds ( ) the defined thresholds (the pixel colour depends on the preamble signal level). Coverage consists
of several independent layers whose visibility in the workspace can be managed. There are as many layers as defined
thresholds. Each layer corresponds to an area where the preamble signal level from the best server exceeds a defined
threshold.
Path Loss (dB)

Atoll calculates path losses from cells on each pixel of the cells coverage areas. A pixel of a coverage area is coloured if
the path loss exceeds ( ) the defined minimum thresholds (pixel colour depends on path loss). Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as cell coverage
areas. Each layer shows different path loss levels in the cells coverage area.
448
AT271_TRG_E6
Forsk 2009
Total Losses (dB)

Atoll calculates total losses from cells on each pixel of the cells coverage areas. A pixel of a coverage area is coloured if
total losses exceed ( ) the defined minimum thresholds (pixel colour depends on total losses). Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as cell coverage
areas. Each layer shows different total losses levels in the cells coverage areas.

Atoll calculates preamble signal levels received from cells on each pixel of the cells coverage areas. Where other coverage areas overlap the studied one, Atoll determines the best cell (i.e., the cell with the highest preamble signal level) and
evaluates path loss from this cell. A pixel of a coverage area is coloured if the path loss exceeds ( ) the defined thresholds
(pixel colour depends on path loss). Coverage consists of several independent layers whose visibility in the workspace
can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the path loss
from the best server exceeds a defined threshold.

Atoll calculates preamble signal levels received from cells on each pixel of the cells coverage areas. Where coverage
areas overlap the studied one, Atoll determines the best cell (i.e., the cell with the highest preamble signal level) and evaluates total losses from this cell. A pixel of a coverage area is coloured if the total losses exceed ( ) the defined thresholds
(pixel colour depends on total losses). Coverage consists of several independent layers whose visibility in the workspace
can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the total
losses from the best server exceed a defined threshold.
Number of Servers
Atoll evaluates the number of cells that cover a pixel (i.e., the pixel falls within the coverage areas of these cells). The pixel
colour depends on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the
number of servers exceeds ( ) a defined threshold.

On each pixel of a cells coverage area, the coverage corresponds to the pixels where the received preamble signal level
from this cell fulfils signal conditions defined in Conditions tab with different cell edge coverage probabilities. Coverage
the number of cells.

On each pixel of each cell coverage area, the coverage corresponds to the pixels where the best preamble signal level
received fulfils signal conditions defined in Conditions tab. There is one coverage area per cell edge coverage probability.
9.2.3
Signal Quality Analysis Coverage Predictions

The following coverage predictions are based on the received preamble, traffic, or pilot signal levels and noise, and take
into account the receiver characteristics ( L
Mi
, G
Mi
Mi
Mi
, L Ant , and L Body ) when calculating the required parameter.
Signal Quality Analysis (DL)

Signal Quality Analysis (UL)
For these calculations, Atoll calculates the received signal level and noise at each pixel for the signal type being studied,
i.e., preamble, traffic, or pilot. Each pixel within the calculation area of TXi(ic) is considered a non-interfering receiver. The
properties of the non-interfering probe receiver are set by selecting a terminal, mobility, and service.
These coverage predictions can be performed at any stage of network planning because they do not depend on the traffic
input. Therefore, these calculations are of special interest before and during the deployment stage of the network to study
the coverage footprint of the system.
The coverage prediction conditions, and

page 449), the calculation of the coverage parameter ("Coverage Parameter Calculation" on page 450), and the display
options ("Coverage Display" on page 450) of the coverage predictions.
9.2.3.1

These coverage predictions are all best server coverage predictions, i.e., the coverage area of each cell comprises the
pixels where the cell is the best server. Best server for each pixel is calculated as explained in "Best Server Determination"
on page 471.
Forsk 2009
AT271_TRG_E6
449
9.2.3.2
Coverage Parameter Calculation

The following parameters are calculated for the Signal Quality Analysis (DL) coverage prediction.
TX i ic
Best Preamble Signal Level (DL) (dBm): C Preamble as explained in "Preamble Signal Level Calculation" on
page 466.
TX i ic
Best Pilot Signal Level (DL) (dBm): C Pilot
as explained in "Traffic and Pilot Signal Level Calculation (DL)" on
page 473.
TX i ic
Best Traffic Signal Level (DL) (dBm): C Traffic as explained in "Traffic and Pilot Signal Level Calculation (DL)"
on page 473.
TX i ic
Preamble C/N Level (DL) (dB): CNR Preamble as explained in "Preamble C/N Calculation" on page 470.
Pilot C/N Level (DL) (dB): CNR Pilot
TX i ic
Traffic C/N Level (DL) (dB):
as explained in "Traffic and Pilot C/N Calculation (DL)" on page 482.
TX i ic
CNR Traffic
as explained in "Traffic and Pilot C/N Calculation (DL)" on page 482.
The following parameters are calculated for the Signal Quality Analysis (UL) coverage prediction.
Mi
Signal Level (UL) (dBm): C UL as explained in "Traffic Signal Level Calculation (UL)" on page 484.
C/N Level (UL) (dB): CNR UL as explained in "Traffic C/N Calculation (UL)" on page 487.
9.2.3.3
Coverage Display
9.2.3.3.1
Coverage Resolution
Mi
9.2.3.3.2
Signal Quality Analysis (DL) Display Types

It is possible to display the Signal Quality Analysis (DL) coverage prediction with colours depending on the following
display options.
Best Preamble Signal Level (DL) (dBm)

Atoll calculates preamble signal levels received from the best serving cells on each pixel of their coverage areas. A pixel
of a coverage area is coloured if the preamble signal level exceeds ( ) the defined thresholds (the pixel colour depends
on the preamble signal level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the preamble signal
level from the best server exceeds a defined threshold.
Best Pilot Signal Level (DL) (dBm)

Atoll calculates pilot signal levels received from the best serving cells on each pixel of their coverage areas. A pixel of a
coverage area is coloured if the pilot signal level exceeds ( ) the defined thresholds (the pixel colour depends on the pilot
signal level). Coverage consists of several independent layers whose visibility in the workspace can be managed. There
are as many layers as defined thresholds. Each layer corresponds to an area where the pilot signal level from the best
server exceeds a defined threshold.
Best Traffic Signal Level (DL) (dBm)

Atoll calculates traffic signal levels received from the best serving cells on each pixel of their coverage areas. A pixel of a
coverage area is coloured if the traffic signal level exceeds ( ) the defined thresholds (the pixel colour depends on the
traffic signal level). Coverage consists of several independent layers whose visibility in the workspace can be managed.
There are as many layers as defined thresholds. Each layer corresponds to an area where the traffic signal level from the
best server exceeds a defined threshold.
Preamble C/N Level (DL) (dB)

Atoll calculates preamble C/N levels received from the best serving cells on each pixel of their coverage areas. A pixel of
a coverage area is coloured if the preamble C/N level exceeds ( ) the defined thresholds (the pixel colour depends on
the preamble C/N level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the preamble C/N
level from the best server exceeds a defined threshold.
450
AT271_TRG_E6
Forsk 2009
Pilot C/N Level (DL) (dB)

Atoll calculates pilot C/N levels received from the best serving cells on each pixel of their coverage areas. A pixel of a
coverage area is coloured if the pilot C/N level exceeds ( ) the defined thresholds (the pixel colour depends on the pilot
C/N level). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are
as many layers as defined thresholds. Each layer corresponds to an area where the pilot C/N level from the best server
exceeds a defined threshold.
Traffic C/N Level (DL) (dB)

Atoll calculates traffic C/N levels received from the best serving cells on each pixel of their coverage areas. A pixel of a
coverage area is coloured if the traffic C/N level exceeds ( ) the defined thresholds (the pixel colour depends on the traffic
C/N level). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are
as many layers as defined thresholds. Each layer corresponds to an area where the traffic C/N level from the best server
exceeds a defined threshold.
9.2.3.3.3
Signal Quality Analysis (UL) Display Types

It is possible to display the Signal Quality Analysis (UL) coverage prediction with colours depending on the following
display options.
Signal Level (UL) (dBm)

Atoll calculates uplink signal levels received from each pixel, of the coverage areas of the best serving cells, at the cells.
A pixel of a coverage area is coloured if the uplink signal level exceeds ( ) the defined thresholds (the pixel colour
depends on the uplink signal level). Coverage consists of several independent layers whose visibility in the workspace can
be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the uplink signal
level at the best server exceeds a defined threshold.
C/N Level (UL) (dB)

Atoll calculates uplink C/N levels received from each pixel, of the coverage areas of the best serving cells, at the cells. A
pixel of a coverage area is coloured if the uplink C/N level exceeds ( ) the defined thresholds (the pixel colour depends
on the uplink C/N level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the uplink C/N level
at the best server exceeds a defined threshold.
9.2.4
Calculations on Subscriber Lists

When calculations are performed on a list of subscribers by running the Automatic Server Allocation, Atoll calculates the
path loss again for the subscriber locations and heights because the subscriber heights can be different from the global
receiver height used for calculating the path loss matrices.
Atoll calculates the following parameters for each subscriber in the list whose Lock Status is set to None.
Serving Base Station and Reference Cell as described in "Best Server Determination" on page 471.
Atoll calculates the following parameters for each subscriber in the list that has a serving base station assigned and whose
Lock Status is set to None or Server.
Azimuth ( ): Angle with respect to the north for pointing the subscriber terminal towards its serving base station.
Mechanical Downtilt ( ): Angle with respect to the horizontal for pointing the subscriber terminal towards its
serving base station.
Atoll calculates the following parameters for each subscriber in the list that has a serving base station assigned, using the
properties of the default terminal and service.
Received Preamble Power (DL) (dBm) as described in "Preamble Signal Level Calculation" on page 466.
Received Traffic Power (DL) (dBm) as described in "Traffic and Pilot Signal Level Calculation (DL)" on page 473.
Traffic Total Noise (I+N) (DL) (dBm) as described in "Traffic C/(I+N) Calculation (DL)" on page 483.
Traffic C/(I+N) (DL) (dB) as described in "Traffic C/(I+N) Calculation (DL)" on page 483.
Bearer (DL) as described in "Bearer Determination" on page 496.
Permutation Zone (DL) as described in "Permutation Zone Selection (WiMAX 802.16e)" on page 471.
BLER (DL): Downlink block error rate read from the BLER vs. CINR Traffic graph available in the reception equip-
Forsk 2009
TX i ic
ment assigned to the terminal used by the subscriber.

Diversity Mode (DL): Antenna diversity mode supported by the permutation zone assigned to the subscriber.
Peak MAC Channel Throughput (DL) (kbps) as described in "Channel Throughput Calculation" on page 496.
Effective MAC Channel Throughput (DL) (kbps) as described in "Channel Throughput Calculation" on
page 496.
Received Power (UL) (dBm) as described in "Traffic Signal Level Calculation (UL)" on page 484.
Total Noise (I+N) (UL) (dBm) as described in "Noise Rise Calculation (UL)" on page 487.
C/(I+N) (UL) (dB) as described in "Traffic C/(I+N) Calculation (UL)" on page 490.
Bearer (UL) as described in "Bearer Determination" on page 496.
Permutation Zone (UL) as described in "Permutation Zone Selection (WiMAX 802.16e)" on page 471.
AT271_TRG_E6
451
9.2.5
Mi
BLER (UL): Uplink block error rate read from the BLER vs. CINR UL graph available in the reception equipment
assigned to the serving cell of the subscriber.

Diversity Mode (UL): Antenna diversity mode supported by the permutation zone assigned to the subscriber.
Transmission Power (UL) as described in "Traffic C/(I+N) Calculation (UL)" on page 490.
Number of Used Subchannels (UL) as described in "Traffic C/(I+N) Calculation (UL)" on page 490.
Peak MAC Channel Throughput (DL) (kbps) as described in "Channel Throughput Calculation" on page 496.
Effective MAC Channel Throughput (DL) (kbps) as described in "Channel Throughput Calculation" on
page 496.
Monte Carlo Simulations

The simulation process is divided into two steps.
Generating a realistic user distribution as explained in "Generating a Realistic User Distribution" on page 452.
Atoll generates user distributions as part of the Monte Carlo algorithm based on traffic data. The resulting user
distribution complies with the traffic database and maps selected when creating simulations.
9.2.5.1
Scheduling and Radio Resource Management as explained under "Simulation Process" on page 455.

During each simulation, Atoll performs two random trials. The first random trial generates the number of users and their
activity status as explained in the following sections depending on the type of traffic input.
"Simulations Based on Raster Traffic Maps, Vector Traffic Maps, and Subscriber Lists" on page 452.
"Simulations Based on Traffic Map Based on Transmitters and Services" on page 454.
Once all the user characteristics have been determined, a second random trial is performed to obtain their geographical
locations weighted according to the clutter classes, and whether they are indoor or outdoor according to the percentage
of indoor users per clutter class defined for the traffic maps.
9.2.5.1.1
Simulations Based on Raster Traffic Maps, Vector Traffic Maps, and Subscriber Lists
a list of user profiles with an associated mobility type and a given density, i.e., number of users of a user profile per km.
Vector traffic maps are based on user profiles. Each polygon or line of the map is assigned a density of users with a given
user profile and mobility type. If the map is composed of points, each point is assigned a number of users with given user
profile and mobility type.
Fixed subscribers listed in subscriber lists have a user profile assigned to each of them.
The user profile models the behaviour of the different user categories. Each user profile contains a list of services and their
associated parameters describing how these services are accessed by the user.
The number of users of each user profile is calculated from the surface area (SEnv) of each environment class map (or
each polygon) and the user profile density (DUP).
N Users = S Env D UP
Notes:
In case of vector traffic map composed of lines, the number of users of each user profile is
calculated from the line length (L) and the user profile density (DUP) (users per km):
The number of users is an input when the vector traffic map is composed of points.
N Users = L D UP
At any given instant, Atoll calculates the probability for a user being active in the uplink and in the downlink according to
the service usage characteristics described in the user profiles, i.e., the number of voice calls or data sessions, the average
duration of each voice call, or the volumes of the data exchanged in the uplink and the downlink in each data session.
Voice Service (v)

User profile parameters for voice type services are:
-
The user terminal equipment used for the service (from the Terminals table).
The average number of calls per hour N Call .
The average duration of a call (seconds) D Call .
N Call D Call
Calculation of the service usage duration per hour ( p 0 : probability of a connection): p 0 = -------------------------------3600
Calculation of the number of users trying to access the service v ( n v ): n v = N Users p 0
The activity status of each user depends on the activity periods during the connection, i.e., the uplink and downlink
UL
DL
activity factors defined for the voice type service v, f Act and f Act .
452
AT271_TRG_E6
Forsk 2009

UL
DL

UL
UL
DL
Probability of being active in the uplink: p Active = f Act 1 f Act

DL
DL
UL
Probability of being active in the downlink: p Active = f Act 1 f Act

UL + DL
UL
DL
Probability of being active in the uplink and downlink both: p Active = f Act f Act
Number of inactive users: n v Inactive = n v p Inactive
UL
UL
Number of users active in the uplink: n v Active = n v p Active

DL
DL
Number of users active in the downlink: n v Active = n v p Active

UL + DL
UL + DL
Number of users active in the uplink and downlink both: n v Active = n v p Active
Therefore, a user can be either active on both links, inactive on both links, active on UL only, or active on DL only.
Data Service (d)

User profile parameters for data type services are:
-
The user terminal equipment used for the service (from the Terminals table).
The average number of data calls per hour N Call .
The average data volume (in kBytes) transferred in the downlink V
The average throughputs in the downlink
Calculation of activity probabilities: f
UL
DL
TP Average
DL
and the uplink
and the uplink V
UL
TP Average
UL
during a session.
for the service d.
UL
DL
N Call V 8
N Call V 8
DL
and f
= ---------------------------------------------= ---------------------------------------------UL
DL
TP Average 3600
TP Average 3600
Probability of being inactive: p Inactive = 1 f
UL
UL
1 f
Probability of being active in the uplink: p Active = f

DL
UL
Probability of being active in the downlink: p Active = f
DL
1 f
DL
DL
1 f
UL
UL + DL
Probability of being active in the uplink and downlink both: p Active = f
UL
DL
Calculation of number of users:

Number of inactive users: n v Inactive = N Users p Inactive
UL
UL
Number of users active in the uplink: n d Active = N Users p Active

DL
DL
Number of users active in the downlink: n d Active = N Users p Active

UL + DL
UL + DL
Number of users active in the uplink and downlink both: n d Active = N Users p Active
Calculation of the number of active users trying to access the service d (nd):
UL
DL
UL + DL
n d = n d Active + n d Active + n d Active

Inactive users are not taken into account.
Note:
Forsk 2009
The user distribution per service and the activity status distribution between the users are
average distributions. And the service and the activity status of each user are randomly
drawn in each simulation. Therefore, if you calculate several simulations at once, the
average number of users per service and average numbers of inactive, active on UL, active
on DL and active on UL and DL users, respectively, will correspond to calculated
distributions. But if you check each simulation, the user distribution between services as
well as the activity status distribution between users is different in each of them.
AT271_TRG_E6
453
9.2.5.1.2
Simulations Based on Traffic Map Based on Transmitters and Services

Traffic maps based on transmitters and services are also referred to as live traffic maps. Live traffic data from the OMC is
spread over the best server coverage areas of the transmitters included in the traffic map. Either throughput demands per
service or the number of active users per service (uplink or downlink) are assigned to the coverage areas of each transmitter.
For each transmitter TXi and each service s,
Traffic Map Based on Transmitters and Services (Throughputs)

Atoll calculates the number of active users of each service s on UL and DL in the coverage area of TXi as follows:
N
UL
UL
DL
TP Cell
TP Cell
DL
= -------------------------- and N
= -------------------------UL
DL
TP Average
TP Average
UL
Where TP Cell is the total uplink throughput demand defined in the map for any service s for the coverage area of
DL
the transmitter TXi, TP Cell is the total downlink throughput demand defined in the map for any service s for the
UL
coverage area of the transmitter TXi, TP Average is the average uplink throughput demand of the service s, and
DL
TP Average is the average downlink throughput demand of the service s.
Traffic Map Based on Transmitters and Services (# Active Users)

UL

and N
coverage area using the service s.
DL
values, i.e., the number of active users on UL and DL in the TXi
At any given instant, Atoll calculates the probability for a user being active in the uplink and in the downlink as follows:
Voice Service (v)

Users active in the uplink and downlink both are included in the N
to accurately determine the number of active users in the uplink
UL
and N
UL
( n v Active
DL
values. Therefore, it is necessary

DL
), in the downlink ( n v Active ), and both
UL + DL
( n v Active ). As for the other types of traffic maps, Atoll considers both active and inactive users for voice services.
The activity status of each user depends on the activity periods during the connection, i.e., the uplink and downlink
UL
DL
activity factors defined for the voice type service v, f Act and f Act .
UL
DL

UL
UL
DL
Probability of being active in the uplink: p Active = f Act 1 f Act

DL
DL
UL
Probability of being active in the downlink: p Active = f Act 1 f Act

UL + DL
Probability of being active in the uplink and downlink both: p Active
UL
DL
= f Act f Act
Calculation of the number of active users trying to access the voice service v:
We have: N
UL
UL
UL + DL
= p Active + p Active n v and N
DL
DL
UL + DL
= p Active + p Active n v
Where, nv is the total number of active users in the TXi coverage area using the service v.
UL
UL + DL
DL
UL + DL
N p Active
N p Active
UL + DL
- -------------------------------------------
Number of users active in the uplink and downlink both: n v Active = Min ------------------------------------------UL + DL
DL
UL + DL
p UL
Active + p Active p Active + p Active

UL
Number of users active in the uplink: n v Active = N

DL
UL
Number of users active in the downlink: n v Active = N

UL
DL
UL + DL
n v Active
DL
UL + DL
n v Active
UL + DL
And, n v = n v Active + n v Active + n v Active

Calculation of the number of inactive users attempting to access the service v:
nv
Number of inactive users: n v Inactive = ------------------------------ p Inactive
1 p Inactive
454
AT271_TRG_E6
Forsk 2009
Data Service (d)

Here, Atoll considers all the users as active. Activity probabilities are not calculated.
Calculation of the number of users attempting to access the service d:
If N
UL
DL
UL + DL
n d Active = N
UL
UL
n d Active = 0
DL
n d Active = N
If N
UL
DL
UL
DL
DL
UL + DL
n d Active = N
DL
DL
n d Active = 0
UL
n d Active = N
UL
nd is the total number of active users in the TXi coverage area using the service d:
UL
DL
UL + DL
n d = n d Active + n d Active + n d Active

Note:
9.2.5.2
The activity status distribution between users is an average distribution. In fact, in each
simulation, the activity status of each user is randomly drawn. Therefore, if you calculate
several simulations at once, average numbers of inactive, active on UL, active on DL and
active on UL and DL users correspond to the calculated distribution. But if you check each
simulation, the activity status distribution between users is different in each of them.
Simulation Process
WiMAX cells include intelligent schedulers and radio resource management features for regulating network traffic loads,
optimising spectral efficiency, and satisfying the QoS demands of the users. Each Monte Carlo simulation in the Atoll
WiMAX BWA module is a snap-shot of the network with resource allocation carried out over a duration of 1 second. The
number of WiMAX frames in 1 second depends on the selected frame duration, D Frame . The steps of this algorithm are
listed below.
The simulation process can be summed up into the following iterative steps.
For each simulation, the simulation process,
1. Generates mobiles according to the input traffic data as explained in "Generating a Realistic User Distribution" on
page 452.
2. Resets the cell loads to initial values, i.e., 100 % downlink traffic load, 0 % uplink traffic load, 0 dB uplink noise
rise, 0 % segmentation usage ratio, and 0 % AAS usage ratio.
-
TX i ic
TX i ic
TX i ic
TX i ic
Cell transmission powers and offsets ( P Preamble , P Traffic , P Pilot , and P Idle Pilot ) are set to the
values defined by the user.
Mi
Mobile transmission power is set to the maximum mobile power ( P Max ).
Cell loads ( TL DL
TX i ic
TX i ic
, TL UL
TX i ic
, NR UL
, SU
TX i ic
, and AU
TX i ic
) are set to their current values in the
Cells table.
3. Determines the best servers for all the mobiles generated for the simulation as explained in "Best Server Determination" on page 471.
Forsk 2009
AT271_TRG_E6
455
Figure 9.1: WiMAX Simulation Algorithm

For each iteration k, the simulation process,
4. Determines the downlink and uplink traffic C/(I+N) for each of these mobiles as explained in "Traffic C/(I+N) Calculation (DL)" on page 483 and "Traffic C/(I+N) Calculation (UL)" on page 490 respectively.
The C/(I+N) is calculated in different ways depending on whether a smart antenna has been assigned to a transmitter or not.
-
Without smart antennas:

The downlink traffic loads of cells are used to calculate the downlink interference, i.e., the interference received from each interfering cell in the downlink is weighted by its downlink traffic load. The uplink traffic loads
of interfering mobiles are used to calculate the uplink interference, i.e., the interference received from each
interfering mobile in the uplink is weighted by its uplink traffic load.
With smart antennas:

Victim and Interfering Mobiles:
In WiMAX simulations, the terms victim and interfering mobiles are used for mobiles served by the victim and
interfering cells respectively. In the downlink, victim mobiles receive interfering signals from interfering cells.
In the uplink, victim cells receive interfering signals from interfering mobiles.
456
AT271_TRG_E6
Forsk 2009
Figure 9.2: Victim and Interfering Mobiles

Atoll assumes that a different beam is formed for each victim mobile.
Time-Slot Scenario:
For each victim mobile, instead of weighting the interference by traffic loads of the interfering mobiles or cells,
a Time-Slot Scenario (TSS) is generated. A time-slot scenario is a list of interfering mobiles per victim mobile.
In other words, a time-slot scenario generated for any victim mobile contains at most one (1) interfering mobile
in each interfering cell. Each time-slot scenario can generate either one (1) or no (0) interfering mobile in each
interfering cell for each victim mobile. One and only one time-slot scenario is generated per victim mobile. And,
there are as many time-slot scenarios generated as there are victim mobiles generated during the simulation.
For a given time-slot scenario, the probability that an interfering mobile is present in an interfering cell depends
on the traffic loads of the potential interfering mobiles in the interfering cell. All mobiles in an interfering cell
are potential interfering mobiles. Which one (1), if any, of these potential interfering mobiles is included in the
time-slot scenario is determined by weighting their probability of selection by the mobile traffic loads.
Interference Averaging Over Iterations:
The interferences calculated for each iteration are averaged for all the iterations in a simulation.
This process considers the probabilities of collision between slots used by victim and interfering mobiles,
which depend upon their transmission times, or in other words, their traffic loads. The averaging of interferences over all the iterations in a simulation gives a realistic average interference at the end of the simulation.
The above process is the same for uplink and downlink. In this way, Atoll simulates the simultaneous connections of victim and interfering mobiles, and considers the effect of the smart antenna, because beamforming
is performed for all victim and interfering mobiles for all time-slot scenarios.
Each iteration starts with the initial conditions which are the results from the previous iteration, i.e., interference
information are input to each iteration. The averaging of interferences over all the iterations in a simulation is
performed in a successive manner. This is done because each time-slot scenario may have a large number
of interfering mobiles for each victim mobile.
Successive averaging means that the interference information input to an iteration takes into account the output interference information of the preceding iteration as well as the weighted input interference information of
all the previously carried out iterations. If I represents the interference information, the interference information
input to the kth iteration can be given by:
n k
I in
n k 1
n k 2
I out + k 2 I in
= ----------------------------------------------------------------------1 + k 2
Where k is the iteration number, n is any given victim mobile, and is a stability factor used to help simulations converge quickly. The stability factor is currently set to 0.2. The effect of the stability factor can be understood by the following figure.
Figure 9.3: Simulation Convergence Stability Factor
Forsk 2009
AT271_TRG_E6
457

5. Determines the mobiles which are within the service areas of their best serving cells as explained in "Service Area
Calculation" on page 471.
6. Determines the permutation zone assigned to each mobile as explained in "Permutation Zone Selection (WiMAX
802.16e)" on page 471.
7. Determines the best available WiMAX bearer at the mobile as explained in "Bearer Determination" on page 496.
8. Determines the channel throughputs at the mobile as explained in "Channel Throughput Calculation" on page 496.
9. Performs radio resource management and scheduling to determine the amount of resources to allocate to each
mobile according to the QoS and service demands of each mobile using the selected scheduler as explained in
"Scheduling and Radio Resource Management" on page 443.
10. Calculates the user throughputs after allocating resources to each mobile as explained in "User Throughput Calculation" on page 504.
11. Updates the traffic loads, and noise rise values of all the cells according to the resources in use and the total
resources as follows:
Calculation of Traffic Loads:
Atoll calculates the traffic loads for all the cells TXi(ic).
TX i ic
TL DL
Mi
%RDL
TX i ic
and TL UL
Mi
Mi
%RUL
Mi
Calculation of Uplink Noise Rise:

For each victim cell TXi(ic), the uplink noise rise is calculated and updated by considering each interfering mobile
Mj as explained in "Noise Rise Calculation (UL)" on page 487.
Calculation of Segmentation Usage (WiMAX 802.16e):
Atoll calculates the segmentation usages for all the cells as follows:
SU
Mi
TX i ic
M
PZ
%R DL
Mi
PZDL = Seg
= Seg
DL
= -----------------------------------------------------------------------TX i ic
TL DL
Mi
Where
Mi
Mi
%R DL
is the sum of the percentages of the downlink cell resources allocated to

i
PZ DL = Seg
M
i
PZ DL = Seg
mobiles served by the segmented permutation zone.

Calculation of AAS Usage:
Atoll calculates the AAS usages for all the cells as follows:
Mi
AU
Mi
TX i ic
Where
AAS
= ------------------------------------------TX i ic
TL DL
AAS
Mi
Mi
%R DL
%R DL
AAS
is the sum of the percentages of the downlink cell resources allocated to mobiles served
AAS
by the smart antenna equipment.

12. Performs the convergence test to see whether the differences between the current and the new loads are within
the convergence thresholds.
The convergence criteria are evaluated at the end of each iteration k, and can be written as follows:
TX i ic
TL DL
TX i ic
TL UL
TX i ic
NR UL
TX i ic
TX i ic
Max TL DL
TL DL
k
All TX ic
k 1
TX i ic
TX i ic
Max TL UL
TL UL
k 1
All TX i ic
TX i ic
TX i ic
Max NR UL
NR UL
k
All TX ic
i
TX i ic
Let TL DL
TX i ic
Req
, TL UL
k 1
TX i ic
Req
, and NR UL
Req
be the simulation convergence thresholds defined when
creating the simulation. Atoll stops the simulation in the following cases.
458
AT271_TRG_E6
Forsk 2009

Convergence: Simulation has converged between iteration k - 1 and k if:
TX i ic
TL DL
TX i ic
TL DL
TX i ic
Req
AND TL UL
TX i ic
TL UL
TX i ic
Req
AND NR UL
TX i ic
NR UL
Req
No convergence: Simulation has not converged even after the last iteration, i.e., k = Max Number of Iterations
defined when creating the simulation, if:
TX i ic
TL DL
TX i ic
TL DL
TX i ic
Req
OR TL UL
TX i ic
TL UL
TX i ic
Req
OR NR UL
TX i ic
NR UL
Req
13. Repeats the above steps (from step 4) for the iteration k+1 using the new calculated loads as the current loads.
Simulation Results
At the end of the simulation process, the main results obtained are:
Downlink Traffic Loads

Uplink Traffic Loads
Uplink Noise Rise received at the main antenna
Angular Distributions of Downlink Traffic Power Density for cells with smart antennas
Angular Distributions of Uplink Noise Rise for cells with smart antennas
These results can be used as input for C/(I+N)-based coverage predictions.

In addition to the above parameters, the simulations also list the connection status of each mobile. Mobiles can be rejected
due to:
No Service: If the mobile does not have any best serving cell (step 3.), if the mobile is not located in the service
area of its best server (step 5.), or if the mobile is not able to access a bearer in the direction of its activity (step 7.),
i.e., UL, DL, or UL+DL.
Scheduler Saturation: If the mobile is not in the list of mobiles selected for scheduling (step 9.)
Resource Saturation: If all the cell resources are used up before allocation to the mobile (step 9.)
Connected mobiles (step 9.) can be:
9.2.6
Connected UL: If a mobile active in UL is allocated resources in UL.

Connected DL: If a mobile active in DL is allocated resources in DL.
Connected DL+UL: If a mobile active in UL+DL is allocated resources in UL+DL.
C/(I+N)-Based Coverage Predictions

The following coverage predictions are based on the received traffic or pilot signal levels, total noise, and the traffic and
pilot interference.
Coverage by C/(I+N) Level (DL)

Coverage by Best Bearer (DL)
Coverage by Channel Throughput (DL)
Coverage by C/(I+N) Level (UL)
Coverage by Best Bearer (UL)
Coverage by Channel Throughput (UL)
These coverage predictions take into account the receiver characteristics ( L
Mi
,G
Mi
Mi
Mi
, L Ant , and L Body ) when calculating
the required parameter. For these calculations, Atoll calculates the received signal level, noise, and interference at each
pixel. Each pixel within the calculation area of TXi(ic) is considered a non-interfering receiver. The properties of the noninterfering probe receiver are set by selecting a terminal, mobility, and service.
The downlink coverage predictions are based on the downlink traffic loads of the cells, and the uplink coverage predictions
are based on the uplink noise rise values. These parameters can either be calculated by Atoll during the Monte Carlo simulations, or set by the user manually for all the cells.
The coverage prediction conditions, and

The minimum thresholds at the receiver are defined in the Display parameters.
page 459), the calculation of the coverage parameter ("Coverage Parameter Calculation" on page 460), and the display
options ("Coverage Display" on page 460) of the coverage predictions.
9.2.6.1

These coverage predictions are all best server coverage predictions, i.e., the coverage area of each cell comprises the
pixels where the cell is the best server. Best server for each pixel is calculated as explained in "Best Server Determination"
on page 471.
Forsk 2009
AT271_TRG_E6
459
9.2.6.2
Coverage Parameter Calculation

The following parameters are calculated for the Coverage by C/(I+N) Level (DL) coverage prediction.
TX i ic
Traffic C/(I+N) Level (DL) (dB): CINR Traffic as explained in "Traffic C/(I+N) Calculation (DL)" on page 483.
Traffic Total Noise (I+N) (DL) (dBm): I + N DL
TX i ic
as explained in "Traffic C/(I+N) Calculation (DL)" on
page 483.
TX i ic
Preamble C/(I+N) Level (DL) (dB): CINR Preamble as explained in "Preamble C/(I+N) Calculation" on page 470.
Preamble Total Noise (I+N) (DL) (dBm): I + N Preamble as explained in "Preamble C/(I+N) Calculation" on
TX i ic
page 470.
The following parameters are calculated for the Coverage by Best Bearer (DL) coverage prediction.
Mi
Best Bearer (DL): B DL as explained in "Bearer Determination" on page 496.
The following parameters are calculated for the Coverage by Channel Throughput (DL) coverage prediction.
Mi
Peak MAC Throughput (DL) (kbps): CTP P DL as explained in "Channel Throughput Calculation" on page 496.
Effective MAC Throughput (DL) (kbps): CTP E DL as explained in "Channel Throughput Calculation" on
Mi
page 496.
Mi
Application Throughput (DL) (kbps): CTP A DL as explained in "Channel Throughput Calculation" on

page 496.
The following parameters are calculated for the Coverage by C/(I+N) Level (UL) coverage prediction.
Mi
C/(I+N) Level (UL) (dB): CINR UL as explained in "Traffic C/(I+N) Calculation (UL)" on page 490.
Total Noise (I+N) (UL) (dBm): I + N UL
TX i ic
as explained in "Noise Rise Calculation (UL)" on page 487.
The following parameters are calculated for the Coverage by Best Bearer (UL) coverage prediction.
Mi
Best Bearer (UL): B UL as explained in "Bearer Determination" on page 496.
The following parameters are calculated for the Coverage by Channel Throughput (UL) coverage prediction.
Mi
Peak MAC Throughput (UL) (kbps): CTP P UL as explained in "Channel Throughput Calculation" on page 496.
Effective MAC Throughput (UL) (kbps): CTP E UL as explained in "Channel Throughput Calculation" on
Mi
page 496.
Mi
Application Throughput (UL) (kbps): CTP A UL as explained in "Channel Throughput Calculation" on

page 496.
9.2.6.3
Coverage Display
9.2.6.3.1
Coverage Resolution
9.2.6.3.2
Coverage by C/(I+N) Level (DL) Display Types

It is possible to display the Coverage by C/(I+N) Level (DL) coverage prediction with colours depending on the following
display options.
Traffic C/(I+N) Level (DL) (dB)

Atoll calculates traffic C/(I+N) levels received from the best serving cells on each pixel of their coverage areas. A pixel of
a coverage area is coloured if the traffic C/(I+N) level exceeds ( ) the defined thresholds (the pixel colour depends on the
traffic C/(I+N) level). Coverage consists of several independent layers whose visibility in the workspace can be managed.
There are as many layers as defined thresholds. Each layer corresponds to an area where the traffic C/(I+N) level from
the best server exceeds a defined threshold.
Traffic Total Noise (I+N) (DL) (dBm)

Atoll calculates traffic total noise (I+N) levels received from the interfering cells on each pixel of the coverage areas of the
best serving cells. A pixel of a coverage area is coloured if the traffic total noise (I+N) level exceeds ( ) the defined thresholds (the pixel colour depends on the traffic total noise (I+N) level). Coverage consists of several independent layers whose
460
AT271_TRG_E6
Forsk 2009

visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to
an area where the traffic total noise (I+N) level from the interfering cells exceeds a defined threshold.
Preamble C/(I+N) Level (DL) (dB)

Atoll calculates preamble C/(I+N) levels received from the best serving cells on each pixel of their coverage areas. A pixel
of a coverage area is coloured if the preamble C/(I+N) level exceeds ( ) the defined thresholds (the pixel colour depends
on the preamble C/(I+N) level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the preamble C/
(I+N) level from the best server exceeds a defined threshold.
Preamble Total Noise (I+N) (DL) (dBm)

Atoll calculates preamble total noise (I+N) levels received from the interfering cells on each pixel of the coverage areas of
the best serving cells. A pixel of a coverage area is coloured if the preamble total noise (I+N) level exceeds ( ) the defined
thresholds (the pixel colour depends on the preamble total noise (I+N) level). Coverage consists of several independent
layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer
corresponds to an area where the preamble total noise (I+N) level from the interfering cells exceeds a defined threshold.
9.2.6.3.3
Coverage by Best Bearer (DL) Display Types

It is possible to display the Coverage by Best Bearer (DL) coverage prediction with colours depending on the following
display options.
Best Bearer (DL)

Atoll determines the best bearer available on each pixel of the coverage areas of the best serving cells. A pixel of a coverage area is coloured if a bearer is available (the pixel colour depends on the available bearer). Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as available bearers. Each layer corresponds to an area covered by an available bearer.
9.2.6.3.4
Coverage by Channel Throughput (DL) Display Types

It is possible to display the Coverage by Channel Throughput (DL) coverage prediction with colours depending on the
following display options.
Peak MAC Throughput (DL) (kbps)

Atoll calculates peak MAC channel throughputs available on each pixel of the coverage areas of the best serving cells. A
pixel of a coverage area is coloured if the peak MAC channel throughput exceeds ( ) the defined thresholds (the pixel
colour depends on the peak MAC channel throughput). Coverage consists of several independent layers whose visibility
in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area
where the peak MAC channel throughput exceeds a defined threshold.
Effective MAC Throughput (DL) (kbps)

Atoll calculates effective MAC channel throughputs available on each pixel of the coverage areas of the best serving cells.
A pixel of a coverage area is coloured if the effective MAC channel throughput exceeds ( ) the defined thresholds (the
pixel colour depends on the effective MAC channel throughput). Coverage consists of several independent layers whose
an area where the effective MAC channel throughput exceeds a defined threshold.
Application Throughput (DL) (kbps)

Atoll calculates application level channel throughputs available on each pixel of the coverage areas of the best serving
cells. A pixel of a coverage area is coloured if the application level channel throughput exceeds ( ) the defined thresholds
(the pixel colour depends on the application level channel throughput). Coverage consists of several independent layers
whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the application level channel throughput exceeds a defined threshold.
9.2.6.3.5
Coverage by C/(I+N) Level (UL) Display Types

It is possible to display the Coverage by C/(I+N) Level (UL) coverage prediction with colours depending on the following
display options.
C/(I+N) Level (UL) (dB)

Atoll calculates uplink C/(I+N) levels received at the best serving cells from each pixel of their coverage areas. A pixel of
a coverage area is coloured if the uplink C/(I+N) level exceeds ( ) the defined thresholds (the pixel colour depends on the
uplink C/(I+N) level). Coverage consists of several independent layers whose visibility in the workspace can be managed.
There are as many layers as defined thresholds. Each layer corresponds to an area where the uplink C/(I+N) level from
the pixels at the best serving cells exceeds a defined threshold.
Forsk 2009
AT271_TRG_E6
461
Total Noise (I+N) (UL) (dBm)

Atoll calculates total noise (I+N) levels received at the best serving cells from each pixel of their coverage areas. A pixel
of a coverage area is coloured if the total noise (I+N) level exceeds ( ) the defined thresholds (the pixel colour depends
on the total noise (I+N) level). Coverage consists of several independent layers whose visibility in the workspace can be
managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the total noise (I+N)
level from the pixels at the best serving cells exceeds a defined threshold.
9.2.6.3.6
Coverage by Best Bearer (UL) Display Types

It is possible to display the Coverage by Best Bearer (UL) coverage prediction with colours depending on the following
display options.
Best Bearer (UL)

Atoll determines the best bearer available on each pixel of the coverage areas of the best serving cells. A pixel of a coverage area is coloured if a bearer is available (the pixel colour depends on the available bearer). Coverage consists of
several independent layers whose visibility in the workspace can be managed. There are as many layers as available bearers. Each layer corresponds to an area covered by an available bearer.
9.2.6.3.7
Coverage by Channel Throughput (UL) Display Types

It is possible to display the Coverage by Channel Throughput (UL) coverage prediction with colours depending on the
following display options.
Peak MAC Throughput (UL) (kbps)

Atoll calculates peak MAC channel throughputs available on each pixel of the coverage areas of the best serving cells. A
pixel of a coverage area is coloured if the peak MAC channel throughput exceeds ( ) the defined thresholds (the pixel
colour depends on the peak MAC channel throughput). Coverage consists of several independent layers whose visibility
in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area
where the peak MAC channel throughput exceeds a defined threshold.
Effective MAC Throughput (UL) (kbps)

Atoll calculates effective MAC channel throughputs available on each pixel of the coverage areas of the best serving cells.
A pixel of a coverage area is coloured if the effective MAC channel throughput exceeds ( ) the defined thresholds (the
pixel colour depends on the effective MAC channel throughput). Coverage consists of several independent layers whose
an area where the effective MAC channel throughput exceeds a defined threshold.
Application Throughput (UL) (kbps)

Atoll calculates application level channel throughputs available on each pixel of the coverage areas of the best serving
cells. A pixel of a coverage area is coloured if the application level channel throughput exceeds ( ) the defined thresholds
(the pixel colour depends on the application level channel throughput). Coverage consists of several independent layers
whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the application level channel throughput exceeds a defined threshold.
9.3
Calculation Algorithms
The following sections describe all the calculation algorithms used in point analysis, calculation of coverage predictions,
calculations on subscriber lists, and Monte Carlo simulations.
9.3.1
Co- and Adjacent Channel Overlaps Calculation

A WiMAX network can consist of cells that use different channel bandwidths. Therefore, the start and the end of all the
channel bandwidths may not coincide exactly. Channel bandwidths of different cells can overlap channel bandwidths of
other cells with different ratios.
462
AT271_TRG_E6
Forsk 2009
Figure 9.4: Co-Channel and Adjacent Channel Overlaps

The following sections describe how the co- and adjacent channel overlaps are calculated between the channel used by
any studied cell TXi(ic) and any other cell of the network TXj(jc). In terms of interference calculation, the studied cell can
be considered a victim of interference received from the other cells that might be interfering the studied cell.
TX i ic
If the studied cell is assigned a channel number N Channel , it receives co-channel interference on channel number
TX i ic
TX i ic
TX i ic
N Channel , and adjacent channel interference on the adjacent channel numbers, i.e., N Channel 1 and N Channel + 1 .
In order to calculate the co- and adjacent channel overlaps between two channels, it is necessary to calculate the start
and end frequencies of both channels (explained in "Conversion From Channel Numbers to Start and End Frequencies"
on page 463). Once the start and end frequencies are known for the studied and other cells, the co- and adjacent overlaps,
and the total overlap ratio are calculated as respectively explained in:
9.3.1.1
"Co-Channel Overlap Calculation" on page 464

"Adjacent Channel Overlap Calculation" on page 464
"Total Overlap Ratio Calculation" on page 465
Conversion From Channel Numbers to Start and End Frequencies

Input
TX i ic
TX j jc
F Start FB and F Start FB : Start frequency of the frequency band assigned to the cells TXi(ic) and TXj(jc).
F Start FB can be the start frequency of a TDD frequency band ( F Start FB TDD ), or the uplink or the downlink
start frequency of an FDD frequency band ( F Start FB FDD UL or F Start FB FDD DL ).
TX i ic
TX j jc
N Channel and N Channel : Channel numbers assigned to cells TXi(ic) and TXj(jc).
For FDD networks, Atoll considers that the same channel number is assigned to a cell in the downlink and uplink,
i.e., the channel number you assign to a cell is considered for uplink and downlink both.
TX i ic
TX j jc
W Channel and W Channel : Bandwidths of the channels assigned to cells TXi(ic) and TXj(jc).
Calculations
The following formulas are used to convert the channel numbers into start and end frequencies:
For cell TXi(ic):
TX i ic
F Start
TX i ic
F End
TX i ic
TX i ic
TX i ic
TX i ic
TX i ic
TX i ic
TX j jc
TX j jc
= F Start FB + N Channel W Channel

TX i ic
= F Start FB + N Channel W Channel + W Channel
For cell TXj(jc):

TX j jc
F Start
Forsk 2009
TX j jc
= F Start FB + N Channel W Channel
AT271_TRG_E6
463

TX j jc
TX j jc
TX j jc
TX j jc
TX j jc
= F Start FB + N Channel W Channel + W Channel
F End
Output
9.3.1.2
TX i ic
F Start
F End
TX i ic
TX j jc
and F Start : Start frequencies for the cells TXi(ic) and TXj(jc).
TX j jc
and F End
: End frequencies for the cells TXi(ic) and TXj(jc).
Co-Channel Overlap Calculation

Input
TX i ic
F Start
TX j jc
and F Start : Start frequencies for the cells TXi(ic) and TXj(jc) as calculated in "Conversion From Channel
Numbers to Start and End Frequencies" on page 463.
TX i ic
F End
TX j jc
and F End
: End frequencies for the cells TXi(ic) and TXj(jc) as calculated in "Conversion From Channel
Calculations
Atoll first verifies that co-channel overlap exists between the cells TXi(ic) and TXj(jc).
Co-channel overlap exists if:
TX i ic
TX j jc
F Start F End
TX i ic
AND F End
TX j jc
F Start
Otherwise there is no co-channel overlap.

Atoll calculates the bandwidth of the co-channel overlap as follows:
TX i ic TX j jc
W CCO
TX j jc
TX i ic
= Min F End F End
TX j jc
TX i ic
Max F Start F Start
The co-channel overlap ratio is given by:

TX i ic TX j jc
r CCO
TX i ic TX j jc
W CCO
= -------------------------------------TX i ic
W Channel
Output
9.3.1.3
TX i ic TX j jc
r CCO
: Co-channel overlap ratio between the cells TXi(ic) and TXj(jc).
Adjacent Channel Overlap Calculation

Input
TX i ic
F Start
TX j jc
and F Start : Start frequencies for the cells TXi(ic) and TXj(jc) as calculated in "Conversion From Channel
TX i ic
F End
TX j jc
and F End
: End frequencies for the cells TXi(ic) and TXj(jc) as calculated in "Conversion From Channel
TX i ic
W Channel : Bandwidth of the channel assigned to the studied cell TXi(ic).
Calculations
Atoll first verifies that adjacent channel overlaps exist between (the lower-frequency and the higher-frequency adjacent
channels of) the cells TXi(ic) and TXj(jc).
Adjacent channel overlap exists on the lower-frequency adjacent channel if:
TX i ic
TX i ic
TX j jc
F Start W Channel F End
TX i ic
TX j jc
AND F Start F Start
Adjacent channel overlap exists on the higher-frequency adjacent channel if:

TX i ic
F End
TX j jc
F End
TX i ic
AND F End
TX i ic
TX j jc
+ W Channel F Start
Otherwise there is no adjacent channel overlap.

Atoll determines the adjacent channel overlap ratio as follows:
464
AT271_TRG_E6
Forsk 2009

Bandwidth of the lower-frequency adjacent channel overlap:
TX i ic TX j jc
W ACO
TX j jc
TX i ic
TX j jc
TX i ic
TX i ic
= Min F End F Start Max F Start F Start W Channel
The lower-frequency adjacent channel overlap ratio is given by:

TX i ic TX j jc
r ACO
TX i ic TX j jc
W ACO
L
= -------------------------------------TX i ic
W Channel
Bandwidth of the higher-frequency adjacent channel overlap:

TX i ic TX j jc
W ACO
TX j jc
TX i ic
= Min F End F End
TX i ic
TX j jc
TX i ic
+ W Channel Max F Start F End
The higher-frequency adjacent channel overlap ratio is given by:

TX i ic TX j jc
r ACO
TX i ic TX j jc
W ACO
H
= -------------------------------------TX i ic
W Channel
The adjacent channel overlap ratio is given by:

TX i ic TX j jc
r ACO
TX i ic TX j jc
= r ACO
TX i ic TX j jc
+ r ACO
Output
9.3.1.4
TX i ic TX j jc
r ACO
: Adjacent channel overlap ratio between the cells TXi(ic) and TXj(jc).
FDD TDD Overlap Ratio Calculation

There are many different interference scenarios possible in a WiMAX network depending on the type of duplexing used
by the cells of the network. The most common interference scenarios are FDD-only and TDD-only interferences. However,
co-existing FDD and TDD cells may also exist, and interfere each other. Atoll models the co-existence of FDD and TDD
cells in a network by determining the FDD TDD overlap ratio as follows:
Input
TDD
r DL Frame : Downlink subframe ratio defined in the Global Parameters.
Calculations
The FDD TDD overlap ratio is calculated as follows depending on the frequency bands assigned to the cells TXi(ic) and
TXj(jc):
Frequency Band
TX ic TX jc
i
j
Overlap Ratio r FDD
TDD
TXi(ic)
TXj(jc)
TDD
TDD
TDD
FDD
FDD
TDD
r DL Frame
-------------------------100
FDD
FDD
TDD
Output
9.3.1.5
TX i ic TX j jc
r FDD TDD
: FDD TDD overlap ratio between the cells TXi(ic) and TXj(jc).
Total Overlap Ratio Calculation

Input
TX i ic TX j jc
r CCO
: Co-channel overlap ratio between the cells TXi(ic) and TXj(jc) as calculated in "Co-Channel
Overlap Calculation" on page 464.
TX i ic TX j jc
r ACO
: Adjacent channel overlap ratio between the cells TXi(ic) and TXj(jc) as calculated in "Adjacent
Channel Overlap Calculation" on page 464.
Forsk 2009
AT271_TRG_E6
465
TX i ic TX j jc
r FDD TDD
: FDD TDD overlap ratio between the cells TXi(ic) and TXj(jc) as calculated in "FDD TDD
Overlap Ratio Calculation" on page 465.

TX i ic
f ACS FB : Adjacent channel suppression factor defined for the frequency band of the cell TXi(ic).
W Channel and W Channel : Bandwidths of the channels assigned to the cells TXi(ic) and TXj(jc).
TX i ic
TX j jc
Calculations
The total overlap ratio is:
TX i ic TX j jc
rO
TX i ic
f ACS FB
---------------------------
TX i ic TX j jc
TXi ic TXj jc TX i ic TX j jc
10
+ r ACO
10
r CCO
r FDD TDD
TX i ic
TX j jc
TX i ic
TX j jc
TX i ic
f ACS FB
TX i ic
---------------------------
TX i ic TX j jc W Channel
TX i ic TX j jc TX i ic TX j jc
10
+ r ACO
10
---------------------- r CCO
r FDD TDD
TX j jc
W Channel
TX i ic
W Channel
- is used to normalise the transmission power of the interfering cell TXj(jc). This means
The multiplicative factor ----------------------TX j jc
W Channel
TX j jc
that if the interfering cell transmits at X dBm over a bandwidth of W Channel , and it interferes over a bandwidth less than
TX i ic
W Channel
, the interference from this cell should not be considered at X dBm but less than that. The factor ----------------------TX j jc
W Channel
TX j jc
W Channel
TX j jc
TX j jc
converts X dBm over W Channel to Y dBm (which is less than X dBm) over less than W Channel .
Output
9.3.2
TX i ic TX j jc
rO
: Total co- and adjacent channel overlap ratio between the cells TXi(ic) and TXj(jc).
Preamble Signal Quality Calculations

Calculation of preamble signal quality includes the calculation of the received preamble signal level, and the noise and
interference on the preamble. The following sections also describe how the received preamble signal level, the noise and
interference are calculated over the preamble, and how preamble C/N and C/(I+N) ratios are calculated in Atoll. Refer to
the section corresponding to the details you need.
9.3.2.1
"Preamble Signal Level Calculation" on page 466.

"Preamble Noise Calculation" on page 467.
"Preamble C/N Calculation" on page 470.
"Preamble Interference Calculation" on page 469.
"Preamble C/N Calculation" on page 470.
"Preamble C/(I+N) Calculation" on page 470.
Preamble Signal Level Calculation

Input
TX i ic
P Preamble : Preamble transmission power of the cell TXi(ic).
E SA : Number of antenna elements defined for the smart antenna equipment used by the transmitter TXi.
TX i
TX i
TX i
: Transmitter antenna gain for the antenna used by the transmitter TXi.
: Total transmitter losses for the transmitter TXi ( L
TX i
= L Total DL ).
L Model : Loss on the transmitter-receiver path (path loss) calculated using a propagation model.
L Ant : Antenna attenuation (from antenna patterns) calculated for the antenna used by the transmitter TXi.
M Shadowing Model : Shadowing margin based on the Model standard deviation.
TX i
This parameter is taken into account when the option "Shadowing taken into account" is selected.
M Shadowing Model = 0 when calculating the received preamble signal level from interfering cells TXj(jc).
466
AT271_TRG_E6
Forsk 2009
L Indoor : Indoor losses taken into account when the option "Indoor coverage" is selected.
These losses are defined for each clutter class.
Mi
: Receiver terminal losses for the pixel, subscriber, or mobile Mi.
Mi
Mi
L Ant
: Receiver terminals antenna gain for the pixel, subscriber, or mobile Mi.
: Receiver terminals antenna attenuation calculated for the pixel, subscriber, or mobile Mi.
The antenna used for the calculations at the pixel, subscriber, or mobile Mi is pointed towards its best serving cell.
Mi
Mi
This means that L Ant = 0 for calculating the signal level. Whereas, L Ant is calculated in the direction of the interfering cells from the antenna patterns of the antenna used by Mi, while the antenna is pointed towards Mis best
serving cell.
Mi
L Body : Body loss defined for the service used by the pixel, subscriber, or mobile Mi.
Note:
L
Mi
, G
Mi
Mi
Mi
, L Ant , and L Body are not considered in the calculations performed for the point
analysis tool and the basic, preamble signal level based coverage predictions.
Calculations
The received preamble signal level (dBm) from any cell TXi(ic) is calculated for a pixel, subscriber, or mobile Mi as follows:
TX i ic
TX i ic
C Preamble = EIRP Preamble L Path M Shadowing Model L Indoor + G
Mi
Mi
Mi
Mi
L Ant L Body
Where EIRP is the effective isotropic radiated power of the cell calculated as follows:
TX i ic
TX i ic
Without smart antenna equipment: EIRP Preamble = P Preamble + G
With smart antenna equipment: EIRP Preamble = P Preamble + G
TX i ic
TX i ic
TX i
TX i
L
TX i
TX i
TX i
+ 10 Log E SA
L Path is the path loss (dB) calculated as follows:

TX i

Furthermore, the total losses between the cell and the pixel, subscriber, or mobile Mi can be calculated as follows:
L Total = L Path + L
TX i
+ L Indoor + M Shadowing Model G
TX i
+L
Mi
Mi
Mi
Mi
+ L Ant + L Body
Output
9.3.2.2
TX i ic
C Preamble : Received preamble signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
L Path : Path loss between the cell TXi(ic) and the pixel, subscriber, or mobile Mi.
L Total : Total losses between the cell TXi(ic) and the pixel, subscriber, or mobile Mi.
Preamble Noise Calculation

For determining the preamble C/N and C/(I+N), Atoll calculates the preamble noise over the bandwidth used by the cell.
The used bandwidth depends on the number of subcarriers used by the preamble. In WiMAX 802.16d, the number of
subcarriers used by the preamble is the same as the number of subcarriers used by the cell over the rest of the WiMAX
frame. But, in WiMAX 802.16e, the number of subcarriers used by the preamble can be different from the number of
subcarriers used by the permutation zones.
The preamble noise comprises thermal noise and the noise figure of the equipment. The thermal noise density depends
on the temperature, i.e., it remains constant for a given temperature. However, the value of the thermal noise varies with
the used bandwidth.
Input
K: Boltzmanns constant.
T: Temperature in Kelvin.
N SCa Preamble : Number of used subcarriers defined in the Global Parameters (WiMAX 802.16d) or the number
TX i ic
of subcarriers used by the preamble defined for the frame configuration of the cell TXi(ic) (WiMAX 802.16e).
TX i ic
N SCa Total : Total number of subcarriers defined in the Global Parameters (WiMAX 802.16d) or for the frame configuration of the cell TXi(ic) (WiMAX 802.16e).
Forsk 2009
AT271_TRG_E6
467

TX i ic
F Sampling : Sampling frequency for the cell TXi(ic) as calculated in "Calculation of Sampling Frequency" on
page 493.
nf
Mi
: Noise figure of the terminal used for calculations by the pixel, subscriber, or mobile Mi.
Calculations
The power spectral density of thermal noise is calculated as follows:
n 0 = 10 Log 10 K T 1000 = 174 dBm/Hz
The thermal noise over the preamble for a cell is calculated as:
TX i ic
n 0 Preamble
TX i ic
TX ic
N SCa Preamble
i
= n 0 + 10 Log 10 F Sampling --------------------------------------TX i ic
N SCa Total
The preamble noise is the sum of the thermal noise and the noise figure of the terminal used for the calculations by the
pixel, subscriber, or mobile Mi.
TX i ic
TX i ic
n Preamble = n 0 Preamble + nf
Mi
Effect of Segmentation (WiMAX 802.16e):

If you select the Segmentation Support check box in the Frame Configurations table for the frame configuration
used by the cell, it means that the preamble is segmented, and one of the three preamble carrier sets is used for
transmission.
Preamble
1
The preamble segmenting factor is f Segment = --- , which means that each preamble carrier set uses 1/3rd of the
3
1 total number of preamble subcarriers. The power transmitted over a segmented preamble has ---------------------times the
Preamble
f Segment
spectral density of the power transmitted over the entire channel bandwidth. When calculating the C/N and C/(I+N)
1
- due to this power concentration is equivalent to a reduction in the noise
ratios, the increase in power by ---------------------Preamble
f Segment
Preamble
level by f Segment . This power concentration due to segmentation on the C/N and C/(I+N) results in an increase
in the coverage footprint of the preamble. Hence, if segmentation is used, the thermal noise at the pixel,
Preamble
subscriber, or mobile Mi covered by the preamble is reduced by a factor of f Segment . Which means that the thermal noise over the preamble for a cell is calculated as:
TX i ic
n 0 Preamble
TX i ic
TX ic
N SCa Preamble Preamble
i
- f Segment
= n 0 + 10 Log 10 F Sampling --------------------------------------TX i ic
N SCa Total
The following table shows the different types of subcarriers and their numbers for preamble transmission in
WiMAX 802.16e.
N SCa Total
128
512
1024
2048
Guard Subcarriers
N SCa Preamble
All
1 (54)
107
1 (54)
35
0.3271
Left
10
Right
Total
10
20
f Segment
None
36
0.3364
None
36
0.3364
All
1 (214)
428
None
143
0.3341
42
41
83
1 (214)
142
0.3318
None
143
0.3341
All
1 (426)
851
1 (426)
283
0.3325
86
86
172
None
284
0.3337
None
284
0.3337
All
1 (852)
1703
1 (852)
567
0.3329
None
568
0.3335
None
568
0.3335
172
172
344
468
Preamble
DC Subcarrier
Segment
AT271_TRG_E6
Forsk 2009
Output
TX i ic
9.3.2.3
n Preamble : Preamble noise for the cell TXi(ic).
Preamble Interference Calculation

The interference received by any cell TXi(ic) from other cells TXj(jc) for a pixel, subscriber, or mobile Mi can be defined as
the preamble signal level received from cells TXj(jc) depending on the overlap that exists between the channels used by
the cells TXi(ic) and TXj(jc), and (for WiMAX 802.16e) whether the preambles of the two cells are segmented or not.
Input
TX j jc
C Preamble : Preamble signal level received from any interfering cell TXj(jc) (as calculated in "Preamble Signal
Level Calculation" on page 466) at the pixel, subscriber, or mobile Mi covered by the cell TXi(ic).
TX i ic TX j jc
rO
: Total channel overlap ratio between the cells TXi(ic) and TXj(jc) as calculated in "Co- and Adjacent
Channel Overlaps Calculation" on page 462.

TX i ic
N Seg
TX j jc
and N Seg
: Segment numbers assigned to the cells TXi(ic) and TXj(jc) calculated from their respective
TX i ic
TX j jc
preamble indexes ( n Preamble and n Preamble ) as follows:

n Preamble
N Seg
0 to 31, 96, 99, 102, 105, 108, 111
32 to 63, 97, 100, 103, 106, 109, 112
64 to 95, 98, 101, 104, 107, 110, 113
Calculations
The received preamble interference (dBm) from any cell TXj(jc) is calculated for a pixel, subscriber, or mobile Mi as follows:
TX j jc
TX j jc
TX i ic TX j jc
I Preamble = C Preamble + f O
TX i ic TX j jc
Where f O
TX i ic TX j jc
+ f Seg
is the interference reduction factor due to channel overlap between the cells TXi(ic) and TXj(jc),
TX i ic TX j jc
fO
TX i ic TX j jc
= 10 Log r O
TX i ic TX j jc
And, f Seg
is the interference reduction factor due to preamble segmentation (WiMAX 802.16e), calculated as
follows:
TX i ic TX j jc
f Seg
TX i ic TX j jc
= 10 Log p Collision
TX i ic TX j jc
The probability of preamble subcarrier collision p Collision
between the cells TXi(ic) and TXj(jc) is given by the follow-
ing table:
TX i ic TX j jc
TX i ic
TX j jc
Not Segmented
Not Segmented
Not Segmented
Segmented
1/3
Segmented
Not Segmented
p Collision
TX i ic
Segmented
Segmented
0 if N Seg
TX i ic
1 if N Seg
TX j jc
N Seg
TX j jc
= N Seg
Output
TX j jc
I Preamble : Preamble interference received from any interfering cell TXj(jc) at the pixel, subscriber, or mobile Mi
covered by a cell TXi(ic).
Forsk 2009
AT271_TRG_E6
469
9.3.2.4
Preamble C/N Calculation

Input
TX i ic
C Preamble : Received preamble signal level from the cell TXi(ic) as calculated in "Preamble Signal Level Calculation" on page 466.
TX i ic
n Preamble : Preamble noise for the cell TXi(ic) as calculated in "Preamble Noise Calculation" on page 467.
Calculations
The preamble C/N for a cell TXi(ic) is calculated as follows for any pixel, subscriber, or mobile Mi:
TX i ic
TX i ic
TX i ic
CNR Preamble = C Preamble n Preamble
Output
9.3.2.5
TX i ic
CNR Preamble : Preamble C/N from the cell TXi(ic) at any pixel, subscriber, or mobile Mi.
Preamble C/(I+N) Calculation

The carrier signal to interference and noise ratio is calculated in three steps. First Atoll calculates the received preamble
signal level from the studied cell (as explained in "Preamble Signal Level Calculation" on page 466) at the pixel, subscriber
or mobile under study. Secondly, Atoll calculates the interference received at the same studied pixel, subscriber, or mobile
from all the interfering cells (as explained in "Preamble Interference Calculation" on page 469). Interference from each cell
is weighted according to the co- and adjacent channel overlap between the studied and the interfering cells, and (in WiMAX
802.15e) the probabilities of subcarrier collision if segmentation is used. Finally, Atoll takes the ratio of the preamble signal
level, and the sum of the total interference from all interfering cells and the noise (as calculated in "Preamble Noise Calculation" on page 467).
The receiver terminal is always considered to be pointed towards its best server. In the case of NLOS between the receiver
and the best server, Atoll does not try to find the direction of the strongest signal, the receiver is oriented towards the best
server just as in the case of LOS.
Input
TX i ic
C Preamble : Preamble signal level received from the cell TXi(ic) at a pixel, subscriber, or mobile Mi as calculated
in "Preamble Signal Level Calculation" on page 466.
TX i ic
n Preamble : Preamble noise for the cell TXi(ic) as calculated in "Preamble Noise Calculation" on page 467.
I Preamble : Preamble interference received from any cell TXj(jc) at a pixel, subscriber, or mobile Mi covered by a
TX j jc
cell TXi(ic) as calculated in "Preamble Interference Calculation" on page 469.
M Shadowing C I : Shadowing margin based on the C/I standard deviation.

TX i ic
CNR Preamble : Preamble C/N from the cell TXi(ic) at a pixel, subscriber, or mobile Mi as calculated in "Preamble
C/N Calculation" on page 470.
Calculations
The preamble C/(I+N) for a cell TXi(ic) is calculated as follows at any pixel, subscriber, or mobile Mi:
TX i ic
CINR Preamble
TX i ic
C Preamble
TX jc
j
I Preamble
TX i ic
------------------------ 10 Log
10
n Preamble M Shadowing C I
10
All TX j jc
The Preamble Total Noise (I+N) for a cell TXi(ic) is calculated as follows for any pixel, subscriber, or mobile Mi:
TX j jc
TX i ic
I + N Preamble
TX i ic
I Preamble
n Preamble
--------------------------------------------------
10
10
10
+ 10
= 10 Log
All TX j jc
Output
470
TX i ic
CINR Preamble : Preamble C/(I+N) from the cell TXi(ic) at a pixel, subscriber, or mobile Mi.
AT271_TRG_E6
Forsk 2009

TX i ic
I + N Preamble : Preamble total noise from the interfering cells TXj(jc) at the pixel, subscriber, or mobile Mi covered
by a cell TXi(ic).
9.3.3
Best Server Determination

In WiMAX, best server refers to a cell ("serving transmitter"-"reference cell" pair) from which a pixel, subscriber, or mobile
TX j jc
Mi gets the best preamble signal level ( C Preamble ).

This calculation also determines whether the pixel, subscriber, or mobile Mi is within the coverage area of any transmitter
or not.
Input
TX i ic
C Preamble : Preamble signal level received from any cell TXi(ic) at a pixel, subscriber, or mobile Mi as calculated
in "Preamble Signal Level Calculation" on page 466 using the terminal and service parameters ( L
Mi
, G
Mi
Mi
, L Ant ,
Mi
and L Body ) of Mi.
Calculations
The best server of any pixel, subscriber, or mobile Mi, BS M , is the cell from which the received preamble signal level is
i
the highest among the preamble signal levels received from all the cells. The best server is determined as follows:
BS M = TX i ic
i
TX ic
i
C Preamble =
Best
All TX ic
i
TXi ic
C Preamble
Output
9.3.4
BS M : Best serving cell of the pixel, subscriber, or mobile Mi.

i
Service Area Calculation

In WiMAX, a pixel, subscriber, or mobile Mi can be covered by a cell (as calculated in "Best Server Determination" on
page 471) but can be outside the service area. A pixel, subscriber, or mobile Mi is said to be within the service area of its
best serving cell TXi(ic) if the preamble C/N from the cell at the pixel, subscriber, or mobile is greater than or equal to the
preamble quality threshold defined for the cell.
Input
TX i ic
CNR Preamble : Preamble C/N from the cell TXi(ic) at a pixel, subscriber, or mobile Mi as calculated in "Preamble
TX i ic
T Preamble : Preamble quality threshold defined for the cell TXi(ic).
Calculations
A pixel, subscriber, or mobile Mi is within the service area of its best serving cell TXi(ic) if:
TX i ic
TX i ic
CNR Preamble T Preamble
Output
9.3.5
True: If the calculation criterion is satisfied.

False: Otherwise.
Permutation Zone Selection (WiMAX 802.16e)

In order to be able to calculate the traffic C/(I+N), and therefore the throughputs, a permutation zone is assigned to each
pixel, subscriber, or mobile Mi located within the service area (as calculated in "Service Area Calculation" on page 471) of
its best serving cell. The permutation zone assigned to Mi is one which covers Mi in terms of distance and preamble C/N,
and accepts user speeds equal to or higher than Mis speed selected for the calculation.
A pixel, subscriber, or mobile Mi which is unable to get a permutation zone is considered to be outside the service area.
Forsk 2009
AT271_TRG_E6
471
Input
TX i ic
d Max PZ : Maximum distance covered by a permutation zone of a cell TXi(ic).
CNR Req PZ : Minimum preamble C/N required at the pixel, subscriber, or mobile Mi to connect to a permutation
TX i ic
zone of a cell TXi(ic).

TX i ic
Speed Max PZ : Maximum speed supported by a permutation zone of a cell TXi(ic).
TX i ic
CNR Preamble
Mobility M i : Speed of the pixel, subscriber, or mobile Mi.
M i TX i ic
: Distance between the pixel, subscriber, or mobile Mi and a cell TXi(ic).

: Preamble C/N from the cell TXi(ic) as calculated in "Preamble C/N Calculation" on page 470.
Calculations
Mi is assigned the permutation zone with the highest priority among the permutation zones whose selection criteria Mi
satisfies. Mi satisfies the selection criteria of a permutation zone if:
The distance between Mi and TXi(ic) is less than or equal to the maximum distance covered by the permutation
zone:
d
M i TX i ic
TX i ic
d Max PZ
The preamble C/N at Mi is better than or equal to the minimum preamble C/N required by the permutation zone:
TX i ic
TX i ic
CNR Preamble CNR Req PZ
The mobility of Mi is less than or equal to the maximum mobile speed supported by the permutation zone:
TX i ic
Mobility M i Speed Max PZ

Therefore, the permutation zones assigned to a pixel, subscriber, or mobile Mi in the downlink and uplink are:
TX ic
Mi
i
PZ DL = Highest Priority PZ DL
TX ic
Mi
i
PZ UL = Highest Priority PZ UL
M TX ic
TX ic
TX ic
TX ic
i
i
i
i
d i
d Max PZ AND CNRPreamble CNRReq PZ
TX ic
i
AND Mobility Mi SpeedMax PZ
M TX ic
TX ic
TX ic
TX ic
i
i
i
i
d i
d Max PZ AND CNRPreamble CNRReq PZ
TX ic
i
AND Mobility Mi SpeedMax PZ
If N (> 1) permutation zones satisfy the distance, speed, and signal quality criteria, and all have the same priority, the
permutation zone assigned to the pixel, subscriber, or mobile will be the first in the list of permutation zones (frame configuration) among these N zones.
Output
9.3.6
Mi
Mi
PZ DL and PZUL : Downlink and uplink permutation zones assigned to the pixel, subscriber, or mobile Mi.
Traffic and Pilot Signal Quality Calculations

Traffic and pilot subcarriers can be transmitted with different transmission powers than the preamble power of a cell. Moreover, traffic and pilot subcarriers do not undergo the same interference and noise as the preamble. The following sections
describe how traffic and pilot signal levels, noise and interference are calculated on the downlink and uplink.
472
"Traffic and Pilot Signal Level Calculation (DL)" on page 473.

"Noise Calculation (DL)" on page 474.
"Traffic and Pilot Interference Calculation (DL)" on page 475.
"Traffic and Pilot C/N Calculation (DL)" on page 482.
"Traffic C/(I+N) Calculation (DL)" on page 483.
"Traffic Signal Level Calculation (UL)" on page 484.
"Noise Calculation (UL)" on page 485.
"Traffic Interference Calculation (UL)" on page 486.
"Traffic C/N Calculation (UL)" on page 487.
"Traffic C/(I+N) Calculation (UL)" on page 490.
AT271_TRG_E6
Forsk 2009
9.3.6.1
Traffic and Pilot Signal Level Calculation (DL)

Input
TX i ic
P Preamble : Preamble transmission power of the cell TXi(ic).
P Traffic : Traffic power offset of the cell TXi(ic).
P Pilot
TX i ic
TX i ic
TX i
: Pilot power offset of the cell TXi(ic).
Without smart antenna equipment: G
With smart antenna equipment: G

mobile Mi, i.e., G
TX i
TX i
TX i
is the transmitter antenna gain, i.e., G
TX i
TX i
= G Ant .
is the smart antenna gain in the direction of the pixel, subscriber, or
= G SA . Where is the direction in which Mi is located. For more information on the
calculation of G SA , refer to section "Optimum Beamformer Model" on page 505.

L
TX i

TX i
L Ant
TX i
= L Total DL ).
L Path : Path loss ( L Path = L Model +
).
TX i
M Shadowing Model = 0 when calculating the received traffic and pilot signal levels for C/(I+N) calculations.
Mi
Mi
Mi
L Ant
The antenna used for the calculations at the pixel, subscriber, or mobile Mi is pointed towards its best serving cell.
Mi
This means that L Ant = 0 for calculating the traffic and pilot signal level.
Mi
Calculations
The received traffic and pilot signal levels (dBm) from any cell TXi(ic) are calculated for a pixel, subscriber, or mobile Mi
as follows:
TX i ic
TX i ic
C Traffic = EIRP Traffic L Path M Shadowing Model L Indoor + G

TX i ic
TX i ic
C Pilot
= EIRP Pilot
L Path M Shadowing Model L Indoor + G
Mi
Mi
L
L
Mi
Mi
Mi
Mi
Mi
Mi
L Ant L Body and

L Ant L Body
TX i ic
TX i ic
EIRP Traffic = P Traffic + G

TX i ic
TX i ic
With P Traffic and P Pilot

TX i ic
TX i ic
TX i
TX i
TX i ic
and EIRP Pilot
TX i ic
= P Pilot
+G
TX i
TX i
being the traffic and pilot transmission powers of the cell TXi(ic) calculated as follows:
TX i ic
TX i ic
P Traffic = P Preamble P Traffic and P Pilot
TX i ic
TX i ic
= P Preamble P Pilot
Output
Forsk 2009
TX i ic
C Traffic : Received traffic signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
C Pilot
TX i ic
: Received pilot signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
AT271_TRG_E6
473
9.3.6.2
Noise Calculation (DL)

For determining the traffic and pilot C/N and C/(I+N), Atoll calculates the downlink noise over the channel bandwidth used
by the cell. The used bandwidth depends on the number of used subcarriers. In WiMAX 802.16e, the numbers of subcarriers used by each permutation zone can be different.
The downlink noise comprises thermal noise and the noise figure of the equipment. The thermal noise density depends
on the temperature, i.e., it remains constant for a given temperature. However, the value of the thermal noise varies with
the used bandwidth.
Input
N SCa Used : Number of used subcarriers defined in the Global Parameters (WiMAX 802.16d).
N SCa Total : Total number of subcarriers defined in the Global Parameters (WiMAX 802.16d).
N SCa Used : Number of subcarriers used by the downlink permutation zone of a WiMAX 802.16e cell TXi(ic)
PZ DL
assigned to Mi.
TX i ic
N SCa Total : Total number of subcarriers defined for the frame configuration of a WiMAX 802.16e cell TXi(ic).
TX i ic
page 493.
nf
Mi
: Noise figure of the terminal used for calculations by the pixel, subscriber, or mobile Mi.
Calculations
n 0 = 10 Log 10 K T 1000 = 174 dBm/Hz
The thermal noise for a cell is calculated as:
TX i ic
TX i ic
N SCa Used
WiMAX 802.16d: n 0 DL = n 0 + 10 Log 10 F Sampling ------------------------------
N SCa Total
M
WiMAX 802.16e (DL):
TX i ic
n 0 DL
PZ DL
N SCa Used
TX i ic
-
= n 0 + 10 Log 10 F Sampling ----------------------------TX i ic
N SCa Total
The downlink noise is the sum of the thermal noise and the noise figure of the terminal used for the calculations by the
pixel, subscriber, or mobile Mi.
TX i ic
n DL
TX i ic
= n 0 DL + nf
Mi
Effect of Segmentation (WiMAX 802.16e):

used by the cell, it means that the first downlink PUSC permutation zone is segmented. All other zones are pooled
together to form a non-segmented zone.
The segmenting factor, f Segment , is calculated from the number of secondary subchannel groups assigned to the
permutation zone in the Permutation Zones table.
3 PSG + 2 SSG
f Segment = ---------------------------------------------------15
Where, PSG is the number of primary subchannel groups, which is always 1, and SSG is the number of used
secondary subchannel groups, which can be 0, 1, 2, or 3. Segmenting factors for different numbers of secondary
subchannel groups can be.
474
Number of SSG
f Segment
1/5
1/3
7/15
3/5
AT271_TRG_E6
Forsk 2009

f Segment represents the fraction of the channel bandwidth used by a segment. The power transmitted over a
1
segment has ---------------------- times the spectral density of the power transmitted over the entire channel bandwidth.
f Segment
1
When calculating the downlink C/N and C/(I+N) ratios, the increase in power by ---------------------- due to this power
f Segment
concentration is equivalent to a reduction in the noise level by f Segment . Hence, if segmentation is used, the thermal noise power at the pixel, subscriber, or mobile Mi covered by the segmented permutation zone is reduced by
a factor of f Segment . Which means that the thermal noise for the a segment of the channel used by a cell is calculated as:
M
TX i ic
n 0 DL
PZ DL
N SCa Used
TXi ic
- f Segment
N SCa Total
Output
9.3.6.3
TX i ic
n DL
: Downlink noise for the cell TXi(ic).
Traffic and Pilot Interference Calculation (DL)

The interference received by at any pixel, subscriber, or mobile, served by a cell TXi(ic) from other cells TXj(jc) can be
defined as the traffic and pilot signal levels received from interfering cells TXj(jc) depending on the overlap that exists
between the channels used by the cells TXi(ic) and TXj(jc), on the traffic loads of the interfering cells TXj(jc), and whether
the cells use segmentation or not. Moreover, the interference can come from cells using classic as well as smart antennas.
The calculation can be divided into the two parts.
9.3.6.3.1
"Traffic and Pilot Interference Signal Levels Calculation (DL)" on page 475.
"Effective Traffic and Pilot Interference Calculation (DL)" on page 478.
Traffic and Pilot Interference Signal Levels Calculation (DL)

The traffic and pilot signal levels received from interfering cells TXj(jc) at a pixel, subscriber, or mobile Mi, covered by a
cell TXi(ic), are calculated in a different manner than the traffic and pilot signal levels from the studied cell TXi(ic). This
section explains how these interfering signals are calculated.
Input
TX j jc
P Preamble : Preamble transmission power of the cell TXj(jc).
P Pilot
P Traffic : Traffic power offset of the interfering cell TXj(jc).
P Idle Pilot : Idle pilot power offset of the interfering cell TXj(jc).
L Path : Path loss ( L Path = L Model + L Ant ).
L Ant : Antenna attenuation (from antenna patterns) calculated for the antenna used by the transmitter TXj.
TX j jc
: Pilot power offset of the interfering cell TXj(jc).
TX j jc
TX j jc
TX j
: Total transmitter losses for the transmitter TXj ( L
TX j
= L Total DL ).
TX j
TX j
Mi
Mi
Mi
L Ant
Mi
Mi
When calculating L Ant , Mis antenna is pointed towards its best serving cell TXi(ic). L Ant is calculated in the direction of TXj(jc) from the antenna patterns of the antenna used by Mi while the antenna is pointed towards TXi(ic).
Forsk 2009
Mi
TL DL
TX j jc
: Downlink traffic load of the interfering cell TXj(jc).
AT271_TRG_E6
475

Traffic loads can either be calculated using Monte Carlo simulations, or entered manually for each cell. Calculation
of traffic loads is explained in "Simulation Process" on page 455.
AU
TX j jc
: AAS usage ratio of the interfering cell TXj(jc).
AAS usage ratios are calculated using Monte Carlo simulations as explained in "Simulation Process" on page 455.
TX j jc
N SCa Used : Number of used subcarriers defined for the first downlink permutation zone in the frame configuration
assigned to the interfering cell TXj(jc).
TX j jc
N SCa Data : Number of data subcarriers defined for the first downlink permutation zone in the frame configuration
assigned to the interfering cell TXj(jc).
Calculations
WiMAX cells can transmit different powers on pilot (NUsed NData) and data (NData) subcarriers for the part of the frame
with traffic, and a different pilot power for the part of the frame that does not have traffic bursts. Data subcarriers are off
during the empty part of the frame. Therefore, the interference received from a cell depends on the traffic load and the
different powers of the cell, i.e., pilot, traffic, and idle pilot powers.
Monte Carlo simulations and coverage prediction calculations present different scenarios for interference calculations in
the case of smart antennas.
Monte Carlo Simulations:

In the case of Monte Carlo simulations, the interferer is either using the transmitter antenna or the smart antenna
at any given moment. So, for each interfered pixel, subscriber, or mobile, Atoll already knows the type of the interference source. Therefore, the interference received from any cell TXj(jc) can be given by:
TX j jc
TX j jc
Without smart antennas: I Total
TX j jc
I Idle
I Non AAS
------------------
-------------------------10
10
= 10 Log 10
+ 10
TXj jc
TX j jc
With smart antennas: I Total
I AAS
----------------10
= 10 Log 10
Coverage Predictions:
In the case of coverage prediction calculations, the interferer could either be transmitting using the transmitter
antenna, or using the smart antenna, or it could be empty, or not transmitting.Therefore, the interference received
from any cell TXj(jc) can be given by:
TX j jc
TX j jc
I Total
TX j jc
TX j jc
I Idle
I AAS
INon AAS
-----------------------------------
-------------------------10
10
10
= 10 Log 10
+ 10
+ 10
Where, the three components of the interference are:

TX j jc
I Non AAS : interference from the loaded part of the frame transmitted using the transmitter antenna,
I AAS
I Idle
TX j jc
TX j jc
: interference from the loaded part of the frame transmitted using the smart antenna,
: interference from the empty, or idle, part of the frame.
The above components of the interference are calculated as follows:

The interference from the loaded part of the frame transmitted using the transmitter antenna is calculated
as follows:
The received interfering traffic and pilot signal levels (dBm) from any cell TXj(jc) are calculated for a pixel,
subscriber, or mobile Mi as follows:
TX j jc
TX j jc
I Traffic = EIRP Traffic L Path L Indoor + G

TX j jc
I Pilot
TX j jc
= EIRP Pilot
L Path L Indoor + G
Mi
Mi
L
L
Mi
Mi
Mi
Mi
Mi
Mi
L Ant L Body
L Ant L Body
TX j jc
TX j jc
EIRP Traffic = P Traffic + G

TX j jc
TX j jc
With P Traffic and P Pilot
476
TX j
TX j
TX j jc
and EIRP Pilot
TX j jc
= P Pilot
+G
TX j
TX j
being the traffic and pilot transmission powers of the cell TXj(jc) calculated as follows:
AT271_TRG_E6
Forsk 2009

TX j jc
TX j jc
TX j jc
P Traffic = P Preamble P Pilot

And, G
TX j
TX j jc
and P Pilot
TX j jc
TX j jc
= P Preamble P Pilot
TX j
= G Ant , i.e., the transmitter antenna gain for the antenna used by the transmitter TXj.
The interference from the loaded part of the frame transmitted using the transmitter antenna is given as:
TX j jc
TX j jc
I Pilot
ITraffic
TX j jc
TX jc
TX j jc
------------------ N j
N SCa Data
TX j jc
TX j jc
----------------10
SCa Used N SCa Data
- + 10 10 ----------------------------------------------------------------
10
= 10 Log TL DL
1 AU
----------------------------TX j jc
TX j jc
N SCa Used
N SCa Used
TX j jc
I Non AAS
The interference from the loaded part of the frame transmitted using the smart antenna is calculated as
follows:
The received interfering traffic signal level (dBm) from any cell TXj(jc) is calculated for a pixel, subscriber, or mobile
Mi as follows:
TX j jc
TX j jc
I AAS
= EIRP AAS
L Path L Indoor + G
Mi
Mi
Mi
Mi
L Ant L Body
TX j jc
TX j jc
EIRP AAS
= P Traffic + G
TX j
TX j
TX j jc
With P Traffic being the traffic transmission power of the cell TXj(jc) calculated as follows:
TX j jc
TX j jc
TX j jc
P Traffic = P Preamble P Pilot

And, G
-
TX j
is the smart antenna gain in the direction of the victim mobile Mi, calculated as follows:
Monte Carlo Simulations:

G
TX j
= G SA is calculated in the direction of the victim mobile Mi, while the smart antenna at the interfering
cell has a beam formed in the direction of an interfering mobile Mj. is the direction, with respect to the azimuth of the cell TXj(jc), in which the victim mobile Mi is located. For more information on the calculation of
G SA , refer to section "Optimum Beamformer Model" on page 505. Victim and interfering mobiles are generated by a time-slot scenario as explained in "Simulation Process" on page 455.
TX i ic
In the figure below, G SA is calculated from the victim cell TXi(ic) to calculate C Traffic , and G SA is calTX j jc
culated from the interfering cell TXj(jc) to calculate I AAS
. In other words, a beam is formed from the victim
cell towards the victim mobile, and another beam is formed by the interfering cell towards the interfering mobile.
Figure 9.5: Downlink C/(I+N) calculation in Simulations

-
Coverage Predictions:
G
TX j
= G SA is calculated in the direction of the victim mobile Mi from the angular distribution of the down-
link traffic power density available at the end of the simulations. The angular distribution of the downlink traffic
power density, which represents the average array correlation matrix, is calculated during Monte Carlo simulations and can be stored in the Cells table. is the direction in which the victim pixel or subscriber Mi is located. For more information on the calculation of G SA , refer to section "Optimum Beamformer Model" on
page 505.
TX i ic
In the figure below, G SA is calculated from the victim cell TXi(ic) to calculate C Traffic , and G SA is calTX j jc
culated from the interfering cell TXj(jc) to calculate I AAS
. In other words, a beam is formed from the victim
cell towards the victim pixel or subscriber, and the interference is calculated from the simulation results.
Forsk 2009
AT271_TRG_E6
477
Figure 9.6: Downlink C/(I+N) calculation in Coverage Predictions

The average array correlation matrix (at the interfering cell) are given by:
J
R Avg =
j pj Rj
j=1
Where R Avg is the average array correlation matrix, J is the number of served mobiles during the simulation,
j is the probability of presence of the mobile j, p j is the EIRP transmitted towards the mobile j, and R j is the
array correlation matrix for the mobile j.
The probability of presence of the mobile j is the ratio between the downlink resources provided to the mobile
j and the total amount of available downlink resources. For example, if a mobile has been granted 10% of the
number of available slots in the downlink subframe, its probability of presence is 10%.
The gain of the interfering signal, G SA , transmitted in the direction of each pixel is given by:
H
G SA = g n S R Avg S
Where S is the steering vector in the direction (probe mobile/pixel), H denotes the Hilbert transformation,
R Avg is the average array correlation matrix, and g n is the gain of the nth antenna element in the direction
of .
The interference from the empty, or idle, part of the frame transmitted using the transmitter antenna is
The received interfering pilot signal level (dBm) from any cell TXj(jc) is calculated for a pixel, subscriber, or mobile
Mi as follows:
TX j jc
TX j jc
I Idle Pilot = EIRP Idle Pilot L Path L Indoor + G
Mi
Mi
Mi
Mi
L Ant L Body
TX j jc
TX j jc
EIRP Idle Pilot = P Idle Pilot + G
TX j
TX j
TX j jc
With P Idle Pilot being the idle pilot transmission power of the cell TXj(jc) calculated as follows:
TX j jc
TX j jc
TX j jc
P Idle Pilot = P Preamble P Idle Pilot

And, G
TX j
TX j
= G Ant , i.e., the transmitter antenna gain for the antenna used by the transmitter TXj.
The interference from the empty, or idle, part of the frame transmitted using the transmitter antenna is given as:
TX j jc
TX j jc
I Idle
I Idle Pilot
TX jc
TX j jc
- N j
TX j jc
-------------------------
10
SCa Used N SCa Data
-
---------------------------------------------------------------= 10 Log 1 TL DL 10
TX j jc
N SCa Used
Output
9.3.6.3.2
TX j jc
I Total : Interference received at the pixel, subscriber, or mobile Mi from any interfering cell TXj(jc).
Effective Traffic and Pilot Interference Calculation (DL)

The effective downlink traffic and pilot interference received at a pixel, subscriber, or mobile Mi covered by a cell TXi(ic)
from interfering cells TXj(jc) depends on the co- and adjacent channel overlap that exists between the channel used by
the studied cell and the interfering cells, and the segmentation parameters of the studied and interfering cells (WiMAX
802.16e). The first downlink PUSC zone can be segmented at the studied and the interfering cells. The probability of
subcarrier collision depends on the lengths of the segmented zones and on the subchannel groups used at both sides.
478
AT271_TRG_E6
Forsk 2009
Input
TX j jc
I Total : Interference received at the pixel, subscriber, or mobile Mi from any interfering cell TXj(jc) as calculated
in "Traffic and Pilot Interference Signal Levels Calculation (DL)" on page 475.
TX i ic TX j jc
rO
Calculations
The total traffic and pilot interference (dBm) from any cell TXj(jc) is calculated for a pixel, subscriber, or mobile Mi as
follows:
TX j jc
I DL
TX j jc
TX i ic TX j jc
= I Traffic + f O
TX i ic TX j jc
+ f Seg
Calculations for the interference reduction factors due to channel overlapping, downlink traffic load, and segmentation are
explained below:
Interference reduction due to the co- and adjacent channel overlap between the studied and the interfering
cells:
Interference reduction due to the co- and adjacent channel overlap between the cells TXi(ic) and TXj(jc) is calculated as follows:
TX i ic TX j jc
fO
TX i ic TX j jc
= 10 Log r O
Interference reduction due to segmentation (WiMAX 802.16e):

that you are using, it means that the first zone in the downlink, i.e., the DL PUSC zone, is segmented. All other
zones are pooled together to form a group of non-segmented zones. There are two effects of segmentation:
a. Power concentration, which means that the spectral density of the power transmitted over one segment is
higher than the spectral density of the same power transmitted over the entire channel bandwidth. The effect
of power concentration is visible when calculating the downlink C/(I+N). The power transmitted over a seg1
mented zone has ---------------------- times the spectral density of the power transmitted over the entire channel bandf Segment
1
width. When calculating the C/(I+N) ratio, the increase in power by ---------------------- is equivalent to decreasing the
f Segment
noise and interference by f Segment . Hence, if segmentation is used, the interference received at the pixel, subscriber, or mobile Mi covered by the segmented zone is reduced by a factor of f Segment .
b. Collision probability between the subcarriers used by the subchannels belonging to the segment of the studied
cell and the subcarriers used by other sectors, segmented or not. The following paragraphs explain how the
collision probability is calculated.
The Segmentation Usage (SU) ratio is the percentage of the total downlink traffic load present in the segmented downlink PUSC zone. For example, if the downlink traffic load is 80 %, and the segmentation usage ratio
is 50 %, then this means that the downlink traffic load of the segmented zone is 40 % (i.e., 50 % of 80 %), and
the downlink traffic load of the non-segmented zones is 40 %.
In coverage predictions, Atoll uses the segmentation usage ratios stored in the cell properties for determining
the interference. In simulations, Atoll resets the segmentation usage ratios for all the cells to 0, and then calculates the segmentation usage ratios according to the traffic loads of the mobiles allocated to the segmented
zone and in the non-segmented zones.
Figure 9.7: Segmentation

Atoll determines the switching point between the segmented and the non-segmented zones using the segmentation usage ratio.
Forsk 2009
AT271_TRG_E6
479

The switching points between the segmented and non-segmented zones of the victim and interfering cells,
TXi(ic) and TXj(jc) respectively, are calculated as follows:
SP
SP
TX i ic
TX i ic
SU
= ------------------------------------------------------------------------------------------------ and
TX i ic
TX i ic
TX i ic
SU
+ f Segment 1 SU
TX j jc
SU
= -----------------------------------------------------------------------------------------------TX j jc
TX j jc
TX j jc
SU
+ f Segment 1 SU
TX j jc
Where, SP is the switching point between the segmented and the non-segmented zones, SU is the segmentation usage ratios of the cells, and f Segment is segmenting factor, which gives the bandwidth used by a segment.
The segmenting factor, f Segment , is calculated from the number of secondary subchannel groups assigned to
the first downlink PUSC permutation zone in the Permutation Zones table.
3 PSG + 2 SSG
f Segment = ---------------------------------------------------15
Where, PSG is the number of primary subchannel groups, which is always 1, and SSG is the number of secondary subchannel groups, which is 1 for FFT sizes 128 and 512, and can be 0, 1, 2, or 3 for FFT sizes 1024
and 2048. Segmenting factors for different numbers of secondary subchannel groups are:
SSG
f Segment
1/5
1/3
7/15
3/5
Note:
The multiplicative coefficients of 3 and 2 are derived from the ratio of the numbers of
subchannels that belong to the primary and to the secondary subchannel gourps. For
example, for the FFT size of 1024 (or 2048), each primary subchannel group contains 6 (or
12) subchannels, and each secondary subchannel group contains 4 (or 8) subchannels,
which gives the ratio of 3:2. And, the denominator of 15 = 3 x 3 + 2 x 3.
If the segmentation usage ratio is set to 0, it means that the segmented zone does not exist. Setting SU to 0
gives SP = 0, and setting SU to 1 gives SP = 1 (or 100%), which shows how the switching point varies with
the segmentation usage ratio.
Derivation of the switching point formula:
The Segmentation Usage ratio is used to partition the total downlink traffic load into
segmented and non-segmented zones. Therefore, the switching point formula is derived
from the equation:
TX i ic
TX i ic
TX i ic
TX i ic
SU
TL DL
1 SU
TL DL
--------------------------------------------------------------------------------- = ------------------------------------------------------------------TX i ic
TX i ic
TX i ic
TX i ic
TX i ic
SP
f Segment W Channel
1 SP
W Channel
With cells using segmentation, there can be four different interference scenarios.
-
480
Between the segmented zone of the victim and the segmented zone of the interferer.
Between the segmented zone of the victim and the non-segmented zone of the interferer.
Between the non-segmented zone of the victim and the segmented zone of the interferer.
Between the non-segmented zone of the victim and the non-segmented zone of the interferer.
AT271_TRG_E6
Forsk 2009
Figure 9.8: Segmentation Interference Scenarios

Therefore, Atoll calculates the probabilities of collision for each scenario and weights the total interference according to the total collision probability. The probability of collision p Coll for each scenario is given by the following formula:
3 PSG Com + 2 SSG Com
p Coll = --------------------------------------------------------------------------------TX i ic
TX i ic
3 PSG
+ 2 SSG
Where, PSGCom is the number of primary subchannel groups common in TXi(ic) and TXj(jc), which can be
either 0 or 1, SSGCom is the number of secondary subchannel groups common in TXi(ic) and TXj(jc),
TX i ic
TX i ic
PSG
is the number of
is the number of primary subchannel groups in the cell TXi(ic), and SSG
secondary subchannel groups in the cell TXi(ic). The numbers of primary and secondary subchannel groups
for different segments are:
Segment
Number of Primary
Subchannel Groups (PSG)
Number of Secondary
Subchannel Groups (SSG)
1 (Group 0)
1 (Group 2)
1 (Group 4)
Where m = 0, 1, 2, or 3 for FFT sizes 1024 and 2048, and m = 1 for FFT sizes 128 and 512.
The numbers of primary and secondary subchannel groups for the different cases are given in the table below:
TX i ic
TX i ic
TX j jc
p Coll
TX i ic
TX j jc
NN
Not
segmented
Not
segmented
NS
Not
segmented
Segmented
p Coll
SN
Segmented
Not
segmented
SS
Segmented
Segmented
p Coll
p Coll
p Coll
PSG
TX j jc
PSG Com
SSG Com
m in TXj(jc)
1 if same
segment
0 otherwise
m in TXi(ic)
m in TXi(ic)
SSG
PSG
SSG
m in TXj(jc)
The segment numbers and the cell permutation base numbers (Cell PermBase) are determined from the cells
preamble index. The mapping between the preamble index, the segment number, and Cell PermBase is available in the IEEE specifications. This mapping is performed in Atoll as follows:
Preamble Index ( PI )
Range: 0 to 113
Cell PermBase ( PB )
Range: 0 to 31
Forsk 2009
AT271_TRG_E6
PI 96
96 PI 114
PI Modulo 32
PI 96
481
Segment Number ( N Seg )
PI
Floor ------
32
Range: 0, 1, 2
PI 96 Modulo 3
There can be 2 cases for calculating the total probability of collision.

Case 1: If the pixel, subscriber, or mobile Mi is covered by the segmented zone of TXi(ic), the total collision
probability for the pixel, subscriber, or mobile Mi is calculated as follows:
TX i ic TX j jc
p Collision
TX j jc
TX i ic
SS
p Coll
If SP
SP
TX j jc
TX i ic
TX j jc
= SS
SN
+ p Coll SP
SP
TX jc
TX i ic
p Coll SP
--------------------------------------------------------------------------------------------------------------------------- If SP j
SP
TX i ic
SP
Case 2: If the pixel, subscriber, or mobile Mi is covered by the non-segmented zone of TXi(ic), the total
collision probability for the pixel, subscriber, or mobile Mi is calculated as follows:
TX i ic TX j jc
p Collision
TX j jc
TX i ic
NN
p Coll
If SP
SP
TX j jc
TX j jc
TX i ic
NS
= NN
+ p Coll SP
SP
TX j jc
TX i ic
Coll 1 SP
p
----------------------------------------------------------------------------------------------------------------------------------------- SP
If SP
TX i ic
1 SP
The interference reduction factor due to segmentation for the pixel, subscriber, or mobile Mi is calculated as
follows:
TX i ic TX j jc
f Seg
TX i ic TX j jc
= 10 Log 10 p Collision
Output
TX j jc
I DL
: Effective downlink traffic and pilot interference received at the pixel, subscriber, or mobile Mi from any
interfering cell TXj(jc).
9.3.6.4
Traffic and Pilot C/N Calculation (DL)

Input
TX i ic
C Traffic : Received traffic signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi as calculated in
TX i ic
C Pilot
: Received pilot signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi as calculated in

TX i ic
n DL
: Downlink noise for the cell TXi(ic) as calculated in "Noise Calculation (DL)" on page 474.
CNR Preamble : Preamble C/N from the cell TXi(ic) at the pixel, subscriber, or mobile Mi as calculated in "Preamble
TX i ic

TX i ic
T AMS : AMS threshold defined for the cell TXi(ic).
N Ant TX : Number of MIMO transmission (downlink) antennas defined for the cell TXi(ic).
N Ant RX : Number of MIMO reception (downlink) antennas defined for the terminal used by the pixel, subscriber,
TX i ic
Mi
or mobile Mi.
Mobility M i : Mobility used for the calculations.
G STTD : Downlink STTD/MRC gain from the MIMO Configurations table for the number of transmission antennas,
DL
TX i ic
Mi
N Ant TX , the number of reception antennas, N Ant RX , and user mobility, Mobility M i .
DL
G STTD : Downlink STTD/MRC gain offset defined for the clutter class of the pixel, subscriber, or mobile Mi.
Calculations
The traffic and pilot C/N for a cell TXi(ic) are calculated as follows for any pixel, subscriber, or mobile Mi:
TX i ic
TX i ic
TX i ic
CNR Traffic = C Traffic n DL

TX i ic
CNR Pilot
482
TX i ic
= C Pilot
TX i ic
n DL
AT271_TRG_E6
Forsk 2009

MIMO STTD/MRC Gain:
If the permutation zone (WiMAX 802.16e) or the cell (WiMAX 802.16d) supports STTD/MRC or AMS, the downlink
STTD/MRC gain corresponding to the numbers of transmission and reception antennas, and the STTD/MRC gain
offset defined for the clutter class of the user are applied to the downlink C/N.
Therefore, the traffic and pilot C/N calculated above become:
In case of STTD/MRC:
TX i ic
TX i ic
DL
DL
DL
DL
DL
DL
if
CNR Preamble T AMS
DL
DL
if
CNR Preamble T AMS
CNR Traffic = CNR Traffic + G STTD + G STTD

TX i ic
CNR Pilot
TX i ic
= CNR Pilot
+ G STTD + G STTD
In case of AMS:
TX i ic
TX i ic
CNR Traffic = CNR Traffic + G STTD + G STTD

TX i ic
CNR Pilot
TX i ic
= CNR Pilot
+ G STTD + G STTD
TX i ic
TX i ic
TX i ic
TX i ic
Output
9.3.6.5
TX i ic
CNR Traffic : Traffic C/N from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
CNR Pilot
TX i ic
: Pilot C/N from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
Traffic C/(I+N) Calculation (DL)

The carrier signal to interference and noise ratio is calculated in three steps. First Atoll calculates the received signal level
from the studied cell (as explained in "Traffic and Pilot Signal Level Calculation (DL)" on page 473) at the pixel, subscriber,
or mobile under study. Secondly, Atoll calculates the interference received at the same studied pixel, subscriber, or mobile
from all the interfering cells (as explained in "Traffic and Pilot Interference Calculation (DL)" on page 475). Interference
from each cell is weighted according to the co- and adjacent channel overlap between the studied and the interfering cells,
the traffic loads of the interfering cells, and (in WiMAX 802.16e) the probabilities of subcarrier collision if segmentation is
used. Finally, Atoll takes the ratio of the signal level and the sum of the total interference from other cells and the downlink
noise (as calculated in "Noise Calculation (DL)" on page 474).
The receiver terminal is always considered to be oriented towards its best server, except when the "Lock Status" is set to
"Server+Orientation" for a subscriber in a subscriber list and its azimuth and tilt manually edited. In the case of NLOS
between the receiver and the best server, Atoll does not try to find the direction of the strongest signal, the receiver is
oriented towards the best server just as in the case of LOS.
Input
TX i ic
C Traffic : Received traffic signal level from the cell TXi(ic) at the pixel, subscriber, or mobile Mi as calculated in
TX i ic
n DL
TX j jc
I DL
: Downlink noise for the cell TXi(ic) as calculated in "Noise Calculation (DL)" on page 474.
: Effective downlink traffic and pilot interference from any cell TXj(jc) calculated for a pixel, subscriber, or
mobile Mi covered by a cell TXi(ic) as explained in "Traffic and Pilot Interference Calculation (DL)" on page 475.
TX i ic

TX i ic
TX i ic
Mi
or mobile Mi.
G STTD : Downlink STTD/MRC gain from the MIMO Configurations table for the number of transmission antennas,
DL
TX i ic
Mi
Forsk 2009
DL
G STTD : Downlink STTD/MRC gain offset defined for the clutter class of the pixel, subscriber, or mobile Mi.
AT271_TRG_E6
483
Calculations
The traffic C/(I+N) for a cell TXi(ic) is calculated as follows for any pixel, subscriber, or mobile Mi:
TX jc
TX i ic
j
I DL
n DL
TX i ic
TX i ic
------------------------------------
CINR Traffic = C Traffic 10 Log
+
10
10 M Shadowing C I
10
10
All TX j jc
The Traffic Total Noise (I+N) for a cell TXi(ic) is calculated as follows for any pixel, subscriber, or mobile Mi:
TX jc
j
I +
TX i ic
N DL
TX ic
i
I DL
n DL
-------------------------------------
10
10
10
+ 10
= 10 Log
All
TX
jc
MIMO STTD/MRC Gain:

If the permutation zone (WiMAX 802.16e) or the cell (WiMAX 802.16d) supports STTD/MRC or AMS, the downlink
offset defined for the clutter class of the user are applied to the downlink C/(I+N).
Therefore, the traffic C/(I+N) calculated above become:
TX i ic
TX i ic
DL
DL
TX i ic
DL
DL

In case of AMS:
TX i ic
if
TX i ic
TX i ic
CNR Preamble T AMS
Output
TX i ic
CINR Traffic : Traffic C/(I+N) from the cell TXi(ic) at the pixel, subscriber, or mobile Mi.
I + N DL
TX i ic
: Traffic Total noise from the interfering cells TXj(jc) at the pixel, subscriber, or mobile Mi covered by
a cell TXi(ic).
9.3.6.6
Traffic Signal Level Calculation (UL)

Input
Mi
P Max : Maximum transmission power of the terminal used by the pixel, subscriber, or mobile Mi without power control.
Mi
P Eff : Effective transmission power of the terminal used by the pixel, subscriber, or mobile Mi after power control
as calculated in "Traffic C/(I+N) Calculation (UL)" on page 490.
TX i
E SA : Number of antenna elements defined for the smart antenna equipment used by the transmitter TXi.
TX i
TX i
Without smart antenna equipment: G
With smart antenna equipment: G

G
TX i
= G SA = 10
TX i
Log E SA
is the transmitter antenna gain, i.e., G
TX i
TX i
TX i
= G Ant .
is the uplink smart antenna beam forming gain, i.e.,
. For more information on the calculation of G SA , refer to section "Optimum
Beamformer Model" on page 505.
TX i

TX i
L Ant
TX i
= L Total UL ).
L Path : Path loss ( L Path = L Model +
).
TX i
M Shadowing Model = 0 when calculating the received signal level from interfering mobiles.
484
AT271_TRG_E6
Forsk 2009

L
Mi
Mi
Mi
L Ant
Mi
When calculating L Ant , Mis antenna is pointed towards its best serving cell. If TXi(ic) is the best serving cell for
Mi
Mi
Mi
Mi, then L Ant = 0 for calculating C UL . If TXi(ic) is not the best serving cell for Mi, then L Ant is calculated in the
direction of TXi(ic) from the antenna patterns of the antenna used by Mi while the antenna is pointed towards Mis
best serving cell.
Mi
Calculations
The received traffic signal level (dBm) from a pixel, subscriber, or mobile Mi at its serving cell TXi(ic) is calculated as
follows:
Mi
Mi
C UL = EIRP UL L Path M Shadowing Model L Indoor + G
TX i
TX i
Mi
Mi
L Ant L Body
Where EIRP is the effective isotropic radiated power of the terminal calculated as follows:
Mi
EIRP UL = P
With P
Mi
Mi
+G
Mi
Mi
Mi
= P Max without power control at the start of the calculations, and is the P
Mi
Mi
= P Eff after power control.
Output
9.3.6.7
Mi
C UL : Received uplink signal level from the pixel, subscriber, or mobile Mi at a cell TXi(ic).
Noise Calculation (UL)

For determining the uplink C/N and C/(I+N), Atoll calculates the uplink noise over the channel bandwidth used by the cell.
The used bandwidth depends on the number of used subcarriers. In WiMAX 802.16e, the numbers of subcarriers used by
each permutation zone can be different.
The uplink noise comprises thermal noise and the noise figure of the equipment. The thermal noise density depends on
the temperature, i.e., it remains constant for a given temperature. However, the value of the thermal noise varies with the
used bandwidth.
Input
N SCa Used : Number of used subcarriers defined in the Global Parameters (WiMAX 802.16d).
N SCa Total : Total number of subcarriers defined in the Global Parameters (WiMAX 802.16d).
N SCa Used : Number of subcarriers used by the uplink permutation zone of a WiMAX 802.16e cell TXi(ic)
Mi
PZ UL
assigned to Mi.
TX i ic
TX i ic
page 493.
nf
TX i ic
: Noise figure of the cell TXi(ic).
Calculations
n 0 = 10 Log 10 K T 1000 = 174 dBm/Hz
The thermal noise for a cell is calculated as:
N SCa Used
TX i ic
TX i ic
WiMAX 802.16d: n 0 UL = n 0 + 10 Log 10 F Sampling ------------------------------
N SCa Total
Forsk 2009
AT271_TRG_E6
485

M
WiMAX 802.16e (UL):
PZ UL
N SCa Used
TX i ic
-
N SCa Total
TX i ic
n 0 UL
The uplink noise is the sum of the thermal noise and the noise figure of the cell TXi(ic).
TX i ic
n UL
TX i ic
= n 0 UL + nf
TX i ic
Output
9.3.6.8
TX i ic
n UL
: Uplink noise for the cell TXi(ic).
Traffic Interference Calculation (UL)

The uplink traffic interference is only calculated in Monte Carlo simulations. In coverage predictions, the uplink noise risevalues already available in simulations on in the Cells table are used.
The interference received by a cell TXi(ic) from an interfering mobiles covered by a cell TXj(jc) can be defined as the uplink
signal level received from interfering mobiles Mj depending on the overlap that exists between the channels used by the
cells TXi(ic) and TXj(jc), on the traffic loads of the interfering mobiles Mj.
The calculation of uplink interference can be divided into two parts:
9.3.6.8.1
Calculation of the uplink interference from an individual interfering mobile as calculated in "Traffic Interference
Signal Levels Calculation (UL)" on page 486.
Calculation of the uplink noise rise which represents the total uplink interference from all the interfering mobiles as
calculated in "Noise Rise Calculation (UL)" on page 487.
Traffic Interference Signal Levels Calculation (UL)

Input
Mj
C UL : Uplink signal level received at a cell TXi(ic) from an interfering mobile Mj covered by a cell TXj(jc) as calculated in "Traffic Signal Level Calculation (UL)" on page 484.
TX i ic TX j jc
rO
Mj
TL UL : Uplink traffic load of the interfering mobile Mj.

Traffic loads are calculated during Monte Carlo simulations as explained in "Simulation Process" on page 455.
Calculations
The uplink interference received at a cell TXi(ic) from an interfering mobile Mj covered by a cell TXj(jc) is calculated as
follows:
Mj
Mj
TX i ic TX j jc
I UL = C UL + f O
Mj
+ f TL UL
Calculations for the interference reduction factors due to channel overlapping and uplink traffic load are explained below:
Interference reduction due to the co- and adjacent channel overlap between the studied and the interfering
cells:
Interference reduction due to the co- and adjacent channel overlap between the cells TXi(ic) and TXj(jc) is calculated as follows:
TX i ic TX j jc
fO
TX i ic TX j jc
= 10 Log r O
Interference reduction due to interfering mobiles traffic load:

The interference reduction factor due to the interfering mobiles uplink traffic load is calculated as follows:
-
Without smart antenna: All the mobiles present in other cells TXj(jc) that are transmitting in uplink contribute
to the interference received by TXi(ic) in uplink.
Mj
Mj
f TL UL = 10 Log 10 TL UL
-
486
With smart antenna: A set of interfering mobiles is generated for each mobile being served by the victim cell
TXi(ic). The probability of generating a mobile as an interferer depends on its uplink traffic load (see "TimeSlot Scenario:" on page 457). As the traffic load has already been taken into account for generating the list of
interfering mobiles,
AT271_TRG_E6
Forsk 2009

Mj
f TL UL = 0
Output
9.3.6.8.2
Mj
I UL : Uplink interference signal level received at a cell TXi(ic) from an interfering mobile Mj covered by a cell TXj(jc).
Noise Rise Calculation (UL)

The uplink noise rise is defined as the ratio of the total uplink interference received by any cell TXi(ic) from interfering
mobiles Mj present in the coverage areas of other cells TXj(jc) to the uplink noise of the cell TXi(ic). More specifically, it is
the ratio (I+N)/N.
Input
Mj
I UL : Uplink interference signal levels received at a cell TXi(ic) from interfering mobiles Mj covered by other cells
TXj(jc).
TX i ic
n UL
: Uplink noise for the cell TXi(ic).
Calculations
The uplink noise for the cell TXi(ic) is calculated as follows:
TX i ic
NR UL
M
TX ic
j
i
I UL
n UL
------------------------TX i ic
10
10
= 10 Log 10
10 + 10
n UL
All M j
All TX jc
With smart antenna:

The angular distribution of the uplink noise rise is calculated during Monte Carlo simulations and can be stored in
the Cells table in order to be used in coverage predictions. The angular distribution of the uplink noise rise is given
by:
2
I UL + n I
NR UL = ----------------------------------2
n I
The Total Noise (I+N) for a cell TXi(ic) is calculated as follows for any pixel, subscriber, or mobile Mi:
I +
TX i ic
N UL
M
TX ic
j
i
I UL
n UL
------------------------
10
10
10 + 10
= 10 Log 10
All
M
j
All TX jc
Output
9.3.6.9
TX i ic
NR UL
I +
TX i ic
or NR UL
TX i ic
N UL
: Uplink noise rise or the angular distribution of the uplink noise rise for the cell TXi(ic).
: Total Noise for a cell TXi(ic) calculated for any pixel, subscriber, or mobile Mi.
Traffic C/N Calculation (UL)

Input
Mi
C UL : Received uplink signal level from the pixel, subscriber, or mobile Mi at its serving cell TXi(ic) as calculated
in "Traffic Signal Level Calculation (UL)" on page 484.
TX i ic
n UL
: Uplink noise for the cell TXi(ic) as calculated in "Noise Calculation (UL)" on page 485.
TX i ic
Forsk 2009
TX i ic
AT271_TRG_E6
487

Mi
N Ant TX : Number of MIMO transmission (uplink) antennas defined for the terminal used by the pixel, subscriber,
or mobile Mi.
TX i ic
N Ant RX : Number of MIMO reception (uplink) antennas defined for the cell TXi(ic)..
G STTD : Uplink STTD/MRC gain from the MIMO Configurations table for the number of transmission antennas,
UL
TX i ic
Mi
UL
G STTD : Uplink STTD/MRC gain offset defined for the clutter class of the pixel, subscriber, or mobile Mi.
N SC UL : Number of subchannels per channel (WiMAX 802.16d).
T B Lowest : Bearer selection threshold of the lowest bearer in the reception equipment assigned to the cell TXi(ic).
T B Highest : Bearer selection threshold of the highest bearer in the reception equipment assigned to the cell
TX i ic
TX i ic
TXi(ic) or defined in the properties of the service being accessed by the pixel, subscriber, or mobile Mi.
The bearer assigned to the pixel, subscriber, or mobile Mi is determined as explained in "Bearer Determination"
on page 496.
Mi
PZ UL
N SC
: Number of subchannels per channel defined for the uplink permutation zone assigned to the pixel, sub-
scriber, or mobile Mi as calculated in "Permutation Zone Selection (WiMAX 802.16e)" on page 471.
Mi
P Max : Maximum transmission power of the terminal used by the pixel, subscriber, or mobile Mi.
P Min : Minimum transmission power of the terminal used by the pixel, subscriber, or mobile Mi.
M PC : Power control margin defined in the Global Parameters.
Mi
Calculations
The uplink C/N from a pixel, subscriber, or mobile Mi at its serving cell TXi(ic) is calculated as follows:
Mi
Mi
TX i ic
CNR UL = C UL n UL
MIMO STTD/MRC Gain:

If the permutation zone (WiMAX 802.16e) or the cell (WiMAX 802.16d) supports STTD/MRC or AMS, the uplink
offset defined for the clutter class of the user are applied to the uplink C/N.
Therefore, the uplink C/N calculated above becomes:
Mi
Mi
UL
UL
Mi
UL
UL

In case of AMS:
Mi
if
TX i ic
TX i ic
CNR Preamble T AMS
Uplink Subchannelisation (WiMAX 802.16d):

Subchannelisation decreases the bandwidth used by a mobile hence increasing the power spectral density for
transmission, and decreasing the noise and interference received. WiMAX terminals can perform subchannelisation in uplink to improve uplink coverage. In WiMAX 802.16d networks, if a terminal is unable to connect to its serving cell using all 16 subchannels, it can use less number of subchannels (8, 4, 2, or 1) in the uplink in order to
concentrate its transmission power on a smaller bandwidth, hence increasing its power spectral density and
increasing transmission range.
The uplink C/N is calculated above using the number of subchannels per channel set in the Global Parameters,
which is 16 by default. The number of subchannels in uplink is provided by the IEEE specifications.
N SC UL = 16 8 4 2 or 1
If the uplink C/N from the pixel, subscriber, or mobile Mi is not enough to get the lowest bearer, i.e.,
Mi
TX i ic
CNR UL T B Lowest , Atoll performs subchannelisation as follows:

Mi
TX i ic
Until CNR UL T B Lowest OR N SC UL 1 , Atoll improves the uplink C/N by adding the subchannelisation gain
to it.
488
AT271_TRG_E6
Forsk 2009

Mi
Mi
CNR UL = CNR UL + G SC UL and reducing the number of subchannels used in the uplink by half.
The value of G SC UL is calculated by determining the number of used subcarriers as follows:
Subchannelisation From
Number of Used Subcarriers
G SC UL
16 to 8
200 to 100.
(192 Data + 8 Pilot to 96 Data + 4 Pilot)
200
10 Log ---------- = 3 dB
100
8 to 4
100 to 50.
100
10 Log ---------- = 3 dB
50
4 to 2
50 to 25.
50
10 Log ------ = 3 dB
25
2 to 1
25 to 13.
25
10 Log ------ = 2.84 dB
13
Even if after performing subchannelisation, the uplink C/N from the pixel, subscriber, or mobile Mi is not enough
TX i ic
Mi
to get a bearer in the uplink, i.e., CNR UL T B Lowest , the pixel, subscriber, or mobile Mi is considered as covered
by the cell TXi(ic) in the uplink.
If you want to turn off subchannelisation in uplink, you can set the number of subchannels per channel to 1 in the
Global Parameters.
Uplink Subchannelisation (WiMAX 802.16e):
transmission, and decreasing the noise and interference received. The uplink C/N calculated above is given for
the number of subchannels defined for the permutation zone assigned to the pixel, subscriber, or mobile Mi in the
uplink.
In WiMAX 802.16e networks, subchannelisation is performed for all the pixels, subscribers, or mobiles in the
uplink. Slots are allocated to users first in the time domain and then in the frequency domain.
Figure 9.9: Uplink Subchannelization in WiMAX 802.16e Networks (Example)

The number of subchannels in uplink is provided by the IEEE specifications for each permutation zone. The uplink
C/N for any pixel, subscriber, or mobile Mi is determined for the entire channel bandwidth. To determine the
number of subchannels required to maintain the highest available bearer at the location of the pixel, subscriber,
or mobile, Atoll performs subchannelisation as follows:
M
Mi
CNR UL
Mi
CNR UL
PZULi
N SC
-
+ 10 Log 10 ------------------- N Mi
SC UL
PZMi M
Mi
UL
i
Where N SC UL = Min N SC N SC UL
is the number of uplink subchannels after subchanneli

B UL = B UL Highest
Mi
Mi
sation with which the pixel, subscriber, or mobile Mi can get the highest available bearer. The highest available
Mi
bearer at the location of the pixel, subscriber, or mobile Mi ( B UL Highest ) is the lowest one among:
Forsk 2009
AT271_TRG_E6
489

-
The highest bearer of the reception equipment of the serving cell of the pixel, subscriber, or mobile.
The highest bearer of the service accessed by the pixel, subscriber, or mobile.
After subchannelisation, the number of uplink subchannels allocated to the pixel, subscriber, or mobile Mi is the
number of subchannels required to get the highest available bearer in the uplink.
Uplink Power Control:
Atoll performs power control in the uplink if the bearer available in the uplink for the pixel, subscriber, or mobile Mi
is the highest bearer. The highest bearer can be either the highest bearer defined in the properties of the service
used by Mi, or the highest bearer listed in the reception equipment of the serving cell TXi(ic). The pixel, subscriber,
or mobile Mi reduces its transmission power so that the uplink C/N from it at its cell is enough to get the highest
bearer.
If with P
Mi
Mi
Mi
TX i ic
= P Max AND CNR UL T B Highest + M PC , the transmission power of Mi is reduced to determine the
effective transmission power from the pixel, subscriber, or mobile Mi as follows:

Mi
Mi
Mi
TX i ic
Mi
P Eff = Max P Max CNR UL T B Highest + M PC P Min

Mi
Mi
CNR UL is calculated again using P Eff .
Output
9.3.6.10
Mi
CNR UL : Uplink C/N from a pixel, subscriber, or mobile Mi at it serving cell TXi(ic).
Traffic C/(I+N) Calculation (UL)

The carrier signal to interference and noise ratio is calculated in three steps. First, Atoll calculates the received signal level
from each pixel, subscriber, or mobile at its serving cell using the effective power of the terminal used by the pixel,
subscriber, or mobile as explained in "Traffic Signal Level Calculation (UL)" on page 484. Secondly, Atoll calculates the
uplink carrier to noise ratio as explained in "Traffic C/N Calculation (UL)" on page 487. And, finally, determines the uplink
C/(I+N) by dividing the previously calculated uplink C/N by the uplink noise rise value of the cell as calculated in "Noise
Rise Calculation (UL)" on page 487.
The uplink noise rise can be set by the user manually for each cell or calculated using the Monte Carlo simulations.
The receiver terminal is always considered to be oriented towards its best server, except when the "Lock Status" is set to
"Server+Orientation" for a subscriber in a subscriber list and its azimuth and tilt manually edited. In the case of NLOS
between the receiver and the best server, Atoll does not try to find the direction of the strongest signal, the receiver is
oriented towards the best server just as in the case of LOS.
Input
Mi
CNR UL : Uplink C/N from a pixel, subscriber, or mobile Mi at it serving cell TXi(ic).

TX i ic
NR UL
TX i ic
CNR Preamble
TX i ic
or NR UL
: Uplink noise rise or the angular distribution of the uplink noise rise for the cell TXi(ic).
: Preamble C/N from the cell TXi(ic) at the pixel, subscriber, or mobile Mi as calculated in "Preamble

TX i ic
Mi
or mobile Mi.
TX i ic
N Ant RX : Number of MIMO reception (uplink) antennas defined for the cell TXi(ic)..
G STTD : Uplink STTD/MRC gain from the MIMO Configurations table for the number of transmission antennas,
UL
TX i ic
Mi
490
UL
G STTD : Uplink STTD/MRC gain offset defined for the clutter class of the pixel, subscriber, or mobile Mi.
N SC UL : Number of subchannels per channel (WiMAX 802.16d).
T B Lowest : Bearer selection threshold of the lowest bearer in the reception equipment assigned to the cell TXi(ic).
TX i ic
AT271_TRG_E6
Forsk 2009
TX i ic
T B Highest : Bearer selection threshold of the highest bearer in the reception equipment assigned to the cell
TXi(ic) or defined in the properties of the service being accessed by the pixel, subscriber, or mobile Mi.
The bearer assigned to the pixel, subscriber, or mobile Mi is determined as explained in "Bearer Determination"
on page 496.
Mi
PZ UL
N SC
: Number of subchannels per channel defined for the uplink permutation zone assigned to the pixel, sub-
scriber, or mobile Mi as calculated in "Permutation Zone Selection (WiMAX 802.16e)" on page 471.
Mi
P Max : Maximum transmission power of the terminal used by the pixel, subscriber, or mobile Mi.
P Min : Minimum transmission power of the terminal used by the pixel, subscriber, or mobile Mi.
M PC : Power control margin defined in the Global Parameters.
Mi
Calculations
The uplink C/(I+N) for any pixel, subscriber, or mobile Mi at a cell TXi(ic) is calculated as follows:

Mi
TX i ic
Mi
CINR UL = CNR UL NR UL
M Shadowing C I
With smart antenna:

-
Monte Carlo simulations: The uplink C/(I+N) is calculated as described in the section "Optimum Beamformer
Model" on page 505. Victim and interfering mobiles are generated by a time-slot scenario as explained in
"Simulation Process" on page 455.
Coverage predictions: CINR UL = CNR UL NR UL
Mi
TX i ic
Mi
M Shadowing C I
MIMO STTD/MRC Gain:

If the permutation zone (WiMAX 802.16e) or the cell (WiMAX 802.16d) supports STTD/MRC or AMS, the uplink
offset defined for the clutter class of the user are applied to the uplink C/(I+N).
Therefore, the uplink C/(I+N) calculated above becomes:
Mi
Mi
UL
UL
Mi
UL
UL
Mi
Mi
UL
UL
CINR UL = CINR UL + G STTD + G STTD and CINR UL = CINR UL + G STTD + G STTD

In case of AMS:
Mi

Mi
Mi
UL
if
UL
TX i ic
TX i ic
CNR Preamble T AMS

if
TX i ic
and
TX i ic
CNR Preamble T AMS
Uplink Subchannelisation (WiMAX 802.16d):

transmission, and decreasing the noise and interference received. WiMAX terminals can perform subchannelisation in uplink to improve uplink coverage. In WiMAX 802.16d networks, if a terminal is unable to connect to its serving cell using all 16 subchannels, it can use less number of subchannels (8, 4, 2, or 1) in the uplink in order to
concentrate its transmission power on a smaller bandwidth, hence increasing its power spectral density and
increasing transmission range.
The uplink C/(I+N) is calculated above using the number of subchannels per channel set in the Global Parameters,
which is 16 by default. The number of subchannels in uplink is provided by the IEEE specifications.
N SC UL = 16 8 4 2 or 1
If the uplink C/(I+N) from the pixel, subscriber, or mobile Mi is not enough to get the lowest bearer, i.e.,
Mi
TX i ic
CINR UL T B Lowest , Atoll performs subchannelisation as follows:

Mi
TX i ic
Until CINR UL T B Lowest OR N SC UL 1 , Atoll improves the uplink C/(I+N) by adding the subchannelisation
gain to it.
Mi
Mi
CINR UL = CINR UL + G SC UL and reducing the number of subchannels used in the uplink by half.
The value of G SC UL is calculated by determining the number of used subcarriers as follows:
Forsk 2009
AT271_TRG_E6
491
Subchannelisation From
Number of Used Subcarriers
G SC UL
16 to 8
200 to 100.
200
10 Log ---------- = 3 dB
100
8 to 4
100 to 50.
100
10 Log ---------- = 3 dB
50
4 to 2
50 to 25.
50
10 Log ------ = 3 dB
25
2 to 1
25 to 13.
25
10 Log ------ = 2.84 dB
13
Even if after performing subchannelisation, the uplink C/(I+N) from the pixel, subscriber, or mobile Mi is not enough
TX i ic
Mi
to get a bearer in the uplink, i.e., CINR UL T B Lowest , the pixel, subscriber, or mobile Mi is considered as
covered by the cell TXi(ic) in the uplink.
If you want to turn off subchannelisation in uplink, you can set the number of subchannels per channel to 1 in the
Global Parameters.
Uplink Subchannelisation (WiMAX 802.16e):
transmission, and decreasing the noise and interference received. The uplink C/(I+N) calculated above is given
for the number of subchannels defined for the permutation zone assigned to the pixel, subscriber, or mobile Mi in
the uplink.
In WiMAX 802.16e networks, subchannelisation is performed for all the pixels, subscribers, or mobiles in the
uplink. Slots are allocated to users first in the time domain and then in the frequency domain.
Figure 9.10: Uplink Subchannelization in WiMAX 802.16e Networks (Example)

The number of subchannels in uplink is provided by the IEEE specifications for each permutation zone. The uplink
C/(I+N) for any pixel, subscriber, or mobile Mi is determined for the entire channel bandwidth. To determine the
number of subchannels required to maintain the highest available bearer at the location of the pixel, subscriber,
or mobile, Atoll performs subchannelisation as follows:
M
Mi
CINR UL
Mi
CINR UL
PZULi
N SC
-
+ 10 Log 10 ------------------- N Mi
SC UL
PZMi M
Mi
UL
i
Where N SC UL = Min N SC N SC UL
is the number of uplink subchannels after subchanneli

i
i
B UL = B UL Highest
M
sation with which the pixel, subscriber, or mobile Mi can get the highest available bearer. The highest available
Mi
bearer at the location of the pixel, subscriber, or mobile Mi ( B UL Highest ) is the lowest one among:
-
492
The highest bearer of the reception equipment of the serving cell of the pixel, subscriber, or mobile.
The highest bearer of the service accessed by the pixel, subscriber, or mobile.
AT271_TRG_E6
Forsk 2009

After subchannelisation, the number of uplink subchannels allocated to the pixel, subscriber, or mobile Mi is the
number of subchannels required to get the highest available bearer in the uplink.
Uplink Power Control:
Atoll performs power control in the uplink if the bearer available in the uplink for the pixel, subscriber, or mobile Mi
is the highest bearer. The highest bearer can be either the highest bearer defined in the properties of the service
used by Mi, or the highest bearer listed in the reception equipment of the serving cell TXi(ic). The pixel, subscriber,
or mobile Mi reduces its transmission power so that the uplink C/(I+N) from it at its cell is enough to get the highest
bearer.
If with P
Mi
Mi
Mi
TX i ic
= P Max AND CINR UL T B Highest + M PC , the transmission power of Mi is reduced to determine the
effective transmission power from the pixel, subscriber, or mobile Mi as follows:

Mi
Mi
TX i ic
Mi
Mi
P Eff = Max P Max CINR UL T B Highest + M PC P Min

Mi
Mi
CINR UL is calculated again using P Eff .
Output
9.3.7
Mi
Mi
CINR UL or CINR UL : Uplink C/(I+N) from a pixel, subscriber, or mobile Mi at it serving cell TXi(ic).
N SC UL : Number of subchannels used by the pixel, subscriber, or mobile Mi in the uplink after subchannelisation.
P Eff : Effective transmission power of the terminal used by the pixel, subscriber, or mobile Mi.
Mi
Mi

Channel throughputs are calculated in three steps.
9.3.7.1
Calculation of uplink and downlink total resources in a cell as explained in "Calculation of Cell Capacity (Total
Resources)" on page 493.
Determination of the uplink and downlink bearers as explained in "Bearer Determination" on page 496.
Calculation of channel throughputs as explained in "Channel Throughput Calculation" on page 496.
Calculation of Cell Capacity (Total Resources)

Cell capacity is the total amount of resources available in a cell. The total amount of resources in a cell is the number of
modulation symbols that can be used for data transfer in each frame. The cell capacity can be calculated separately for
the downlink and the uplink subframes. The following sections describe how the cell capacities are calculated for TDD and
FDD networks.
9.3.7.1.1
Calculation of Sampling Frequency

Input
TX i ic
f Sampling : Sampling factor defined for the frequency band of the cell TXi(ic).
W Channel : Channel bandwidth of the cell TXi(ic).
TX i ic
Calculations
Atoll determines the sampling frequency as follows:
TX i ic
TX i ic
W Channel 10
F Sampling = Floor f Sampling ---------------------------------------- 8000

8000
Output
9.3.7.1.2
TX i ic
F Sampling : Sampling frequency for the cell TXi(ic).
Calculation of Symbol Duration

Input
TX i ic
page 493.
Forsk 2009
AT271_TRG_E6
493

TX i ic
r CP : Cyclic prefix ratio defined for the network in the Global Parameters.
Calculations
From the sampling frequency, Atoll determines the inter-subcarrier spacing.
F
TX i ic
TX i ic
F Sampling 10
= -----------------------------------------TX i ic
N SCa Total
Atoll calculates the useful symbol duration.

TX i ic
1
D Sym Useful = ---------------------TX i ic
F
And, the duration of the cyclic prefix.

r CP
D CP = -------F
Adding the Cyclic prefix ratio to the useful symbol duration, Atoll determines the total symbol duration.
TX i ic
TX i ic
D Symbol = D Sym Useful + D CP
Output
9.3.7.1.3
TX i ic
D Symbol : Total symbol duration of one modulation symbol for a cell TXi(ic).
Calculation of Cell Capacity - TDD Networks

Input
D Frame : Frame duration.
D TTG : TTG duration.
D RTG : RTG duration.
D Symbol : Total symbol duration of one modulation symbol for a cell TXi(ic) as calculated in "Calculation of Symbol
TDD
TDD
TX i ic
Duration" on page 493.

TDD
r DL Frame : DL ratio.
O Fixed : Downlink fixed overhead.
O Variable : Downlink variable overhead.
O Fixed : Uplink fixed overhead.
O Variable : Uplink variable overhead.
N SCa Data : Number of data subcarriers for a WiMAX 802.16d cell TXi(ic).
N SCa Data : Number of data subcarriers of the downlink permutation zone of a WiMAX 802.16e cell TXi(ic)
DL
DL
UL
UL
TX i ic
M
PZ DL
assigned to Mi.
Mi
PZ UL
N SCa Data : Number of data subcarriers of the uplink permutation zone of a WiMAX 802.16e cell TXi(ic) assigned
to Mi.
Calculations
The downlink and the uplink subframes of a TDD frame are separated in time by the TTG and the RTG time guards.
First of all, Atoll calculates the useful frame duration by removing the TTG and RTG from the frame duration:
Used
TDD
TDD
D Frame = D Frame D TTG D RTG

Then, Atoll calculates the frame duration in terms of number of symbol durations:
D Used
TX i ic
Frame
N SD Used Frame = Floor ------------------- TXi ic
D Symbol
494
AT271_TRG_E6
Forsk 2009

Next, Atoll calculates the downlink and uplink cell capacities as follows:
Downlink Subframe:
Atoll calculates the number of symbol durations in the downlink subframe excluding the fixed overhead defined in the
Global Parameters:
TX i ic
TX i ic
TDD
DL
N SD DL Subframe = RoundUp N SD Used Frame r DL Frame O Fixed

The RoundUp function rounds a float value up to the nearest integer value.
The total number of symbols in the downlink subframe after removing the variable overhead is:
DL
TX i ic
TX i ic
TX i ic
O Variable
TX i ic
= N Sym DL Subframe = Floor N SD DL Subframe N SCa Data 1 ----------------------
100
TX i ic
i
TX i ic
PZ DL
O Variable
TX i ic
= N Sym DL Subframe = Floor N SD DL Subframe N SCa Data 1 ----------------------
100
WiMAX 802.16d: R DL
WiMAX 802.16e: R DL
DL
Uplink Subframe:
Atoll calculates the number of symbol durations in the uplink subframe excluding the fixed overhead defined in the Global
Parameters:
TX i ic
TX i ic
TDD
UL
N SD UL Subframe = RoundDown N SD Used Frame 1 r DL Frame O Fixed

The RoundDown function rounds a float value down to the nearest integer value.
The total number of symbols in the uplink subframe after removing the variable overhead is:
UL
TX i ic
TX i ic
TX i ic
O Variable
TX i ic
= N Sym UL Subframe = Floor N SD UL Subframe N SCa Data 1 ----------------------
100
TX i ic
i
TX i ic
PZ UL
O Variable
TX i ic
= N Sym UL Subframe = Floor N SD UL Subframe N SCa Data 1 ----------------------
100
WiMAX 802.16d: R UL
WiMAX 802.16e: R UL
UL
Output
9.3.7.1.4
TX i ic
R DL
R UL
TX i ic
TX i ic
= N Sym DL Subframe : Amount of downlink resources in the cell TXi(ic).

TX i ic
= N Sym UL Subframe : Amount of uplink resources in the cell TXi(ic).
Calculation of Cell Capacity - FDD Networks

The cell capacity calculation is the same for downlink and uplink subframes in FDD networks. Therefore, the symbol X is
used to represent DL or UL in the expressions below.
Input
D Symbol : Total symbol duration of one modulation symbol for a cell TXi(ic) as calculated in "Calculation of Symbol
TX i ic
Duration" on page 493.

X
O Fixed : Downlink or uplink fixed overhead.
O Variable : Downlink or uplink variable overhead.
N SCa Data : Number of data subcarriers for a WiMAX 802.16d cell TXi(ic).
PZ X
N SCa Data
TX i ic
M
: Number of data subcarriers of the downlink or uplink permutation zone of a WiMAX 802.16e cell
TXi(ic) assigned to Mi.
Calculations
There are no transmit and receive time guards in FDD systems. Therefore, the downlink and the uplink subframe durations
are the same as the frame duration.
X
D Subframe = D Frame
The subframe durations in terms of the number of symbol durations excluding the fixed overheads are:
Forsk 2009
AT271_TRG_E6
495

DX
TX i ic
Subframe
- OX
N SD X Subframe = Floor ------------------------Fixed
TXi ic
D Symbol
The total numbers of symbols in the downlink or uplink subframes after removing the variable overheads are:
X
TX i ic
TX i ic
TX i ic
O Variable
TXi ic
= N Sym X Subframe = Floor N SD X Subframe N SCa Data 1 ----------------------
100
TX i ic
i
TX i ic
PZ X
O Variable
TXi ic
= N Sym X Subframe = Floor N SD X Subframe N SCa Data 1 ----------------------
100
WiMAX 802.16d: R X
WiMAX 802.16e: R X
Output
9.3.7.2
TX i ic
TX i ic
RX
= N Sym X Subframe : Amount of downlink or uplink resources in the cell TXi(ic).
Bearer Determination
The second step in calculating the channel throughput for a cell at any pixel, subscriber, or mobile Mi within its service area
is to determine which bearer it is able to get in the downlink and in the uplink.
Input
TX i ic
CINR Traffic : Traffic C/(I+N) from the cell TXi(ic) at a pixel, subscriber, or mobile Mi as calculated in "Traffic C/(I+N)
Calculation (DL)" on page 483.
Mi
CINR UL : Uplink C/(I+N) from a pixel, subscriber, or mobile Mi at it serving cell TXi(ic) as calculated in "Traffic C/
(I+N) Calculation (UL)" on page 490.
Mi
T B : Bearer selection thresholds of the bearers defined in the reception equipment used by Mis terminal.
TB
B DL Highest Service : Highest downlink bearer defined in the properties of the service used by the pixel, sub-
TX i ic
: Bearer selection thresholds of the bearers defined in the reception equipment used by the cell TXi(ic).
Mi
scriber, or mobile Mi.
Mi
B UL Highest Service : Highest uplink bearer defined in the properties of the service used by the pixel, subscriber, or mobile Mi.
Calculations
The bearer available for data transfer for the pixel, subscriber, or mobile Mi is either the highest bearer from among the
bearers whose selection thresholds are less than the traffic C/(I+N) at Mi, or the highest bearer defined for the service
being accessed by Mi, whichever is lower.
Mi
B DL = Min Best Bearers
T B CINRTraffic
Mi
B UL = Min Best Bearers
TB
Mi
TX ic
i
TX i ic
CINR UL
Mi
B DL Highest Service and
Mi
B UL Highest Service
Output
9.3.7.3
Mi
B DL : Bearer assigned to the pixel, subscriber, or mobile Mi in the downlink.
B UL : Bearer assigned to the pixel, subscriber, or mobile Mi in the uplink.
Mi

Channel throughputs are calculated for the entire channel resources allocated to the pixel, subscriber, or mobile Mi.
Input
TX i ic
R DL
: Amount of downlink resources in the cell TXi(ic) as calculated in "Calculation of Cell Capacity (Total
496
AT271_TRG_E6
Forsk 2009

TX i ic
R UL

M : Bearer efficiency (bits/symbol) of the bearer assigned to the pixel, subscriber, or mobile Mi in the downlink
: Amount of uplink resources in the cell TXi(ic) as calculated in "Calculation of Cell Capacity (Total
B DL
in "Bearer Determination" on page 496.

M : Bearer efficiency (bits/symbol) of the bearer assigned to the pixel, subscriber, or mobile Mi in the uplink in
i
B UL
"Bearer Determination" on page 496.

PZDL : Permutation zone assigned to the pixel, subscriber, or mobile Mi in the downlink.
Mi
f Segment : Segmenting factor for the first downlink PUSC zone as calculated in "Effective Traffic and Pilot Interference Calculation (DL)" on page 478.
TX i ic
N Ant RX : Number of MIMO reception (uplink) antennas defined for the cell TXi(ic).
TX i ic
Mi
or mobile Mi.
Mi
or mobile Mi.
f SM Gain : SM Gain Factor defined for the clutter class where the pixel, subscriber, or mobile Mi is located.
G SM : Max MIMO gain for a given number of transmission and reception antennas read from the Max MIMO Gain
Max
vs. C/(I+N) graph from the MIMO Configurations table.

TX i ic
CNR Preamble : Preamble C/N the cell TXi(ic) as calculated in "Preamble C/N Calculation" on page 470.
T AMS : Adaptive MIMO switch threshold defined for the cell TXi(ic).
BLER B DL : Downlink block error rate read from the BLER vs. CINR Traffic graph available in the reception
TX i ic
TX i ic
Mi
equipment assigned to the terminal used by the pixel, subscriber, or mobile Mi.
Mi
Mi
BLER B UL : Uplink block error rate read from the BLER vs. CINR UL graph available in the reception equipment
assigned to the cell TXi(ic).
Mi
f TP Scaling : Throughput scaling factor defined in the properties of the service used by the pixel, subscriber, or
mobile Mi.
Mi
TP Offset : Throughput offset defined in the properties of the service used by the pixel, subscriber, or mobile Mi.
Calculations
Downlink:
TX i ic
Peak MAC Channel Throughput:
Mi
CTP P DL
R DL
Mi
DL
= -----------------------------------D Frame
In the above formula, the actual value of D Frame is used to calculate the channel throughput for coverage predictions, while D Frame = 1 sec for Monte Carlo simulations.
Segmentation (WiMAX 802.16e):
Mi
If the permutation zone assigned to the pixel, subscriber, or mobile Mi is the first downlink PUSC zone ( PZ DL = 0 )
and it is segmented, the channel throughput is calculated as:
TX i ic
Mi
CTP P DL
R DL
DL
- f Segment
= -----------------------------------D Frame
MIMO - Spatial Multiplexing Gain:

If the permutation zone assigned to the pixel, subscriber, or mobile Mi (WiMAX 802.16e) or the cell (WiMAX
802.16d) supports SM or AMS, spatial multiplexing gain is applied to the bearer efficiency.
Forsk 2009
AT271_TRG_E6
497

Max
Atoll reads the Max MIMO Gain, G SM , from the Max MIMO Gain graphs for the number of transmission antennas,
TX i ic
Mi
N Ant TX , the number of reception antennas, N Ant RX , and user mobility, Mobility M i . It then applies this
spatial multiplexing gain to the bearer efficiency taking into account the SM Gain Factor defined for the clutter class
of the user Mi as follows:
In case of SM:
B DL
In case of AMS:
Max
B DL
Mi
Max
B DL
Mi
B DL
Mi
if
TX i ic
TX i ic
CNR Preamble T AMS
Mi
Mi
Effective MAC Channel Throughput: CTP E DL = CTP P DL 1 BLER B DL
Mi
Mi
Mi
f TP Scaling
Application Level Channel Throughput: CTP A DL = CTP E DL ----------------------------- TP Offset
100
Mi
Uplink:
TX i ic
Peak MAC Channel Throughput:
Mi
CTP P UL
R UL
Mi
UL
= -----------------------------------D Frame
In the above formula, the actual value of D Frame is used to calculate the channel throughput for coverage predictions, while D Frame = 1 sec for Monte Carlo simulations.
MIMO - Spatial Multiplexing Gain:
If the permutation zone assigned to the pixel, subscriber, or mobile Mi (WiMAX 802.16e) or the cell (WiMAX
802.16d) supports SM or AMS, spatial multiplexing gain is applied to the bearer efficiency.
Max
Atoll reads the Max MIMO Gain, G SM , from the Max MIMO Gain graphs for the number of transmission antennas,
TX i ic
Mi
N Ant TX , the number of reception antennas, N Ant RX , and user mobility, Mobility M i . It then applies this
spatial multiplexing gain to the bearer efficiency taking into account the SM Gain Factor defined for the clutter class
of the user Mi as follows:
In case of SM:
B UL
In case of AMS:
Max
B UL
B UL
Max
B UL
Mi
if
TX i ic
TX i ic
CNR Preamble T AMS
Mi
Mi
Effective MAC Channel Throughput: CTP E UL = CTP P UL 1 BLER B UL
Mi
Mi
f TP Scaling
Mi
Application Level Channel Throughput: CTP A UL = CTP E UL ----------------------------- TP Offset
100
Mi
Output
9.3.8
Mi
CTP P DL : Downlink peak MAC channel throughput at the pixel, subscriber, or mobile Mi.
CTP E DL : Downlink effective MAC channel throughput at the pixel, subscriber, or mobile Mi.
CTP A DL : Downlink application level channel throughput at the pixel, subscriber, or mobile Mi.
CTP P UL : Uplink peak MAC channel throughput at the pixel, subscriber, or mobile Mi.
CTP E UL : Uplink effective MAC channel throughput at the pixel, subscriber, or mobile Mi.
CTP A UL : Uplink application level channel throughput at the pixel, subscriber, or mobile Mi.
Mi
Mi
Mi
Mi
Mi
Scheduling and Radio Resource Management

Atoll WiMAX BWA module includes a number of scheduling methods which can be used for scheduling and radio resource
allocation during Monte Carlo simulations. These resource allocation algorithms are explained in "Scheduling and Radio
Resource Allocation" on page 499 and the calculation of user throughputs is explained in "User Throughput Calculation"
on page 504.
498
AT271_TRG_E6
Forsk 2009
9.3.8.1
Scheduling and Radio Resource Allocation

Input
TX i ic
N Users Max : Maximum number of users defined for the cell TXi(ic).
QoS
Mi
Mi
: QoS class of the service (UGS, ErtPS, rtPS, nrtPS, or Best Effort) accessed by a mobile Mi.
: Priority of the service accessed by a mobile Mi.
TPD Min DL : Downlink minimum throughput demand for the service accessed by a mobile Mi.
TPD Min UL : Uplink minimum throughput demand for the service accessed by a mobile Mi.
TPD Max DL : Downlink maximum throughput demand for the service accessed by a mobile Mi.
TPD Max UL : Uplink maximum throughput demand for the service accessed by a mobile Mi.
Mi
Mi
Mi
Mi
TX i ic
Mi
equipment assigned to the terminal used by the mobile Mi.
Mi
Mi
BLER B UL : Uplink block error rate read from the BLER vs. CINR UL graph available in the reception equipment
assigned to the cell TXi(ic).
Mi
f TP Scaling : Throughput scaling factor defined in the properties of the service used by the mobile Mi.
TP Offset : Throughput offset defined in the properties of the service used by the mobile Mi.
CTP P DL : Downlink peak MAC channel throughput at the mobile Mi as calculated in "Channel Throughput Cal-
Mi
Mi
culation" on page 493.
Mi
CTP P UL : Uplink peak MAC channel throughput at the mobile Mi as calculated in "Channel Throughput Calculation" on page 493.
QoS
f Bias : Bias factor defined for the Biased (QoS Class) scheduling method.
Calculations
The following calculations are described for any cell TXi(ic) containing the users Mi for which it is the best server.
Mobile Selection:
TX i ic
The scheduler selects N Users mobiles for the scheduling and RRM process. If the Monte Carlo user distribution has
TX i ic
generated a number of users which is less than N Users Max , the scheduler keeps all the mobiles generated for the cell
TXi(ic).
TX i ic
TX i ic
TX i ic
N Users = Min N Users Max N Users Generated

Sel
For a cell, mobiles M i
TX i ic
N Users are selected for RRM by the scheduler.
Calculation of Actual Minimum and Maximum Throughput Demands:

Depending on the selected target throughput of the scheduler assigned to the cell TXi(ic), the actual minimum and maximum throughput demands can be considered as the peak MAC, effective MAC, or application throughput. Therefore:
Target Throughput = Peak MAC Throughput

Sel
Sel
Mi
Sel
Mi
Sel
Mi
Mi
TPD Min DL , TPD Min UL , TPD Max DL , and TPD Max UL are used for the RRM process as defined in the
service properties.
Target Throughput = Effective MAC Throughput

Sel
Sel
Downlink:
Mi
TPD Min DL
Sel
Mi
Sel
Sel
Uplink:
Forsk 2009
Mi
TPD Min UL
Mi
Sel
Mi
TPD Min DL
TPD Max DL
, TPD Max DL = --------------------------------------------------= --------------------------------------------------Sel
Sel
Mi
1 BLER B
1 BLER B Mi
DL
DL
Mi
Sel
Sel
Mi
Mi
TPD Min UL
TPD Max UL
TPD Max UL = --------------------------------------------------= --------------------------------------------------Sel
Sel
Mi
1 BLER B
1 BLER B Mi
UL
UL
Target Throughput = Application Level Throughput

AT271_TRG_E6
499

Sel
Sel
Mi
TPD Min DL
Downlink:
Mi
Sel
Sel
Uplink:
Mi
TPD Min UL
Sel
Mi
Mi
Sel
Mi
Mi
TPD Min DL + TP Offset
TPD Max DL + TP Offset
= -------------------------------------------------------------------------------------- , TPD Max DL = -------------------------------------------------------------------------------------Sel
Sel
Mi
Mi
1 BLER B
f
1 BLER B Mi f Mi
DL TP Scaling
DL TP Scaling
Mi
Sel
Mi
Mi
Sel
Mi
Mi
TPD Min UL + TP Offset
TPD Max UL + TP Offset
= -------------------------------------------------------------------------------------- , TPD Max UL = -------------------------------------------------------------------------------------Sel
Sel
Mi
Mi
f
1 BLER B
1 BLER B Mi f Mi
UL TP Scaling
UL TP Scaling
Resource Allocation for Minimum Throughput Demands:

Sel
1. For the QoS classes UGS, ErtPS, rtPS, and nrtPS, Atoll sorts the M i
TX i ic
N Users in order of decreasing service
Sel
priority, p
Mi
:
Sel
Sel
Mi
QoS
Sel
Mi
Sel
UGS
Mi
Mi
... n > p
=n
Sel
Mi
> 0 ...
Sel
Mi
=0
Sel
ErtPS
Mi
=n
Sel
... n > p
Mi
> 0 ...
Sel
rtPS
Mi
=0
Sel
Mi
... n > p
=n
Sel
Mi
> 0 ...
Sel
Mi
=0
Sel
nrtPS
Mi
=n
Sel
N1
... n > p
Mi
> 0 ...
Sel
TX i ic
Mi
=0
TX i ic
Where N N Users , if there are some Best Effort users, or N = N Users if there are no Best Effort users selected.
Sel
2. Starting with M i
Sel
= 1 up to M i
= N , Atoll allocates the downlink and uplink resources required to satisfy each
users minimum throughput demands in downlink and uplink as follows:

Sel
Sel
Mi
R Min DL
Sel
Mi
Mi
Sel
Mi
TPD Min DL
TPD Min UL
= ------------------------------ and R Min UL = -----------------------------Sel
Sel
Mi
Mi
CTP P DL
CTP P UL
3. Atoll stops the resource allocation in downlink or uplink,

Sel
Mi
RMin DL = 100 % , i.e., the resources available in downlink have been used up for
When/If in downlink
Sel
Mi
satisfying the minimum throughput demands of the mobiles.

-
When/If in uplink
Sel
Mi
R Min UL = 100 % , i.e., the resources available in uplink have been used up for satis-
Sel
Mi
fying the minimum throughput demands of the mobiles.

4. Mobiles which are active UL+DL must be able to get their minimum throughput demands in both UL and DL in
order to be considered connected UL+DL. If an active UL+DL mobile is only able to get its minimum throughput
demand in one direction, it is rejected, and the resources, that were allocated to it in the one direction in which it
was able to get a throughput, are allocated to other mobiles.
500
AT271_TRG_E6
Forsk 2009
5. If
Sel
Mi
R Min DL 100 % or
Sel
Sel
Mi
R Min UL 100 % , and all the minimum throughput resources demanded by the
Sel
Mi
Mi
mobiles have been allocated, Atoll goes to the next step for allocating resources to satisfy the maximum
throughput demands.
The remaining cell resources available for the next step are:
TX i ic
Downlink: R Rem DL = 100 %
Sel
Mi
R Min DL
Sel
Mi
Sel
TX i ic
Mi
RMin UL
Uplink: R Rem UL = 100 %
Sel
Mi
Resource Allocation for Maximum Throughput Demands:

For each mobile, the throughput demands remaining once the minimum throughput demands have been satisfied are the
difference between the maximum and the minimum throughput demands:
Sel
Sel
Mi
Sel
Mi
Mi
Downlink: TPD Rem DL = TPD Max DL TPD Min DL

Sel
Sel
Mi
Sel
Mi
Mi
Uplink: TPD Rem UL = TPD Max UL TPD Min UL

For the remaining throughput demands of the mobiles belonging to the QoS classes ErtPS, rtPS, nrtPS, and Best Effort,
the following resource allocation methods are available:
1. Proportional Fair:
The goal of this scheduling method is to distribute resources among users fairly in such a way that, on the average,
each user gets the highest possible throughput that it can get under the radio conditions at its location.
Sel
Let the total number of users belonging to the QoS classes ErtPS, rtPS, nrtPS, and Best Effort, be N M i
a. Atoll divides the remaining resources in the cell into equal parts for each user:
TX i ic
TX i ic
R Rem DL
R Rem UL
------------------------ and -----------------------N
N
b. Atoll converts the remaining throughput demands of all the users to their respective remaining resource demands:
Sel
Sel
Mi
RD Rem DL
Sel
Mi
Mi
Sel
Mi
TPD Rem DL
TPD Rem UL
and RD Rem UL = --------------------------------= --------------------------------Sel
Sel
Mi
Mi
CTP P DL
CTP P UL
Remaining resource demands of a user are given by the ratio between its remaining throughput demands and
the peak channel throughputs at the users location.
c. The resources allocated to each user by the Proportional Fair scheduling method for satisfying its maximum
throughput demands are:
TX i ic
TX i ic
Sel
Sel
Sel
Sel
Mi
Mi
R Rem DL
Mi
Mi
R Rem UL
R Max DL = Min RD Rem DL ------------------------- and R Max UL = Min RD Rem UL -------------------------

N
N
Each user gets either the resources it needs to achieve its maximum throughput demands or an equal share
from the remaining resources of the cell, whichever is smaller.
d. Atoll stops the resource allocation in downlink or uplink,
Sel
TX i ic
Mi
RMax DL = RRem DL , i.e., the resources available in downlink have been used
When/If in downlink
Sel
Mi
up for satisfying the maximum throughput demands of the mobiles.

-
When/If in uplink
Sel
Mi
TX i ic
R Max UL = R Rem UL , i.e., the resources available in uplink have been used up for
Sel
Mi
satisfying the maximum throughput demands of the mobiles.

e. If the resources allocated to a user satisfy its maximum throughput demands, this user is removed from the
list of remaining users.
f.
Forsk 2009
Atoll recalculates the remaining resources as follows:
AT271_TRG_E6
501

TX i ic
Sel
Mi
R Rem DL = 100 %
R Min DL
Sel
Sel
TX i ic
Mi
R Max DL and R Rem UL = 100 %
Sel
Mi
Sel
Mi
R Min UL
Sel
Mi
Sel
Mi
R Max UL
Sel
Mi
Mi
g. Atoll repeats the all the above steps for the users whose maximum throughput demands have not been satisTX i ic
TX i ic
fied until either R Rem DL = 0 and R Rem UL = 0 , or all the maximum throughput demands are satisfied.
2. Proportional Demand:
The goal of this scheduling method is to allocate resources to users weighted according to their remaining throughput demands. Therefore, the user throughputs for users with high throughput demands will be higher than those
with low throughput demands. In other words, this scheduler distributes channel throughput between users proportionally to their demands.
a. Atoll converts the remaining throughput demands of all the users to their respective remaining resource demands:
Sel
Sel
Mi
RD Rem DL
Sel
Mi
Mi
Sel
Mi
TPD Rem DL
TPD Rem UL
and RD Rem UL = --------------------------------= --------------------------------Sel
Sel
Mi
Mi
CTP P DL
CTP P UL
b. Atoll calculates the amount effective remaining resources for the cell of each user to distribute among the users as follows:
TX ic
TX i ic
i
R Eff Rem DL = Min R Rem DL
Sel
TX ic
Mi
TX i ic
i
RD Rem DL and R Eff Rem UL = Min R Rem UL
Sel
Sel
Mi
RD Rem UL
Sel
Mi
Mi
c. The resources allocated to each user by the Proportional Demand scheduling method for satisfying its maximum throughput demands are:
Sel
Sel
Mi
R Max DL
Sel
Mi
Mi
Sel
RD Rem DL
RD Rem UL
Mi
TX i ic
- and R Max
-------------------------------------- UL = R Eff Rem UL --------------------------------------Sel
Sel
TX i ic
R Eff Rem DL
Mi
Mi
RDRem DL
RDRem UL
Sel
Sel
Mi
Mi
3. Biased (QoS Class):

The goal of this scheduling method is to distribute resources among users of each QoS class fairly in such a way
that, on the average, each user gets the highest possible throughput that it can get under the radio conditions at
its location. The resources available for allocation to users of each QoS class depend on a bias factor. The QoS
Class Bias Factor controls the amount of resources available for each QoS class.
Calculation of the Remaining Resources per QoS Class:
QoS
The bias factor f Bias represents the bias in terms of resources allocated to 1 user of a QoS class with rank r to
the resources allocated to 1 user of a QoS class with rank r1:
QoS
Sel
Sel
Mi
Sel
Mi
Mi
f Bias
R Max rtPS
R Max nrtPS
R Max ErtPS
= 1 + ---------- = -----------------------------= ------------------------------ = -----------------------------Sel
Sel
Sel
100
Mi
Mi
Mi
R Max rtPS
R Max nrtPS
R Max BE
The ranks of QoS classes are:
QoS Class
QoS Class Rank r QoS
ErtPS
rtPS
nrtPS
Best Effort
The resources available for the users of each QoS class from among the remaining resources is calculated as
follows:
r
TX i ic
R QoS DL
TX i ic
R Rem DL
All QoS
502
1 QoS
1 QoS
N QoS ---
N QoS ---

TX i ic
TX i ic
and R QoS UL = R Rem UL ----------------------------------------------------------- -----------------------------------------------------------r
r
1 QoS
1 QoS
N QoS ---
N QoS ---

All QoS
AT271_TRG_E6
Forsk 2009

Resource Allocation:
Once the remaining resources available for the users of each QoS class have been determined, the allocation of
resources within each QoS class is performed as for the proportional fair scheduler.
Sel
Let the number of users belonging to a QoS class N QoS M i
a. Atoll divides the remaining resources of the QoS class into equal parts for each user:
TX i ic
TX i ic
R QoS DL
R QoS UL
------------------------ and -----------------------N QoS
N QoS
b. Atoll converts the remaining throughput demands of all the users to their respective remaining resource demands:
Sel
Sel
Mi
RD Rem DL
Sel
Mi
Mi
Sel
Mi
TPD Rem DL
TPD Rem UL
= --------------------------------and RD Rem UL = --------------------------------Sel
Sel
Mi
Mi
CTP P DL
CTP P UL
c. The resources allocated to each user by the Biased scheduling method for satisfying its maximum throughput
demands are:
TX i ic
TX i ic
Sel
Sel
Sel
Sel
Mi
Mi
R QoS DL
Mi
Mi
R QoS UL
R Max DL = Min RD Rem DL ------------------------ and R Max UL = Min RD Rem UL ------------------------

N
N QoS
QoS
Each user gets either the resources it needs to achieve its maximum throughput demands or an equal share
from the remaining resources of the QoS class, whichever is smaller.
d. Atoll stops the resource allocation for a QoS class in downlink or uplink,
-
Sel
TX i ic
Mi
When/If in downlink
R Max DL = R QoS DL , i.e., the resources available in downlink for the QoS class
Sel
Mi
have been used up for satisfying the maximum throughput demands of the mobiles.
Sel
TX i ic
Mi
RMax UL = RQoS UL , i.e., the resources available in uplink for the QoS class have
When/If in uplink
Sel
Mi
been used up for satisfying the maximum throughput demands of the mobiles.
e. If the resources allocated to a user satisfy its maximum throughput demands, this user is removed from the
list of remaining users.
f.
Atoll recalculates the remaining resources as follows:

TX i ic
R QoS DL = 100 %
Sel
Mi
R Min DL
Sel
Sel
TX i ic
Mi
R Max DL and R QoS UL = 100 %
Sel
Mi
Sel
Mi
Mi
Sel
Mi
R Min UL
Sel
Mi
R Max UL
Sel
Mi
g. Atoll repeats the all the above steps for the users of the QoS class whose maximum throughput demands have
TX i ic
TX i ic
not been satisfied until either R QoS DL = 0 and R QoS UL = 0 , or all the maximum throughput demands are
satisfied.
4. Max Aggregate Throughput:
The goal of this scheduling method is to achieve the maximum aggregate throughput for the cells. This is done by
allocating the as much resources as needed to mobiles with high C/(I+N) conditions. As mobiles with high C/(I+N)
can get higher bearers, and therefore require less amount of resources, more mobiles can therefore be allocated
resources in the same frame, and the end-throughput for each cell will be the highest compared to other types of
schedulers.
Sel
a. Atoll sorts the M i
TX i ic
N Users in order of decreasing downlink or uplink traffic C/(I+N), depending on whether
the allocation is being performed for the downlink or for the uplink.
b. Starting with the mobile with the highest rank, Atoll allocates the downlink and uplink resources required to
satisfy each users remaining throughput demands in downlink and uplink as follows:
Sel
Sel
Mi
R Max DL
Mi
Sel
Sel
Mi
Mi
TPD Rem DL
TPD Rem UL
and R Max UL = --------------------------------= --------------------------------Sel
Sel
Mi
Mi
CTP P DL
CTP P UL
c. Atoll stops the resource allocation in downlink or uplink,
Forsk 2009
AT271_TRG_E6
503
When/If in downlink
Sel
TX i ic
Mi
R Max DL = R Rem DL , i.e., the resources available in downlink have been used
Sel
Mi
up for satisfying the maximum throughput demands of the mobiles.

Sel
TX i ic
Mi
RMax UL = RRem UL , i.e., the resources available in uplink have been used up for
When/If in uplink
Sel
Mi
satisfying the maximum throughput demands of the mobiles.

Total Amount of Resources Assigned to Each Selected Mobile:
Sel
Atoll calculates the amounts of downlink and uplink resources allocated to each individual mobile M i
(which can also
be referred to as the traffic loads of the mobiles) as follows:

Sel
Mi
Downlink: TL DL
Sel
Mi
Uplink: TL UL
Sel
Mi
= R DL
Sel
Mi
= R UL
Sel
Sel
Mi
Mi
= R Min DL + R Max DL
Sel
Sel
Mi
Mi
= R Min UL + R Max UL
Output
Sel
9.3.8.2
Mi
TL DL
Sel
Mi
Sel
= R DL : Downlink traffic load or the amount of downlink resources allocated to the mobile M i
Sel
Mi
TL UL
Sel
Mi
Sel
= R UL : Uplink traffic load or the amount of uplink resources allocated to the mobile M i
User Throughput Calculation

User throughputs are calculated for the percentage of resources allocated to each mobile selected by the scheduling for
Sel
RRM during the Monte Carlo simulations, M i
Input
Sel
Mi
Sel
R DL : Amount of downlink resources allocated to the mobile M i
as calculated in "Scheduling and Radio
Resource Allocation" on page 499.

Sel
Mi
Sel
R UL : Amount of uplink resources allocated to the mobile M i
as calculated in "Scheduling and Radio Resource
Allocation" on page 499.

Sel
Mi
Sel
CTP P DL : Downlink peak MAC channel throughput at the mobile M i
as calculated in "Channel Throughput
Calculation" on page 493.

Sel
Mi
Sel
CTP P UL : Uplink peak MAC channel throughput at the mobile M i
as calculated in "Channel Throughput Cal-
culation" on page 493.

Sel
Mi
TX i ic
Sel
equipment assigned to the terminal used by the mobile M i
Sel
Mi
Mi
BLER B UL : Uplink block error rate read from the BLER vs. CINR UL graph available in the reception equip
ment assigned to the cell TXi(ic).
f TP Scaling : Throughput scaling factor defined in the properties of the service used by the mobile M i
Sel
Mi
Sel
Sel
Mi
Sel
TP Offset : Throughput offset defined in the properties of the service used by the mobile M i
Calculations
Downlink:
Sel
Sel
Mi
Mi
Sel
Mi
Peak MAC User Throughput: UTP P DL = R DL CTP P DL
Mi
Mi
Mi
Effective MAC User Throughput: UTP E DL = UTP P DL 1 BLER B DL
Sel
Sel
Sel
Sel
Sel
504
Application Level User Throughput:
Mi
UTP A DL
AT271_TRG_E6
Sel
Mi
UTP E DL
Mi
Sel
Mi
f TP Scaling
----------------------------- TP Offset
100
Forsk 2009

Uplink:
Sel
Sel
Mi
Mi
Sel
Mi
Peak MAC User Throughput: UTP P UL = R UL CTP P UL
Mi
Mi
Mi
Effective MAC User Throughput: UTP E UL = UTP P UL 1 BLER B UL
Sel
Sel
Sel
Sel
Sel
Application Level User Throughput:
Mi
UTP A UL
Sel
Mi
UTP E UL
Mi
Sel
Mi
f TP Scaling
----------------------------- TP Offset
100
Output
Sel
Mi
Sel
UTP P DL : Downlink peak MAC user throughput at the pixel, subscriber, or mobile M i
Sel
UTP E DL : Downlink effective MAC user throughput at the pixel, subscriber, or mobile M i
Sel
Mi
UTP A DL : Downlink application level user throughput at the pixel, subscriber, or mobile M i
UTP P UL : Uplink peak MAC user throughput at the pixel, subscriber, or mobile M i
Sel
Sel
9.3.9
Sel
Mi
Mi
Sel
Sel
Mi
.
Sel
UTP E UL : Uplink effective MAC user throughput at the pixel, subscriber, or mobile M i
Sel
Mi
Sel
UTP A UL : Uplink application level user throughput at the pixel, subscriber, or mobile M i

This section describes how the optimum beamformer smart antenna model works. The smart antenna performs simple
beam forming in the downlink to steer the main beam towards the served user, and uses the Minimum Mean Square Error
algorithm in the uplink in order to cancel interference.
9.3.9.1
Introduction
Adaptive antenna systems use more than one antenna elements, along with smart signal processing, to locate and track
various types of signals, to dynamically minimize interference, and maximize useful signal reception. The signal processor
dynamically applies weights to each element of the adaptive antenna system to create array patterns in real-time to maximize the output C/(I+N).
Figure 9.11: Linear Adaptive Array System

The antenna patterns created for downlink transmission have a main beam pointed in the direction of the useful signal. In
the uplink, in addition to the main beam pointed in the direction of the useful signal, there can also be one or more nulls in
the direction of the interfering signals. If the adaptive antenna system is using E SA antenna elements, it is possible to
create E SA 1 nulls to cancel E SA 1 interfering signals. In a mobile environment where the interference is not stationary, the antenna patterns are adjusted so that the nulls remain in the direction of the moving interference. A system using
adaptive antennas adjusts the weights on each antenna element to achieve such a pattern.
The optimum beamformer supports linear adaptive array systems, such as the one shown in Figure 9.11: on page 505.
Forsk 2009
AT271_TRG_E6
505

In the following explanations, we assume:
9.3.9.2
There are a total of E SA elements in the adaptive antenna system.
is the angle of arrival for the useful signal.
is the angle at which we want to calculate the smart antenna gain.

d is the distance between two adjacent antenna elements.
Downlink
Figure 9.12: Downlink Beamforming

The downlink algorithm models a conventional beam former. The smart antenna processor applies complex weights, w n ,
to each antenna element in order to form a beam towards the served user. The magnitude of these complex weights is set
to 1. The beamforming is performed using only the phase of the complex weights. The steering vector, S , representing
the complex weights for forming a beam towards the served user, i.e., at the angle of arrival is given by:
S = 1 e
2
j ------- d sin
2
j ------- 2d sin
... e
T
2
j ------- E SA 1 d sin
Where the notation T represents the transpose of a matrix.

Therefore, the complex weight at any nth antenna element can be given by:
wn = e
2
j ------- nd sin
j n sin
In Atoll, d = --- , therefore, w n = e

.
2
The smart antenna gain in any direction can be given by:

H
G SA = g n S R S
Where the notation H represents the Hilbert transform, which is the complex conjugate transpose of a matrix, g n is the
gain of the nth antenna element in the direction , and R is the array correlation matrix for a given user direction , given
by:
H
R = S S
For the direction of the served user, i.e., , the smart antenna gain is calculated as follows:
H
G SA = g n S R S = g n S S S S = g n E SA
The smart antenna gain includes the gain of the beamforming as well as the gain of power combination.
The smart antenna gain in dB will be G SA = 10 Log G SA .
The smart antenna is able to form the beam only in the horizontal plane, therefore, the vertical pattern is assumed to
remain the same.
506
AT271_TRG_E6
Forsk 2009
Power Combination Gain

Cell transmission power is fed to each antenna element of the smart antenna system. Since each element transmits the
same input power, this results in a gain due to power combination, i.e., the powers fed to each antenna element are
combined for transmission.

During Monte Carlo simulations, Atoll generates a time-slot scenario for each victim mobile and calculates the downlink
C/(I+N) as described in the section "Traffic and Pilot Interference Signal Levels Calculation (DL)" on page 475. However,
as it calculates the smart antenna gains for each victim mobile in a cells coverage area, it averages the array correlation
matrix R over all the iterations in order to generate an angular distribution of the downlink traffic power density, which is
a combination of signal power and angles.
The average array correlation matrix is given by:
J
R Avg =
j pj Rj
j=1
Where R Avg is the average downlink array correlation matrix, J is the number of served mobiles during the simulation, j
is the probability of presence of the mobile j, p j is the EIRP transmitted towards the mobile j, and R j is the array correlation
matrix for the mobile j.
The probability of presence of the mobile j is the ratio between the downlink resources provided to the mobile j and the
total amount of available downlink resources. For example, if a mobile has been granted 10 % of the number of available
slots in the downlink subframe, its probability of presence is 10 %.
9.3.9.3
Uplink
The uplink models the adaptive Minimum Mean Square Error algorithm which optimizes the useful signal as well as
cancels the interference from the most interfering E SA 1 interfering mobiles. The optimal beam forming method used in
Atoll overcomes the limitations of a null steering beam former and maximizes the output C/(I+N).
A simple null steering beam former requires the knowledge of the directions of interference sources, and the estimated
weights do not maximize the output C/(I+N). Whereas, an optimal beam former does not require knowledge of directions
and power levels of interference to maximize the output C/(I+N). It only requires the direction of the useful signal in order
to calculate the optimum C/(I+N).
Figure 9.13: Uplink Adaptive Algorithm

represent the vector of E complex weights for the beam former. w
is given by:
Let w
SA
= R1 S
w
Where S is the steering vector in the direction of the served user, . , which is a constant value for a given useful
signal, represents the smart antenna gain in the uplink given by the equation:
E SA
= ---------------------------------H
1
S RN S
Forsk 2009
AT271_TRG_E6
507

1
R N is the inverse of the total noise correlation matrix. The total noise correlation matrix is the sum of the thermal noise
correlation matrix R n , and the interference correlation matrix R I , given by:
J
2
RN = Rn + RI = n I +
pj Sj Sj
j=1
J
Where R n =
2
n
pj Sj Sj
I and R I =
j=1
2
n is the thermal noise power. I is the identity matrix. p j is the power received by one element of the smart antenna from
the jth interfering mobile. S j is the steering vector in the direction of the jth interfering mobile, . J is the total number of
interfering mobiles.
The total noise power, including thermal noise and interference from all uplink interferers, received by a cell is given by:
= 2 S H R1 S
P
N
And, the total power received from the served user is given by:
= p 2 S H R 1 S 2
P
Where p is the power received by one element of the smart antenna from the served user.
The C/(I+N) in the uplink is then calculated by:
2
H
2
1
p S RN S
H
P
1
CINR UL = ------- = --------------------------------------------------------------= p S RN S
H
2
1
P
N
S RN S
From the above equation, we can determine the uplink smart antenna beam forming gain in the direction of the served
1
user. C UL can be calculated from the above equation by considering the interference and noise to be null, i.e., R N = I .
This gives:
H
C UL = p S I S = p E SA
From the above equation, the uplink smart antenna beam forming gain equals the number of smart antenna elements, i.e.,
G SA = E SA .

During Monte Carlo simulations, Atoll generates a time-slot scenario for each victim mobile and calculates the uplink C/
(I+N) as described above.
1
The inverse noise correlation matrix R N for each iteration k includes the effect of the matrix calculated for the previous
iteration. The interference power and its direction is stored at the end of each simulation. Hence, Atoll is able to calculate
an average of the smart antenna interference-cancellation effect. The result is the angular distribution of the uplink noise
rise, which is calculated from the inverse of the noise correlation matrix obtained at the end of the last iteration of a Monte
Carlo simulation. This angular distribution of the uplink noise rise can be stored in the Cells table. The average of the
inverse noise correlation matrices is calculated as follows:
K
1
RN
Avg
1
= ----
K
RN
1
k
k=1
1
Where R N
Avg
is the average of the inverse noise correlation matrices of all the iterations from k = 1 to K, and R N
is
the inverse noise correlation matrix of the kth iteration.

The interference can be isolated from the thermal noise and can be calculated for any direction using the formula.
E SA
2
I UL = ------------------------------------------ n
H
1
S RN
S
Avg
Where I UL is the interfering signal in the direction , ESA is the number of smart antenna elements, S is the steering
2
vector in the direction , and n is the thermal noise power, with I being an identity matrix.
The angular distribution of the uplink noise rise is given by:
508
AT271_TRG_E6
Forsk 2009

2
I UL + n
NR UL = ---------------------------2
n
9.4
Automatic Allocation Algorithms

The following sections describe the algorithms for automatic neighbour and preamble index planning.
9.4.1

The intra-technology neighbour allocation algorithm takes into account the cells of all the TBC transmitters. It means that
the cells of all the TBC transmitters of your .atl document are potential neighbours.
The cells to be allocated will be called TBA cells. They must fulfil the following conditions:
They are active,

Only TBA cells are assigned neighbours.

Note:
We assume a reference cell TXi(ic) and a candidate neighbour cell TXj(jc). When automatic allocation starts, Atoll checks
the following conditions:
1. The distance between both cells must be less than the user-definable maximum inter-site distance. If the distance
is discarded.
2. The calculation options,
-
Force Co-site Cells as Neighbours: If selected, Atoll adds all the cells located on the same site as the reference cell to the candidate neighbour list. The weight of this constraint can be defined. It is used to calculate
the rank of each neighbour, and its importance.
Force Adjacent Cells as Neighbours: If selected, Atoll adds all the cells geographically adjacent to the reference cell to the candidate neighbour list. The weight of this constraint can be defined. It is used to calculate
the rank of each neighbour, and its importance.
Determination of Adjacent Cells: Geographically adjacent cells are determined on the basis of their best
server coverage areas. A candidate neighbour cell TXi(ic) is considered adjacent to the reference cell TXi(ic)
if there exists at least one pixel of TXj(jc)s best server coverage area where TXi(ic) is the second best server.
The ranking of adjacent neighbour cells increases with the number of such pixels. Adjacent cells are sorted in
the order of decreasing ranking.
Figure 9.14: Determination of Adjacent Cells

-
Force Neighbour Symmetry: If selected, Atoll adds the reference cell to the candidate neighbour list of the
its candidate neighbour.
A symmetric neighbour relation is allowed only if the neighbour list of the reference cell is not already full. If
TXj(jc) is a neighbour of TXi(ic) but TXi(ic) is not a neighbour of TXj(jc), there can be two possibilities:
i.
The neighbour list of TXj(jc) is not full, Atoll will add TXi(ic) to the end of the list.
ii. The neighbour list of TXj(jc) is full, Atoll will not be able to add TXi(ic) to the list, so it will also remove TXj(jc)
from the neighbour list of TXi(ic).
-
Forsk 2009
Force Exceptional Pairs: This option enables you to force/forbid some neighbour relations. Exceptional pairs
are pairs of cells which will always or never be neighbours of each other.
AT271_TRG_E6
509

If you select "Force exceptional pairs" and "Force symmetry", Atoll considers the constraints between exceptional pairs in both directions so as to respect symmetry condition. On the other hand, if neighbourhood relationship is forced in one direction and forbidden in the other, symmetry cannot be respected. In this case, Atoll
-
Delete Existing Neighbours: If selected, Atoll deletes all the current neighbours and carries out a new neighbour allocation. If not selected, the existing neighbours are kept in the list.
3. The coverage areas of TXi(ic) and TXj(jc) must have an overlap ( S TX ic S TX jc ).

i
Here S TX ic is the surface area covered by the cell TXi(ic) that comprises all the pixels where:
The received preamble signal level is greater than or equal to the preamble signal level threshold. The
TX i ic
received preamble signal level ( C Preamble ) and the preamble signal level threshold are calculated from
TX i ic
TX i ic
CNR Preamble and T Preamble , respectively, by adding the value of the noise to them.
TX i ic
TX i ic
S TX ic is the surface area covered by TXi(ic) within C Preamble + HO Start and C Preamble + HO End .
HO Start is the margin with respect to the best preamble signal level at which the handover starts, and
HO End is the margin with respect to the best preamble signal level at which the handover ends.
S TX jc is the coverage area where the candidate cell TXj(jc) is the best server.
Note:
When
the
For calculating the overlapping coverage areas, Atoll uses the service with the lowest body
loss, the terminal that has the highest difference between gain and losses, and the
shadowing margin calculated using the defined cell edge coverage probability, if the option
is selected. The service and terminal are selected such that the selection gives the largest
possible coverage areas for the cells.
dialogue of the Predictions folder.
above
conditions
are
met,
Atoll
calculates
the
percentage
of
the
coverage
area
overlap
S TX ic S TX jc
i
j
- 100 ), and compares this value with the % Min Covered Area. TXj(jc) is considered a neighbour of
( -------------------------------------------S TX ic
i
S TX ic S TX jc
i
j
- 100 % Min Coverage Area .
TXi(ic) if -------------------------------------------S TX ic
i
Figure 9.15: Overlapping Zones

Next, Atoll calculates the importance of the automatically allocated neighbours. Atoll sorts the neighbours in decreasing
order of importance in order to keep the ones with high importance. If the maximum number of neighbours to be allocated
to each cell is exceeded, Atoll keeps the ones with high importance.
The importance (%) of neighbours depends on the reason of allocation:
510
Neighbour Cause
When
Importance
Value
Existing neighbour
Only if the Delete Existing Neighbours option is not selected

Existing
importance
Exceptional pair
Only if the Force Exceptional Pairs option is selected
100 %
AT271_TRG_E6
Forsk 2009

Co-site cell
Only if the Force Co-site Cells as Neighbours option is

selected
Importance
Function (IF)
Adjacent cell
Only if the Force Adjacent Cells as Neighbours option is

selected
Importance
Function (IF)

coverage conditions
Only if the % Min Covered Area is exceeded
Importance
Function (IF)
relationship
Only if the Force Neighbour Symmetry option is selected
Importance
Function (IF)
The importance is evaluated using an Importance Function (IF), which takes into account the following factors:
Co-site factor (C): a Boolean,

Adjacency factor (A): the percentage of adjacency,
Overlapping factor (O): the percentage of overlapping
The minimum and maximum importance assigned to each of the above factors can be defined.
Factor
Min
Importance
Default Value
Max
Importance
Default Value

Min(O)
1%
Max(O)
30 %
Min(A)
30 %
Max(A)
60 %
Co-site factor (C)
Min(C)
60 %
Max(C)
100 %
The Importance Function is evaluated as follows:
Neighbour Cause
Co-site
Importance Function
Adjacent
IF with With Default Values
no
no
1 % + 29 %(O)
no
yes
Min(A)+Delta(A){Max(O)(O)+(100 %-Max(O))(A)}
30 % + 30 %{30 %(O) + 70 %(A)}
yes
yes
Min(C)+Delta(C){Max(O)(O)+(100 %-Max(O))(A)}
60 % + 40 %{30 %(O )+ 70 %(A)}
Where Delta(x) = Max(x) - Min(x)

Notes:
by neighbour cause. With the default values for minimum and maximum importance fields,
neighbours will be ranked in this order: co-site neighbours, adjacent neighbours, and
neighbours allocated based on coverage overlapping.
If ranges of the IF factors overlap each other, the neighbours may not be ranked by
neighbour cause.
The ranking between neighbours from the same category depends on (A) and (O) factors.
The default value of Min(O) = 1 % ensures that neighbours selected for symmetry will have
an importance greater than 0 %. With a value of Min(O) = 0 %, neighbours selected for
symmetry, will have an importance greater than 0 % only if there is some overlapping.
In the results, Atoll lists only the cells for which it finds new neighbours. Therefore, if a transmitter has already reached its
maximum number of neighbours before starting the allocation, it will not appear in the results table.
9.4.2
Automatic Preamble Index Allocation

IEEE 802.16e defines 114 preamble indexes. Each preamble index, from 0 to 113, contains the following information:
Segment number (0, 1, or 2),

DL PermBase (0 to 31) for the obligatory first DL PUSC zone, and
A pseudo-noise sequence transmitted using the subcarriers corresponding to the preamble carrier set.
The downlink subframe can be divided into a 3-segment structure, and includes a preamble which begins the transmission
(the first symbol of the downlink transmission). The preamble subcarriers are divided into 3 carrier sets. There are three
possible groups consisting of a carrier set each which may be used by any segment. These are defined by allocation of
different subcarriers to each one of them. The subcarriers are modulated using a BPSK modulation with a specific PseudoNoise (PN) sequence.
Preamble carrier sets are defined using equation below:
PreambleCarrierSet n = n + k 3
Where PreambleCarrierSetn gives the subcarriers used by the preamble, n is the number of the preamble carrier set
indexed 0, 1, or 2, k is a running index from 0 to 567 for FFT 2048, from 0 to 283 for FFT 1024, from 0 to 142 for FFT 512,
and from 0 to 35 for FFT 128.
In a WiMAX 802.16e network, each base station transmits a different PN sequence, out of the 114 available, on the preamble carrier set. A mobile trying to connect to the network scans all the preamble subcarriers, listens to all the preambles
Forsk 2009
AT271_TRG_E6
511

(i.e., PN sequences) from all the base stations it can receive, and compares the PN sequences it is receiving with the 114
stored in its memory in order to detect the preamble index from the PN sequence.
It selects the base station as its server whose preamble it receives with either the highest signal level or the highest C/
(I+N). Once the best server is known, its PN sequence is used to identify its transmission. The PN sequence of the best
server gives the preamble index, which in turn gives the segment number, and the IDCell (DL PermBase of the first DL
PUSC zone, referred to as Cell PermBase in Atoll). Therefore, the mobile knows which subcarriers to listen to for the FCH,
DCD, UCD, DL-MAP, and UL-MAP.
As can be understood from the above description, if all the cells in the network transmit the same preamble index, the
network will have 100% interference on downlink preambles, and it will be impossible for a mobile to identify different cells.
Cell search and selection will be impossible. Therefore, it is important to intelligently allocate preamble indexes to cells so
as to reduce preamble interference, and allow easy recognition of cells by mobiles.
The following describes the automatic allocation method for preamble indexes in a WiMAX 802.16e network, which takes
into account the distance between transmitters, the frequency plan of the network (i.e., co- and adjacent channel interference probabilities), and the neighbour relations.
The preamble index allocation algorithm takes into account the cells of all the TBC transmitters. The cells to be allocated
will be called TBA cells. They must fulfil the following conditions:
They are active,

Their status is not set to locked,
Note:
9.4.2.1
Constraint and Relationship Weights

The automatic preamble index allocation algorithm is based on a cost-based function which takes into account the following constraints, in the order of priority:
1. Same preamble index,
Assigned weight PI = 0.6
2. Same segment number,
Assigned weight Seg = 0.38
3. Same cell permbase,
Assigned weight PB = 0.02
Note that the sum of the weights assigned to the above constraints is 1.
The above constraints are studied between each TBA cell and its related cells. Atoll calculates the cost between each individual TBA and related cell, and then the global cost for the TBA cell.
Related cells of a TBA cell are:
9.4.2.2
Its neighbours, if the check box "Take Neighbours into Account" is selected,
Assigned weight Neighbour = 0.35
Cells within the cells (or the default) minimum reuse distance,
Assigned weight Dis tan ce = 0.35
Calculation of Cost Between TBA and Related Cells

Atoll calculates the constraint violation level between the TBA cell TXi(ic) and its related cell TXj(jc) as follows:
VL
TX i ic TX j jc
TX i ic TX j jc
= rO
TX i ic TX j jc
Where r O
PI
Seg
PB
PI p Coll + Seg p Coll + PB p Penalty
is the total channel overlap ratio between the TXi(ic) and TXj(jc) as calculated in "Co- and Adjacent
Channel Overlaps Calculation" on page 462, PI , Seg , and PB are the weights assigned to the preamble index,
segment number, and cell permbase constraints, and,
512
PI
the preamble index collision probability is given by p Coll = 1
1
Seg
the segment number collision probability is given by p Coll =
AT271_TRG_E6
if PI
if PI
TX i ic
TX i ic
= PI
PI
TX i ic
if N Seg
TX i ic
if N Seg
TX j jc
TX j jc
TX j jc
= N Seg
TX j jc
, and
N Seg
Forsk 2009
the cell permbase penalty is given by
PB
p Penalty
1 if PB TXi ic PB TXj jc
=
TX i ic
TX j jc
PB
0.001 if PB
0 Otherwise
AND Site
AND Site
TX i ic
TX i ic
= Site
Site
TX j jc
TX j jc
if the cell permbase allocation
PB
strategy is set to "Same Cell PermBase per Site", and by p Penalty = 0 if the cell permbase allocation strategy is
set to "Free".
The cell permbase penalty models the cell permbase constraint.
Next, Atoll calculates the importance of the relation between the TBA cell and its related cell.
TX i ic TX j jc
TX i ic TX j jc
Total
= Neighbour Neighbour
TX i ic TX j jc
Where Neighbour
TX i ic TX j jc
Dis tan ce
TX i ic TX j jc
+ Dis tan ce Dis tan ce
is the importance of the relationship between the TBA and its related neighbour cell, and
is the importance of the relationship between the TBA and its related cell with respect to the distance
between them.
TX i ic TX j jc
Neighbour
is calculated during automatic neighbour allocation by Atoll as explained in "Automatic Neighbour Alloca-
tion" on page 509. For manual neighbour allocation, this value is equal to 1.
TX i ic TX j jc
Dis tan ce
is calculated by the preamble allocation algorithm as follows:
TX i ic TX j jc
Dis tan ce
D Reuse
2
= Log -------------------------------------
D TXi ic TXj jc
--------------------------------------------------------------2
Log D Reuse
if D
TX i ic TX j jc
Otherwise
Where D Reuse is the minimum reuse distance, either defined for the TBA cell individually or set for all the cells in the automatic allocation dialogue, and D
TXj(jc) calculated as follows:
D
D
TX i ic TX j jc
TX i ic TX j jc
joining them. d
= d
TX i ic TX j jc
TX i ic TX j jc
is the weighted distance between the TBA cell TXi(ic) and its related cell
1 + x cos cos 2
is weighted according to the orientations of the TBA and its related cell with respect to the straight line
TX i ic TX j jc
is the distance between the two cells considering any offsets with respect to the site locations.
TX i ic TX j jc
x is set to 15 % so that the maximum variation in D

due to the azimuths does not exceed 60 %. and
are, of course, calculated from the azimuths of the two cells as shown in Figure 9.16: on page 513.
Figure 9.16: Weighted Distance Between Cells

The above formula implies that two cells facing each other will have a shorter effective distance between them than the
real distance, and two cells pointing in opposite directions will have a greater effective distance.
The importance of the distance relation is explained in Figure 9.17: on page 514. This figure shows that cells that are
located near (based on the effective distance which is weighted by the orientations of the cells) have high importance,
which is interpreted as a high cost, and cells that are located far have low importance. Cells that are further than the reuse
distance do not have any cost related to the distance relation.
Forsk 2009
AT271_TRG_E6
513
Figure 9.17: Importance Based on Distance Relation

From the constraint violation level and the total importance of the relation between the TBA and its related cell, Atoll calculates the quality reduction factor for the pair as follows:
QRF
TX i ic TX j jc
1
=
TX i ic TX j jc
TX i ic TX j jc
Total
1 VL
TX i ic TX j jc
if r O
=0
OR VL
TX i ic TX j jc
=0
Otherwise
The quality reduction factor is a measure of the cost of an individual relation.

The total cost of the current preamble index allocation for any TBA cell is given as follows, considering all the cells with
which the TBA cell has relations:
TX i ic
$ Total
= 1
QRF
TX i ic TX j jc
TX j jc
And, the total cost of the current preamble index allocation for the entire network is simply the sum of the total TBA cell
costs calculated above, i.e.,
$ Total =
TX i ic
$ Total
TX i ic
9.4.2.3
Automatic Allocation Algorithm

The automatic preamble index allocation algorithm is an iterative algorithm which:
9.5
Calculates the cost (as described above) of the current preamble index allocation,
Allocates new preamble indexes to cells in order to reduce the costs, and calculates the cost again,
Memorises the different allocation plans in order to determine the best allocation, i.e., which provides the lowest
total cost.
Glossary of WiMAX Terms

Figure 9.18: on page 515 shows a TDD WiMAX frame along with its different components and explains the significance
of different terms in the Atoll WiMAX module.
Atoll allows you to set network level parameters which are common to all the transmitters and cells in the entire network.
These parameters are used in coverage predictions as well as during Monte Carlo simulations by the radio resource
management and scheduling algorithms. The global WiMAX parameters include.
514
Symbol: A symbol is the smallest resource unit that can be allocated to a user in WiMAX networks.
Symbol Duration (SD): The symbol duration is the length of each symbol in the frame. The length of a frame, i.e.,
the frame duration, can be expressed in terms of the number of symbol durations in the frame. It is referred to as
OFDM symbol in the IEEE 802.16 specifications.
Frame duration: It is the length of a WiMAX frame in milliseconds. You can choose from a list of frame durations
defined in the IEEE 802.16 specifications.
Cyclic prefix ratio: The total symbol duration in WiMAX comprises the useful part of the symbol, carrying the data
bits, and a CRC part, which is a portion of the useful data part repeated at the beginning of each symbol. The cyclic
prefix is the method used by WiMAX to counter inter-symbol interference (ISI). The cyclic prefix and the orthogonality of subcarriers ensure that there is negligible intra-cell interference in WiMAX.
AT271_TRG_E6
Forsk 2009
Figure 9.18: WiMAX Frame
Forsk 2009
Fixed and variable overheads: The fixed and variable overheads in the uplink and downlink subframes are used
to model the preamble and other time domain overheads such as broadcast messages including DL-MAP, ULMAP, UCD, and DCD, and the FCH, in downlink, and Ranging and Bandwidth Request messages in the uplink.
The preamble is always one symbol duration long and can be modelled using the fixed overhead, while other messages whose lengths vary according to either the frame duration or the channel width can be modelled using the
variable overhead. This is the reason why the fixed overheads are available in terms of symbol durations (SD) and
the variable overheads in terms of percentages of the uplink and downlink subframes. Variable overheads are percentages of the DL and the UL subframe sizes, after the fixed overheads have been removed from the subframe
size.
DL subframe ratio (TDD only): It is the percentage of the entire frame duration which corresponds to the downlink
subframe. In FDD networks, the downlink and uplink subframes cover entire frame durations in different frequency
bands.
Transmit and receive time guards (TDD only): Transmit and receive time guards are also time domain overheads, i.e., these are portions of the frame which cannot be used for data transfer. You can enter TTG and RTG
times in milliseconds. These time guards allow the receiver and the transmitter to switch between transmit and
receive modes.
Subchannels: A subchannel is a group of subcarriers. A channel can be divided into a number of subchannels.
You can set the number of these subchannels at the network level in Atoll.
Subcarriers (or tone): The entire channel contains a number of subcarriers which compose the upper and lower
guard bands, the pilot subcarriers, and the data subcarriers. The guards, pilots, and the DC subcarrier can not be
used for data transfer. The total thermal noise over the entire channel bandwidth is calculated according to the
number of used subcarriers out of the total number of subcarriers. The used subcarriers are the data and the pilot
subcarriers. The data transfer capacity of a channel is calculated by considering the data subcarriers only.
User: A general term used interchangeably with subscribers, mobiles, and receivers.
Subscriber: Term used for users with fixed geographical coordinates.
Mobile: Term used to indicate the users generated and distributed during simulations. These users have, among
other parameters, definite services, terminal types, and mobility types assigned for the duration of the simulations.
Receiver: A probe mobile, with the minimum required parameters needed for the computation of path loss, used
for propagation loss and raster coverage predictions.
Bearer: Defines a Modulation and Coding Scheme (MCS) used to carry data over the channel.
Peak MAC Throughput: The maximum MAC layer throughput (user or channel) that can be achieved at a given
location using the best possible WiMAX bearer. This throughput is the raw data rate without considering the effects
of retransmission due to errors and higher layer coding and encryption.
Effective MAC Throughput: The net MAC layer throughput (user or channel) that can be achieved at a given
location using the best possible WiMAX bearer calculated taking into account the reduction of throughput due to
retransmissions due to errors.
Application Throughput: The application layer throughput (user or channel) that can be achieved at a given location using the best possible WiMAX bearer calculated taking into account the reduction of throughput due to PDU/
SDU header information, padding, encryption, coding, and other types of overhead.
Channel Throughputs: Peak MAC, effective MAC or application level throughputs achieved at a given location
using the best possible WiMAX bearer with the entire channel resources.
User Throughputs: Peak MAC, effective MAC or application level throughputs achieved at a given location using
the best possible WiMAX bearer with the amount of resources allocated to a user by the scheduler.
Uplink Noise Rise: The uplink noise rise is the engineering metric that provides an idea of the uplink interference.
Uplink noise rise is defined as the ratio of the total uplink interference power to the background noise power.
Smart Antenna: Smart antenna systems refers to a system of antenna arrays with smart signal processing algorithms that are used to identify the direction of arrival (DOA) of the signal, and use it to calculate beamforming
vectors, to track and locate the antenna beam on the mobile.
Adaptive Antenna Systems (AAS): AAS is a type of smart antenna systems. Adaptive antenna systems use
more than one antenna elements along with smart signal processing in order to locate and track various types of
AT271_TRG_E6
515
516
signals to dynamically minimize interference and maximize useful signal reception. The signal processor applies
dynamic weighting to each element of the adaptive antenna system hence creating dynamic array patterns in realtime to maximise the output C/(I+N).
Multiple Input Multiple Output (MIMO): Multiple Input Multiple Output systems use more than one transmission
and reception antennas for increasing system capacity by either using space-time transmit diversity or spatial multiplexing.
Spatial Multiplexing (SM): Uses more than one transmission antenna to send different signals (data streams) on
each antenna. The receiver can also have more than one antenna for receiving different signals. Using spatial
multiplexing with M transmission and N reception antennas, the throughput over the transmitter-receiver link can
be theoretically increased M or N times, depending on which is smaller, M or N. SM improves the throughput
(channel capacity) for a given C/(I+N), and is used for the regions of a cell that have sufficient C/(I+N) conditions.
SM is often referred to as MIMO.
Space-Time Transmit Diversity (STTD): Uses more than one transmission antenna to send the same signal on
all antennas. The signals are constructively combined (using optimum selection or maximum ratio combining,
MRC) at the receiver to extract the useful signal. As the receiver gets more than one copy of the useful signal, the
signal level at the receiver after combination of all the copies is much more resistant to interference than a single
signal would be. Therefore, STTD/MRC improves the C/(I+N) at the receiver. It is often used for the regions of a
cell that have bad C/(I+N) conditions. STTD is also known as STC (Space Time Coding) and STBC (Space-Time
Block Codes).
Adaptive MIMO Switch (AMS): AMS is a technique for switching from spatial multiplexing to space-time transmit
diversity as the C/(I+N) conditions get worse than a given threshold. AMS can be used in cells to provide spatial
multiplexing gains to users that have better C/(I+N) conditions than a given AMS threshold, and space-time
transmit diversity to users that have worse C/(I+N) conditions than the threshold. AMS provides the optimum solution using both MIMO features to their best.
AT271_TRG_E6
Forsk 2009
Chapter 10
Repeaters and Remote Antennas
This chapter describes how repeaters and remote antennas are modelled in Atoll. It also provides information
on the differences in modelling for different project types.
Atoll
518
AT271_TRG_E6
Forsk 2009
Chapter 10: Repeaters and Remote Antennas
10
Repeaters and Remote Antennas

A repeater receives, amplifies, and re-transmits the radiated or conducted RF carrier both in downlink and uplink. It has a
donor side and a server side. The donor side receives the signal from a donor transmitter or repeater and the server side
re-transmits the received signal.
The link between the donor transmitter/repeater and the repeater may be chosen between three types:
Over the Air

Fibre link
Microwave link
For the air link, a user-defined or computed propagation loss is involved whereas user-defined link losses must be defined
for the other two cases.
Repeater modelling focuses on the additional coverage they provide to transmitters or repeaters. A repeater is modelled
in the same manner for GSM (GSM GPRS EGPRS) and CDMA (UMTS HSPA, TD-SCDMA, IS-95 cdmaOne, and
CDMA2000) networks, except that in the GSM networks Atoll deals with EIRP while in the CDMA it is concerned with a
total gain. We assume that all the TRXs of GSM donor transmitters and all the carriers of CDMA donor transmitters are
amplified.
Remote antennas enable you to place antennas at locations that would normally require very long feeder cables. A remote
antenna is connected to the base station via optic fibre. The main difference from a repeater is that a remote antenna
generates its own cell whereas a repeater extends the coverage of an existing cell.
10.1
Modelling Repeaters
We assume in this part that the repeater receives a signal from a donor transmitter.
10.1.1
CDMA Documents
10.1.1.1
Over the Air
10.1.1.1.1
Signal Level Received From Repeaters

Atoll calculates the signal level received from a repeater Rpk on a carrier ic as follows:
Rpk
Txd
Rpk
Rpk
P rec ic = P pilot ic + G total Air DL L path M Shadowing L indoor (in dB)
Figure 10.1: CDMA Documents - Over the Air Repeater

Where,
Txd
P pilot ic is the pilot power of the donor transmitter on the carrier ic,
Rpk
G total Air DL is the output downlink total gain of repeater linked to a donor transmitter with an air link. This gain will be
evaluated considering the path loss between the donor transmitter and the donor side of the repeater (see after),
Rpk
L path is the path loss between the repeater Rpk and the receiver,
M Shadowing is the shadowing margin,
Indoor coverage is selected in point analysis or coverage predictions.
The total gain can be either user-specified or directly calculated by Atoll from the link budget.
On each pixel, the resulting received signal on a carrier ic is calculated as follows:
Txd Rpk
P rec
Forsk 2009
Txd
Rpk
ic = P rec ic + P rec ic (not in dB6)
AT271_TRG_E6
519
10.1.1.1.2
Gain Automatic Calculation

Two types of repeater total gains can be calculated: Total Uplink and Downlink gains. The gains are evaluated from the
basis of the donor BTS to the output of the repeater coverage antenna.
Downlink total gain

The Downlink total gain is calculated as follows:
Txd
Rpk
G total Air DL =
Txd
Txd Rpk
G ant L total DL L model

Rpk
G amp
Rpk coverage side

G ant
Rpk donor side
+ G ant
Rpk donor side
L feeder Rec
Rpk coverage side

L feeder Tr
(in dB)
Figure 10.2: Over the Air Repeater - Downlink Total Gain

Where,
Txd
G ant is the gain of the donor transmitter antenna,

Txd
L total DL corresponds to the total downlink losses of the donor transmitter (user-defined or calculated considering
transmitter equipment characteristics - see Transmitter radio equipment part),
Txd Rpk
L model
corresponds to path losses between the donor transmitter and the repeater. They are either user-
defined or calculated using the selected propagation model. If you do not select a propagation model, the propagation losses between the donor transmitter and the repeater are calculated using the ITU 526-5 propagation
model,
Rpk donor side
G ant
Rpk donor side
L feeder Rec
is the gain of the repeaters donor side antenna,

refers to the losses of the repeater donor side due to feeders (see Transmitter radio equipment
part). They involve the cable length at reception on the donor side and the loss per metre,
Rpk
G amp is the amplification gain of the repeater,

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr
is the gain of the repeater coverage side antenna,

corresponds to the losses of the repeater coverage side due to feeders (see Transmitter radio
equipment part). They involve the cable length at transmission on the coverage side and the loss per metre.
Uplink total gain

The Uplink total gain is calculated as follows:
Rpk
G total Air UL =
Txd
Txd
Txd Rpk
G ant L total UL L model

Rpk
G amp +
Rpk coverage side

G ant
Rpk donor side
+ G ant
Rpk donor side
L feeder Tr
Rpk coverage side
(in dB)
L feeder Rec
6.
Formula cannot be directly calculated from components stated in dB and, therefore, must be converted in linear
values.
520
AT271_TRG_E6
Forsk 2009
Figure 10.3: Over the Air Repeater - Uplink Total Gain

Where,
Txd

Txd
L total UL corresponds to the total uplink losses of the donor transmitter (user-defined or calculated considering
transmitter equipment characteristics - see Transmitter radio equipment part),
Txd Rpk
L model
corresponds to path losses between the donor transmitter and the repeater. They are either user-
defined or calculated using the selected propagation model. If you do not select a propagation model, the propagation losses between the donor transmitter and the repeater are calculated using the ITU 526-5 propagation
model,
Rpk donor side
is the gain of the repeaters donor side antenna,
Rpk donor side
refers to the losses of the repeater donor side due to feeders (see Transmitter radio equipment
G ant
L feeder Tr
part). They involve the cable length at transmission on the donor side and the loss per metre,
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Rec

equipment part). They involve the cable length at reception on the coverage side and the loss per metre.
10.1.1.1.3
Donor Side Parameter Automatic Calculation

Azimuth
This is the angle at which the donor antenna is situated with respect to the North at the repeater side. This angle is measured clock-wise as shown in the figure below.
This parameter can be recalculated and applied using the Calculate button on the Donor Side tab. This is the absolute
horizontal angle at which the donor-side antenna of the repeater should be pointed in order to be aligned with the donor
antenna.
Figure 10.4: Angle from North (Azimuth)
Mechanical Downtilt
This is the tilt angle for the repeaters donor-side antenna, which ensures that it points towards the donor antenna in the
vertical plane. As a general rule, downtilt angles are considered positive and uptilt angles negative.
Forsk 2009
AT271_TRG_E6
521
Figure 10.5: Positive/Negative Mechanical Downtilt

Since this parameter depends on the difference of heights/altitudes between the donor transmitter and the repeater, it can
be automatically calculated in the repeaters Donor side properties. If the height/altitude of the antenna is modified, the
corresponding tilt angle can be found out and applied using the Calculate button. The following example may further clarify
the viewpoint:
Figure 10.6: Tilt Angle Computation

The tilt angle in the above figure would be:
Rpk donor side
T ant
Rpk donor side
Txd
H ant
H ant
-
= atan -------------------------------------------------------------Txd Rpk
As obvious, this angle will be negative for uptilts and positive for downtilts of the antenna. Here,
D
Txd Rpk
is the distance between the donor transmitter antenna and the repeater antenna,
Txd
H ant is the height/altitude of the donor transmitter antenna,

Rpk donor side
is the height/altitude of the repeaters donor-side antenna, and
Rpk donor side
is the calculated tilt angle for the repeaters donor-side antenna.
H ant
T ant
In the case of the above figure, the donor-side antenna at the repeater should have a tilt angle of X to point directly at the
transmitter antenna.
10.1.1.2
Microwave Link
10.1.1.2.1

Rpk
Txd
Rpk
Rpk
P rec ic = P pilot ic + G total MW DL L path M Shadowing L indoor (in dB)
Figure 10.7: CDMA Documents - Microwave Link Repeater

Where,
Txd
Rpk
G total MW DL is the output downlink total gain of repeater linked to a donor transmitter with an microwave link. This gain
will be evaluated considering the link loss between the donor transmitter and the donor side of the repeater (see after),
522
AT271_TRG_E6
Forsk 2009

Rpk
Txd Rpk
P rec
10.1.1.2.2
Txd
Rpk

In that case, the total UL and DL gains are evaluated without considering the donor antenna parameters (gain and total
TX losses) and the repeater donor-side antenna parameters (gain and feeder losses).
Downlink total gain

Rpk
Txd Rpk
G total MW DL = L MW
Rpk
Rpk coverage side
+ G amp + G ant
Rpk coverage side
L feeder Tr
(in dB)
Figure 10.8: Microwave Link Repeater - Downlink Total Gain

Where,
Txd Rpk
L MW
is the user-defined microwave link loss between the donor transmitter and the repeater,
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr

Uplink total gain

Rpk
Txd Rpk
G total MW UL = L MW
Rpk
Rpk coverage side
+ G amp + G ant
Rpk coverage side
L feeder Rec
(in dB)
Figure 10.9: Microwave Link Repeater - Uplink Total Gain

Where,
Txd Rpk
L MW
Forsk 2009
AT271_TRG_E6
523

Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Rec

10.1.1.3
Fibre Link
10.1.1.3.1

Rpk
Txd
Rpk
Rpk
P rec ic = P pilot ic + G total Fibre DL L path M Shadowing L indoor (in dB)
Figure 10.10: CDMA Documents - Fibre Link Repeater

Where,
Txd
Rpk
G total Fibre DL is the output downlink total gain of repeater linked to a donor transmitter with an optical fibre link. This
gain will be evaluated considering the fibre loss between the donor transmitter and the donor side of the repeater (see
after),
Rpk
L path is the link loss between the repeater Rpk and the receiver,
Txd Rpk
P rec
10.1.1.3.2
Txd
Rpk

In that case, the total UL and DL gains are evaluated without considering the donor antenna parameters (gain and total
TX losses) and the repeater donor-side antenna parameters (gain and feeder losses).
Downlink total gain

Rpk
Txd Rpk
G total Fibre DL = L Fibre
524
Rpk
Rpk coverage side
+ G amp + G ant
AT271_TRG_E6
Rpk coverage side
L feeder Tr
(in dB)
Forsk 2009
Figure 10.11: Fibre Link Repeater - Downlink Total Gain

Where,
Txd Rpk
L Fibre
is the user-defined fibre link loss between the donor transmitter and the repeater,
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr

Uplink total gain

Rpk
Txd Rpk
G total Fibre UL = L Fibre
Rpk
Rpk coverage side
+ G amp + G ant
Rpk coverage side
L feeder Rec
(in dB)
Figure 10.12: Fibre Link Repeater - Uplink Total Gain

Where,
Txd Rpk
L Fibre
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Rec

10.1.1.4
Appendices
10.1.1.4.1
Automatic Controls
Atoll performs systematic controls when creating a new repeater:
On the amplification gain,

On the downlink power.
Amplification Gain
The amplification gain of the repeater must be consistent with the limits of its equipment. Atoll checks that:
min
Rpk
max
G amp G amp G amp

Where
Forsk 2009
AT271_TRG_E6
525

Rpk
G amp is the amplification gain of the repeater.

min
G amp is the minimum amplification gain allowed by the equipment.

max
G amp is the maximum amplification gain allowed by the equipment.
Downlink Power
The downlink power after amplification must be consistent with the limit of the equipment installed on the repeater. Atoll
checks that:
Txd
Rpk
P pilot ic + G total DL P
max
Rpk coverage side
+ G ant
Rpk coverage side
L feeder Tr
Where
Txd
P pilot ic is the pilot power of the donor transmitter on the carrier ic. When there are several carriers on a transmitter, Atoll
takes the highest pilot power on all the carriers.
Rpk coverage side
G ant
P
max
is the gain of the repeater coverage side antenna
is the maximum downlink power allowed by the equipment.
Rpk coverage side
L feeder Tr
corresponds to the transmission losses due to feeders installed on the repeater coverage side (see
Transmitter radio equipment part). They involve the cable length at transmission on the coverage side and the loss per
metre.
10.1.1.4.2
Carrier Power and Interference Calculation

This section explains how Atoll calculates the received carrier power and interference when a transmitter has a connected
repeater.
A mobile receiver receives signal from the donor transmitter as well as its repeater. Similarly, the signal from the mobile
is received at the donor transmitter as well as its repeater. In practice, when a mobile receiver is in the vicinity of the donor
transmitter, the signal to/from the repeater would be very weak due to high pathloss between the repeater and the mobile
receiver. Similarly, when the mobile receiver is located in the vicinity of the repeater, the signal to/from the donor transmitter would be very weak due to the same reason.
Atoll does not differentiate between the mobile receiver being in the transmitter coverage area or being in its repeater
coverage area. Atoll adds the signals received from the donor transmitter and its repeater to generate a combined pathloss
matrix that is associated with the donor transmitter and includes the effect of its repeater.
Calculation of Total Path Loss

The total pathloss, L total , is calculated by computing a downlink budget. If we take the case of a CDMA project, without
considering any shadowing margin or indoor loss, the power received from the donor transmitter, Txd on a carrier ic, at
the mobile receiver can be stated as (for a link over the air):
Txd
Txd
P pilot ic G ant
Txd
P rec ic = ----------------------------------------------------Txd
Txd Rx
L total DL L path
Where,
Txd
P rec ic is the carrier power received at the receiver from the donor transmitter on a carrier ic (in W)
Txd
P pilot ic is the pilot power of the donor transmitter on the carrier ic (in W)
Txd
G ant is the donor transmitter antenna gain.

Txd
L total DL is the transmission feeder loss of the donor transmitter.

Txd Rx
L path
is the path loss between the donor transmitter and the mobile receiver.
Similarly, the power received at the mobile receiver from the repeater, Rpk, is:
Txd
Rpk
P pilot ic G total Air DL

Rpk
P rec ic = -------------------------------------------------------------------Rpk Rx
L path
Where,
Rpk
P rec ic is the carrier power received at the mobile receiver from the repeater on a carrier ic (in W)
Txd
P pilot ic is the pilot power of the donor transmitter on the carrier ic (in W)
526
AT271_TRG_E6
Forsk 2009

Rpk
G total Air DL is the output downlink total gain of repeater linked to a donor transmitter with an air link.
Rpk Rx
is the path loss between the repeater and the mobile receiver
L path
So, the total carrier power received at the mobile receiver is:
Txd Rpk
P rec
Txd
Rpk
G ant
G total Air DL
Txd
Rpk
Txd
ic = P rec ic + P rec ic = P pilot ic ------------------------------------------------------
- + ----------------------------------Txd Rx
Rpk Rx
L Txd
L path
total DL
path
Since,
Txd
Txd
P pilot ic G ant
L total = --------------------------------------------------------------Txd Rpk
Txd
L total DL P rec
ic
Therefore,
Txd
Txd
P pilot ic G ant
L total = ---------------------------------------------------------------------------------------------------------------------------------------------------------------Txd
Rpk
G total Air DL
G ant
Txd
Txd
L total DL P pilot ic ------------------------------------------------------- + ------------------------------------
Txd Rx
Rpk Rx
L Txd
L path
total DL L path
Hence,
Txd
G ant
L total = ----------------------------------------------------------------------------------------------------------------------------------Txd
Rpk
G ant
G total Air DL
Txd
- + ----------------------------------L total DL ------------------------------------------------------
Txd Rx
Rpk Rx
L Txd
L path
total DL L path
This total path loss depends on the location of the mobile receiver in realistic network scenarios. As a mobile in the donor
transmitter/repeater coverage area is likely to be far from the repeater/donor transmitter coverage area, the respective
pathloss value will be very large. This implies that we can study the two cases separately without influencing the results
much.
Case 1: Receiver in Donor Transmitter Coverage Area
Rpk Rx
L path
Rpk
G total Air DL
- can be ignored. This implies that:
is likely to be very high, so the term ----------------------------------Rpk Rx
L path
Txd Rx
L total = L path
Considering this total pathloss value, the total received power in the uplink and in the downlink can be stated as:
Txd
Txd
Txd
Txd
P pilot ic G ant
P pilot ic G ant
Txd
= -----------------------------------------------------P rec DL ic = ---------------------------------------------Txd
Txd
Txd Rx
L total DL L total
L total DL L path
Rx
Txd
Rx
Txd
P output ic G ant
P output ic G ant
Txd
P rec UL ic = -------------------------------------------------- = -----------------------------------------------------Txd
Txd
Txd Rx
L total UL L total
L total UL L path
Where,
Rx
P output ic is the transmitted power from the mobile terminal on the carrier ic (in W)
Txd
L total UL is the reception feeder loss of the transmitter
Case 2: Receiver in Repeater Coverage Area
Txd Rx
L path
Txd
G ant
- can be ignored. This implies that:
is likely to be very high, so the term -----------------------------------------------------Txd
Txd Rx
L total DL L path
Txd
Txd
G ant
G ant
= ---------------------------------------------------------------L total = --------------------------------------------------------------------------Rpk
Rpk
Txd
G
G
Txd
total Air DL
total Air DL L total DL
-
-------------------------------------------------------------- L total DL ---------------------------------- Rx
Rpk Rx
L Rpk
L path
path
Txd
Txd
Txd
Rpk
P pilot ic G ant
P pilot ic G total
Txd
P rec DL ic = ---------------------------------------------= ----------------------------------------------Txd
Rpk Rx
L total DL L total
L path
Rx
Rpk
Txd
Rx
Txd
P output ic G total L total DL
P output ic G ant
Txd
- ------------------------ = --------------------------------------------------P rec UL = -------------------------------------------------Rpk Rx
Txd
Txd
L path
L total UL
L total UL L total
Where,
Forsk 2009
AT271_TRG_E6
527

Rx
P output ic is the transmitted power from the mobile terminal (in W)

Txd
L total UL is the reception feeder loss of the transmitter
Calculation of Eb/Nt Uplink

In the uplink, the quality level at the transmitter on a traffic channel is:
C W
E
-----b-
= ---- ---- N t UL
I R
Where,
C is the carrier power received from the mobile terminal (in W)
I is the total interference (in W)
W is the spreading bandwidth (Hz)
R is the effective service data rate in the uplink (bits/s)
(W/R is the service processing gain in the uplink)
C and I are both evaluated at the same reference point, which is the entry of BTS using the following formulas.
Rx
Txd
P output G ant
C = P total UL = ----------------------------------------Txd
L total UL L total
I = I total + N 0
Where,
I total is the sum of the signals received from mobile terminals inside the same cell and those outside (in W)
N 0 is the transmitter equipment thermal noise (in W)
Therefore, for each mobile terminal Rxi,
I total =
Rxi
Rxi
Txd
P output G ant
-
----------------------------------------Rxi
L Txd
total UL L total
And,
N 0 = NF
Txd
KTW
Where,
NF
Txd
is the noise figure of the transmitter equipment at the reference point, i.e. the entry of the BTS
K is Boltzman constant
T is the ambient temperature (in K)
Hence
N 0 = NF
10.1.1.4.3
BTS
KTW
Consideration of Repeater Noise Figure

You can define and assign a repeater equipment to each repeater. In addition to the allowed ranges of gains and powers
allowed to each repeater, these equipment contain a noise figure which is applied to the repeater they are assigned to.
This noise figure has an impact on the donor transmitter total reception losses. For information, see "Transmitter Radio
Equipment" on page 124.
10.1.2
GSM Documents
10.1.2.1
Over the Air
10.1.2.1.1

Atoll evaluates the signal level received from a repeater Rpk on a TRX type tt as follows:
Rpk
Rpk
Rpk
P rec tt = EIRP Air tt P tt L path M Shadowing L indoor + G ant
528
AT271_TRG_E6
Rx
L Rx
(in dB)
Forsk 2009
Figure 10.13: GSM Documents - Over the Air Repeater

Where,
Rpk
EIRP Air tt is the effective isotropic radiated power of the repeater on the TRX type tt linked to a donor transmitter over
the air. This value will be evaluated considering the path loss between the donor transmitter and the donor side of the
repeater (see after),
P tt is the power offset defined for the selected TRX type,
L Rx is the receiver loss,
G ant
Rx
Rpk
Total gain and EIRP can be either user-specified or directly calculated by Atoll from the link budget.
On each pixel, the resulting received signal on a TRX type tt is calculated as follows:
Txd Rpk
P rec
10.1.2.1.2
Txd
Rpk
tt = P rec tt + P rec tt (not in dB6)
EIRP Automatic Calculation

In GSM projects, an EIRP is calculated and repeated by the repeater. This EIRP is evaluated from the basis of the donor
BTS to the output of the repeater coverage antenna. In this technology, the UL part of the repeater is not modelled.
EIRP is calculated as follows:
Rpk
EIRP Air =
Txd
Rpk
Txd
Txd
Txd Rpk
+ G ant L total DL L model
G amp +
Rpk coverage side

G ant
Rpk donor side
+ G ant
Rpk donor side
L feeder Rec
Rpk coverage side
(in dB)
L feeder Tr
Figure 10.14: Over the Air Repeater - EIRP

Where,
P
Txd
is the power of the donor transmitter,
Txd
Forsk 2009
AT271_TRG_E6
529

Txd
L total DL corresponds to the total downlink losses of the donor transmitter (user-defined or calculated considering transmitter equipment characteristics),
Txd Rpk
L model
corresponds to path losses between the donor transmitter and the repeater. They are either user-defined or
calculated using the selected propagation model. If you do not select a propagation model, the propagation losses
between the donor transmitter and the repeater are calculated using the ITU 526-5 propagation model,
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr

corresponds to the losses of the repeater coverage side due to feeders (see Transmitter radio equip-
ment part). They involve the cable length at transmission on the coverage side and the loss per metre.
10.1.2.1.3
Donor Side Parameter Automatic Calculation

Azimuth
This is the angle at which the donor antenna is situated with respect to the North at the repeater side. This angle is measured clock-wise as shown in the figure below.
This parameter can be recalculated and applied using the Calculate button on the Donor Side tab. This is the absolute
horizontal angle at which the donor-side antenna of the repeater should be pointed in order to be aligned with the donor
antenna.
Figure 10.15: Angle from North (Azimuth)
Mechanical Downtilt
This is the tilt angle for the repeaters donor-side antenna, which ensures that it points towards the donor antenna in the
vertical plane. As a general rule, downtilt angles are considered positive and uptilt angles negative.
Figure 10.16: Positive/Negative Mechanical Downtilt

Since this parameter depends on the difference of heights/altitudes between the donor transmitter and the repeater, it can
be automatically calculated in the repeaters Donor side properties. If the height/altitude of the antenna is modified, the
corresponding tilt angle can be found out and applied using the Calculate button. The following example may further clarify
the viewpoint:
Figure 10.17: Tilt Angle Computation

The tilt angle in the above figure would be:
Rpk donor side
T ant
530
Rpk donor side
Txd
H ant
H ant
-
= atan -------------------------------------------------------------Txd Rpk
AT271_TRG_E6
Forsk 2009

As obvious, this angle will be negative for uptilts and positive for downtilts of the antenna. Here,
D
Txd Rpk
is the distance between the donor transmitter antenna and the repeater antenna,
Txd
H ant is the height/altitude of the donor transmitter antenna,

Rpk donor side
is the height/altitude of the repeaters donor-side antenna, and
Rpk donor side
is the calculated tilt angle for the repeaters donor-side antenna.
H ant
T ant
In the case of the above figure, the donor-side antenna at the repeater should have a tilt angle of X to point directly at the
transmitter antenna.
10.1.2.2
Microwave Link
10.1.2.2.1

Rpk
Rpk
Rpk
P rec tt = EIRP MW tt P tt L path M Shadowing L indoor + G ant
Rx
L Rx (in dB)
Figure 10.18: GSM Documents - Microwave Link Repeater

Where,
Rpk
EIRP MW tt is the effective isotropic radiated power of the repeater on the TRX type tt linked to a donor transmitter
through a microwave link. This value will be evaluated considering the link loss between the donor transmitter and the
donor side of the repeater (see after),
G ant
Rx
Rpk
Txd Rpk
P rec
10.1.2.2.2
Txd
Rpk

Rpk
EIRP MW = P
Txd
Txd Rpk
L MW
Rpk
Rpk coverage side
+ G amp + G ant
Rpk coverage side
L feeder Tr
(in dB)
Where,
P
Txd
Txd Rpk
L MW
Forsk 2009
AT271_TRG_E6
531

Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr

Figure 10.19: Microwave Link Repeater - EIRP
10.1.2.3
Fibre Link
10.1.2.3.1

Rpk
Rpk
Rpk
P rec tt = EIRP Fibre tt P tt L path M Shadowing L indoor + G ant
Rx
L Rx (in dB)
Where,
Rpk
EIRP Fibre tt is the effective isotropic radiated power of the repeater on the TRX type tt linked to a donor transmitter
through an optical fibre link. This value will be evaluated considering the fibre loss between the donor transmitter and the
donor side of the repeater (see after),
G ant
Rx
Rpk
Figure 10.20: GSM Documents - Fibre Link Repeater

Txd Rpk
P rec
10.1.2.3.2
Txd
Rpk

532
AT271_TRG_E6
Forsk 2009

Rpk
EIRP Fibre = P
Txd
Txd Rpk
L Fibre
Rpk
Rpk coverage side
+ G amp + G ant
Rpk coverage side
L feeder Tr
(in dB)
Figure 10.21: Fibre Link Repeater - EIRP

Where,
P
Txd
Txd Rpk
L Fibre
Rpk

Rpk coverage side
G ant
Rpk coverage side
L feeder Tr
10.1.2.4
Appendices
10.1.2.4.1
Automatic Controls
Atoll performs systematic controls when creating a new repeater:
On the amplification gain,

On the downlink power.
Amplification Gain
The amplification gain of the repeater must be consistent with the limits of its equipment. Atoll checks that:
Rpk
min
max
G amp G amp G amp

Where
Rpk
G amp is the amplification gain of the repeater.

min
G amp is the minimum amplification gain allowed by the equipment.

max
G amp is the maximum amplification gain allowed by the equipment.
Downlink Power
The downlink power after amplification must be consistent with the limit of the equipment installed on the repeater. Atoll
checks that:
EIRP
Rpk
tt P
max
Rpk coverage side
+ G ant
Rpk coverage side
L feeder Tr
Where
EIRP
Rpk
tt is the effective isotropic radiated power of the repeater on the TRX type tt.
Rpk coverage side
G ant
P
max
is the gain of the repeater coverage side antenna.
is the maximum downlink power allowed by the equipment.
Rpk coverage side
L feeder Tr
corresponds to the transmission losses due to feeders installed on the repeater coverage side (see
Transmitter radio equipment part). They involve the cable length at transmission on the coverage side and the loss per
metre.
Forsk 2009
AT271_TRG_E6
533
10.2
Modelling Remote Antennas

Atoll models remote antennas in a similar way to repeaters. A remote antenna is connected to the base station via an optic
fibre. The main difference from a repeater is that a remote antenna covers its own cell whereas a repeater extends the
coverage of an existing cell. In the case of a remote antenna, the donor transmitter does not have any main antenna. In
addition, it is assumed that a remote antenna, as opposed to a repeater, does not have any equipment and generates no
amplification gain nor noise.
10.2.1
CDMA Documents
10.2.1.1

Atoll calculates the signal level received from a remote antenna Ra on a carrier ic as follows:
Ra
Txd
Ra
Ra
P rec ic = P pilot ic + G total Fibre DL L path M Shadowing L indoor (in dB)
Figure 10.22: CDMA Documents - Remote Antenna Signal Level

Where,
Txd
Ra
G total Fibre DL is the output downlink total gain of the remote antenna of a donor transmitter with an optical fibre link.
This gain will be evaluated considering the fibre loss between the donor transmitter and the donor side of the repeater (see
after),
Ra
L path is the link loss between the repeater Ra and the receiver,
Since the donor transmitter antenna is not defined, the resulting received signal on a carrier ic is calculated as follows:
Txd Ra
P rec
10.2.1.2
Ra
ic = P rec ic (not in dB6)

The downlink and uplink total gain are calculated as follows:
Ra
Downlink total gain

Txd Ra
G total Fibre DL = L Fibre
Ra coverage side
+ G ant
Ra coverage side
L feeder Tr
(in dB)
Figure 10.23: Remote Antennas - Downlink Total Gain

Where,
534
AT271_TRG_E6
Forsk 2009

Txd Ra
L Fibre
is the user-defined fibre link loss between the donor transmitter and the remote antenna,
Ra coverage side
G ant
Ra coverage side
L feeder Tr
is the gain of the remote antenna coverage side,

corresponds to the losses of the remote antenna coverage side due to feeders (see Transmitter radio
Uplink total gain
Ra
Txd Ra
G total Fibre UL = L Fibre
Ra coverage side
+ G ant
Ra coverage side
L feeder Rec
(in dB)
Figure 10.24: Remote Antennas - Uplink Total Gain

Where,
Txd Ra
L Fibre
Ra coverage side
G ant
Ra coverage side
L feeder Rec

10.2.2
GSM Documents
10.2.2.1

Atoll calculates the signal level received from a remote antenna Ra on a TRX type tt as follows:
Ra
Ra
Ra
P rec tt = EIRP Fibre tt P tt L path M Shadowing L indoor + G ant
Rx
L Rx (in dB)
Figure 10.25: GSM Documents - Remote Antenna Signal Level

Where,
Ra
EIRP Fibre tt is the effective isotropic radiated power of the remote antenna on the TRX type tt linked to a donor transmitter through an optical fibre link. This value will be evaluated considering the fibre loss between the donor transmitter
and the donor side of the remote antenna (see after),
G ant
Rx
Rpk
L path is the path loss between the remote antenna Ra and the receiver,
Forsk 2009
AT271_TRG_E6
535

Since the donor transmitter antenna is not defined, the resulting received signal on a TRX type tt is calculated as follows:
Txd Ra
P rec
10.2.2.2
Ra
tt = P rec tt (not in dB6)

Ra
EIRP Fibre = P
Txd
Txd
Txd Ra
+ L total DL L Fibre
Ra coverage side
+ G ant
Ra coverage side
L feeder Tr
(in dB)
Where,
P
Txd
Txd Ra
L Fibre
Ra coverage side
G ant
Ra coverage side
L feeder Tr

Figure 10.26: Remote Antennas - EIRP
536
AT271_TRG_E6
Forsk 2009
Index
Index
Symbols
Inter-carrier neighbours 255, 347
.agd format 36, 38
.bil file 43, 44
.bil format 35, 36, 37, 43
.bmp file 46
.bmp format 35, 36, 37, 46
.bmp raster data encoding 48
.bmw file 37, 46
.bpw file 37, 49
.bpw sample 49
.clc file 66
.clc format 65
.clc sample 66
.cs file 27
.dbf file 58, 60, 61, 63
.dbf format 50, 58, 61
.dbf header 58, 62
.dct file 67
.dct sample 67
.dxf format 37, 50
.ecw format 37, 51
.grc format 51
.grd format 51
.hdr file 43
.im0 format 64
.im0 Sample 65
.im1 format 68
.im1 sample 68
.im2 format 69
.im2 sample 69
.los file 60, 61
.mid format 37
.mif format 37, 50
.mnu format 56
.mnu sample 56
.pts file 63, 64
.shp format 37, 50
.shx format 50
.tab format 50
.tfw file 45
.tfw sample 46
.tif format 35, 36, 37, 45
.wld file 49
.wld sample 49
Numerics
3-D interpolation of horizontal and vertical patterns 109
8PSK modulation 166, 168, 169, 170, 171, 172
A
AAS 515
AAS usage 435, 458
Absolute Spot Ht 87
Acknowledgement gain 286
Acl calculation 96
Active set 192, 199, 234, 258, 321, 335, 339, 342, 343, 350
Active set determination 204
Active set management 192, 289
Activity factor 163
Activity factor for voice services 435
Activity status 193, 194, 226, 291
Adaptive Antenna Systems 515
Forsk 2009
Adaptive antenna systems 408, 505

Adaptive MIMO Switch 498, 516
Adaptive Modulation and Coding 209
Additional Ec/Io conditions 259, 351
Additional electrical downtilt 111
Adjacence criterion 157, 254, 347
Adjacency factor 159, 511
Adjacent channel suppression factor 434
Adjacent PN-Clusters per site allocation 352
Adjacent PN-Clusters Per Site Strategy 357
Adjacent transmitter 159, 257, 349
Admission Control 217
Admission control 202, 299, 304, 311, 316
Admission control in the R99 part 226
A-DPCH Eb/Nt DL 248
A-DPCH Eb/Nt UL 248
AEDT 111
AHRxeff calculation 96
Algorithm based on coverage overlapping 268, 359, 427
Algorithm based on distance 268, 358, 427
Algorithm initialization 200, 298, 303, 310
All servers 161, 166, 176, 184, 274, 447
Allocate Carriers Identically 266
Allocate carriers identically 259, 352
Allocation examples 264
Allocation for a group of cells 258, 350
Allocation for a group of transmitters 160
Allocation strategies 260, 264, 266, 352, 418
Allocation strategy 267, 426
Altitude 77, 81
Ambient temperature 187, 277, 284, 368, 435, 528
AMS 497, 498, 516
AMS threshold 497
Analysis based on a specific carrier 242, 338
Analysis based on all carriers 338
Analysis on a specific carrier 247, 346
Analysis on all carriers 247, 346
Angular distribution of the uplink noise rise 508
Antenna 38
Antenna attenuation 73, 110, 131, 183, 273, 372, 373, 384, 475
Antenna attenuation calculation 109
Antenna coordinates 76
Antenna diversity gain 124, 126
Antenna height 76
Antenna pattern 76
Antenna pattern 3-D interpolation 110
Antenna pattern smoothing 111
Application level channel throughput 498
Application level user throughput 505
Application throughput 250, 460, 515
Application throughput/timeslot study display type 173
ArcView Grid file description 52
ArcView Grid file sample 52
ArcView Grid format 52
AS 192, 289
AS analysis tab 232, 319
Atoll.ini file 45, 111, 123, 127, 193, 201, 290, 377
Automatic allocation description 259, 267, 351, 358, 361, 418, 426
Automatic cdmaOne-CDMA2000 neighbour allocation 360
Automatic GSM/TD-SCDMA neighbour allocation 426
Automatic GSM-CDMA neighbour allocation 357
Automatic GSM-UMTS neighbour allocation 267
Automatic neighbour allocation 160, 254, 346
Automatic repeater calculation 520, 523, 524, 529, 531, 532, 534, 536
Automatic SPM Calibration 91
Average application throughput per mobile 251
AT271_TRG_E6
537

Average array correlation matrix 409, 507
Average call duration 291
Average MAC throughput per mobile 251
Average noise level 247
Average noise rise 247, 346
Average packet switched traffic demand 152
Average RLC throughput 250
Average RLC throughput per mobile 251
Average throughput/timeslot 173
Azimuth 76, 109, 127, 268, 358, 361, 427, 451
B
Bar graph and pilot sub-menu 232, 320
Base band hopping 162
BBH 162
BCCH 39
Beam forming 408, 505
Bearer 211, 434, 451
Bearer efficiency 434, 497
Bearer selection threshold 435
BER 174, 179
Best bearer 460, 461, 462
Best cell edge coverage probability (%) 186, 276, 449
Best coding schemes 173
Best Effort 501
Best Idle Mode Reselection Criterion (C2) 134
Best server 193, 289
Best server path loss (dB) 185, 275, 449
Best server total losses (dB) 185, 276, 449
Best signal level 273
Best signal level (dBm) 374
Best signal level (in dBm, dBV, dBV/m) 185, 275, 448
Best signal level and a margin 132, 184, 274, 447
Best signal level of the highest priority HCS layer and a margin 161
Best signal level of the highest priority layer and a margin 167, 176
Best signal level per HCS layer and a margin 161, 166, 176
Best throughput/timeslot 173
Best traffic signal level 450
Biased (QoS Class) scheduler 502
BIL samples 44
Bilinear interpolation 77, 80, 96, 275
Bin 74
Bin size 113
Bin vertices 35
Binary tree 317
Bit Error Rate 174
Bit power received at terminal on carrier ic 190
Bit power received at transmitter on carrier ic used by terminal 190
Bit received power at terminal for FCH on carrier ic 279
Bit received power at terminal for FCH+SCH on carrier ic 279
Bit received power at terminal for SCH on carrier ic 279
Bit received power at transmitter for FCH on carrier ic 280
Bit received power at transmitter for SCH on carrier ic 280
Bit received power at transmitter for SCH+FCH on carrier ic 280
Bit received power at transmitter on carrier ic 286
BLER 451, 497, 499, 504
BLER percentage 173
Block error rate 497, 499, 504
Block error rate computation 173
Blocking Probability 152
Blocking probability 147, 149, 150, 154, 156
Body loss 189, 279, 286, 367, 436
Boltzman constant 187, 277, 284, 368, 435, 528
Break distance 86
Broadcast 82
BSIC format 31
BTS 124
BTS Noise Factor 366
BTS noise figure 124, 125, 126
Buddy algorithm 227, 317
538
C
C/(I+N) 451, 461, 462
C/(I+N) level 170, 171, 460
C/I level 165, 167, 169, 170, 341
C/I ratio 160, 310
C/I threshold 66, 68
C/N 170, 171
C/N level 450
Calculate 76
Calculate / Force Calculation comparison 76
Calculate All 76
Calculate or Force Calculation? 77
Calculation area 75
Calculation area determination 75
Calculation area management 77
Calculation areas 76
Calculation bin 74, 83, 86
Calculation criteria 131, 183, 273, 373
Calculation of azimuth and tilt angles 109
Calculation of Eb/Nt uplink 528
Calculation of inter-transmitter distance 270, 429
Calculation of the required power for DL traffic channel 300, 307
Calculation of the service usage duration per hour 194, 291, 378, 452
Calculation of total path loss 526
Calculation options 167, 176
Calculation radius 75
Calculations based on C 167
Calculations based on C/(I+N) 177
Calculations based on C/I considering thermal noise 168
Calculations Based on C/I Without Considering Thermal Noise 167
Calculations based on C/N 177
Candidate neighbours 158, 257, 349, 358, 360, 361
Carrier power 147, 153, 162, 166
Carrier power received at the mobile receiver from the repeater 526
Carrier power received from the mobile terminal 528
Carrier selection mode 201, 202, 203, 299, 304, 311
carrier selection mode 201, 202, 299, 304, 311
Carrier to interference and noise ratio 168, 177
Carrier to interference ratio 162, 167
Carrier to interference ratio calculation 161
Carrier-to-interference distribution 147
Carrier-to-interference ratio 153
CDMA2000 82
CDMA2000 1xEV-DO 327, 335, 340
CDMA2000 1xEV-DO power/rate control simulation algorithm 310
CDMA2000 1xEV-DO services 292, 296
CDMA2000 1xEV-DO users 320
CDMA2000 1xRTT 324, 331
CDMA2000 1xRTT power control simulation algorithm 302
Cell 39
Cell edge coverage probability 73, 77, 114, 115, 116, 117, 158, 233, 242, 243,
244, 255, 321, 339, 340, 342, 344, 347, 359
Cell edge coverage probability (%) 186, 276, 374, 376, 449
Cell FCH power for a traffic channel on carrier ic 278
Cell other common channel power (except CPICH and SCH) 278
Cell pilot power 188, 273, 278
Cell priority 261, 353
Cell size 82
Cell synchro channel power 188, 189
Cell synchronisation channel power 278
Cell type 39
Central meridian 53
Channel bandwidth 434
Channel element management 317
Channel elements management 227
Channel throughput 515
Chip power received at terminal 190, 279
Chip rate 187, 277, 285
Chip received power at transmitter 280
Circuit quality indicators 174
Circuit switched service 194, 197, 378, 381
Circuit switched traffic 155
Circuit switched traffic demand 150, 151, 155
Circuit switched traffic overflow rate 155
Cirquit quality indicators studies 174
AT271_TRG_E6
Forsk 2009
Index
Cirquit quality indicators study display types 179
Clearance angle 96
Clustered allocation 260, 418, 419, 424
Clustered strategy 265
Clutter 36, 76, 80, 83, 113
Clutter class 78, 83
Clutter classes 36, 43, 45, 46, 113
Clutter classes file 44, 46, 49, 53
Clutter classes file sample 54
Clutter determination 78
Clutter height 78, 80, 90
Clutter height file 78
Clutter heights 36, 43, 45
Clutter maps 36
Cn calculation 95
Coding rate 434
Coding scheme based on C with ILA 169
Coding Scheme Based on C Without ILA 168
Coding scheme based on C/I with ILA and thermal noise 171
Coding scheme based on C/I with ILA and without thermal noise 170
Coding scheme based on C/I without ILA and thermal noise 169
Coding scheme based on C/I without ILA and with thermal noise 170
Coding schemes 172
Collision 164
Collision probability for BBH and SFH modes 164
Collision probability for non hopping mode 164
Colour by application throughput 461, 462
Colour by best bearer 461, 462
Colour by C/(I+N) level 461
Colour by effective MAC throughput 461, 462
Colour by peak MAC throughput 461, 462
Colour by preamble signal level 450
Colour by total noise 462
Colour by traffic C/N level 451
Colour by traffic signal level 450
Colour by uplink C/N level 451
Colour by uplink signal level 451
Colour per Application Throughput 253
Colour per application throughput 250
Colour per application throughput per mobile 251
Colour per average noise level 247, 346
Colour per average noise rise 247, 346
Colour per average RLC throughput 250
Colour per C/I 341
Colour per Cell Edge Coverage Probability 252
Colour per cell edge coverage probability 243, 244, 339, 340, 342, 344
Colour per CQI 249
Colour per data rate 340, 343, 344
Colour per effective quality level (Effective Eb/Nt) 244, 246, 340, 342, 344
Colour per HS-PDSCH Ec/Nt 249
Colour per HS-SCCH Ec/Nt 249
Colour per HS-SCCH power 249
Colour per MAC Rate 252
Colour per MAC rate 249
Colour per MAC throughput 249
Colour per MAC throughput per mobile 251
Colour per max A-DPCH Eb/Nt DL 248
Colour per max A-DPCH Eb/Nt UL 248
Colour per maximum noise level 247, 346
Colour per maximum noise rise 247, 346
Colour per maximum quality level (Max Eb/Nt) 246, 342, 344
Colour per maximum quality level (max Eb/Nt) 244, 340
Colour per minimum noise level 247, 346
Colour per minimum noise rise 247, 346
Colour per Minimum RLC Throughput 253
Colour per mobility 242, 243, 246, 339, 342, 343
Colour per noise level 247, 346
Colour per noise rise 248, 346
Colour per probability 242, 244, 246, 339, 340, 342, 343
Colour per quality level (Ec/Io) 243, 339
Colour per quality margin (Eb/Nt margin) 244, 246, 340, 342, 344
Colour per quality margin (Ec/Io margin) 243, 339
Colour per rate 341
Colour per Required E-DPDCH Ec/Nt 252
Colour per required power 245, 246, 340, 343, 344
Forsk 2009
Colour per required power margin 245, 246, 340, 343, 344
Colour per Required Terminal Power 252
Colour per RLC Peak Rate 253
Colour per RLC peak rate 250
Colour per RLC peak throughput 250
Colour per RLC peak throughput per mobile 251
Colour per service 244, 246, 339, 342, 343
Colour per transmitter 242, 243, 245, 338, 339, 342, 343
Colour per UL soft handover gain 246
Common channels power (except CPICH and SCH) 188
Complex smart antenna weight 371, 445
Complex weights 410, 507
Compressed mode 205, 206, 235, 236, 239
Computation zone 75, 157, 254, 259, 351, 418, 509, 512
Connector transmission loss 125, 126, 127
Control of radio resource limits (MAC indexes and site channel elements) 314
Control of radio resource limits (OVSF codes, cell power, channel elements)
206
Control of radio resource limits (Walsh codes, cell power and site channel elements) 301, 308
Convergence 220, 302, 310, 316
Convergence criteria 220
Convergence criterion 302, 309, 316
Coordinate system 127
Coordinate system file formats 27
Coordinate systems 25
Coordinate systems in Atoll 26
Corrected standard loss 95
Correction for hilly regions in case of LOS 89
Corrections in the Hata formula 83
Correlated MSAs m and n 164
Correlation coefficient 122
Correlation coefficient determination 120, 123
Co-site cell 268, 358, 360, 361, 427
Co-site factor 159, 511
Co-site transmitter 159, 257, 349
Cost-Hata 83
Cost-Hata model 80
Coverage area determination 165, 166, 175, 447
Coverage area display 165
Coverage by best bearer 459, 461, 462
Coverage by C/(I+N) level 459, 460, 461
Coverage by C/I level study 160
Coverage by channel throughput 459, 461, 462
Coverage display 172, 179, 185, 275, 448, 450, 460
Coverage plot reports 37
Coverage resolution 448, 450, 460
Coverage studies 131, 160, 184, 241, 274, 338
Coverage study conditions 77
Coverage study display options 77
Coverage study resolution 77
Coverage Study Scenarios 168
CPICH quality 209
CQI 209, 238, 249
CQI based on CPICH quality 209
CQI based on HS-PDSCH quality 213
CQI study based on C/(I+N) with ILA 179
CQI study based on C/(I+N) without ILA 178
CQI study based on C/N with ILA 178
CQI study based on C/N without ILA 178
Cyclic prefix ratio 433, 514
D
Data Erlangs 147
Data rate 340, 343
Data Rate Control gain 286
Data service users 291, 294
Data services 453, 455
Datum 25
Datum codes 28
DC subcarrier 434
Dedicated packet switched timeslots 151
Delay 149, 150, 152, 154, 156
Delete existing neighbours 158, 160, 255, 268, 347, 358, 359, 361, 427
AT271_TRG_E6
539

Delete existing neighbours option 270, 360, 362, 428
DEM 35
Description of coordinate systems 25
Determination of Mis Best server 298, 303, 311
Determination of Mis best server 201
Determination of the active set 299, 304, 312
Deterministic clutter 91
Deygout construction 106, 107
Deygout method 84, 94, 98, 105
Deygout method with correction 107
Diffraction 85, 87, 90, 95
Diffraction calculation method 82
Diffraction loss 83, 90, 98, 104
Diffraction losses 78
Digital elevation model 35
Digital Terrain Model 35, 43, 45, 46, 77
Digital Terrain Model file 44
Dimensioning 150
Dimensioning model 150, 154
Display BSIC format 31
Display coordinate system 26
Display length units 31
Display power units 30
Display types 185, 275, 448
Distance criterion 262, 353, 420, 423
Distributed per cell 418, 419, 424
Distributed per cell allocation 260
Distributed per site 419, 425
Distributed per site allocation 260
Distributed per site strategy 266
Distributed PN-clusters per site allocation 352
Distributed PN-Clusters Per Site Strategy 357
Distributed strategy 265
Diversity mode 451
DL Ec/Io 115
DL gain due to availability of several pilot signals at the mobile 189, 279, 286
DL macro-diversity gain 115, 123
DL macro-diversity gain evaluation 121
DL min power 201, 202, 203, 298, 303, 311
DL minimum power 242, 320
DL noise rise 346
DL subframe ratio 515
DL total noise 346
Donor side parameters 521, 530
Donor transmitter antenna gain 520, 521, 526, 529
Donor transmitter antenna height/altitude 522, 531
Donor transmitter antenna-to-repeater antenna distance 522, 531
Donor transmitter feeder losses 526
Donor transmitter pilot power 519, 522, 526, 534
Donor transmitter power 529, 531, 533, 536
Donor transmitter total downlink losses 520, 521
Donor transmitters total downlink losses 530
Donor transmitters 519
Donor transmitter-to-mobile receiver path loss 526
Donor transmitter-to-repeater path loss 520, 521, 523, 525, 530, 531, 533, 535,
536
Downlink activity factor on FCH 278
Downlink average cell rate 285
Downlink Average Rate 341
Downlink budget 526
Downlink Data Rate 341
Downlink extra-cellular interference at terminal on carrier ic 191, 192, 221, 280,
281, 283, 287, 288, 289
Downlink extra-cellular interference based on pilot at terminal on carrier ic 287
Downlink extra-cellular interference based on traffic at terminal on carrier ic
287, 289
Downlink FCH nominal rate 277
Downlink intra-cellular interference at terminal on carrier ic 191, 280, 287, 288
Downlink load factor 228
Downlink load factor calculation 228, 318
Downlink load factor on carrier ic 221, 283, 289
Downlink load factor per cell 228, 318
Downlink load factor per mobile 229, 319
Downlink macro-diversity gain 234, 321
Downlink maximum rate supplied to the terminal 285
Downlink noise rise 346
540
Downlink power control 205, 300, 306

Downlink rake efficiency factor 236, 322, 323, 325, 327
Downlink rake receiver efficiency factor 187, 277
Downlink rate control 315
Downlink reuse factor on a carrier ic 221, 283
Downlink SCH bit rate 277
Downlink service area analysis 243, 339
Downlink service processing gain on FCH 278
Downlink service processing gain on SCH 278
Downlink soft handover gain 237, 324
Downlink soft handover gain for FCH channel on carrier ic 281
Downlink soft handover gain for SCH channel on carrier ic 281
Downlink soft handover gain on FCH 327
Downlink soft handover gain on SCH 327
Downlink sub-menu 234, 321
Downlink total noise analysis 246, 345
Downlink total power 232
Downlink traffic channel quality 339
Downlink traffic data quality 324, 340
Downtilt 76, 109
DTM 35, 76, 77, 80, 83, 89, 91, 98
DTM file 53
DTM sample 53
DTX 163, 166
E
Earth curvature 104
Eb/Io target on downlink 186
Eb/Io target on downlink FCH 276
Eb/Io target on downlink SCH 277
Eb/Nt 115, 119, 244, 246
Eb/Nt margin 244, 246, 340, 342, 344
Eb/Nt max 236, 323
Eb/Nt max for each cell of active set 236, 239
Eb/Nt max for the first f (number of fingers) cells of active set 322, 323, 324
Eb/Nt max on FCH and Eb/Nt max on SCH 327
Eb/Nt max on FCH for the first f (number of fingers) cells of active set 325
Eb/Nt max on SCH for the first f (number of fingers) cells of active set 326
Eb/Nt target 228, 234, 322
Eb/Nt target for FCH channel on uplink 277
Eb/Nt target for SCH channel on uplink 277
Eb/Nt target on FCH 324
Eb/Nt target on SCH 324
Eb/Nt target on uplink 186, 288, 367
Ec/Io 123, 193, 243, 255, 259, 320, 339
Ec/Io activation threshold 235, 236, 239
Ec/Io evaluation 232
Ec/Io margin 243, 255, 339
Ec/Io target on downlink for active set members 276, 284
Ec/Io target on downlink for the best server 186, 276, 284
Effective C/I 328
Effective Eb/Nt 237, 240, 244, 246, 323, 330, 340, 342, 344
Effective Eb/Nt FCH and SCH 334
Effective Eb/Nt on FCH and Eb/Nt on SCH 327
Effective MAC channel throughput 451, 498
Effective MAC throughput 460, 515
Effective MAC user throughput 505
Effective pilot quality level 327
Effective quality level 340, 342, 344
Effective rate 328
Effective rate of traffic overflow 151
Effective receiver antenna height 89
Effective service data rate in the uplink 528
Effective traffic channel 335
Effective traffic channel quality 240, 327
Effective traffic overflow rate 152, 155
Effective transmitter antenna height 86, 95
EIRP 74, 117, 131, 183, 273, 372, 373, 467, 473
Electrical tilt 110
Ellipsoid 25
Ellipsoid codes 29
Enhanced slope at receiver 87
Environment class 36
AT271_TRG_E6
Forsk 2009
Index
Environment traffic maps 36, 43, 45
Epstein-Peterson method 107
Erceg-Greenstein (SUI) path loss formula 97
Erceg-Greenstein (SUI) propagation model 96, 98
Erdas Imagine 35, 36, 37, 38
Erlang 147
Erlang B 147, 151, 155
Erlang C 147, 151, 155
Erlangs 37, 151
ErtPS 500, 501
Exceptional pair 159, 257, 268, 349, 358, 360, 361, 427
Exceptional pair criterion 263, 355, 422, 423
Exceptional pairs 259, 351
Existing neighbour 159, 257, 268, 349, 358, 360, 361, 427
Exponential weighting function 90
External neighbour allocation 358
External neighbour allocation algorithm 361
Externalised propagation results format 58
F
Fading 121
Far from transmitter 86
Fast fading 113
Fast link adaptation 209, 248
Fast link adaptation modelling 248
FAX-CCITT3 45
FCH 322
FCH activity factor 291, 293
FCH nominal rate 293
FCH nominal rates 294
FCH rates 291, 294
FDD 434
Feeder 124
Feeder length 125, 127
Feeder loss per metre 125, 127
Feeder noise figure 125, 126
Feeder reception loss 124, 125, 126
Feeder transmission loss 125, 126, 127
Feeder UL gain 125, 126
FER 174, 179
FFT size 434
Fibre repeater link 524, 532
Field strength 95
File formats 43
Focus zone 157, 254, 259, 351, 418, 509, 512
Force adjacent cells as neighbours 254, 347
Force adjacent transmitters as neighbours 157
Force co-site cells as neighbours 254, 268, 347, 359, 361, 427
Force co-site transmitters as neighbours 157
Force exceptional pairs 158, 254, 268, 347, 350, 358, 359, 361, 427
Force neighbour symmetry 158, 254, 347, 350
Force symmetry 350
Frame duration 433, 497, 514
Frame Erasure Rate 174
Free space loss 95, 104
Frequency 76, 83
Frequency band 60, 82
Frequency hopping mode 162, 163
Fresnel zone 104
Fresnel zone index 105
Friis' equation 125, 126
G
Gaussian 118, 120, 122
Gaussian distribution 113
General prediction studies 131, 183, 273
Generating a realistic user distribution 193, 290, 377, 452
Generic geographic data 43, 45, 46
Generic raster header file (.wld) 49
Geodetic datum 25
Geographic coordinate system 25
Geographic data 35
Forsk 2009
Geographic data formats 37

Geographic file resolution 76
Geographic layer order 76
Geographic profile extraction 78
Georeferenced formats 37
Global allocation for all cells 254, 347
GPRS EGPRS coverage studies 166
GPRS/EGPRS coding schemes study display types 172
GPRS/EGPRS Studies Based on C With ILA 169
GPRS/EGPRS Studies Based on C Without ILA 168
GPRS/EGPRS studies based on C/I with ILA and thermal noise 171
GPRS/EGPRS studies based on C/I with ILA and without thermal noise 170
GPRS/EGPRS studies based on C/I without ILA and thermal noise 169
GPRS/EGPRS studies based on C/I without ILA and with thermal noise 170
Grade of Service 151
Greenwich 25
Grid resolution 76
Ground altitude determination 77
Ground elevation 86, 89, 96
Ground height 96
Ground slope at receiver 86
GSM GPRS EGPRS documents 39, 131
GSM1800 82
GSM900 82
Guard subcarriers 434
H
Half-rate circuit switched traffic 151
Half-rate traffic 151
Half-rate traffic ratio 151
Handoff 199, 204, 235, 236, 238, 240, 305, 312, 322, 323, 325, 327, 332, 334,
335, 337, 383
Handover 329, 330
Handover end 158
Handover start 158
HARQ 341, 345
Hata calculations 83
Hata formula 83
HCS 161, 166, 175
Height above average profile 86
Height above ground 86
Hexagonal design 39
Hierarchical cell structure layer 161
Hierarchical knife-edge sorting 105
Histogram 66
Hopping sequence 164
Horizontal pattern 110
HSDPA 209, 212
HSDPA application throughput 223
HSDPA Bearer 209
HSDPA bearer allocation process 208
HSDPA bearer selection 211, 215
HSDPA Part of the Algorithm 207
HSDPA power allocation 207, 390
HSDPA power dynamic allocation 224
HSDPA prediction studies 248
HS-DSCH 209
HSN 164
HSPA Related Simulation Results 222
HS-PDSCH 212
HS-PDSCH CQI 211, 215
HS-PDSCH CQI Determination 215
HS-PDSCH Ec/Nt 237, 249
HS-PDSCH quality calculation 210, 213
HS-PDSCH quality update 213
HS-SCCH Ec/Nt 237, 249
HS-SCCH power 249
HSUPA application throughput 223
HSUPA Bearer 217
HSUPA bearer allocation process 218
HSUPA Bearer Selection 218
HSUPA Part of the Algorithm 216
HSUPA Prediction Study 252
HSUPA User Equipment Categories 217
AT271_TRG_E6
541

Hybrid Automatic Repeat Request 341, 345
I
Ideal Link Adaptation (ILA) 168, 169, 170, 171
Ideal link adaptation (ILA) 178
Identity matrix 508
Idle power gain 285
IEEE working group 802.16 97
Image files 55
Inter-carrier neighbour 254
Inter-carrier neighbours 346
Inter-carrier power sharing 231
Interfered areas study 160
Interfered transmitter 65, 66, 68
Interfered TRX type 65, 66, 68
Interference 74, 163, 166, 302, 308, 314, 383
Interference calculation 163
Interference correlation matrix 372, 446
Interference histograms formats 64
Interference prediction studies 160
Interference studies 74
Interference tab 166
Interfering transmitter 65, 66, 68
Internal BSIC format 31
Internal coordinate system 26
Internal length units 31
Internal power units 30
Inter-technology neighbour allocation 357
Inter-technology neighbours 362
Inter-transmitter distance 267, 358, 426
Intra-carrier neighbour 254
Intra-carrier neighbours 255, 346, 347
Intra-technology neighbour allocation 157
Intra-technology neighbour allocation algorithm 254, 346
Invalid path loss matrices 76
Inverse noise correlation matrix 411, 508
Io (Best server) 234, 321
IS-95 cdmaOne 321, 328
IS-95 cdmaOne and CDMA2000 1xRTT 339
IS-95 cdmaOne and CDMA2000 1xRTT services 291, 293
IS-95 cdmaOne and CDMA2000 1xRTT users 320
IS-95 cdmaOne and CDMA2000 documents 273
IS-95 cdmaOne and CDMA2000 prediction studies 319
IS-95 cdmaOne power control simulation algorithm 297
Iteration 199, 383
Iterations 220, 302, 303, 310, 316
ITU 1546 82
ITU 370-7 82
ITU 370-7 calculations 95
ITU 370-7 formula 95
ITU 526-5 82
ITU 526-5 calculations 95
ITU 526-5 formula 95
ITU 529-3 82
ITU Rec. 370-7 96
ITU Rec. 526-5 104
ITU-R P.1546-2 98
ITU-R P.370-7 95
ITU-R P.526-5 model 95
J
JD factor 366
K
Key Performance Indicators calculation 154
Knife-edge diffraction 104
KPI 150
KPIs 152
542
L
Lambert Conformal-Conic method 25
Land use 36
Large city 83
Length units 31
Line of sight 86, 87, 94
Linear regression 88
Link budget 117
Live traffic maps 36
Load 147
Load factor 207, 299, 318
Load rise 316
Load saturation 226, 316
Logarithmic weighting function 90
Log-normal distribution 113
LOS 87, 88
Losses due to clutter 90
LZW 45
M
MAC rate 249
MAC throughput 249
Macro diversity 186
Macrocell 82
Macro-diversity gain 115, 233, 239, 240
Main antenna 127
MAIO 162, 164
MAL 162, 164
MAS 166
Matrix validity 76
Max Aggregate Throughput scheduler 503
Max BER 179
Max C/I level 165
Max Cell power 188
Max cell power 285
Max FER 179
Max MOS 179
Max terminal power 238, 328, 335
Max terminal power on FCH and SCH 331
Maximum allowable blocking probability 151
Maximum BLER percentage 174
Maximum cell power 273, 278
Maximum Eb/Nt 342, 344
Maximum inter-site distance 157, 347
Maximum loading factor 187, 277, 285
Maximum noise level 247, 346
Maximum noise rise 247, 346
Maximum number of HSDPA users 207
Maximum number of MAC indexes available per cell (59) 284
Maximum number of neighbours 267, 358, 360, 362, 426
maximum number of neighbours allowed for each cell 159, 258, 350
Maximum number of OVSF codes available per cell (512) 187, 277
Maximum number of TRXs per transmitter 151
Maximum permissible delay 151
Maximum power allowed on FCH 324, 326
Maximum power allowed on traffic channel 235
Maximum quality level 340, 342, 344
Maximum reduction factor 148
Maximum terminal power 436
Maximum terminal power allowed 328
Maximum throughput demand 436
Maximum traffic channel power per transmitter allowed on FCH 321
Mean Opinion Score 174
Mechanical downtilt 451
Meridian 25
Microwave repeater link 522, 531
Millington method 107
MIMO 435, 497, 498, 516
Min C/I level 165
Minicell 82
Minimum covered area 158
Minimum dedicated packet switched timeslots per transmitter 151
Minimum Mean Square Error 408, 505
AT271_TRG_E6
Forsk 2009
Index
Minimum noise level 247, 346
Minimum noise rise 247, 346
Minimum percentage of covered traffic 158
Minimum pilot quality level on uplink 284
Minimum power allowed on FCH 300, 306
Minimum power allowed on SCH 306
Minimum power allowed on traffic channel 205, 299, 304, 312
Minimum required throughput per user 151
Minimum signal level 158
Minimum terminal power 436
Minimum throughput demand 435
Minimum throughput reduction factor 147
Minimum throughput reduction factor calculation 153
MMSE Smart Antenna Model 445
Mobile allocation list 162
Mobile receiver 166, 174
Mobile Station Allocation 162
Mobility 200, 234, 239, 242, 243, 246, 298, 303, 336, 338, 339, 342, 343, 345,
449, 459
Model standard deviation 73, 113, 233, 321
Modelling the transmitter-repeater link 534
Modulation 434
Monte-Carlo algorithm 193, 290
Monte-Carlo simulation 117
Monte-Carlo simulator 37
MOS 174, 179
MSA 162, 164
MUD factor 186, 276, 284
Multi user gain 285
Multipath interference 113
Multiple Input Multiple Output 516
Multiple knife-edges 105
Multi-resolution profile management 80
Number of inactive users 194, 198, 291, 292, 296, 378, 382, 453, 454
Number of MAC indexes used by the cell 284
Number of mobiles connected to transmitter Sj on carrier ic 289
Number of OVSF codes used by the cell 187, 277, 284
Number of rake fingers 279, 321
Number of required TRXs per transmitter 151
Number of servers 186, 276, 449
Number of shared timeslots 154
Number of subchannels per channel 434
Number of used subchannels 452
Number of users active on DL 194, 198, 291, 292, 296, 378, 382, 453, 454
Number of users active on UL 194, 198, 291, 292, 296, 378, 382, 453, 454
Number of users active on UL and DL 194, 198, 291, 292, 293, 294, 296, 378,
382, 453, 454
O
Okumura-Hata 83
One cluster per site allocation 260
One cluster per site strategy 266
One histogram per line format 64
One SYNC_CL code per site 418
One SYNC_DL code per site 419, 425
One value per line with dictionary file format 65
One value per line with transmitter name repeated format 68
Only co-channel and adjacent values format 69
Optimum Beamformer 408
Orthogonality factor 186, 276, 284, 368
Other geographic data 37
Over the air repeater link 519, 528
Overhead 515
Overlapping factor 159, 511
OVSF code management 226
OVSF codes 227
N
Near cells 260, 352, 419
Near transmitter 86
Neighbour allocation 157
Neighbour priority 159, 257, 349, 361
Neighbourhood cause 159, 257, 268, 349, 358, 360, 361, 427
Neighbourhood criterion 262, 354, 421, 423
Neighbout importance 158
Neighbout Importance Function 159
Network dimensioning 37
Network dimensioning engine 150
Network dimensioning process 150
Network dimensioning steps 151
Network regulation mechanism 199, 290, 297, 302, 383
NH 162
Noise 177
Noise correlation matrix 372, 446
Noise figure 125, 126
Noise figure of the transmitter equipment 528
Noise level 247, 346
Noise rise 247, 248, 346
Noise rise on downlink 221, 283, 289
Noise rise on uplink 221, 283, 286, 289
Noise Rise Scheduling 218
Noise rise scheduling 218
Non correlated MSAs m and n 164
Non hopping 162
nrtPS 500, 501
Number of active users on UL and DL 197, 381, 454
Number of calls per hour 291
Number of cells in active set 234, 321
Number of channel elements available for a site on downlink 186, 277
Number of channel elements available for a site on uplink 186, 277
Number of channel elements of a site consumed by users on downlink 186, 277
Number of channel elements of a site consumed by users on uplink 186, 277
Number of dedicated circuit switched timeslots 154
Number of dedicated packet switched timeslots 154
Number of EVDO channel elements available for a site on uplink and downlink
284
Number of HS-SCCH channels 207
Forsk 2009
P
Packbit 45
Packet queue length 149
Packet switched service 194, 198, 379, 382
Packet switched traffic 155
Packet switched traffic demand 150, 152, 157
Packet switched traffic load 147, 152
Packet switched traffic overflow 156
Packet traffic overflow 154
Parameters used for CDMA2000 1xEV-DO modelling 283
Parameters used for IS-95 cdmaOne and CDMA2000 1xRTT modelling 276
Path loss 43, 73, 83, 89, 113, 127, 131, 183, 189, 273, 279, 286, 369, 372, 436
Path loss (dB) 185, 275, 448
Path loss calculations 77
Path loss matrices 74, 76, 183, 274, 350, 360, 373, 447
path loss matrix 60
Path loss matrix resolution 275
Peak MAC channel throughput 451, 498, 499, 504
Peak MAC throughput 460, 515
Peak MAC user throughput 505
Percentage of covered area 160
Percentage of delayed circuit switched traffic 155
Percentage of max transmitter power used 283
Percentage of maximum transmitter power used. 187, 277
Percentage of traffic covered in the overlapping area 160
Permutation zone 434, 451, 497
Pilot burst transmitted by the transmitter on carrier ic 285
Pilot bust received at terminal from a transmitter on carrier ic 286
Pilot power 183, 233
Pilot power percentage 279
Pilot quality 193, 234, 242, 320, 321
Pilot quality level 315
Pilot quality level at terminal on carrier ic 287
Pilot quality level on uplink 311, 336, 345
Pilot reception analysis 232, 242, 338
Pilot RSCP activation threshold 235, 236, 239
Pilot signal 347
Pilot subcarriers 434
AT271_TRG_E6
543

Pixel size 35
Planet 35, 36, 37
Planet format 52
Plot resolution 185, 275
PN offset allocation 350
PN offset per cell allocation 352
PN Offset per Cell Strategy 356
PN sequence 350
PN-cluster size 351
Point analysis 131, 166, 183, 232, 273, 319, 372
Point index 87
Population 37
Population density vector maps 36
Population maps 43, 45, 46
Power Backoff 168, 169, 170, 171, 172
Power combination gain 409, 507
Power control 310, 320
Power control simulation 193, 199, 302, 383
Power control simulations 258, 350
Power pooling between transmitters 301, 309
Power received at the receiver from the donor transmitter 526
Power spectral density of thermal noise 436
Power units 30
Preamble C/N 497
Preamble index 435
Predicted path loss 116
Prediction studies 76
Presentation of the algorithm 298, 303, 311
Primary scrambling code allocation 258, 417
Prime meridian 25
Principal knife-edge 106
Priority determination 261, 353
Probability density function (pdf) 116
Probability of being active on DL 194, 197, 198, 291, 292, 293, 295, 296, 378,
381, 382, 453, 454
Probability of being active on UL 194, 197, 198, 291, 292, 293, 295, 296, 378,
381, 382, 453, 454
Probability of being active on UL and DL 194, 197, 198, 291, 292, 293, 295,
296, 378, 381, 382, 453, 454
Probability of being connected 194, 291, 378, 452
Probability of being inactive 194, 198, 291, 292, 293, 295, 296, 378, 381, 382,
453, 454
Probability of Having a Certain RLC Peak Rate 252
Profile extraction 73
Profile extraction methods 78
Profile extraction mode 82
Profile resolution 87
Profile tab 183, 273, 372
Projection 26, 53
Projection coordinate system 25, 26
Projection method codes 29
Projection parameter indices 30
Propagation model 73, 76, 113
propagation model 60
Propagation model appendices 104
Propagation model parameters 76
Propagation models 82
Proportional Demand scheduler 502
Proportional Fair scheduler 501
Pseudo Noise sequence 350
Q
QoS class 435
QoS class bias factor 436, 502
QoS class rank 502
Quality level 205, 299, 300, 306, 339
Quality level at terminal for FCH using carrier ic due to combination of all transmitters of the active set 281
Quality level at terminal for SCH using carrier ic due to combination of all transmitters of the active set 281
Quality level at terminal on a traffic channel from one transmitter for a FCH
channel on carrier ic 281
Quality level at terminal on a traffic channel from one transmitter for a SCH
channel on carrier ic 281
Quality level at terminal on a traffic channel from one transmitter on carrier ic
544
191
Quality level at terminal on pilot for carrier ic 191, 280
Quality level at transmitter on a traffic channel for a FCH channel on carrier ic
282, 288
Quality level at transmitter on a traffic channel for a SCH channel on carrier ic
282
Quality level at transmitter on a traffic channel for carrier ic 192
Quality margin 339, 340, 342, 344
R
R99 200
R99 related simulation results 220
Radial calculation mode 84, 98
Radial profile extraction 78
Radio data 38
Radio equipment 124
Radio resource management 317
Random carrier selection mode 201, 202, 203, 299, 304, 311
Random shadowing error drawn during Monte-Carlo simulation 190, 279, 286
Raster images 43, 45, 46
Raster maps 36
Rayleigh fading 113
Receive Time Guard 434
Received preamble power 451
Received signal level 158
Received traffic power 451
Receiver 82
Receiver antenna 76
Receiver antenna gain 74, 117, 183, 273, 374, 384, 529, 531, 535
Receiver antenna height 83
Receiver clearance 91, 94
Receiver gain 76
Receiver height 76, 94, 96
Receiver height in metre 61
Receiver in donor transmitter coverage area 527
Receiver in repeater coverage area 527
Receiver location 82
Receiver loss 529, 531, 535
Receiver losses 74, 76, 117, 131, 183, 273, 374, 384
Reception tab 183, 274, 373, 447
REDT 111
Reduction Factor 149, 150, 152, 153
Regression line 88, 89
Relation between RLC/MAC and application throughputs 173
Reliability level 153
Remote Antennas 39
Remote electrical downtilt 109, 111
Repeater 38
repeater
definition (UMTS) 519
Repeater amplifier gain 520, 521, 523, 524, 525, 526, 530, 532, 533
Repeater coverage side antenna gain 520, 521, 523, 524, 525, 526, 533, 535
Repeater EIRP 529, 531, 535
Repeater noise figure 528
Repeater output global amplification gain 519, 522, 527, 534
Repeaters coverage side antenna gain 530, 532, 533
Repeaters coverage side feeder losses 520, 521, 523, 524, 525, 530, 532,
533, 535, 536
Repeaters donor side antenna azimuth 521, 530
Repeaters donor side antenna gain 520, 521
Repeaters donor side antennas mechanical Downtilt 521, 530
Repeaters donor side feeder losses 520, 521
Repeaters donor-side antenna height/altitude 522, 531
Repeaters donor-side antenna tilt 522, 531
Repeaters and Remote Antennas 519
Repeater-to-mobile receiver path loss 527
Repeater-to-receiver path loss 519, 523, 529
Required C/I 327
Required power 245, 246, 299, 318, 340, 343, 344
Required power margin 245, 246, 340, 343, 344
Required quality 318
Required quality level on uplink 336
Required rate 327
Required terminal power 236, 238, 239, 328
AT271_TRG_E6
Forsk 2009
Index
Required terminal power on FCH and SCH 331
Required terminal power to achieve Eb/Nt target at transmitter for FCH on carrier ic 282
Required terminal power to achieve Eb/Nt target at transmitter for SCH on carrier ic 282
Required terminal power to achieve Eb/Nt target at transmitter on carrier ic 288
Required terminal power with ACK 335
Required terminal power without ACK 337
Required transmitter FCH traffic channel power to achieve Eb/Nt target at terminal on carrier ic 281
Required transmitter SCH traffic channel power to achieve Eb/Nt target at terminal on carrier ic 281
Required transmitter traffic channel power to achieve Eb/Nt target at terminal
on carrier ic 191, 281
Resolution 74, 80, 81
Resource control 317
Restricted to neighbours 290
Reuse distance 259, 351, 418
Rho factor 436
RLC Peak Rate 238
RLC peak rate 250
RLC peak throughput 250
RLC/MAC throughput/timeslot study display type 173
Round trip time 149
rtPS 500, 501
Rural area 83
S
Sample Values for SPM Formula Parameters 92
Sampling factor 434
Scanned images 37
Scarmbling code allocation examples 423
SCH rate 293, 306
SCH rate factor 277, 292, 295
Scheduling algorithms 215
Scrambling code allocation order 420
Scrambling code allocation process 260, 352
Scrambling code domain criterion 263, 422, 423
Scrambling codes 258, 417
Scrambling codes domains 259, 351
Second best signal level and a margin 132, 184, 274, 448
Second best signal level per HCS layer and a margin 161, 166, 176
Secondary antenna 127
Secondary antenna gain 127
Secondary antenna index 127
Secondary antennas 127
Secondary knife-edge 106
Segment 434
Segmentation 478
Segmentation usage 435, 458
Segmenting factor 497
Semi-deterministic clutter 91
Sequential carrier selection mode 201, 202, 299, 304, 311
Served circuit switched traffic 151, 152, 154, 155
Served packet switched traffic 151, 154, 156
Service 200, 234, 239, 244, 246, 298, 303, 338, 339, 342, 343, 449, 459
Service area determination 161, 184, 274
Service coding factor on downlink 187
Service downlink effective bit rate 187, 188
Service downlink nominal bit rate 187, 367
Service downlink process gain 188
Service priority 301, 302, 308, 315
Service processing gain 528
Service uplink effective bit rate 187, 188
Service uplink nominal bit rate 187, 188, 367
Service uplink process gain 188
Services 324, 326, 333
SFH 162
Shadow fading 98
Shadowing 113
Shadowing error 118
Shadowing error pdf (n signals) 118, 121
Shadowing error pdf (one signal) 116
Shadowing margin 73, 121, 123, 131, 165, 183, 189, 233, 273, 279, 286, 321,
369, 372, 374, 384, 436, 519, 523, 529, 531, 534, 536
Forsk 2009
Shadowing margin calculation in predictions 116

Shadowing model 113
Signal level 131, 167, 183, 273, 372, 373, 450
Signal level (in dBm, dBV, dBV/m) 185, 275, 448
Signal level at the receiver 74
Signal level received from a repeater on a carrier (CDMA) 519, 522, 534
Signal level received from a repeater on a TRX type (GSM) 528, 532, 535
Signal level received on a carrier 274
Signal level study 74
Signal Quality Analysis 449
Signal strength 113
Signal to noise ratio 119
Simulation 455
Simulations 193, 226, 243, 248, 290, 377
Simulations based on raster traffic and vector traffic maps 193, 290, 378, 452
Simulations based on traffic map per service and per transmitter 197, 293, 381,
454
Single colour 242, 243, 245, 338, 339, 342, 343
Site 38
Site altitude 77
Site position 76
Site priority 264, 355
Slope at receiver between 0 and distance min 86
Slow fading 113
SM 497, 498, 516
Smart Antenna 515
Smart antenna 371, 445
Smoothing angle 111
Smoothing factor 111
Soft handoff 118, 121, 205, 238, 240, 312, 318
Soft handover 329, 330, 332, 334, 335, 337
Soft handover gain on downlink 191
Soft handover gain on uplink 192
Soft/soft handoff 205, 300, 305, 312
Softer handoff 120, 204, 239, 240, 300, 305, 312, 317
Softer handover 332, 334, 336, 338
Softer/soft handoff 205, 300, 305, 312
Softer-soft handoff 239, 240
Softer-soft handover 329, 330, 332, 334, 336, 338
Softer-softer handoff 239, 240
Softer-softer handover 332, 334, 336, 338
Soft-soft handoff 239, 240
Soft-soft handover 329, 330, 332, 334, 336, 337
Space-Time Transmit Diversity 516
Spatial Multiplexing 497, 498, 516
Spatial Multiplexing gain factor 497
Speed of light 105
Spheroid 53
SPM 82, 85, 94, 111
SPM calculations 86
SPM formula 85
Spot Ht 86
Spreading angle 368
Spreading Bandwidth 187, 277, 284
Spreading bandwidth 528
Stability factor 457
Standard deviation 116, 117, 118
Standard meridian 25
Standard Propagation Model 80, 85, 111
Stanford University Interim model 97
Start frequency 434
Station 39
Steering vector 371, 409, 410, 445, 506, 508
STTD 516
Subcarrier 515
Subcell 39
Subchannel 515
Subchannels 434
Suburban area 83
SUI terrain types 97
Supported geographic data formats 37
Symbol 514
Symmetric neighbourhood relationship 159, 257, 349
Synthesised frequency hopping 162
Systematic profile extraction 79
AT271_TRG_E6
545
T
Take into account Covered Area 158
Take into account Covered Traffic 158
Target throughput 499
Target traffic overflow rate 151
TBA cells 417
TBA transmitters 157
TBC transmitter 74
TBC transmitters 433
TBF 149
TBF blocking 149
TCH 39
TCH_INNER 39
TDD 433, 515
Terminal 153, 236, 241, 325, 327, 331, 338, 449, 459
Terminal FCH power transmitted in carrier ic 278, 285
Terminal gain 189, 279, 286, 368, 436
Terminal loss 189, 279, 286, 368, 436
Terminal Noise Factor 187, 277, 284, 368
Terminal noise figure 436
Terminal power transmitted 189
Terminal power transmitted on carrier ic 285
Terminal SCH power transmitted on carrier ic 278
Terrain clearance 95
Terrain profile 87, 89
Text data 52
Text data files 55
Thermal noise 170, 171, 232, 234, 238, 240, 319, 321, 329, 330, 332, 333, 334,
335, 337, 508
Thermal noise at terminal 187, 277, 285, 368
Thermal noise at transmitter 187, 277, 285, 368
Thermal noise ratio 168
Throughput 171, 172
Throughput based on C with ILA 169
Throughput Based on C Without ILA 168
Throughput based on interpolation between C/N and C/(I+N) with ILA 172
Throughput based on interpolation between C/N and C/(I+N) without ILA 171
Throughput based on worst case between C and C/I with ILA 170
Throughput based on worst case between C and C/I without ILA 170
Throughput offset 436, 497, 499, 504
Throughput reduction factor 154, 156
Throughput scaling factor 436, 497, 499, 504
Throughput/timeslot 173
Tilt 127
Tilt angle 109
Time-domain overhead 433
Timeslot capacity 149
Timeslot configuration 151
Time-slot scenario 457
TMA 124
TMA noise figure 125, 126
TMA reception gain 125, 126
TMA transmission loss 125, 126, 127
Total effective interference based on traffic at terminal on carrier ic 289
Total FCH power on carrier ic 278
Total interference 528
Total loss between transmitter and receiver 233
Total losses 131, 183, 273, 372
Total losses (dB) 185, 275, 449
Total noise 204, 205, 232, 299, 300, 306, 312, 329, 330, 332, 333, 334, 335,
337
Total noise (I+N) 460
Total noise at transmitter on carrier ic 192, 221, 282, 283, 288, 289
Total noise based on pilot received at terminal on carrier ic 287
Total noise based on traffic received at terminal on carrier ic 287, 289
Total number of EVDO channel elements of a site consumed by users on uplink
and downlink 284
Total power on a carrier ic 200
Total power on carrier ic 298, 303
Total power received 233
Total power received at terminal from a transmitter on carrier ic 190
Total power received at terminal from traffic channels of a transmitter on carrier
ic 190
Total power received at transmitter from extra-cellular terminals using carrier ic
192, 221
546
Total power received at transmitter from intra-cellular terminals using carrier ic

192, 221
Total power transmitted by the terminal on carrier ic 280
Total received interference at transmitter on carrier ic 192, 221, 281, 283, 288,
289
Total received noise at terminal on carrier ic 191, 280
Total received power at terminal from a transmitter on carrier ic 279
Total received power at terminal from traffic channels of a transmitter on carrier
ic 279
Total received power at transmitter from extracell terminals using carrier ic 281,
283, 287, 289
Total received power at transmitter from intracell terminals using carrier ic 281,
283, 287, 289
Total SCH power on carrier ic 278
Total traffic demand 155
Traffic analysis 37, 147, 151
Traffic burst received at terminal from a transmitter on carrier ic 286
Traffic burst transmitted by the transmitter on carrier ic 285
Traffic C/N level 450
Traffic channel availibility on downlink 321
Traffic channel quality 235, 236, 238, 243, 245, 322, 323, 329
Traffic channel quality on FCH 324, 325, 327, 328
Traffic channel quality on SCH 324, 325, 327
Traffic data 36
Traffic data channel gain 286
Traffic demand 151
Traffic density maps 43, 45, 46
Traffic load 147, 149, 151, 155, 156, 163, 166, 435
Traffic loads 458
Traffic Map Based on Transmitters and Services (# Active Users) 381, 454
Traffic Map Based on Transmitters and Services (Throughputs) 381, 454
Traffic maps 160
Traffic overflow 152
Traffic quality level at terminal on carrier ic 287
Traffic quality target on FCH 330
Transmit Time Guard 433
Transmitter 38, 165
Transmitter antenna azimuth 60
Transmitter antenna gain 74, 183, 273, 372, 374, 384
Transmitter antenna height 83
Transmitter antenna mechanical tilt 60
Transmitter equipment thermal noise 528
Transmitter gain 189, 279, 285, 368, 435
Transmitter height in metre 61
Transmitter Identifier 67
Transmitter loss 183, 189, 273, 279, 286, 366, 373, 374, 384, 435
Transmitter losses 127
Transmitter Noise Factor 187, 277, 284, 435
Transmitter power 74, 127
Transmitter priority 264, 355
Transmitter radio equipment 124
Transmitter SCH power for a traffic channel on carrier ic 278
Transmitter total transmitted power on carrier ic 189, 278
Transmitter traffic channel power on carrier ic 188, 189
Transmitter-terminal total loss 190, 279, 286, 369
Transverse Mercator coordinate system 25
Triangular weighting function 90
TRX 39
TRX power offset 529, 531, 535
Typical values for losses per clutter class (SPM) 93
U
UGS 500
UL load factor 316
UL load percentage 232, 338
UL macro-diversity gain 115, 121
UL macro-diversity gain evaluation 118
UL min noise 201, 202, 203, 298, 303, 311
UL minimum noise 242, 320
UL quality gain due to signal diversity in soft handoff 190, 279, 286
UL SHO gain 337, 338
UL soft handover gain 335
UMTS 82
UMTS HSPA documents 183
AT271_TRG_E6
Forsk 2009
Index
UMTS HSPA Prediction studies 232
Uniform weighting function 90
Unit codes 27
Units 30
Units for distances 31
Units for heights 31
Units for offsets 31
Units for reception power 30
Units for transmission power 30
Universal Transverse Mercator method 25
Update of interference on active mobiles 301
Uplink activity factor on FCH 278
Uplink and downlink FCH nominal rates 291
Uplink and downlink interference update 206
Uplink and downlink interference updates 301, 308
Uplink Average Rate 345
Uplink Data Rate 344
Uplink data traffic quality 333
Uplink effective traffic channel quality 331, 335
Uplink FCH nominal rate 278
Uplink interference updates 314
Uplink load factor 226, 238, 240
Uplink load factor control 207, 302, 309, 314
Uplink load factor on carrier ic 221, 283, 289
Uplink macro-diversity gain 329, 330, 332, 334
Uplink noise rise 435, 458, 515
Uplink power control 204, 299, 304, 312
Uplink power control margin 434
Uplink quality 335, 337
Uplink rake receiver efficiency factor 187, 277, 285
Uplink received powers on carrier ic 200, 298, 303
Uplink reuse efficiency factor on carrier ic 221, 283, 289
Uplink reuse factor on carrier ic 221, 283, 289
Uplink SCH bit rate 278
Uplink service area analysis 245, 342
Uplink service effective bit rate 285
Uplink service processing gain on FCH 278, 285
Uplink service processing gain on SCH 278
Uplink service rate 285
Uplink soft handover gain 240, 331
Uplink soft handover gain FCH and SCH 335
Uplink soft handover gain for FCH channel on carrier ic 282, 288
Uplink soft handover gain for SCH channel on carrier ic 282
Uplink sub-menu 238, 328
Uplink traffic channel quality 238, 329, 331
Urban environment 83
Use a maximum of codes 259, 264, 418
Use a maximum of PN offsets 352
User density traffic maps 37
User multiplexing factor 149
User profile 291
User profile traffic maps 36
User profiles 36
User throughput 515
WLL calculations 94
WLL formula 94
Z
Zero mean gaussian random variable 117, 118, 121
Zero-mean unit-variance Gaussian 116
V
Vector data 37, 52
Vector file sample 55
Vector files 54
Vertical Mapper 51
Vertical pattern 110
Victim and interfering mobiles 456
Visibility and distance between the transmitter and the receiver 86
Voice service users 291, 293
Voice services 452, 454
W
Waiting queue 149
Walsh code management 317
WGS84 ellipsoid 25
WiMAX 82
WLL 82, 94, 106
Forsk 2009
AT271_TRG_E6
547
548
AT271_TRG_E6
Forsk 2009
Forsk 2009
AT271_TRG_E6
549

Release 2.7.1
7 rue des briquetiers 31700 Blagnac France
Tel: +33 (0)5 62 74 72 10 Fax: +33 (0)5 62 74 72 11
http://www.forsk.com
AT271_TRG_E6
February 2009

Technical Reference Guide Atoll

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Technical Reference Guide Atoll

Uploaded by

Copyright:

Available Formats

Atoll

RF Planning and Optimisation Software

Technical Reference Guide

Forsk (USA Office) 200 South Wacker Drive

Forsk (China Office) Suite 302, 3/F, West Tower,

Sales and pricing information

Atoll 2.7.1 Technical Reference Guide Release AT271_TRG_E6

Copyright 1997 - 2009 by Forsk

About Technical Reference Guide

Technical Reference Guide

Coordinate Systems and Units ....................................................... 25

Geographic and Radio Data ........................................................... 35

File Formats .................................................................................... 43

Technical Reference Guide

HDR Header File...............................................................................................................................43

One Histogram per Line (.im0) Format............................................................................................. 64

Technical Reference Guide

Free Space Loss....................................................................................................................95

GSM GPRS EGPRS Networks......................................................131

Technical Reference Guide

Service Area Determination ......................................................................................................161

HCS Servers and a Margin....................................................................................................... 176

UMTS HSPA Networks ................................................................. 183

Technical Reference Guide

OVSF Codes Management..................................................................................................226

IS-95 cdmaOne and CDMA2000 Networks ...................................273

Reception Tab .......................................................................................................................... 274

Technical Reference Guide

Uplink Service Area Analysis ....................................................................................................342

Simulations Based on Raster and Vector Traffic Maps ............................................................ 378

Technical Reference Guide

Multi Carrier Network.................................................................................................................425

WiMAX BWA Networks..................................................................433

Coverage by Best Bearer (DL) Display Types .................................................................... 461

Repeaters and Remote Antennas................................................. 519

Technical Reference Guide

Gain Automatic Calculation .................................................................................................523

Digital Terrain Model.................................................................................................................................. 35

Technical Reference Guide

Overlapping Zone for Inter-carrier Neighbours - 1st Case .......................................................................

Angle from North (Azimuth) .....................................................................................................................

Technical Reference Guide

Technical Reference Guide

Chapter 1: Coordinate Systems and Units

Coordinate Systems and Units

Description of Coordinate Systems

The longitude of the natural origin (central meridian)

Basic definitions are presented below.

Geographic Coordinate System

Technical Reference Guide

Projection Coordinate System

Coordinate Systems in Atoll

Projection coordinate system

Projection coordinate system for the Atoll document

Projection Coordinate System

Display Coordinate System

Internal Coordinate Systems

Chapter 1: Coordinate Systems and Units

4230 = "ED50"; 101; 230; 1; "Europe - west"

Technical Reference Guide

Greek Geodetic Reference System 1987

Australian Geodetic Datum 1966

North American Datum 1927

Australian Geodetic Datum 1984