Technical Reference Guide

v e r s i o n 2.8.3
AT283_TRG_E2


Technical Reference Guide
© Forsk 2010 AT283_TRG_E2 3
Contact Information
Atoll 2.8.2 Technical Reference Guide Release AT283_TRG_E2
© Copyright 1997 - 2010 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 the 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.
The 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 (Head Office) 7 rue des Briquetiers
31700 Blagnac
France





www.forsk.com
sales@forsk.com
support@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
Forsk (USA Office) 200 South Wacker Drive
Suite 3100
Chicago, IL 60606
USA




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
Forsk (China Office) Suite 302, 3/F, West Tower,
Jiadu Commercial Building,
No.66 Jianzhong Road,
Tianhe Hi-Tech Industrial Zone,
Guangzhou, 510665,
People’s Republic of China




www.forsk.com.cn
enquiries@forsk.com.cn
+86 20 8553 8938
+86 20 8553 8285
Web
Information and enquiries
Telephone
Fax
Technical Reference Guide
4 AT283_TRG_E2 © Forsk 2010
Table of Contents
© Forsk 2010 AT283_TRG_E2 5
Table of Contents
1 Coordinate Systems and Units ....................................................... 23
1.1 Coordinate Systems............................................................................................................................... 23
1.1.1 Description of Coordinate Systems .................................................................................................. 23
1.1.1.1 Geographic Coordinate System.................................................................................................. 23
1.1.1.2 Datum......................................................................................................................................... 23
1.1.1.3 Meridian...................................................................................................................................... 23
1.1.1.4 Ellipsoid ...................................................................................................................................... 23
1.1.1.5 Projection.................................................................................................................................... 24
1.1.1.6 Projection Coordinate System.................................................................................................... 24
1.1.2 Coordinate Systems in Atoll ............................................................................................................. 24
1.1.2.1 Projection Coordinate System.................................................................................................... 24
1.1.2.2 Display Coordinate System........................................................................................................ 24
1.1.2.3 Internal Coordinate Systems ...................................................................................................... 24
1.1.3 File Formats ..................................................................................................................................... 25
1.1.3.1 Unit Codes.................................................................................................................................. 25
1.1.3.2 Datum Codes.............................................................................................................................. 26
1.1.3.3 Projection Method Codes ........................................................................................................... 27
1.1.3.4 Ellipsoid Codes........................................................................................................................... 27
1.1.3.5 Projection Parameter Indices...................................................................................................... 28
1.1.4 Creating a Coordinate System......................................................................................................... 28
1.2 Units....................................................................................................................................................... 28
1.2.1 Power Units ...................................................................................................................................... 28
1.2.2 Length Units ..................................................................................................................................... 29
1.3 BSIC Format .......................................................................................................................................... 29
2 Geographic and Radio Data ........................................................... 33
2.1 Geographic Data.................................................................................................................................... 33
2.1.1 Data Type......................................................................................................................................... 33
2.1.1.1 Digital Terrain Model (DTM) ....................................................................................................... 33
2.1.1.2 Clutter (Land Use) ...................................................................................................................... 34
2.1.1.2.1 Clutter Classes...................................................................................................................... 34
2.1.1.2.2 Clutter Heights ...................................................................................................................... 34
2.1.1.3 Traffic Data ................................................................................................................................. 34
2.1.1.3.1 User Profile Environment Based Traffic Maps ...................................................................... 34
2.1.1.3.2 User Profile Traffic Maps....................................................................................................... 34
2.1.1.3.3 Sector Traffic Maps............................................................................................................... 34
2.1.1.3.4 User Density Traffic Maps..................................................................................................... 35
2.1.1.4 Vector Data................................................................................................................................. 35
2.1.1.5 Scanned Images......................................................................................................................... 35
2.1.1.6 Population................................................................................................................................... 35
2.1.1.7 Other Geographic Data............................................................................................................... 35
2.1.2 Supported Geographic Data Formats .............................................................................................. 35
2.2 Radio Data............................................................................................................................................. 36
2.2.1 Site ................................................................................................................................................... 36
2.2.2 Antenna............................................................................................................................................ 36
2.2.3 Transmitter ....................................................................................................................................... 36
2.2.4 Repeater........................................................................................................................................... 36
2.2.5 Remote Antenna .............................................................................................................................. 37
2.2.6 Station .............................................................................................................................................. 37
2.2.7 Hexagonal Design............................................................................................................................ 37
2.2.8 GSM GPRS EGPRS Documents ..................................................................................................... 37
2.2.8.1 TRX............................................................................................................................................. 37
2.2.8.2 Subcell ........................................................................................................................................ 37
2.2.8.3 Cell Type..................................................................................................................................... 37
2.2.9 All CDMA, WiMAX, and LTE Documents ......................................................................................... 37
2.2.9.1 Cell.............................................................................................................................................. 37
Technical Reference Guide
6 AT283_TRG_E2 © Forsk 2010
3 File Formats.....................................................................................41
3.1 BIL Format ..............................................................................................................................................41
3.1.1 HDR Header File...............................................................................................................................41
3.1.1.1 Description ..................................................................................................................................41
3.1.1.2 Samples ......................................................................................................................................42
3.1.1.2.1 Digital Terrain Model..............................................................................................................42
3.1.1.2.2 Clutter Classes File................................................................................................................42
3.1.1.2.3 BIL File...................................................................................................................................42
3.2 TIF Format start here..............................................................................................................................42
3.2.1 TFW Header File...............................................................................................................................43
3.2.2 Sample..............................................................................................................................................44
3.2.2.1 Clutter Classes File .....................................................................................................................44
3.3 BMP Format............................................................................................................................................44
3.3.1 BMP File Description.........................................................................................................................44
3.3.1.1 BMP File Structure ......................................................................................................................44
3.3.1.2 BMP Raster Data Encoding ........................................................................................................45
3.3.1.2.1 Raster Data Compression Descriptions.................................................................................46
3.3.2 BPW/BMW Header File Description..................................................................................................47
3.3.3 Sample..............................................................................................................................................47
3.3.3.1 Clutter Classes File .....................................................................................................................47
3.4 PNG Format............................................................................................................................................47
3.4.1 PGW Header File Description...........................................................................................................47
3.5 Generic Raster Header File (.wld) ..........................................................................................................47
3.5.1 WLD File Description ........................................................................................................................48
3.5.2 Sample..............................................................................................................................................48
3.5.2.1 Clutter Classes File .....................................................................................................................48
3.6 DXF Format ............................................................................................................................................48
3.7 SHP Format ............................................................................................................................................48
3.8 MIF Format .............................................................................................................................................48
3.9 TAB Format ............................................................................................................................................49
3.10 ECW Format ...........................................................................................................................................49
3.11 Erdas Imagine Format ............................................................................................................................50
3.12 Planet EV/Vertical Mapper Geographic Data Format .............................................................................50
3.13 ArcView Grid Format ..............................................................................................................................50
3.13.1 ArcView Grid File Description ...........................................................................................................50
3.13.2 Sample..............................................................................................................................................51
3.14 Other Supported Geographic Data File Formats....................................................................................51
3.15 Planet Format .........................................................................................................................................51
3.15.1 DTM File............................................................................................................................................51
3.15.1.1 Description ..................................................................................................................................51
3.15.1.2 Sample ........................................................................................................................................52
3.15.2 Clutter Class Files.............................................................................................................................52
3.15.2.1 Description ..................................................................................................................................52
3.15.2.2 Sample ........................................................................................................................................52
3.15.3 Vector Files .......................................................................................................................................53
3.15.3.1 Description ..................................................................................................................................53
3.15.3.2 Sample ........................................................................................................................................53
3.15.4 Image Files........................................................................................................................................53
3.15.5 Text Data Files..................................................................................................................................54
3.16 MNU Format ...........................................................................................................................................54
3.16.1 Description........................................................................................................................................54
3.16.2 Sample..............................................................................................................................................54
3.17 XML Table Export/Import Format ...........................................................................................................55
3.17.1 Index.xml File....................................................................................................................................55
3.17.2 XML File............................................................................................................................................55
3.18 Externalised Propagation Results Format ..............................................................................................57
3.18.1 DBF File............................................................................................................................................57
3.18.1.1 DBF File Format ..........................................................................................................................57
3.18.1.1.1 DBF Structure........................................................................................................................57
3.18.1.1.2 DBF Header (Variable Size - Depends on Field Count) ........................................................57
3.18.1.1.3 Each DBF Record (Fixed Length)..........................................................................................59
3.18.1.2 DBF File Content .........................................................................................................................59
3.18.2 LOS File............................................................................................................................................60
3.19 Externalised Tuning Files .......................................................................................................................60
3.19.1 DBF File............................................................................................................................................60
3.19.1.1 DBF File Format ..........................................................................................................................60
3.19.1.1.1 DBF Structure........................................................................................................................60
Table of Contents
© Forsk 2010 AT283_TRG_E2 7
3.19.1.1.2 DBF Header (Variable Size - Depends on Field Count)........................................................ 60
3.19.1.1.3 Each DBF Record (Fixed Length) ......................................................................................... 62
3.19.1.2 DBF File Content ........................................................................................................................ 62
3.19.2 PTS File............................................................................................................................................ 62
3.20 Interference Histograms File Formats.................................................................................................... 63
3.20.1 One Histogram per Line (.im0) Format............................................................................................. 63
3.20.1.1 Sample........................................................................................................................................ 63
3.20.2 One Value per Line with Dictionary File (.clc) Format ...................................................................... 64
3.20.2.1 CLC File...................................................................................................................................... 64
3.20.2.1.1 Description............................................................................................................................ 64
3.20.2.1.2 Sample.................................................................................................................................. 65
3.20.2.2 DCT File...................................................................................................................................... 66
3.20.2.2.1 Description............................................................................................................................ 66
3.20.2.2.2 Sample.................................................................................................................................. 66
3.20.3 One Value per Line (Transmitter Name Repeated) (.im1) Format ................................................... 67
3.20.3.1 Sample........................................................................................................................................ 67
3.20.4 Only Co-Channel and Adjacent Values (.im2) Format ..................................................................... 68
3.20.4.1 Sample........................................................................................................................................ 68
3.21 Antenna Pattern Formats....................................................................................................................... 69
3.21.1 2D Antenna Diagram Format ........................................................................................................... 69
3.21.2 Import Format of Text Files Containing 3D Antenna Patterns.......................................................... 70
4 Calculations .................................................................................... 73
4.1 Overview................................................................................................................................................ 73
4.2 Path Loss Matrices................................................................................................................................. 74
4.2.1 Calculation Area Determination........................................................................................................ 75
4.2.1.1 Computation Zone ...................................................................................................................... 75
4.2.2 Calculate / Force Calculation Comparison....................................................................................... 75
4.2.2.1 Calculate..................................................................................................................................... 75
4.2.2.2 Force Calculation........................................................................................................................ 76
4.2.3 Matrix Validity ................................................................................................................................... 76
4.3 Path Loss Calculations........................................................................................................................... 77
4.3.1 Ground Altitude Determination......................................................................................................... 77
4.3.2 Clutter Determination ....................................................................................................................... 77
4.3.2.1 Clutter Class ............................................................................................................................... 78
4.3.2.2 Clutter Height.............................................................................................................................. 78
4.3.3 Geographic Profile Extraction........................................................................................................... 78
4.3.3.1 Extraction Methods..................................................................................................................... 78
4.3.3.1.1 Radial Extraction................................................................................................................... 78
4.3.3.1.2 Systematic Extraction ........................................................................................................... 79
4.3.3.2 Profile Resolution: Multi-Resolution Management...................................................................... 80
4.4 Coverage Predictions............................................................................................................................. 81
4.4.1 Use of Polygonal Zones in Coverage Prediction Reports ................................................................ 81
4.4.2 Filtering Coverage Prediction Exports.............................................................................................. 81
4.4.3 Smoothing Coverage Prediction Exports ......................................................................................... 81
4.4.3.1 Smoothing: Percentage Method ................................................................................................. 81
4.4.3.2 Smoothing: Number of points method ........................................................................................ 82
4.5 Propagation Models ............................................................................................................................... 84
4.5.1 Okumura-Hata and Cost-Hata Propagation Models......................................................................... 85
4.5.1.1 Hata Path Loss Formula............................................................................................................. 85
4.5.1.2 Corrections to the Hata Path Loss Formula................................................................................ 85
4.5.1.3 Calculations in Atoll .................................................................................................................... 85
4.5.2 ITU 529-3 Propagation Model .......................................................................................................... 86
4.5.2.1 ITU 529-3 Path Loss Formula..................................................................................................... 86
4.5.2.2 Corrections to the ITU 529-3 Path Loss Formula ....................................................................... 86
4.5.2.2.1 Environment Correction ........................................................................................................ 86
4.5.2.2.2 Area Size Correction............................................................................................................. 86
4.5.2.2.3 Distance Correction .............................................................................................................. 87
4.5.2.3 Calculations in Atoll .................................................................................................................... 87
4.5.3 Standard Propagation Model (SPM) ................................................................................................ 87
4.5.3.1 SPM Path Loss Formula............................................................................................................. 87
4.5.3.2 Calculations in Atoll .................................................................................................................... 88
4.5.3.2.1 Visibility and Distance Between Transmitter and Receiver................................................... 88
4.5.3.2.2 Effective Transmitter Antenna Height ................................................................................... 88
4.5.3.2.3 Effective Receiver Antenna Height ....................................................................................... 91
4.5.3.2.4 Correction for Hilly Regions in Case of LOS......................................................................... 91
4.5.3.2.5 Diffraction.............................................................................................................................. 92
4.5.3.2.6 Losses due to Clutter ............................................................................................................ 92
Technical Reference Guide
8 AT283_TRG_E2 © Forsk 2010
4.5.3.2.7 Recommendations.................................................................................................................93
4.5.3.3 Automatic SPM Calibration .........................................................................................................93
4.5.3.3.1 General Algorithm..................................................................................................................94
4.5.3.3.2 Sample Values for SPM Path Loss Formula Parameters......................................................94
4.5.3.4 Unmasked Path Loss Calculation ...............................................................................................95
4.5.4 WLL Propagation Model....................................................................................................................96
4.5.4.1 WLL Path Loss Formula..............................................................................................................96
4.5.4.2 Calculations in Atoll .....................................................................................................................96
4.5.4.2.1 Free Space Loss....................................................................................................................96
4.5.4.2.2 Diffraction...............................................................................................................................96
4.5.5 ITU-R P.526-5 Propagation Model....................................................................................................96
4.5.5.1 ITU 526-5 Path Loss Formula .....................................................................................................96
4.5.5.2 Calculations in Atoll .....................................................................................................................97
4.5.5.2.1 Free Space Loss....................................................................................................................97
4.5.5.2.2 Diffraction...............................................................................................................................97
4.5.6 ITU-R P.370-7 Propagation Model....................................................................................................97
4.5.6.1 ITU 370-7 Path Loss Formula .....................................................................................................97
4.5.6.2 Calculations in Atoll .....................................................................................................................97
4.5.6.2.1 Free Space Loss....................................................................................................................97
4.5.6.2.2 Corrected Standard Loss.......................................................................................................97
4.5.7 Erceg-Greenstein (SUI) Propagation Model .....................................................................................98
4.5.7.1 SUI Terrain Types .......................................................................................................................99
4.5.7.2 Erceg-Greenstein (SUI) Path Loss Formula................................................................................99
4.5.7.3 Calculations in Atoll ...................................................................................................................100
4.5.8 ITU-R P.1546-2 Propagation Model................................................................................................100
4.5.8.1 Calculations in Atoll ...................................................................................................................101
4.5.8.1.1 Step 1: Determination of Graphs to be Used.......................................................................101
4.5.8.1.2 Step 2: Calculation of Maximum Field Strength...................................................................101
4.5.8.1.3 Step 3: Determination of Transmitter Antenna Height .........................................................101
4.5.8.1.4 Step 4: Interpolation/Extrapolation of Field Strength ...........................................................101
4.5.8.1.5 Step 5: Calculation of Correction Factors............................................................................103
4.5.8.1.6 Step 6: Calculation of Path Loss..........................................................................................104
4.5.9 Sakagami Extended Propagation Model.........................................................................................104
4.5.10 Appendices .....................................................................................................................................106
4.5.10.1 Free Space Loss .......................................................................................................................106
4.5.10.2 Diffraction Loss..........................................................................................................................106
4.5.10.2.1 Knife-Edge Diffraction..........................................................................................................106
4.5.10.2.2 3 Knife-Edge Deygout Method.............................................................................................107
4.5.10.2.3 Epstein-Peterson Method ....................................................................................................108
4.5.10.2.4 Deygout Method with Correction .........................................................................................108
4.5.10.2.5 Millington Method.................................................................................................................109
4.6 Path Loss Tuning..................................................................................................................................109
4.6.1 Transmitter Path Loss Tuning.........................................................................................................109
4.6.2 Repeater Path Loss Tuning ............................................................................................................110
4.7 Antenna Attenuation Calculation ..........................................................................................................111
4.7.1 Calculation of Azimuth and Tilt Angles............................................................................................111
4.7.2 Antenna Pattern 3-D Interpolation...................................................................................................112
4.7.3 Additional Electrical Downtilt Modelling...........................................................................................113
4.7.4 Antenna Pattern Smoothing............................................................................................................113
4.7.4.1 Smoothing Algorithm.................................................................................................................115
4.8 Shadowing Model .................................................................................................................................115
4.8.1 Shadowing Margin Calculation........................................................................................................119
4.8.1.1 Shadowing Margin Calculation in Predictions ...........................................................................120
4.8.1.2 Shadowing Margin Calculation in Monte-Carlo Simulations......................................................121
4.8.2 Macro-Diversity Gains Calculation..................................................................................................122
4.8.2.1 Uplink Macro-Diversity Gain Evaluation....................................................................................122
4.8.2.1.1 Shadowing Error PDF (n Signals)........................................................................................122
4.8.2.1.2 Uplink Macro-Diversity Gain ................................................................................................124
4.8.2.2 Downlink Macro-Diversity Gain Evaluation ...............................................................................124
4.8.2.2.1 Shadowing Error PDF (n Signals)........................................................................................124
4.8.2.2.2 Downlink Macro-Diversity Gain............................................................................................127
4.9 Appendices...........................................................................................................................................127
4.9.1 Transmitter Radio Equipment .........................................................................................................127
4.9.1.1 UMTS, CDMA2000, TD-SCDMA, WiMAX, and LTE Documents..............................................128
4.9.1.2 GSM Documents .......................................................................................................................129
4.9.2 Secondary Antennas.......................................................................................................................129
Table of Contents
© Forsk 2010 AT283_TRG_E2 9
5 GSM GPRS EDGE Networks ....................................................... 133
5.1 Signal Level Calculations..................................................................................................................... 133
5.1.1 Point Analysis................................................................................................................................. 133
5.1.1.1 Profile Tab ................................................................................................................................ 133
5.1.1.2 Reception Tab .......................................................................................................................... 133
5.1.2 Signal Level-based Coverage Predictions ..................................................................................... 134
5.1.2.1 Service Area Determination...................................................................................................... 134
5.1.2.1.1 All Servers........................................................................................................................... 134
5.1.2.1.2 Best Signal Level and a Margin .......................................................................................... 134
5.1.2.1.3 Second Best Signal Level and a Margin ............................................................................. 134
5.1.2.1.4 Best Signal Level per HCS Layer and a Margin.................................................................. 135
5.1.2.1.5 Second Best Signal Level per HCS Layer and a Margin .................................................... 135
5.1.2.1.6 HCS Servers and a Margin ................................................................................................. 135
5.1.2.1.7 Highest Priority HCS Server and a Margin.......................................................................... 136
5.1.2.1.8 Best Idle Mode Reselection Criterion (C2).......................................................................... 136
5.1.2.2 Coverage Display ..................................................................................................................... 137
5.1.2.2.1 Coverage Resolution .......................................................................................................... 137
5.1.2.2.2 Display Types ..................................................................................................................... 137
5.2 Interference-based Calculations .......................................................................................................... 138
5.2.1 Carrier-to-Interference Ratio Calculation........................................................................................ 138
5.2.2 Point Analysis................................................................................................................................. 141
5.2.3 Interference-based Coverage Predictions...................................................................................... 141
5.2.3.1 Service Area Determination...................................................................................................... 141
5.2.3.2 Coverage Area Determination .................................................................................................. 141
5.2.3.2.1 Interference Condition Satisfied by At Least One TRX....................................................... 141
5.2.3.2.2 Interference Condition Satisfied by The Worst TRX ........................................................... 142
5.2.3.3 Coverage Display ..................................................................................................................... 142
5.2.3.3.1 Coverage Resolution .......................................................................................................... 142
5.2.3.3.2 Display Types ..................................................................................................................... 142
5.3 GPRS/EDGE Calculations ................................................................................................................... 142
5.3.1 Coding Scheme Selection and Throughput Calculation Without Ideal Link Adaptation................. 143
5.3.1.1 Calculations Based on C .......................................................................................................... 143
5.3.1.2 Calculations Based on C/I ........................................................................................................ 143
5.3.1.3 Calculations Based on C/(I+N) ................................................................................................. 144
5.3.2 Coding Scheme Selection and Throughput Calculation With Ideal Link Adaptation...................... 144
5.3.2.1 Calculations Based on C .......................................................................................................... 144
5.3.2.2 Calculations Based on C/I ........................................................................................................ 145
5.3.2.3 Calculations Based on C/(I+N) ................................................................................................. 145
5.3.3 Application Throughput Calculation................................................................................................ 145
5.3.4 BLER Calculation ........................................................................................................................... 146
5.3.5 GPRS/EDGE Coverage Predictions............................................................................................... 146
5.3.5.1 Service Area Determination...................................................................................................... 146
5.3.5.1.1 All Servers........................................................................................................................... 146
5.3.5.1.2 Best Signal Level and a Margin .......................................................................................... 146
5.3.5.1.3 Second Best Signal Level and a Margin ............................................................................. 146
5.3.5.1.4 Best Signal Level per HCS Layer and a Margin.................................................................. 147
5.3.5.1.5 Second Best Signal Level per HCS Layer and a Margin .................................................... 147
5.3.5.1.6 HCS Servers and a Margin ................................................................................................. 147
5.3.5.1.7 Highest Priority HCS Server and a Margin.......................................................................... 147
5.3.5.1.8 Best Idle Mode Reselection Criterion (C2).......................................................................... 148
5.3.5.2 Coverage Display ..................................................................................................................... 148
5.3.5.2.1 Coverage Resolution .......................................................................................................... 148
5.3.5.2.2 Display Types ..................................................................................................................... 148
5.4 Codec Mode Selection and CQI Calculations...................................................................................... 150
5.4.1 Circuit Quality Indicator Calculations.............................................................................................. 152
5.4.2 CQI Calculation Without Ideal Link Adaptation .............................................................................. 152
5.4.2.1 Calculations Based on C/N....................................................................................................... 152
5.4.2.2 Calculations Based on C/(I+N) ................................................................................................. 152
5.4.3 CQI Calculation With Ideal Link Adaptation ................................................................................... 153
5.4.3.1 Calculations Based on C/N....................................................................................................... 153
5.4.3.2 Calculations Based on C/(I+N) ................................................................................................. 153
5.4.4 Circuit Quality Indicators Coverage Predictions ............................................................................. 154
5.4.4.1 Service Area Determination...................................................................................................... 154
5.4.4.1.1 All Servers........................................................................................................................... 154
5.4.4.1.2 Best Signal Level and a Margin .......................................................................................... 154
5.4.4.1.3 Second Best Signal Level and a Margin ............................................................................. 154
5.4.4.1.4 Best Signal Level per HCS Layer and a Margin.................................................................. 154
Technical Reference Guide
10 AT283_TRG_E2 © Forsk 2010
5.4.4.1.5 Second Best Signal Level per HCS Layer and a Margin.....................................................155
5.4.4.1.6 HCS Servers and a Margin..................................................................................................155
5.4.4.1.7 Highest Priority HCS Server and a Margin ..........................................................................155
5.4.4.2 Coverage Display ......................................................................................................................156
5.4.4.2.1 Coverage Resolution ...........................................................................................................156
5.4.4.2.2 Display Types ......................................................................................................................156
5.5 Traffic Analysis .....................................................................................................................................156
5.5.1 Traffic Distribution...........................................................................................................................156
5.5.1.1 Normal Cells (Nonconcentric, No HCS Layer) ..........................................................................156
5.5.1.1.1 Circuit Switched Services ....................................................................................................156
5.5.1.1.2 Packet Switched Services ...................................................................................................157
5.5.1.2 Concentric Cells ........................................................................................................................157
5.5.1.2.1 Circuit Switched Services ....................................................................................................157
5.5.1.2.2 Packet Switched Services ...................................................................................................157
5.5.1.3 HCS Layers ...............................................................................................................................157
5.5.1.3.1 Circuit Switched Services ....................................................................................................157
5.5.1.3.2 Packet Switched Services ...................................................................................................157
5.5.2 Calculation of the Traffic Demand per Subcell................................................................................157
5.5.2.1 User Profile Traffic Maps...........................................................................................................157
5.5.2.1.1 Normal Cells (Nonconcentric, No HCS Layer).....................................................................157
5.5.2.1.2 Concentric Cells...................................................................................................................158
5.5.2.1.3 HCS Layers .........................................................................................................................159
5.5.2.2 Sector Traffic Maps ...................................................................................................................162
5.5.2.2.1 Normal Cells (Nonconcentric, No HCS Layer).....................................................................162
5.5.2.2.2 Concentric Cells...................................................................................................................163
5.5.2.2.3 HCS Layers .........................................................................................................................163
5.6 Network Dimensioning..........................................................................................................................167
5.6.1 Dimensioning Models and Quality Graphs......................................................................................167
5.6.1.1 Circuit Switched Traffic..............................................................................................................168
5.6.1.2 Packet Switched Traffic.............................................................................................................168
5.6.1.2.1 Throughput ..........................................................................................................................168
5.6.1.2.2 Delay....................................................................................................................................170
5.6.1.2.3 Blocking Probability .............................................................................................................170
5.6.2 Network Dimensioning Process ......................................................................................................172
5.6.2.1 Network Dimensioning Engine ..................................................................................................172
5.6.2.1.1 Inputs...................................................................................................................................172
5.6.2.1.2 Outputs ................................................................................................................................172
5.6.2.2 Network Dimensioning Steps ....................................................................................................172
5.6.2.2.1 Step 1: Timeslots Required for CS Traffic ...........................................................................172
5.6.2.2.2 Step 2: TRXs Required for CS Traffic and Dedicated PS Timeslots ...................................173
5.6.2.2.3 Step 3: Effective CS Blocking, Effective CS Traffic Overflow and Served CS Traffic..........173
5.6.2.2.4 Step 4: TRXs to Add for PS Traffic......................................................................................173
5.6.2.2.5 Step 5: Served PS Traffic ....................................................................................................175
5.6.2.2.6 Step 6: Total Traffic Load ....................................................................................................176
5.7 Key Performance Indicators Calculation...............................................................................................176
5.7.1 Circuit Switched Traffic ...................................................................................................................176
5.7.1.1 Erlang B.....................................................................................................................................176
5.7.1.2 Erlang C ....................................................................................................................................176
5.7.1.3 Served Circuit Switched Traffic .................................................................................................177
5.7.2 Packet Switched Traffic...................................................................................................................177
5.7.2.1 Case 1: Total Traffic Demand > Dedicated + Shared Timeslots ...............................................177
5.7.2.1.1 Traffic Load..........................................................................................................................177
5.7.2.1.2 Packet Switched Traffic Overflow........................................................................................177
5.7.2.1.3 Throughput Reduction Factor ..............................................................................................177
5.7.2.1.4 Delay....................................................................................................................................177
5.7.2.1.5 Blocking Probability .............................................................................................................177
5.7.2.1.6 Served Packet Switched Traffic...........................................................................................177
5.7.2.2 Case 2: Total Traffic Demand < Dedicated + Shared Timeslots ...............................................178
5.7.2.2.1 Traffic Load..........................................................................................................................178
5.7.2.2.2 Packet Switched Traffic Overflow........................................................................................178
5.7.2.2.3 Throughput Reduction Factor ..............................................................................................178
5.7.2.2.4 Delay....................................................................................................................................178
5.7.2.2.5 Blocking Probability .............................................................................................................178
5.7.2.2.6 Served Packet Switched Traffic...........................................................................................178
5.8 Neighbour Allocation.............................................................................................................................178
5.8.1 Global Allocation for All Transmitters..............................................................................................179
5.8.2 Allocation for a Group of Transmitters or One Transmitter .............................................................182
Table of Contents
© Forsk 2010 AT283_TRG_E2 11
5.9 AFP Appendices .................................................................................................................................. 182
5.9.1 The AFP Cost Function.................................................................................................................. 182
5.9.1.1 Cost Function............................................................................................................................ 183
5.9.1.2 Cost Components..................................................................................................................... 184
5.9.1.2.1 Separation Violation Cost Component ................................................................................ 184
5.9.1.2.2 Interference Cost Component ............................................................................................. 185
5.9.1.2.3 I_DIV, F_DIV and Other Advanced Cost Parameters......................................................... 187
5.9.2 The AFP Blocked Traffic Cost ........................................................................................................ 187
5.9.2.1 Calculation of New Traffic Loads Including Blocked Traffic Loads........................................... 188
5.9.2.2 Recalculation of CS and PS From Traffic Loads ...................................................................... 189
5.9.2.3 Testing the Blocked Cost Using Traffic Analysis ...................................................................... 190
5.9.3 Interferences .................................................................................................................................. 190
5.9.3.1 Using Interferences................................................................................................................... 190
5.9.3.2 Cumulative Density Function of C/I Levels............................................................................... 190
5.9.3.3 Precise Definition...................................................................................................................... 191
5.9.3.4 Precise Interference Distribution Strategy ................................................................................ 191
5.9.3.4.1 Direct Availability of Precise Interference Distribution to the AFP....................................... 191
5.9.3.4.2 Efficient Calculation and Storage of Interference Distribution............................................. 191
5.9.3.4.3 Robustness of the IM.......................................................................................................... 191
5.9.3.5 Traffic Load and Interference Information Discrimination......................................................... 191
6 UMTS HSPA Networks ................................................................. 195
6.1 General Prediction Studies .................................................................................................................. 195
6.1.1 Calculation Criteria......................................................................................................................... 195
6.1.2 Point Analysis................................................................................................................................. 195
6.1.2.1 Profile Tab ................................................................................................................................ 195
6.1.2.2 Reception Tab .......................................................................................................................... 195
6.1.3 Coverage Studies........................................................................................................................... 196
6.1.3.1 Service Area Determination...................................................................................................... 196
6.1.3.1.1 All Servers........................................................................................................................... 196
6.1.3.1.2 Best Signal Level and a Margin .......................................................................................... 196
6.1.3.1.3 Second Best Signal Level and a Margin ............................................................................. 196
6.1.3.2 Coverage Display ..................................................................................................................... 197
6.1.3.2.1 Plot Resolution.................................................................................................................... 197
6.1.3.2.2 Display Types ..................................................................................................................... 197
6.2 Definitions and Formulas ..................................................................................................................... 198
6.2.1 Inputs.............................................................................................................................................. 198
6.2.2 Ec/I0 Calculation ............................................................................................................................ 204
6.2.3 DL Eb/Nt Calculation...................................................................................................................... 204
6.2.4 UL Eb/Nt Calculation...................................................................................................................... 205
6.3 Active Set Management ....................................................................................................................... 206
6.4 Simulations........................................................................................................................................... 206
6.4.1 Generating a Realistic User Distribution ........................................................................................ 207
6.4.1.1 Simulations Based on User Profile Traffic Maps ...................................................................... 207
6.4.1.1.1 Circuit Switched Service (i) ................................................................................................. 207
6.4.1.1.2 Packet Switched Service (j) ................................................................................................ 208
6.4.1.2 Simulations Based on Sector Traffic Maps............................................................................... 210
6.4.1.2.1 Throughputs in Uplink and Downlink................................................................................... 211
6.4.1.2.2 Total Number of Users (All Activity Statuses) ..................................................................... 211
6.4.1.2.3 Number of Users per Activity Status ................................................................................... 212
6.4.2 Power Control Simulation............................................................................................................... 212
6.4.2.1 Algorithm Initialization............................................................................................................... 213
6.4.2.2 R99 Part of the Algorithm......................................................................................................... 214
6.4.2.3 HSDPA Part of the Algorithm.................................................................................................... 218
6.4.2.3.1 HSDPA Power Allocation.................................................................................................... 218
6.4.2.3.2 Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users............. 219
6.4.2.3.3 HSDPA Bearer Allocation Process ..................................................................................... 219
6.4.2.3.4 Fast Link Adaptation Modelling........................................................................................... 221
6.4.2.3.5 MIMO Modelling.................................................................................................................. 231
6.4.2.3.6 Scheduling Algorithms ........................................................................................................ 231
6.4.2.3.7 Dual-Cell HSDPA................................................................................................................ 233
6.4.2.4 HSUPA Part of the Algorithm.................................................................................................... 236
6.4.2.4.1 Admission Control ............................................................................................................... 237
6.4.2.4.2 HSUPA Bearer Allocation Process ..................................................................................... 239
6.4.2.4.3 Noise Rise Scheduling........................................................................................................ 240
6.4.2.4.4 Radio Resource Control...................................................................................................... 244
6.4.2.5 Convergence Criteria................................................................................................................ 244
Technical Reference Guide
12 AT283_TRG_E2 © Forsk 2010
6.4.3 Results ............................................................................................................................................244
6.4.3.1 R99 Related Results .................................................................................................................244
6.4.3.2 HSPA Related Results ..............................................................................................................246
6.4.3.2.1 Statistics Tab .......................................................................................................................246
6.4.3.2.2 Mobiles Tab .........................................................................................................................247
6.4.3.2.3 Cells Tab..............................................................................................................................250
6.4.3.2.4 Sites Tab..............................................................................................................................252
6.4.4 Appendices .....................................................................................................................................252
6.4.4.1 Admission Control in the R99 Part ............................................................................................252
6.4.4.2 Resources Management ...........................................................................................................252
6.4.4.2.1 OVSF Codes Management..................................................................................................252
6.4.4.2.2 Channel Elements Management .........................................................................................254
6.4.4.2.3 Iub Backhaul Throughput.....................................................................................................254
6.4.4.3 Downlink Load Factor Calculation.............................................................................................255
6.4.4.3.1 Downlink Load Factor per Cell.............................................................................................255
6.4.4.3.2 Downlink Load Factor per Mobile ........................................................................................257
6.4.4.4 Uplink Load Factor Due to One User ........................................................................................257
6.4.4.5 Inter-carrier Power Sharing Modelling.......................................................................................259
6.4.4.6 Best Server Determination in Monte Carlo Simulations - Old Method ......................................259
6.5 UMTS HSPA Prediction Studies...........................................................................................................261
6.5.1 Point Analysis..................................................................................................................................261
6.5.1.1 AS Analysis Tab........................................................................................................................261
6.5.1.1.1 Bar Graph and Pilot Sub-Menu............................................................................................261
6.5.1.1.2 Downlink Sub-Menu.............................................................................................................264
6.5.1.1.3 Uplink Sub-Menu .................................................................................................................268
6.5.2 Coverage Studies............................................................................................................................272
6.5.2.1 Pilot Reception Analysis............................................................................................................272
6.5.2.1.1 Prediction Study Inputs........................................................................................................273
6.5.2.1.2 Study Display Options .........................................................................................................273
6.5.2.2 Downlink Service Area Analysis................................................................................................273
6.5.2.2.1 Prediction Study Inputs........................................................................................................274
6.5.2.2.2 Study Display Options .........................................................................................................274
6.5.2.3 Uplink Service Area Analysis ....................................................................................................275
6.5.2.3.1 Prediction Study Inputs........................................................................................................275
6.5.2.3.2 Study Display Options .........................................................................................................276
6.5.2.4 Downlink Total Noise Analysis ..................................................................................................277
6.5.2.4.1 Study Inputs.........................................................................................................................277
6.5.2.4.2 Analysis on All Carriers........................................................................................................277
6.5.2.4.3 Analysis on a Specific Carrier..............................................................................................278
6.5.2.5 HSDPA Prediction Study...........................................................................................................278
6.5.2.5.1 Prediction Study Inputs........................................................................................................278
6.5.2.5.2 Study Display Options .........................................................................................................278
6.5.2.6 HSUPA Prediction Study...........................................................................................................282
6.5.2.6.1 Prediction Study Inputs........................................................................................................282
6.5.2.6.2 Calculation Options..............................................................................................................283
6.5.2.6.3 Display Options....................................................................................................................283
6.6 Automatic Neighbour Allocation............................................................................................................285
6.6.1 Neighbour Allocation for All Transmitters........................................................................................285
6.6.2 Neighbour Allocation for a Group of Transmitters or One Transmitter............................................289
6.6.3 Importance Calculation ...................................................................................................................289
6.6.3.1 Importance of Intra-carrier Neighbours .....................................................................................289
6.6.3.2 Importance of Inter-carrier Neighbours .....................................................................................290
6.7 Primary Scrambling Code Allocation ....................................................................................................291
6.7.1 Automatic Allocation Description.....................................................................................................291
6.7.1.1 Options and Constraints............................................................................................................291
6.7.1.2 Allocation Process.....................................................................................................................292
6.7.1.2.1 Single Carrier Network.........................................................................................................293
6.7.1.2.2 Multi-Carrier Network...........................................................................................................293
6.7.1.3 Priority Determination................................................................................................................294
6.7.1.3.1 Cell Priority ..........................................................................................................................294
6.7.1.3.2 Transmitter Priority ..............................................................................................................296
6.7.1.3.3 Site Priority ..........................................................................................................................296
6.7.2 Allocation Examples........................................................................................................................296
6.7.2.1 Allocation Strategies and Use a Maximum of Codes ................................................................296
6.7.2.1.1 Strategy: Clustered..............................................................................................................297
6.7.2.1.2 Strategy: Distributed ............................................................................................................298
6.7.2.1.3 Strategy: ‘One Cluster per Site............................................................................................298
6.7.2.1.4 Strategy: ‘Distributed per Site..............................................................................................299
Table of Contents
© Forsk 2010 AT283_TRG_E2 13
6.7.2.2 Allocate Carriers Identically ...................................................................................................... 299
6.8 Automatic GSM-UMTS Neighbour Allocation ...................................................................................... 300
6.8.1 Overview ........................................................................................................................................ 300
6.8.2 Automatic Allocation Description.................................................................................................... 300
6.8.2.1 Algorithm Based on Distance ................................................................................................... 300
6.8.2.2 Algorithm Based on Coverage Overlapping ............................................................................. 301
6.8.2.3 Appendices............................................................................................................................... 303
6.8.2.3.1 Delete Existing Neighbours Option ..................................................................................... 303
6.8.2.3.2 Calculation of Inter-Transmitter Distance............................................................................ 303
7 CDMA2000 Networks ................................................................... 307
7.1 General Prediction Studies .................................................................................................................. 307
7.1.1 Calculation Criteria......................................................................................................................... 307
7.1.2 Point Analysis................................................................................................................................. 307
7.1.2.1 Profile Tab ................................................................................................................................ 307
7.1.2.2 Reception Tab .......................................................................................................................... 308
7.1.3 Coverage Studies........................................................................................................................... 308
7.1.3.1 Service Area Determination...................................................................................................... 308
7.1.3.1.1 All Servers........................................................................................................................... 308
7.1.3.1.2 Best Signal Level and a Margin .......................................................................................... 308
7.1.3.1.3 Second Best Signal Level and a Margin ............................................................................. 308
7.1.3.2 Coverage Display ..................................................................................................................... 309
7.1.3.2.1 Plot Resolution.................................................................................................................... 309
7.1.3.2.2 Display Types ..................................................................................................................... 309
7.2 Definitions and Formulas ..................................................................................................................... 310
7.2.1 Parameters Used for CDMA2000 1xRTT Modelling ...................................................................... 310
7.2.1.1 Inputs........................................................................................................................................ 310
7.2.1.2 Ec/I0 Calculation....................................................................................................................... 314
7.2.1.3 DL Eb/Nt Calculation ................................................................................................................ 314
7.2.1.4 UL Eb/Nt Calculation ................................................................................................................ 315
7.2.1.5 Simulation Results.................................................................................................................... 317
7.2.2 Parameters Used for CDMA2000 1xEV-DO Modelling.................................................................. 318
7.2.2.1 Inputs........................................................................................................................................ 318
7.2.2.2 Ec/I0 and Ec/Nt Calculations .................................................................................................... 321
7.2.2.3 UL Eb/Nt Calculation ................................................................................................................ 322
7.2.2.4 Simulation Results.................................................................................................................... 323
7.3 Active Set Management ....................................................................................................................... 324
7.4 Simulations........................................................................................................................................... 325
7.4.1 Generating a Realistic User Distribution ........................................................................................ 325
7.4.1.1 Number of Users, User Activity Status and User Data Rate..................................................... 325
7.4.1.1.1 Simulations Based on User Profile Traffic Maps................................................................. 325
7.4.1.1.2 Simulations Based on Sector Traffic Maps ......................................................................... 328
7.4.1.2 Transition Flags for 1xEV-DO Rev.0 User Data Rates............................................................. 333
7.4.1.3 User Geographical Position...................................................................................................... 333
7.4.2 Network Regulation Mechanism..................................................................................................... 333
7.4.2.1 CDMA2000 1xRTT Power Control Simulation Algorithm.......................................................... 333
7.4.2.1.1 Algorithm Initialization ......................................................................................................... 334
7.4.2.1.2 Presentation of the Algorithm.............................................................................................. 334
7.4.2.1.3 Convergence Criterion ........................................................................................................ 340
7.4.2.2 CDMA2000 1xEV-DO Power/Data Rate Control Simulation Algorithm.................................... 341
7.4.2.2.1 Algorithm Initialization ......................................................................................................... 342
7.4.2.2.2 Presentation of the Algorithm.............................................................................................. 342
7.4.2.2.3 Convergence Criterion ........................................................................................................ 347
7.4.3 Appendices..................................................................................................................................... 348
7.4.3.1 Admission Control..................................................................................................................... 348
7.4.3.2 Resources Management........................................................................................................... 348
7.4.3.2.1 Walsh Code Management .................................................................................................. 348
7.4.3.2.2 Channel Element Management .......................................................................................... 349
7.4.3.3 Downlink Load Factor Calculation ............................................................................................ 349
7.4.3.3.1 Downlink Load Factor per Cell ............................................................................................ 349
7.4.3.3.2 Downlink Load Factor per Mobile........................................................................................ 351
7.4.3.4 Best Server Determination in Monte Carlo Simulations - Old Method...................................... 351
7.5 CDMA2000 Prediction Studies............................................................................................................. 353
7.5.1 Point Analysis: The AS Analysis Tab ............................................................................................. 353
7.5.1.1 Bar Graph and Pilot Sub-Menu................................................................................................. 353
7.5.1.2 Downlink Sub-Menu.................................................................................................................. 355
7.5.1.2.1 CDMA2000 1xRTT.............................................................................................................. 355
7.5.1.2.2 CDMA2000 1xEV-DO......................................................................................................... 359
Technical Reference Guide
14 AT283_TRG_E2 © Forsk 2010
7.5.1.3 Uplink Sub-Menu.......................................................................................................................360
7.5.1.3.1 CDMA2000 1xRTT ..............................................................................................................360
7.5.1.3.2 CDMA2000 1xEV-DO..........................................................................................................364
7.5.2 Coverage Studies............................................................................................................................367
7.5.2.1 Pilot Reception Analysis............................................................................................................367
7.5.2.2 Downlink Service Area Analysis................................................................................................368
7.5.2.2.1 CDMA2000 1xRTT ..............................................................................................................368
7.5.2.2.2 CDMA2000 1xEV-DO..........................................................................................................370
7.5.2.3 Uplink Service Area Analysis ....................................................................................................371
7.5.2.3.1 CDMA2000 1xRTT ..............................................................................................................371
7.5.2.3.2 CDMA2000 1xEV-DO..........................................................................................................372
7.5.2.4 Downlink Total Noise Analysis ..................................................................................................375
7.5.2.4.1 Analysis on all Carriers........................................................................................................375
7.5.2.4.2 Analysis on a Specific Carrier..............................................................................................376
7.6 Automatic Neighbour Allocation............................................................................................................376
7.6.1 Neighbour Allocation for all Transmitters ........................................................................................376
7.6.2 Neighbour Allocation for a Group of Transmitters or One Transmitter............................................379
7.6.3 Importance Calculation ...................................................................................................................379
7.6.3.1 Importance of Intra-carrier Neighbours .....................................................................................379
7.6.3.2 Importance of Inter-carrier Neighbours .....................................................................................380
7.7 PN Offset Allocation..............................................................................................................................381
7.7.1 Automatic Allocation Description.....................................................................................................381
7.7.1.1 Options and Constraints............................................................................................................381
7.7.1.2 Allocation Process.....................................................................................................................382
7.7.1.2.1 Single Carrier Network.........................................................................................................382
7.7.1.2.2 Multi-Carrier Network...........................................................................................................383
7.7.1.2.3 Difference between Adjacent and Distributed PN-Clusters .................................................383
7.7.1.3 Priority Determination................................................................................................................384
7.7.1.3.1 Cell Priority ..........................................................................................................................384
7.7.1.3.2 Transmitter Priority ..............................................................................................................386
7.7.1.3.3 Site Priority ..........................................................................................................................386
7.7.2 Allocation Examples........................................................................................................................386
7.7.2.1 Strategy: PN Offset per Cell ......................................................................................................387
7.7.2.2 Strategy: Adjacent PN-Clusters Per Site...................................................................................387
7.7.2.3 Strategy: ‘Distributed PN-Clusters Per Site...............................................................................388
7.8 Automatic GSM-CDMA Neighbour Allocation.......................................................................................388
7.8.1 Overview.........................................................................................................................................388
7.8.2 Automatic Allocation Description.....................................................................................................388
7.8.2.1 Algorithm Based on Distance....................................................................................................389
7.8.2.2 Algorithm Based on Coverage Overlapping..............................................................................389
7.8.2.3 Delete Existing Neighbours Option ...........................................................................................391
8 TD-SCDMA Networks....................................................................395
8.1 Definitions and Formulas......................................................................................................................395
8.1.1 Inputs ..............................................................................................................................................395
8.1.2 P-CCPCH Eb/Nt and C/I Calculation ..............................................................................................399
8.1.3 DwPCH C/I Calculation...................................................................................................................400
8.1.4 DL TCH Eb/Nt and C/I Calculation..................................................................................................400
8.1.5 UL TCH Eb/Nt and C/I Calculation..................................................................................................400
8.1.6 Interference Calculation..................................................................................................................401
8.1.7 HSDPA Dynamic Power Calculations.............................................................................................401
8.1.8 Smart Antenna Models....................................................................................................................401
8.1.8.1 Downlink Beamforming .............................................................................................................401
8.1.8.2 Uplink Beamforming..................................................................................................................402
8.1.8.3 Uplink Beamforming and Interference Cancellation (MMSE) ....................................................402
8.2 Signal Level Based Calculations ..........................................................................................................403
8.2.1 Point Analysis..................................................................................................................................403
8.2.1.1 Profile Tab.................................................................................................................................403
8.2.1.2 Reception Tab...........................................................................................................................403
8.2.2 RSCP Based Coverage Predictions................................................................................................404
8.2.2.1 Calculation Criteria....................................................................................................................404
8.2.2.2 P-CCPCH RSCP Coverage Prediction .....................................................................................404
8.2.2.2.1 Coverage Condition.............................................................................................................404
8.2.2.2.2 Coverage Display ................................................................................................................405
8.2.2.3 Best Server P-CCPCH Coverage Prediction.............................................................................405
8.2.2.4 P-CCPCH Pollution Coverage Prediction..................................................................................405
8.2.2.5 DwPCH RSCP Coverage Prediction.........................................................................................406
8.2.2.5.1 Coverage Condition.............................................................................................................406
Table of Contents
© Forsk 2010 AT283_TRG_E2 15
8.2.2.5.2 Coverage Display................................................................................................................ 406
8.2.2.6 UpPCH RSCP Coverage Prediction......................................................................................... 406
8.2.2.6.1 Coverage Condition ............................................................................................................ 406
8.2.2.6.2 Coverage Display................................................................................................................ 406
8.2.2.7 Baton Handover Coverage Prediction ...................................................................................... 407
8.2.2.7.1 Coverage Condition ............................................................................................................ 407
8.2.2.7.2 Coverage Display................................................................................................................ 407
8.2.2.8 Scrambling Code Interference Analysis.................................................................................... 407
8.3 Monte Carlo Simulations...................................................................................................................... 408
8.3.1 Generating a Realistic User Distribution ........................................................................................ 408
8.3.1.1 Simulations Based on User Profile Traffic Maps ...................................................................... 408
8.3.1.1.1 Circuit Switched Service (i) ................................................................................................. 409
8.3.1.1.2 Packet Switched Service (j) ................................................................................................ 409
8.3.1.2 Simulations Based on Sector Traffic Maps............................................................................... 412
8.3.1.2.1 Throughputs in Uplink and Downlink................................................................................... 412
8.3.1.2.2 Total Number of Users (All Activity Statuses) ..................................................................... 412
8.3.1.2.3 Number of Users per Activity Status ................................................................................... 413
8.3.2 Power Control Simulation............................................................................................................... 413
8.3.2.1 Algorithm Initialisation............................................................................................................... 414
8.3.2.2 R99 Part of the Algorithm......................................................................................................... 414
8.3.2.2.1 Determination of Mi’s Best Server (SBS(Mi))...................................................................... 414
8.3.2.2.2 Dynamic Channel Allocation ............................................................................................... 415
8.3.2.2.3 Uplink Power Control .......................................................................................................... 416
8.3.2.2.4 Downlink Power Control...................................................................................................... 418
8.3.2.2.5 Uplink Signals Update......................................................................................................... 420
8.3.2.2.6 Downlink Signals Update.................................................................................................... 420
8.3.2.2.7 Control of Radio Resource Limits (Downlink Traffic Power and Uplink Load) .................... 420
8.3.2.3 HSDPA Part of the Algorithm.................................................................................................... 421
8.3.2.3.1 HSDPA Power Allocation.................................................................................................... 421
8.3.2.3.2 Connection Status and Number of HSDPA Users .............................................................. 423
8.3.2.3.3 HSDPA Admission Control.................................................................................................. 423
8.3.2.3.4 HSDPA Dynamic Channel Allocation.................................................................................. 424
8.3.2.3.5 Ressource Unit Saturation .................................................................................................. 424
8.3.2.4 Convergence Criteria................................................................................................................ 424
8.4 TD-SCDMA Prediction Studies ............................................................................................................ 425
8.4.1 P-CCPCH Reception Analysis (Eb/Nt) or (C/I) ............................................................................... 425
8.4.2 DwPCH Reception Analysis (C/I) ................................................................................................... 426
8.4.3 Downlink TCH RSCP Coverage..................................................................................................... 428
8.4.4 Uplink TCH RSCP Coverage ......................................................................................................... 428
8.4.5 Downlink Total Noise...................................................................................................................... 429
8.4.6 Downlink Service Area (Eb/Nt) or (C/I)........................................................................................... 430
8.4.7 Uplink Service Area (Eb/Nt) or (C/I) ............................................................................................... 432
8.4.8 Effective Service Area (Eb/Nt) or (C/I) ........................................................................................... 433
8.4.9 Cell to Cell Interference.................................................................................................................. 434
8.4.10 UpPCH Interference....................................................................................................................... 435
8.4.11 HSDPA Coverage .......................................................................................................................... 435
8.5 Smart Antenna Modelling..................................................................................................................... 436
8.5.1 Modelling in Simulations................................................................................................................. 437
8.5.1.1 Grid of Beams Modelling .......................................................................................................... 437
8.5.1.2 Adaptive Beam Modelling......................................................................................................... 438
8.5.1.3 Statistical Modelling.................................................................................................................. 439
8.5.1.4 Beamforming Smart Antenna Models....................................................................................... 439
8.5.1.4.1 Downlink Beamforming....................................................................................................... 440
8.5.1.4.2 Uplink Beamforming............................................................................................................ 441
8.5.1.4.3 Uplink Beamforming and Interference Cancellation (MMSE).............................................. 442
8.5.1.5 3rd Party Smart Antenna Modelling.......................................................................................... 444
8.5.2 Construction of the Geographic Distributions................................................................................. 444
8.5.3 Modelling in Coverage Predictions................................................................................................. 445
8.5.4 HSDPA Coverage Prediction ......................................................................................................... 446
8.5.4.1 Fast Link Adaptation Modelling................................................................................................. 446
8.5.4.1.1 CQI Based on P-CCPCH Quality........................................................................................ 446
8.5.4.1.2 CQI Based on HS-PDSCH Quality...................................................................................... 450
8.5.4.2 Coverage Prediction Display Options....................................................................................... 451
8.5.4.2.1 Colour per CQI .................................................................................................................... 451
8.5.4.2.2 Colour per Peak Throughput............................................................................................... 451
8.5.4.2.3 Colour per HS-PDSCH Ec/Nt.............................................................................................. 451
8.6 N-Frequency Mode and Carrier Allocation........................................................................................... 452
8.6.1 Automatic Carrier Allocation........................................................................................................... 452
Technical Reference Guide
16 AT283_TRG_E2 © Forsk 2010
8.7 Neighbour Allocation.............................................................................................................................452
8.7.1 Neighbour Allocation for All Transmitters........................................................................................453
8.7.2 Neighbour Allocation for a Group of Transmitters or One Transmitter............................................456
8.7.3 Importance Calculation ...................................................................................................................456
8.8 Scrambling Code Allocation..................................................................................................................457
8.8.1 Automatic Allocation Description.....................................................................................................457
8.8.1.1 Allocation Constraints and Options ...........................................................................................457
8.8.1.2 Allocation Strategies..................................................................................................................458
8.8.1.3 Allocation Process.....................................................................................................................458
8.8.1.3.1 Single Carrier Network.........................................................................................................459
8.8.1.3.2 Multi-Carrier Network...........................................................................................................460
8.8.1.4 Priority Determination................................................................................................................460
8.8.1.4.1 Cell Priority ..........................................................................................................................460
8.8.1.4.2 Transmitter Priority ..............................................................................................................462
8.8.1.4.3 Site Priority ..........................................................................................................................463
8.8.2 Scrambling Code Allocation Example.............................................................................................463
8.8.2.1 Single Carrier Network ..............................................................................................................463
8.8.2.1.1 Strategy: Clustered..............................................................................................................464
8.8.2.1.2 Strategy: Distributed per Cell ...............................................................................................464
8.8.2.1.3 Strategy: One SYNC_DL Code per Site..............................................................................465
8.8.2.1.4 Strategy: Distributed per Site...............................................................................................465
8.8.2.2 Multi Carrier Network.................................................................................................................465
8.9 Automatic GSM/TD-SCDMA Neighbour Allocation ..............................................................................466
8.9.1 Automatic Allocation Description.....................................................................................................466
8.9.1.1 Algorithm Based on Distance....................................................................................................467
8.9.1.2 Algorithm Based on Coverage Overlapping..............................................................................467
8.9.1.3 Appendices................................................................................................................................469
8.9.1.3.1 Delete Existing Neighbours Option......................................................................................469
8.9.1.3.2 Calculation of Inter-Transmitter Distance ............................................................................469
9 WiMAX BWA Networks..................................................................473
9.1 Definitions and Formulas......................................................................................................................473
9.1.1 Input ................................................................................................................................................473
9.1.2 Co- and Adjacent Channel Overlaps Calculation............................................................................477
9.1.3 Preamble Signal Quality Calculations.............................................................................................477
9.1.3.1 Preamble Signal Level Calculation............................................................................................477
9.1.3.2 Preamble Noise Calculation......................................................................................................478
9.1.3.3 Preamble Interference Calculation............................................................................................478
9.1.3.4 Preamble C/N Calculation.........................................................................................................478
9.1.3.5 Preamble C/(I+N) Calculation....................................................................................................478
9.1.4 Traffic and Pilot Signal Quality Calculations ...................................................................................478
9.1.4.1 Traffic and Pilot Signal Level Calculation (DL) ..........................................................................478
9.1.4.2 Traffic and Pilot Noise Calculation (DL) ....................................................................................479
9.1.4.3 Traffic and Pilot Interference Calculation (DL) ..........................................................................479
9.1.4.4 Traffic and Pilot C/N Calculation (DL) .......................................................................................480
9.1.4.5 Traffic and Pilot C/(I+N) Calculation (DL) ..................................................................................481
9.1.4.6 Traffic Signal Level Calculation (UL) .........................................................................................481
9.1.4.7 Traffic Noise Calculation (UL) ...................................................................................................481
9.1.4.8 Traffic Interference Calculation (UL) .........................................................................................482
9.1.4.9 Traffic C/N Calculation (UL) ......................................................................................................482
9.1.4.10 Traffic C/(I+N) Calculation (UL) .................................................................................................482
9.1.5 Throughput Calculation...................................................................................................................482
9.1.5.1 Calculation of Total Cell Resources ..........................................................................................482
9.1.5.2 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation...........484
9.1.6 Scheduling and Radio Resource Management...............................................................................485
9.1.6.1 User Throughput Calculation.....................................................................................................486
9.1.7 Smart Antenna Models....................................................................................................................487
9.1.7.1 Downlink Beamforming .............................................................................................................487
9.1.7.2 Uplink Beamforming..................................................................................................................487
9.1.7.3 Uplink Beamforming and Interference Cancellation (MMSE) ....................................................488
9.2 Calculation Processes ..........................................................................................................................488
9.2.1 Point Analysis: Profile Tab ..............................................................................................................488
9.2.2 Point Analysis: Reception Tab ........................................................................................................488
9.2.3 Point Analysis: Interference Tab .....................................................................................................489
9.2.4 Preamble Signal Level Coverage Predictions.................................................................................489
9.2.4.1 Coverage Area Determination...................................................................................................490
9.2.4.1.1 All Servers ...........................................................................................................................490
9.2.4.1.2 Best Signal Level and a Margin...........................................................................................490
Table of Contents
© Forsk 2010 AT283_TRG_E2 17
9.2.4.1.3 Second Best Signal Level and a Margin ............................................................................. 490
9.2.4.2 Coverage Display ..................................................................................................................... 490
9.2.4.2.1 Coverage Resolution .......................................................................................................... 490
9.2.4.2.2 Display Types ..................................................................................................................... 490
9.2.5 Effective Signal Analysis Coverage Predictions............................................................................. 491
9.2.5.1 Coverage Area Determination .................................................................................................. 492
9.2.5.2 Coverage Parameter Calculation.............................................................................................. 492
9.2.5.3 Coverage Display ..................................................................................................................... 492
9.2.5.3.1 Coverage Resolution .......................................................................................................... 492
9.2.5.3.2 Effective Signal Analysis (DL) Display Types ..................................................................... 492
9.2.5.3.3 Effective Signal Analysis (UL) Display Types ..................................................................... 493
9.2.6 Calculations on Subscriber Lists .................................................................................................... 494
9.2.7 Monte Carlo Simulations ................................................................................................................ 494
9.2.7.1 Generating a Realistic User Distribution................................................................................... 494
9.2.7.1.1 Simulations Based on User Profile Traffic Maps and Subscriber Lists ............................... 495
9.2.7.1.2 Simulations Based on Sector Traffic Maps ......................................................................... 496
9.2.7.2 Simulation Process................................................................................................................... 498
9.2.8 C/(I+N)-Based Coverage Predictions............................................................................................. 502
9.2.8.1 Coverage Area Determination .................................................................................................. 503
9.2.8.2 Coverage Parameter Calculation.............................................................................................. 503
9.2.8.3 Coverage Display ..................................................................................................................... 504
9.2.8.3.1 Coverage Resolution .......................................................................................................... 504
9.2.8.3.2 Coverage by C/(I+N) Level (DL) Display Types.................................................................. 504
9.2.8.3.3 Coverage by Best Bearer (DL) Display Types .................................................................... 505
9.2.8.3.4 Coverage by Throughput (DL) Display Types..................................................................... 505
9.2.8.3.5 Coverage by Quality Indicator (DL) Display Types ............................................................. 506
9.2.8.3.6 Coverage by C/(I+N) Level (UL) Display Types.................................................................. 506
9.2.8.3.7 Coverage by Best Bearer (UL) Display Types .................................................................... 507
9.2.8.3.8 Coverage by Throughput (UL) Display Types..................................................................... 507
9.2.8.3.9 Coverage by Quality Indicator (UL) Display Types ............................................................. 508
9.3 Calculation Algorithms ......................................................................................................................... 508
9.3.1 Co- and Adjacent Channel Overlaps Calculation........................................................................... 508
9.3.1.1 Conversion From Channel Numbers to Start and End Frequencies ........................................ 509
9.3.1.2 Co-Channel Overlap Calculation .............................................................................................. 510
9.3.1.3 Adjacent Channel Overlap Calculation..................................................................................... 510
9.3.1.4 FDD – TDD Overlap Ratio Calculation ..................................................................................... 511
9.3.1.5 Total Overlap Ratio Calculation................................................................................................ 512
9.3.2 Preamble Signal Level and Quality Calculations............................................................................ 512
9.3.2.1 Preamble Signal Level Calculation........................................................................................... 512
9.3.2.2 Preamble Noise Calculation ..................................................................................................... 514
9.3.2.3 Preamble Interference Calculation ........................................................................................... 515
9.3.2.4 Preamble C/N Calculation ........................................................................................................ 516
9.3.2.5 Preamble C/(I+N) Calculation................................................................................................... 517
9.3.3 Best Server Determination ............................................................................................................. 517
9.3.4 Service Area Calculation................................................................................................................ 518
9.3.5 Permutation Zone Selection (WiMAX 802.16e).............................................................................. 519
9.3.6 Traffic and Pilot Signal Level and Quality Calculations .................................................................. 520
9.3.6.1 Traffic and Pilot Signal Level Calculation (DL) ......................................................................... 520
9.3.6.2 Traffic and Pilot Noise Calculation (DL).................................................................................... 521
9.3.6.3 Traffic and Pilot Interference Calculation (DL).......................................................................... 522
9.3.6.3.1 Traffic and Pilot Interference Signal Levels Calculation (DL).............................................. 523
9.3.6.3.2 Effective Traffic and Pilot Interference Calculation (DL) ..................................................... 527
9.3.6.4 Traffic and Pilot C/N Calculation (DL)....................................................................................... 530
9.3.6.5 Traffic and Pilot C/(I+N) and Bearer Calculation (DL) .............................................................. 532
9.3.6.6 Traffic Signal Level Calculation (UL) ........................................................................................ 534
9.3.6.7 Traffic Noise Calculation (UL)................................................................................................... 535
9.3.6.8 Traffic Interference Calculation (UL)......................................................................................... 536
9.3.6.8.1 Traffic Interference Signal Levels Calculation (UL)............................................................. 536
9.3.6.8.2 Noise Rise Calculation (UL) ................................................................................................ 537
9.3.6.9 Traffic C/N Calculation (UL)...................................................................................................... 537
9.3.6.10 Traffic C/(I+N) and Bearer Calculation (UL) ............................................................................. 541
9.3.7 Throughput Calculation .................................................................................................................. 544
9.3.7.1 Calculation of Total Cell Resources.......................................................................................... 544
9.3.7.1.1 Calculation of Sampling Frequency .................................................................................... 544
9.3.7.1.2 Calculation of Symbol Duration........................................................................................... 545
9.3.7.1.3 Calculation of Total Cell Resources - TDD Networks ......................................................... 545
9.3.7.1.4 Calculation of Total Cell Resources - FDD Networks ......................................................... 547
9.3.7.2 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation.......... 547
Technical Reference Guide
18 AT283_TRG_E2 © Forsk 2010
9.3.8 Scheduling and Radio Resource Management...............................................................................551
9.3.8.1 Scheduling and Radio Resource Allocation ..............................................................................551
9.3.8.2 User Throughput Calculation.....................................................................................................557
9.3.9 Smart Antenna Models....................................................................................................................558
9.3.9.1 Downlink Beamforming .............................................................................................................559
9.3.9.2 Uplink Beamforming..................................................................................................................560
9.3.9.3 Uplink Beamforming and Interference Cancellation (MMSE) ....................................................562
9.4 Automatic Allocation Algorithms ...........................................................................................................563
9.4.1 Automatic Neighbour Allocation......................................................................................................563
9.4.2 Automatic Inter-Technology Neighbour Allocation..........................................................................566
9.4.3 Automatic Frequency Planning.......................................................................................................568
9.4.3.1 Separation Constraint and Relationship Weights......................................................................569
9.4.3.2 Calculation of Cost Between TBA and Related Cells................................................................569
9.4.3.3 AFP Algorithm...........................................................................................................................571
9.4.4 Automatic Preamble Index Allocation .............................................................................................571
9.4.4.1 Constraint and Relationship Weights ........................................................................................572
9.4.4.2 Calculation of Cost Between TBA and Related Cells................................................................573
9.4.4.3 Automatic Allocation Algorithm..................................................................................................575
10 LTE Networks ................................................................................579
10.1 Definitions and Formulas......................................................................................................................579
10.1.1 Input ................................................................................................................................................579
10.1.2 Downlink Transmission Powers Calculation ...................................................................................582
10.1.3 Co- and Adjacent Channel Overlaps Calculation............................................................................584
10.1.4 Signal Level and Signal Quality Calculations..................................................................................585
10.1.4.1 Signal Level Calculation (DL) ....................................................................................................585
10.1.4.2 Noise Calculation (DL) ..............................................................................................................586
10.1.4.3 Interference Calculation (DL) ....................................................................................................586
10.1.4.4 C/N Calculation (DL) .................................................................................................................587
10.1.4.5 C/(I+N) Calculation (DL) ............................................................................................................587
10.1.4.6 Signal Level Calculation (UL) ....................................................................................................588
10.1.4.7 Noise Calculation (UL) ..............................................................................................................589
10.1.4.8 Interference Calculation (UL) ....................................................................................................589
10.1.4.9 Noise Rise Calculation (UL) ......................................................................................................589
10.1.4.10 C/N Calculation (UL) .................................................................................................................590
10.1.4.11 C/(I+N) Calculation (UL) ............................................................................................................590
10.1.5 Throughput Calculation...................................................................................................................590
10.1.5.1 Calculation of Downlink Cell Resources....................................................................................590
10.1.5.2 Calculation of Uplink Cell Resources ........................................................................................591
10.1.5.3 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation...........592
10.1.6 Scheduling and Radio Resource Management...............................................................................593
10.1.6.1 User Throughput Calculation.....................................................................................................594
10.2 Calculation Processes ..........................................................................................................................594
10.2.1 Point Analysis: Profile Tab ..............................................................................................................594
10.2.2 Point Analysis: Reception Tab ........................................................................................................595
10.2.3 Point Analysis: Interference Tab .....................................................................................................595
10.2.4 Downlink Reference Signal Level Coverage Predictions................................................................595
10.2.4.1 Coverage Area Determination...................................................................................................596
10.2.4.1.1 All Servers ...........................................................................................................................596
10.2.4.1.2 Best Signal Level and a Margin...........................................................................................596
10.2.4.1.3 Second Best Signal Level and a Margin..............................................................................596
10.2.4.2 Coverage Display ......................................................................................................................596
10.2.4.2.1 Coverage Resolution ...........................................................................................................596
10.2.4.2.2 Display Types ......................................................................................................................596
10.2.5 Effective Signal Analysis Coverage Predictions..............................................................................597
10.2.5.1 Coverage Area Determination...................................................................................................598
10.2.5.2 Coverage Parameter Calculation ..............................................................................................598
10.2.5.3 Coverage Display ......................................................................................................................598
10.2.5.3.1 Coverage Resolution ...........................................................................................................598
10.2.5.3.2 Effective Signal Analysis (DL) Display Types......................................................................598
10.2.5.3.3 Effective Signal Analysis (UL) Display Types......................................................................600
10.2.6 Calculations on Subscriber Lists.....................................................................................................600
10.2.7 Monte Carlo Simulations.................................................................................................................601
10.2.7.1 Generating a Realistic User Distribution ...................................................................................601
10.2.7.1.1 Simulations Based on User Profile Traffic Maps and Subscriber Lists................................601
10.2.7.1.2 Simulations Based on Sector Traffic Maps..........................................................................603
10.2.7.2 Simulation Process....................................................................................................................604
Table of Contents
© Forsk 2010 AT283_TRG_E2 19
10.2.8 C/(I+N)-Based Coverage Predictions............................................................................................. 607
10.2.8.1 Coverage Area Determination .................................................................................................. 607
10.2.8.2 Coverage Parameter Calculation.............................................................................................. 607
10.2.8.3 Coverage Display ..................................................................................................................... 609
10.2.8.3.1 Coverage Resolution .......................................................................................................... 609
10.2.8.3.2 Coverage by C/(I+N) Level (DL) Display Types.................................................................. 609
10.2.8.3.3 Coverage by Best Bearer (DL) Display Types .................................................................... 610
10.2.8.3.4 Coverage by Throughput (DL) Display Types..................................................................... 610
10.2.8.3.5 Coverage by Quality Indicator (DL) Display Types ............................................................. 611
10.2.8.3.6 Coverage by C/(I+N) Level (UL) Display Types.................................................................. 611
10.2.8.3.7 Coverage by Best Bearer (UL) Display Types .................................................................... 611
10.2.8.3.8 Coverage by Throughput (UL) Display Types..................................................................... 612
10.2.8.3.9 Coverage by Quality Indicator (UL) Display Types ............................................................. 613
10.3 Calculation Algorithms ......................................................................................................................... 613
10.3.1 Downlink Transmission Powers Calculation................................................................................... 613
10.3.2 Co- and Adjacent Channel Overlaps Calculation........................................................................... 617
10.3.2.1 Conversion From Channel Numbers to Start and End Frequencies ........................................ 618
10.3.2.2 Co-Channel Overlap Calculation .............................................................................................. 618
10.3.2.3 Adjacent Channel Overlap Calculation..................................................................................... 619
10.3.2.4 Total Overlap Ratio Calculation................................................................................................ 620
10.3.3 Signal Level and Signal Quality Calculations ................................................................................. 620
10.3.3.1 Signal Level Calculation (DL) ................................................................................................... 620
10.3.3.2 Noise Calculation (DL).............................................................................................................. 623
10.3.3.3 Interference Calculation (DL).................................................................................................... 624
10.3.3.4 C/N Calculation (DL)................................................................................................................. 628
10.3.3.5 C/(I+N) and Bearer Calculation (DL) ........................................................................................ 630
10.3.3.6 Signal Level Calculation (UL) ................................................................................................... 634
10.3.3.7 Noise Calculation (UL).............................................................................................................. 636
10.3.3.8 Interference Calculation (UL).................................................................................................... 636
10.3.3.8.1 Interfering Signal Level Calculation (UL)............................................................................. 637
10.3.3.8.2 Noise Rise Calculation (UL) ................................................................................................ 638
10.3.3.9 C/N Calculation (UL)................................................................................................................. 639
10.3.3.10 C/(I+N) and Bearer Calculation (UL) ........................................................................................ 641
10.3.4 Best Server Determination ............................................................................................................. 644
10.3.5 Service Area Calculation................................................................................................................ 645
10.3.6 Throughput Calculation .................................................................................................................. 646
10.3.6.1 Calculation of Total Cell Resources.......................................................................................... 646
10.3.6.1.1 Calculation of Downlink Cell Resources ............................................................................. 646
10.3.6.1.2 Calculation of Uplink Cell Resources.................................................................................. 648
10.3.6.2 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation.......... 649
10.3.7 Scheduling and Radio Resource Management .............................................................................. 652
10.3.7.1 Scheduling and Radio Resource Allocation.............................................................................. 652
10.3.7.2 User Throughput Calculation.................................................................................................... 658
10.4 Automatic Allocation Algorithms........................................................................................................... 659
10.4.1 Automatic Neighbour Allocation ..................................................................................................... 659
10.4.2 Automatic Inter-Technology Neighbour Allocation ......................................................................... 661
10.4.3 Automatic Frequency Planning ...................................................................................................... 663
10.4.3.1 Separation Constraint and Relationship Weights ..................................................................... 664
10.4.3.2 Calculation of Cost Between TBA and Related Cells ............................................................... 664
10.4.3.3 AFP Algorithm........................................................................................................................... 667
10.4.4 Automatic Physical Cell ID Allocation............................................................................................. 667
10.4.4.1 Constraint and Relationship Weights........................................................................................ 667
10.4.4.2 Calculation of Cost Between TBA and Related Cells ............................................................... 668
10.4.4.3 Automatic Allocation Algorithm................................................................................................. 671
11 Repeaters and Remote Antennas................................................. 675
11.1 UMTS, CDMA2000, TD-SCDMA, WiMAX, and LTE Documents......................................................... 675
11.1.1 Signal Level Calculation................................................................................................................. 675
11.1.2 Downlink Total Gain Calculation .................................................................................................... 677
11.1.3 Uplink Total Gain Calculation......................................................................................................... 678
11.1.4 Repeater Noise Figure ................................................................................................................... 680
11.1.5 Appendix: Carrier Power and Interference Calculation .................................................................. 680
11.2 GSM Documents.................................................................................................................................. 683
11.2.1 Signal Level Calculation................................................................................................................. 683
11.2.2 EIRP Calculation ............................................................................................................................ 684
11.3 Donor-side Parameter Calculations ..................................................................................................... 686
11.3.1 Azimuth .......................................................................................................................................... 686
11.3.2 Mechanical Downtilt ....................................................................................................................... 686
Technical Reference Guide
20 AT283_TRG_E2 © Forsk 2010
Chapter 1
Coordinate Systems and Units
© Forsk 2010 AT283_TRG_E2 23
Chapter 1: Coordinate Systems and Units
1 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.
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.
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)
24 AT283_TRG_E2 3DF 01955 6980 RKZZA© Forsk 2010
Technical Reference Guide
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:
• 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
1.1.2.1 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.
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.
1.1.2.3 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.
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.
Note:
• 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.
© Forsk 2010 AT283_TRG_E2 25
Chapter 1: Coordinate Systems and Units
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.
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:
Examples:
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:
• There can be up to seven projection parameters. These parameters must be ordered according to the parameter
index (see "Projection Parameter Indices" on page 28). 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 = "Name of the system"; Unit Code; Datum Code; Projection Method Code,
Projection Parameters; "Comments"
4230 = "ED50"; 101; 230; 1; "Europe - west"
32045 = "NAD27 / Vermont"; 2; 267; 6, -72.5, 42.5, 500000, 0, 0.9999643; "United
States - Vermont"
Code = "Name of the system"; Unit Code; {Ellipsoid Code, Dx, Dy, Dz, Rx, Ry,
Rz, S}; Projection Method Code, Projection Parameters; "Comments"
Code Cartographic Units Code Geographic Units
0 Metre 100 Radian
1 Kilometre 101 Degree
2 Foot 102 Grad
3 Link 103 ArcMinute
4 Chain 104 ArcSecond
5 Yard
6 Nautical mile
7 Mile
-1 Unspecified -1 Unspecified
26 AT283_TRG_E2 3DF 01955 6980 RKZZA© Forsk 2010
Technical Reference Guide
1.1.3.2 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 Française
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
© Forsk 2010 AT283_TRG_E2 27
Chapter 1: Coordinate Systems and Units
1.1.3.3 Projection Method Codes
1.1.3.4 Ellipsoid Codes
249 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 Française
256 Mahe 1971 902 Nord de Guerre
257 Makassar 903
NAD 1927 Guatemala/Honduras/Salvador
(Panama Zone)
258 European Reference System 1989
Code Datum Code Datum
Code Projection Method Code Projection Method
0 Undefined 8 Oblique Stereographic
1 No projection > Longitude / Latitude 9 New Zealand Map Grid
2 Lambert Conformal Conical 1SP 10 Hotine Oblique Mercator
3 Lambert Conformal Conical 2SP 11 Laborde Oblique Mercator
4 Mercator 12 Swiss Oblique Cylindrical
5 Cassini-Soldner 13 Oblique Mercator
6 Transverse Mercator 14 UTM Projection
7 Transverse Mercator South Oriented
Code Name Major Axis Minor Axis
1 Airy 1830 6377563.396 6356256.90890985
2 Airy Modified 1849 6377340.189 6356034.44761111
3 Australian National Spheroid 6378160 6356774.71919531
4 Bessel 1841 6377397.155 6356078.96261866
5 Bessel Modified 6377492.018 6356173.50851316
6 Bessel Namibia 6377483.865 6356165.38276679
7 Clarke 1858 6378293.63924683 6356617.98173817
8 Clarke 1866 6378206.4 6356583.8
9 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 AT283_TRG_E2 3DF 01955 6980 RKZZA© Forsk 2010
Technical Reference Guide
1.1.3.5 Projection Parameter Indices
1.1.4 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 dialogue 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 dialogue:
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
28 Struve 1860 6378297 6356655.84708038
29 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
Code Name Major Axis Minor Axis
Index Projection Parameter Index Projection Parameter
0 UTM zone number 4 Scale factor at origin
0 Longitude of origin 4 Latitude of 1st parallel
1 Latitude of origin 5 Azimuth of central line
2 False Easting 5 Latitude of 2nd parallel
3 False Northing 6 Angle from rectified to skewed grid
© Forsk 2010 AT283_TRG_E2 29
Chapter 1: Coordinate Systems and Units
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.
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 folder’s 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.
30 AT283_TRG_E2 3DF 01955 6980 RKZZA© Forsk 2010
Technical Reference Guide
Chapter 2
Geographic and Radio Data
© Forsk 2010 AT283_TRG_E2 33
Chapter 2: Geographic and Radio Data
2 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.
Therefore, a n*n bin DTM file requires (n)
2
points (altitude values).
Figure 2.1Digital Terrain Model
Figure 2.2Schematic view of a DTM file
Notes:
• 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.
34 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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.
Therefore, a n*n bin Clutter file requires (n)
2
code values.
2.1.1.2.2 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.
2.1.1.3 Traffic Data
Atoll offers different kinds of traffic data:
2.1.1.3.1 User Profile Environment Based 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 Sector 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
Figure 2.3Clutter Classes
Note:
• The clutter code is the same inside a bin.
Note:
• 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.
© Forsk 2010 AT283_TRG_E2 35
Chapter 2: Geographic and Radio Data
study that defines sector boundaries for the traffic distribution in each sector. In UMTS and 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 and 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, 24-bits/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,
etc.) 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.
2.1.2 Supported Geographic Data Formats
Atoll offers Import/Export filters for the most commonly used geographic data formats. The different filters are:
Notes:
• 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.
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
DTM, Clutter classes and heights, Traffic,
Image, Population, Other data
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,
Image, Population, Other data
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
Vector data, Vector traffic, Population, Other
data
Yes
36 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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), 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.
2.2 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.
Erdas Imagine Import Only
DTM, Clutter classes and heights, Traffic,
Image, Population, Other data
Yes
ArcView Grid Export Only
DTM, Clutter classes and heights, Traffic,
Image, Population, Other data
Yes automatically embedded in
the data file
.agd Both
Vector data, Vector traffic, Population, Other
data
Yes automatically embedded in
the data file
Vertical Mapper
(.grd, .grc)
Both
DTM, Clutter classes and heights, Traffic,
Image, Population, Other data
Yes automatically embedded in
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.
Note:
• 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.
© Forsk 2010 AT283_TRG_E2 37
Chapter 2: Geographic and Radio Data
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:
• 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.
2.2.8.2 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, WiMAX, and LTE 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.
38 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Chapter 3
File Formats
© Forsk 2010 AT283_TRG_E2 41
Chapter 3: File Formats
3 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 User profile environment based traffic maps (8 bits)
• User density traffic 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:
where ‘keyword’ corresponds to an attribute type, and ‘value’ defines the attribute value.
Keywords required by Atoll are described below. Other keywords are ignored.
Four additional keywords may be optionally managed.
which can be :
in some cases, this keyword can be replace by datatype defined as follows:
keywordvalue
nrowsNumber of rows in the image.
ncolsNumber of columns in the image.
nbandsNumber of spectral bands in the image, (1 for DTM data and 8 bit pictures).
nbitsNumber 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, dBµV and DBµV/m).
byteorderByte order in which image pixel values are stored. Accepted values are
M (Motorola byte order) or I (Intel byte order).
layoutMust be ‘bil’.
skipbytesByte to be skipped in the image file in order to reach the beginning
of the image data. Default value is 0.
ulxmapx coordinate of the centre of the upper-left pixel.
ulymapy 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.
pixeltypeType of data read (in addition to the length)
UNSIGNDINT Undefined 8, 16, 24 or 32 bits
SIGNEDINT Integer 16 or 32 bits
FLOAT Real 32 or 64 bits
42 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
It can be:
The other optional keywords are :valueoffset, valuescale and nodatavalue.
By default, integer data types are chosen with respect to the pixel length (nbits).
So, we have
3.1.1.2 Samples
Here, the data is 20m.
3.1.1.2.1 Digital Terrain Model
3.1.1.2.2 Clutter Classes File
3.1.1.2.3 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.
3.2 TIF Format start here
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
datatypeType of data read (in addition to the length)
Un Undefined n bits (8, 16, 24 or 32 bits)
In Integer n bits (16 or 32 bits)
Rn Real n bits (32 or 64 bits)
RGB24 Integer 3 colour components on 24 bits
valueoffsetReal value to be added to the read value (V
read
)
valuescaleScaling factor to be applied to the read value
nodatavalueValue corresponding to “NO DATA”
V V
read
val uescal e val ueoffset + × =
nrows1500
ncols1500
nbands1
nbits8 or 16
byteorderM
layoutbil
skipbytes0
ulxmap975000
ulymap1891000
xdim 20.00
ydim 20.00
nrows1500
ncols1500
nbands1
nbits8
byteorderM
layoutbil
skipbytes0
ulxmap975000
ulymap1891000
xdim 20.00
ydim 20.00
© Forsk 2010 AT283_TRG_E2 43
Chapter 3: File Formats
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.
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:
Please refer to the Administrator Manual for more details about the Atoll.ini file.
Atoll supports the following objects in .tif format:
• Digital Terrain Model (8 or 16 bits)
• Clutter heights (8 or 16 bits)
• Clutter classes and User profile environment based traffic maps (8 bits)
• User density traffic 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)
.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:
Note:
• Atoll also supports .tif files using the Packbit, FAX-CCITT3 and LZW compression modes.
[TiffExport]
PaletteConvention=Gis
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.
• Please refer to the Administrator Manual for more details about the Atoll.ini file.
Line Description
1 x dimension of a pixel in map units
2 amount of translation
3 amount of rotation
4 negative of the y dimension of a pixel in map units
5 x-axis map coordinate of the centre of the upper-left pixel
6 y-axis map coordinate of the centre of the upper-left pixel
Note:
• Atoll does not use the lines 2 and 3 when importing a .tif format geographic file.
44 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.2.2 Sample
3.2.2.1 Clutter Classes File
3.3 BMP Format
This is the MS-Windows standard format. It holds black & white, 16-, 256- and True-colour images. The palletized 16-
colour 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 y-axis 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 .bmp format:
• Digital Terrain Model (8 bits)
• Clutter Heights (8 bits)
• Clutter classes and User density traffic maps (8 bits)
• Raster images (1, 4, 8, 24 bits)
• Population maps (8, 32 bits)
• Other generic geographic data (8, 32 bits)
3.3.1 BMP File Description
A .bmp file contains of the following data structures:
• BITMAPFILEHEADER bmfh Contains some
information about the bitmap file (about the file, not about the bitmap
itself).
• BITMAPINFOHEADER bmih Contains information
about the bitmap (such as size, colours, etc.).
• RGBQUAD aColors[] Contains a colour table.
• BYTE aBitmapBits[] Image data (whose
format is specified by the bmih structure).
3.3.1.1 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.
• BITMAPFILEHEADER (Header - 14 bytes):
• BITMAPINFOHEADER (InfoHeader - 40 bytes):
100.00
0.00
0.00
-100.00
60000.00
2679900.00
Start Size
Name
Description
Generic MS API
1 2 Signature bfType Must always be set to 'BM' to declare that this is a .bmp-file.
3 4 FileSize bfSize Specifies the size of the file in bytes.
7 2 Reserved1 bfReserved1 Unused. Must be set to zero.
9 2 Reserved2 bfReserved2 Unused. Must be set to zero.
11 4 DataOffset bfOffBits
Specifies the offset from the beginning of the file to the bitmap (raster)
data.
Start Size
Name
Description
Generic MS API
© Forsk 2010 AT283_TRG_E2 45
Chapter 3: File Formats
• RGBQUAD array (ColorTable):
• 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).
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.
15 4 Size biSize
Specifies the size of the BITMAPINFOHEADER structure, in
bytes (= 40 bytes).
19 4 Width biWidth Specifies the width of the image, in pixels.
23 4 Height biHeight Specifies the height of the image, in pixels.
27 2 Planes biPlanes
Specifies the number of planes of the target device, must be
set to zero or 1.
29 2 BitCount biBitCount
Specifies the number of bits per pixel.
1 = monochrome pallete. # of colours = 1
4 = 4-bit palletized. # of colours = 16
8 = 8-bit palletized. # of colours = 256
16 = 16-bit palletized. # of colours = 65536
24 = 24-bit palletized. # of colours = 16M
31 4 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 4 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 4 XpixelsPerM biXPelsPerMeter Specifies the the horizontal pixels per meter.
43 4 YpixelsPerM biYPelsPerMeter Specifies the the vertical pixels per meter.
47 4 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 4 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.
Start Size
Name
Description
Generic MS API
1 1 Blue rgbBlue Specifies the blue part of the colour.
2 1 Green rgbGreen Specifies the green part of the colour.
3 1 Red rgbRed Specifies the red part of the colour.
4 1 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.
46 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• 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.
3.3.1.2.1 Raster Data Compression Descriptions
• 4-bit / 16 colour images
• 8-bit / 256 colour images
Encoding type
BitCoun
t
Compressio
n
Remarks
1-bit
B&W images
1 0
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.
4-bit
16 colour images
4 0
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'.
8-bit
256 colour images
8 0
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 0
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.
24-bit
True colour images
24 0
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.
4-bit
16 colour images
4 2
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.
8-bit
256 colour images
8 1
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.
n (Byte 1) 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.
0 0 End-of-line
0 1 End-of-Bitmap
0 2
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.
0 >=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.
n (Byte 1) c (Byte 2) Description
>0 any n pixels of colour number c
0 0 End-of-line
0 1 End-of-Bitmap
0 2
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.
© Forsk 2010 AT283_TRG_E2 47
Chapter 3: File Formats
3.3.2 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:
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 Clutter Classes File
3.4 PNG Format
Portable Network Graphics (PNG) is a bitmapped image format that employs lossless data compression. PNG supports
palette-based (palettes of 24-bit RGB or 32-bit RGBA colors), greyscale, RGB, or RGBA images. PNG was designed for
transferring images on the Internet, not professional graphics, and so does not support other color spaces (such as
CMYK). PNG files nearly always use file extension .PNG or .png.
When exporting (saving) a .png file, an associated file with .pgw extension is created with the same name and in the same
directory as the .png file it refers to. Atoll stores the georeferencing information in this file for future imports of the .png so
that the .pgw file can be used during the import procedure for automatic geo-referencing.
For more information on the PNG file format, see www.w3.org/TR/PNG/.
3.4.1 PGW Header File Description
A PNG World file (.pgw file extension) is a plain text file used by geographic information systems (GIS) to provide
georeferencing information for raster map images in .png format. The world file parameters are:
3.5 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.
0 >=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.
Line Description
1 x dimension of a pixel in map units
2 amount of translation
3 amount of rotation
4 negative of the y dimension of a pixel in map units
5 x-axis map coordinate of the centre of the upper-left pixel
6 y-axis map coordinate of the centre of the upper-left pixel
100.00
0.00
0.00
-100.00
60000.00
2679900.00
Line Description
1 x dimension of a pixel in map units
2 amount of translation
3 amount of rotation
4 negative of the y dimension of a pixel in map units
5 x-axis map coordinate of the centre of the upper-left pixel
6 y-axis map coordinate of the centre of the upper-left pixel
48 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.5.1 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:
3.5.2 Sample
3.5.2.1 Clutter Classes File
3.6 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.
3.7 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.
3.8 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.
Line Description
1 x dimension of a pixel in map units
2 amount of translation
3 amount of rotation
4 negative of the y dimension of a pixel in map units
5 x-axis map coordinate of the centre of the upper-left pixel
6 y-axis map coordinate of the centre of the upper-left pixel
100.00
0.00
0.00
-100.00
60000.00
2679900.00
© Forsk 2010 AT283_TRG_E2 49
Chapter 3: File Formats
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.
3.9 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). The .TAB file must have the
following format:
The fields in bold are described below:
3.10 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:
• 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.)
!table
!version 300
!charset WindowsLatin1
Definition Table
File "raster.bmp"
Type "RASTER"
(ulxmap,ulymap) (0,0) Label "Pt 1",
(llxmap,llymap) (0,nrows) Label "Pt 2",
(lrxmap,lrymap) (ncols,nrows) Label "Pt 3",
(urxmap,urymap) (ncols,0) Label "Pt 4"
Field Description
File "raster.bmp" Name of the raster file (e.g., raster.bmp)
ulxmap x coordinate of the centre of the upper-left pixel in metres
ulymap y coordinate of the centre of the upper-left pixel in metres
llxmap x coordinate of the centre of the lower-left pixel in metres
llymap y coordinate of the centre of the lower-left pixel in metres
lrxmap x coordinate of the centre of the lower-right pixel in metres
lrymap y coordinate of the centre of the lower-right pixel in metres
urxmap x coordinate of the centre of the upper-right pixel in metres
urymap y coordinate of the centre of the upper-right pixel in metres
nrows Number of rows in the image
ncols Number of columns in the image
50 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.11 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.
3.12 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.
3.13 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.
3.13.1 ArcView Grid File Description
The format of this file is as follows:
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.
• Please refer to the Administrator Manual for more details about the Atoll.ini file.
ncols XXXNumber of columns of the grid (XXX columns).
nrows XXXNumber of rows of the grid (XXX rows).
xllcenter XXX OR
xllcorner XXXSignificant value relative to the bin centre or corner.
© Forsk 2010 AT283_TRG_E2 51
Chapter 3: File Formats
3.13.2 Sample
3.14 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 Géographique 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.
3.15 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:
• Digital Terrain Model (8 and 16 bits)
• Clutter class maps (16 bits)
• Raster images (1, 4, 8 and 24 bits)
• Vector data
• Text data
3.15.1 DTM File
3.15.1.1 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:
yllcenter OR
yllcorner XXXSignificant value relative to the bin centre or corner.
cellsize XXXGrid resolution.
nodata_value XXXOptional value corresponding to no data (no information).
//Row 1Top of the raster. Description of the first row. Syntax: ncols number of
values separated by spaces.
:
:
//Row NBottom of the raster.
ncols 303
nrows 321
xllcorner 585300.000000
yllcorner 5615700.000000
cellsize 100.000000
nodata_value 0
...
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
52 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• The projection file provides information about the projection system used. This file is optional. It is an ASCII text
file with four lines maximum.
3.15.1.2 Sample
Index file associated with height file (DTM data):
Projection file associated with height file (DTM data):
3.15.2 Clutter Class Files
3.15.2.1 Description
The Clutter directory consists of three files; the clutter file and two other files detailed below:
• 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.
3.15.2.2 Sample
Menu file associated with the clutter file:
Line Description
Spheroid
Zone
Projection
Central meridian
Latitude and longitude of projection central meridian and equivalent x and y coordinates in meters
(optional)
Note:
• In the associated binary file, the value -9999 corresponds to ‘No data’ which is supported
by Atoll.
sydney1 303900 343900 6227900 6267900 50
Australian-1965
56
UTM
0 153 500000 10000000
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)
Note:
• In the associated binary file, the value -9999 corresponds to ‘No data’ which is supported
by Atoll.
1 open
2 sea
3 inlandwater
4 residential
5 meanurban
6 denseurban
7 buildings
8 village
9 industrial
10 openinurban
11 forest
© Forsk 2010 AT283_TRG_E2 53
Chapter 3: File Formats
3.15.3 Vector Files
3.15.3.1 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:
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:
The fields are separated by spaces.
• The attribute file stores the height and description properties of vector paths. This file is optional.
3.15.3.2 Sample
Index file associated with the vector files
3.15.4 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.
12 parks
13 denseurbanhigh
14 blockbuildings
15 denseblockbuild
16 rural
17 mixedsuburban
Field Type Description
Vector type code Integer > 0 Identification code for the vector type
Vector type name Text (up to 32 characters in length) Name of the vector type
Field Type Description
Vector file name Text (up to 32 characters in length) Name of the vector file
Attribute file name Text (up to 32 characters in length)
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 path points in the vector file
vector file northmin: minimum y-axis coordinate of all
vector path points in the vector file
vector file northmax: maximum y-axis coordinate of all
vector path points
Vector type name Text (up to 32 characters in length)
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.
sydney1.airport313440 333021 6239426 6244784 airport
sydney1.riverlake303900 342704 6227900 6267900 riverlake
sydney1.coastline322837 343900 6227900 6267900 coastline
sydney1.railways303900 336113 6227900 6267900 railways
sydney1.highways303900 325155 6240936 6267900 highways
sydney1.majstreets303900 342770 6227900 6267900 majstreets
sydney1.majorroads303900 342615 6227900 6267900 majorroads
54 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.15.5 Text Data Files
The text data directory consists of:
• The text data files are ASCII text files with the following format:
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:
The fields are separated by spaces.
• The menu file, an ASCII text file, contains the text features. This file is optional.
3.16 MNU Format
3.16.1 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.
Separator used can either be a space character or a tab.
3.16.2 Sample
A .mnu file associated to a clutter classes file:
Airport
637111.188 3094774.00
Airport
628642.688 3081806.25
Field Type Description
File name Text (up to 32 characters in length) 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 Text (up to 32 characters in length) This field is omitted in case no menu file is available.
railwayp.txt -260079 693937 2709348 3528665 Railway_Station
airport.txt -307727 771663 2547275 3554675 Airport
ferryport.txt 303922 493521 2667405 3241297 Ferryport
1 Airport
2 Ferryport
3 Railway_Station
Field Type Description
Class code Integer > 0 Identification code for the clutter (or traffic) class
Class name Text (up to 50 characters in length) Name of the clutter (or traffic) class. It may contain spaces.
0 none
1 open
2 sea
© Forsk 2010 AT283_TRG_E2 55
Chapter 3: File Formats
3.17 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.
3.17.1 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:
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:
A sample extract of the index.xml is given below:
Note that no closing tag </XML_Table> is required.
3.17.2 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:
3 inland_water
4 residential
5 meanurban
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
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")
<Atoll_XML_Config Atoll_File_System="UMTS" Atoll_File_Technology="CDMA"
Atoll_File_Version="2.x.x build xxxx">
<XML_Table XML_File="CustomFields.xml" Atoll_Table="CustomFields" />
<XML_Table XML_File="CoordSys.xml" Atoll_Table="CoordSys" />
...
</Atoll_XML_Config>
<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'>
56 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
The schema definition follows the root tag and is enclosed between the following tags:
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:
<s:Schema id=’RowsetSchema’>
<!-Schema is defined here, using <s:ElementType> and <s:AttributeType> tags ->
</s:Schema>
<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'>
<s:datatype dt:type='r4' dt:maxLength='4' rs:precision='7'
rs:fixedlength='true'/>
</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>
© Forsk 2010 AT283_TRG_E2 57
Chapter 3: File Formats
3.18 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.
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 MS-
Access. 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:
3.18.1.1.1 DBF Structure
3.18.1.1.2 DBF Header (Variable Size - Depends on Field Count)
Note:
• Each transmitter path loss matrix is calculated on the area where calculation radius
intersects the computation zone (see: Computation zone).
FS = FlagShipD3 = dBaseIII+
Fb = FoxBaseD4 = dBaseIV
Fp = FoxProD5 = dBaseV
CL = Clipper
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
If .dbf is not empty
last optional: 0x1a (eof byte)
Byte Size Contents Description Applies for (supported by)
00 1 0x03 plain .dbf FS, D3, D4, D5, Fb, Fp, CL
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 FS, D3, D4, D5, Fb, Fp, CL
0x8B with .dbt memo in D4 format D4, D5
0x8E with SQL table D4, D5
0xF5 with .fmp memo Fp
01 3 YYMMDD Last update digits All
04 4 ulong Number of records in file All
08 2 ushort Header size in bytes All
10 2 ushort Record size in bytes All
12 2 0,0 Reserved All
14 1 0x01 Begin transaction D4, D5
0x00 End Transaction D4, D5
0x00 ignored FS, D3, Fb, Fp, CL
15 1 0x01 Encrypted D4, D5
0x00 normal visible All
16 12 0 (1) multi-user environment use D4,D5
28 1 0x01 production index exists Fp, D4, D5
58 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• Field descriptor array in the .dbf header (32 bytes for each field)
• Field type and size in the .dbf header, field descriptor (1 byte)
0x00 index upon demand All
29 1 n 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
30 2 0,0 reserved All
32 n*32 Field Descriptor, (see next paragraph) all
+1 1 0x0D Header Record Terminator all
Byte Size Contents Description Applies for (supported by)
0 11 ASCI field name, 0x00 termin all
11 1 ASCI field type (see next paragraph) all
12 4 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 1 byte Field length, bin (see next paragraph) all \ FS,CL: for C field type
17 1 byte decimal count, bin all / both used for fld lng
18 2 0,0 reserved all
20 1 byte Work area ID D4, D5
0x00 unused FS, D3, Fb, Fp, CL
21 2 n,n multi-user dBase D3, D4, D5
0,0 ignored FS, Fb, Fp, CL
23 1 0x01 Set Fields D3, D4, D5
0x00 ignored FS, Fb, Fp, CL
24 7 0...0 reserved all
31 1 0x01 Field is in .mdx index D4, D5
0x00 ignored FS, D3, Fb, Fp, CL
Size Type Description/Storage Applies for (supported by)
C 1...n Char
ASCII (OEM code page chars)
rest= space, not \0 term.
all
n = 1...64kb (using deci count) FS
n = 1...32kb (using deci count) Fp, CL
n = 1...254 all
D 8 Date 8 ASCII digits (0...9) in the YYYYMMDD format all
F 1...n Numeric
ASCII digits (-.0123456789)
variable pos. of float.point
n = 1...20
FS, D4, D5, Fp
N 1...n Numeric
ASCII digits (-.0123456789)
fix posit/no float.point
all
n = 1...20 FS, Fp, CL
n = 1...18 D3, D4, D5, Fb
L 1 Logical ASCII chars (YyNnTtFf space) FS, D3, Fb, Fp, CL
ASCII chars (YyNnTtFf?) D4, D5 (FS)
M 10 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
© Forsk 2010 AT283_TRG_E2 59
Chapter 3: File Formats
3.18.1.1.3 Each DBF Record (Fixed Length)
3.18.1.2 DBF File Content
The .dbf file provides information that is needed to check validity of each path loss matrix.
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
2 2 short int binary int max +/- 32767 FS
4 4 long int binary int max +/- 2147483647 FS
8 8 double binary signed double IEEE FS
Byte Size Description Applies for (supported by)
0 1 deleted flag "*" or not deleted " " all
1…n 1…
x-times contents of fields, fixed length, unterminated.
For n, see (2) byte 10…11
All
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 ID
a
.
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
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 radius
b
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 zone
c
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
60 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.18.2 LOS File
The data file is a binary file with a standard row-column structure. Data are stored starting from the southwest to the
northeast corner of the area. The file contains 16-bit signed integer values in the range [-32768; +32767] with a 1/16 dB
precision. "No data" values are represented by +32767.
3.19 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 .
3.19.1 DBF File
dBASE III file (pathloss.dbf) has a standard .dbf format described below. Its content can be checked by opening it in MS-
Access. The format is detailed hereafter.
3.19.1.1 DBF File Format
For general information, the format of .dbf files in any Xbase language is described.
Following notations are used in tables:
3.19.1.1.1 DBF Structure
3.19.1.1.2 DBF Header (Variable Size - Depends on Field Count)
LOCKED Boolean
Locking status
0: path loss matrix is not locked
1: path loss matrix is locked.
INC_ANT 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
a. 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.
b. These coordinates enable Atoll to determine the area of calculation for each transmitter.
c. These coordinates enable Atoll to determine the rectangle including the computation zone.
FS = FlagShipD3 = dBaseIII+
Fb = FoxBaseD4 = dBaseIV
Fp = FoxProD5 = dBaseV
CL = Clipper
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
If .dbf is not empty
last optional: 0x1a (eof byte)
Byte Size Contents Description Applies for (supported by)
00 1 0x03 plain .dbf FS, D3, D4, D5, Fb, Fp, CL
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 FS, D3, D4, D5, Fb, Fp, CL
© Forsk 2010 AT283_TRG_E2 61
Chapter 3: File Formats
• Field descriptor array in the .dbf header (32 bytes for each field)
• Field type and size in the .dbf header, field descriptor (1 byte)
0x8B with .dbt memo in D4 format D4, D5
0x8E with SQL table D4, D5
0xF5 with .fmp memo Fp
01 3 YYMMDD Last update digits All
04 4 ulong Number of records in file All
08 2 ushort Header size in bytes All
10 2 ushort Record size in bytes All
12 2 0,0 Reserved All
14 1 0x01 Begin transaction D4, D5
0x00 End Transaction D4, D5
0x00 ignored FS, D3, Fb, Fp, CL
15 1 0x01 Encrypted D4, D5
0x00 normal visible All
16 12 0 (1) multi-user environment use D4,D5
28 1 0x01 production index exists Fp, D4, D5
0x00 index upon demand All
29 1 n 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
30 2 0,0 reserved All
32 n*32 Field Descriptor, (see next paragraph) all
+1 1 0x0D Header Record Terminator all
Byte Size Contents Description Applies for (supported by)
0 11 ASCI field name, 0x00 termin all
11 1 ASCI field type (see next paragraph) all
12 4 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 1 byte Field length, bin (see next paragraph) all \ FS,CL: for C field type
17 1 byte decimal count, bin all / both used for fld lng
18 2 0,0 reserved all
20 1 byte Work area ID D4, D5
0x00 unused FS, D3, Fb, Fp, CL
21 2 n,n multi-user dBase D3, D4, D5
0,0 ignored FS, Fb, Fp, CL
23 1 0x01 Set Fields D3, D4, D5
0x00 ignored FS, Fb, Fp, CL
24 7 0...0 reserved all
31 1 0x01 Field is in .mdx index D4, D5
0x00 ignored FS, D3, Fb, Fp, CL
Size Type Description/Storage Applies for (supported by)
C 1...n Char
ASCII (OEM code page chars)
rest= space, not \0 term.
all
n = 1...64kb (using deci count) FS
n = 1...32kb (using deci count) Fp, CL
n = 1...254 all
62 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.19.1.1.3 Each DBF Record (Fixed Length)
3.19.1.2 DBF File Content
The .dbf file provides information about the measured transmitters participating in the tuning.
3.19.2 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
D 8 Date 8 ASCII digits (0...9) in the YYYYMMDD format all
F 1...n Numeric
ASCII digits (-.0123456789)
variable pos. of float.point
n = 1...20
FS, D4, D5, Fp
N 1...n Numeric
ASCII digits (-.0123456789)
fix posit/no float.point
all
n = 1...20 FS, Fp, CL
n = 1...18 D3, D4, D5, Fb
L 1 Logical ASCII chars (YyNnTtFf space) FS, D3, Fb, Fp, CL
ASCII chars (YyNnTtFf?) D4, D5 (FS)
M 10 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
2 2 short int binary int max +/- 32767 FS
4 4 long int binary int max +/- 2147483647 FS
8 8 double binary signed double IEEE FS
Byte Size Description Applies for (supported by)
0 1 deleted flag "*" or not deleted " " all
1…n 1…
x-times contents of fields, fixed length, unterminated.
For n, see (2) byte 10…11
All
Field Type Description
TX_NAME Text Name of the transmitter
FILE_NAME Text Name (and optionally, path) of .pts file
AREA_XMIN Float Not used
AREA_XMAX Float Not used
AREA_YMIN Float Not used
AREA_YMAX Float Not used
© Forsk 2010 AT283_TRG_E2 63
Chapter 3: File Formats
• 8 bytes : Tx Position
The list of points contains following 4-uplet for all points
• 4 bytes : X
• 4 bytes : Y
• 4 bytes : Measurement value
• 4 bytes : Incidence angle.
3.20 Interference Histograms File Formats
Interference histograms required by automatic frequency planning tools can be imported and exported.
3.20.1 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:
• 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:
The 4 tab-separated columns are defined in the table below:
3.20.1.1 Sample
Notes:
• 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%).
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
<Column1><tab><Column2><tab><Column3><tab><Column4><newline>
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.
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
# Remark: C/I results do not incorporate power offset values.
64 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.20.2 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).
3.20.2.1 CLC File
3.20.2.1.1 Description
The .clc file consists of two parts:
• The first part is a header used for format identification. It must start with and contain the following lines:
• 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:
The 5 tab-separated columns are defined in the table below:
# 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
Note:
• 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.
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
<Column1><tab><Column2><tab><Column3><tab><Column4><tab><Column5><newline>
© Forsk 2010 AT283_TRG_E2 65
Chapter 3: File Formats
3.20.2.1.2 Sample
Column name Description
Column1 Interfered transmitter
Identification number of the interfered transmitter. If the column is empty,
its value is identical to the one of the line above.
Column2 Interfering transmitter
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:
• The columns 1, 2, and 3 must be defined only in the first line of each histogram.
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
# Remark: C/I results do not incorporate power offset values.
# Fields are:
##------------#------------#------------#-----------#------------------#
#| Interfered | Interfering| Interfered | C/I | Probability |
#| Transmitter| Transmitter| Trx type | Threshold | C/I >= Threshold |
##------------#------------#------------#-----------#------------------#
#
# 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
##---------------------------------------------------------------------#
1 2 TCH_INNER 8 1
9 0.944
10 0.904
11 0.892
14 0.844
15 0.832
16 0.812
17 0.752
22 0.316
25 0.292
1 2 BCCH,TCH 8 1
9 0.944
10 .904
13 0.872
14 0.84
17 0.772
Note:
66 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.20.2.2 DCT File
3.20.2.2.1 Description
The .dct file is divided into two parts:
• The first part is a header used for format identification. It must start with and contain the following lines:
• The second part provides information about transmitters taken into account in AFP.
The lines after the header are considered as comments if they start with the symbol "#". If not, they must have the following
format:
The last four columns describe the interference matrix scope. One transmitter per line is described separated with a tab
character.
3.20.2.2.2 Sample
• If the TCH and BCCH histograms are the same, they are not duplicated. A single record
indicates that the histograms belong to TCH and BCCH both. For example, instead of:
1 2 TCH -9.5 1 - 9 1 - 6 1
1 2 BCCH -9.5 1 - 9 1 - 6 1
We have:
1 2 TCH,BCCH -9.5 1 - 9 1 - 6 1
# Calculation Results Dictionary File.
# Version 1.1, Tab separated format. Commented lines start with #.
<Column1><tab><Column2><newline>
Column name Type Description
Column1 Transmitter name Text Name of the transmitter
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
# 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 |
##-----------#-----------#-----------#-----------#---------#---------#
#
# 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 per HCS layer".
# Margin is 5.
# Cell edge coverage probability is 75%.
# Traffic spreading was Uniform (percentage of interfered area)
##---------------------------#
Site0_0 1 -1 -1 100 100
Site0_1 2 -1 -1 100 100
Site0_2 3 -1 -1 100 100
Site1_0 4 -1 -1 100 100
© Forsk 2010 AT283_TRG_E2 67
Chapter 3: File Formats
3.20.3 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:
• The first part is a header used for format identification. It must start with and contain the following lines:
• 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:
The 5 tab-separated columns are defined in the table below:
3.20.3.1 Sample
Site1_1 5 -1 -1 100 100
Site1_2 6 -1 -1 100 100
Site2_0 7 -1 -1 100 100
Site2_1 8 -1 -1 100 100
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
<Column1><tab><Column2><tab><Column3><tab><Column4><tab><Column5><newline>
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 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.
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
# Remark: C/I results do not incorporate power offset values.
# Fields are:
#------------------------------------------------------------------------
#Transmitter Interferer TRX type C/I Probability
#------------------------------------------------------------------------
#
# 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
Site0_2 Site0_1 BCCH,TCH -9 0.996
Site0_2 Site0_1 BCCH,TCH -6 0.976
68 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
3.20.4 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:
Where the separator (<SEP>) can either be a tab or a semicolon.
The four columns are defined in the table below:
corresponds to the required C/I threshold. This parameter is defined for each subcell.
is the adjacent channel protection level.
3.20.4.1 Sample
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 0.932
Site0_2 Site0_1 BCCH,TCH 1 0.924
Site0_2 Site0_1 BCCH,TCH 4 0.896
Site0_2 Site0_1 BCCH,TCH 7 0.864
Site0_2 Site0_1 BCCH,TCH 8 0.848
Site0_2 Site0_1 BCCH,TCH 9 0.832
Site0_2 Site0_1 BCCH,TCH 10 0.824
...
<Column1><SEP><Column2><SEP><Column3><SEP><Column4><newline>
Column name Description
Column1 Interfered transmitter Name of the interfered transmitter.
Column2 Interfering transmitter Name of the interferer transmitter.
Column3
Co-channel interference
probability
Probability of having
Column4
Adjacent channel
interference probability
Probability of having
C I Max
BCCH TCH ,
C I
req
( ) s
C I Max
BCCH TCH ,
C I
req
( ) F – s
C I
req

F
# Calculation Results Data File.
# Version 1.1, Tab separated format. Commented lines start with #.
# Remark: C/I results do not incorporate power offset values.
# Fields are:
#------------------------------------------------------------------------
#Transmitter Interferer Co-channel Adjacent channel
#------------------------------------------------------------------------
#
# 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 0.226667 0.024
Site0_2 Site0_3 0.27 0.024
© Forsk 2010 AT283_TRG_E2 69
Chapter 3: File Formats
The columns in the sample above are separated with a tab. These columns can also be separated with a semilcolon:
3.21 Antenna Pattern Formats
This section describes the format of the DIAGRAM field of the Antennas table. This field stores the antenna diagrams in
a 2D (angle vs. attenuation) format. This is the format of the contents of the DIAGRAM field of the Antennas table when it
is copied from, pasted to, imported to (from txt, csv, or xls files), and exported from (in txt or csv files) the Antennas table.
Antenna patterns can also be imported in Planet 2D-format antenna files and 3D antenna files. The file format required for
3D antenna file import is described in "Import Format of Text Files Containing 3D Antenna Patterns" on page 70.
3.21.1 2D Antenna Diagram Format
The format of 2D antenna patterns containing co-polar diagrams only can be understood from Figure 3.1 on page 69.
The contents of the DIAGRAM field are formatted as follows:
• Pattern Descriptor 1: Space-separated list of parameters.
- First entry: The number of co-polar diagrams. For example, 2.
- Second entry: First co-polar diagram type = 0 for azimuth (horizontal) diagram.
- Third entry: The elevation angle of the azimuth diagram.
- Fourth entry: The number of angle-attenuation pairs in the first co-polar diagram. For example, 360.
• Co-polar Horizontal Diagram: Horizontal co-polar diagram (the second entry in the preceding descriptor is 0).
The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....
• Pattern Descriptor 2: Space-separated list of parameters.
- First entry: Second co-polar diagram type = 1 for elevation (vertical) diagram.
- Second entry: The azimuth angle of the elevation diagram.
- Third entry: The number of angle-attenuation pairs in the second co-polar diagram. For example, 360.
• Co-polar Vertical Diagram: Vertical co-polar diagram (the first entry in the preceding descriptor is 1). The format
is space-separated angle attenuation pairs. For example, 0 0 1 0.1....
• End: The number cross-polar diagrams = 0.
The format of 2D antenna patterns containing co-polar and cross-polar diagrams can be understood from Figure 3.2 on
page 69.
The contents of the DIAGRAM field are formatted as follows:
• Pattern Descriptor 1: Space-separated list of parameters.
- First entry: The number of co-polar diagrams. For example, 2.
- Second entry: First co-polar diagram type = 0 for azimuth (horizontal) diagram.
Site0_3 Site0_1 0.276 0.02
Site0_3 Site0_2 0.226 0.028
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
Figure 3.12D Antenna Pattern Format Containing Co-polar Diagrams Only
Figure 3.22D Antenna Pattern Format Containing Co-polar and Cross-polar Diagrams
Pattern
End Co-polar Horizontal Diagram Co-polar Vertical Diagram
2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 … 0
Discriptor 1
Pattern
Discriptor 2
Pattern
Co-polar Horizontal Diagram Co-polar Vertical Diagram
2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 …
Discriptor 1
Pattern
Discriptor 2
Pattern
Cross-polar Horizontal Diagram Cross-polar Vertical Diagram
Discriptor 3
Pattern
Discriptor 4
2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 …
70 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
- Third entry: The elevation angle of the azimuth diagram.
- Fourth entry: The number of angle-attenuation pairs in the first co-polar diagram. For example, 360.
• Co-polar Horizontal Diagram: Horizontal co-polar diagram (the second entry in the preceding descriptor is 0).
The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....
• Pattern Descriptor 2: Space-separated list of parameters.
- First entry: Second co-polar diagram type = 1 for elevation (vertical) diagram.
- Second entry: The azimuth angle of the elevation diagram.
- Third entry: The number of angle-attenuation pairs in the second co-polar diagram. For example, 360.
• Co-polar Vertical Diagram: Vertical co-polar diagram (the first entry in the preceding descriptor is 1). The format
is space-separated angle attenuation pairs. For example, 0 0 1 0.1....
• Pattern Descriptor 3: Space-separated list of parameters.
- First entry: The number of cross-polar diagrams. For example, 2.
- Second entry: First cross-polar diagram type = 0 for azimuth (horizontal) diagram.
- Third entry: The elevation angle of the azimuth diagram.
- Fourth entry: The number of angle-attenuation pairs in the first cross-polar diagram. For example, 360.
• Cross-polar Horizontal Diagram: Horizontal cross-polar diagram (the second entry in the preceding descriptor
is 0). The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....
• Pattern Descriptor 4: Space-separated list of parameters.
- First entry: Second cross-polar diagram type = 1 for elevation (vertical) diagram.
- Second entry: The azimuth angle of the elevation diagram.
- Third entry: The number of angle-attenuation pairs in the second cross-polar diagram. For example, 360.
• Cross-polar Vertical Diagram: Vertical cross-polar diagram (the first entry in the preceding descriptor is 1). The
format is space-separated angle attenuation pairs. For example, 0 0 1 0.1....
You may use a 3rd party software or develop a tool to to convert the contents of the DIAGRAM field into binary. In binary,
each antenna is described by a header and a list of value pairs.
The header is defined as follows:
• flag: (Integer, 32 bits) -1 for omni diagrams, 0 for directional
• num: (Short integer, 16 bits) Number of diagrams (0, 1, 2, 3, 4)
• siz0: (Short integer, 16 bits) Size of the first diagram (horizontal co-polar section, elevation = 0°)
• siz1: (Short integer, 16 bits) Size of the second diagram (vertical co-polar section, azimuth = 0°)
• siz2: (Short integer, 16 bits) Size of the third diagram (horizontal cross-polar)
• siz3: (Short integer, 16 bits) Size of the fourth diagram (vertical cross-polar)
• prec: (Short integer, 16 bits) Precision of the following angle values (100)
Then follows the content of each of the defined diagrams, i.e., the diagrams whose sizes (siz0, siz1, siz2, siz3) are not
zero. Each diagram consists of a list of value pairs. The number of value pairs in a list depends on the value of the siz0,
siz1, siz2, and siz3 parameters. For example, siz2 = 5 means there are five value pairs in the third diagram.
The value pairs in each list are:
• ang: (Short integer, 16 bits) The first component of the value pair is the angle in degrees multiplied by 100. For
example, 577 means 5.77 degrees.
• loss: (Short integer, 16 bits) The second component of the value pair is the loss in dB for the given angle ang.
All the lists of value pairs are concatenated without a separator.
3.21.2 Import Format of Text Files Containing 3D Antenna Patterns
Text files containing 3D antenna patterns that may be imported in Atoll must have the following format:
• Header: The text file may contain a header with additional information. When you import the antenna pattern you
can indicate the row number in the file where the header ends and the antenna pattern begins.
• Antenna Pattern: Each row contains three values to describe the 3D antenna pattern. The columns containing
the values can be in any order:
- Azimuth: Allowed range of values is from 0°to 360°. The smallest increment allowed is 1°.
- Tilt: Allowed range of values is from -90°to 90°or from 0°to 180°. The smallest increment allowed is 1°.
- Attenuation: The attenuation in dB.
Chapter 4
Calculations
© Forsk 2010 AT283_TRG_E1 73
Chapter 4: Calculations
4 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:
In any case, prediction is performed in three steps:
1
st
step: First of all, Atoll calculates the path loss ( ), using the selected propagation model.
is the loss on the transmitter-receiver path calculated through the propagation model. value depends on
the selected propagation model.
is the transmitter antenna attenuation (from antenna patterns).
is the receiver antenna attenuation ( ) (from antenna patterns).
2
nd
step: When the option “Shadowing taken into account” is selected, Atoll evaluates a shadowing margin,
, from the user-defined model standard deviation at the receiver and the cell edge coverage probability.
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
extraction
a
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.
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
calculation. However, all the propagation models available in Atoll calculate by
considering transmitter antenna attenuation.
L
path
L
path
L
model
L
ant
Tx
L
ant
Rx
+ + =
L
model
L
model
L
ant
Tx
L
ant
Rx
L
ant
Rx
0 =
L
path
L
path
M
Shadowi ng model –
74 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
3
rd
step: Then, Atoll determines the prediction criterion and displays coverage.
For a signal level study,
The signal level at the receiver ( ) is calculated. We have (in dBm):
Where
is the effective isotropic radiated power of the transmitter.
is the transmitter power.
is the transmitter antenna gain.
are transmitter losses.
is the shadowing margin.
are the indoor losses, taken into account when the option “Indoor coverage” is selected,
are receiver losses.
is the receiver antenna 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.
4.2 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 ( )
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.
Note:
• For a cell edge coverage probability of 50%, the shadowing margin is always zero. In this
case, Atoll still works as above.
Notes:
• In UMTS and CDMA documents, and .
• In UMTS and CDMA documents, Atoll considers that and 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, .
• In GSM_EGPRS documents, receiver is equipped with an antenna with zero gain.
Note:
• 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.
P
Rec
P
Rec
EIRP L
path
– M
Shadowi ng model –
– L
Indoor
– G
ant
Rx
L
Rx
– ( ) + =
EIRP P
Tx
G
ant
Tx
L
Tx
– + =
EIRP
P
Tx
G
ant
Tx
L
Tx
M
Shadowi ng model –
L
Indoor
L
Rx
G
ant
Rx
P
Tx
P
Pi l ot
= L
Tx
L
total DL –
=
G
ant
Rx
L
Rx
L
Tx
L
total DL –
=
L
path
© Forsk 2010 AT283_TRG_E1 75
Chapter 4: Calculations
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).
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.
Figure 4.1Example 1: Single Calculation Area
Figure 4.2Example 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
76 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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.
4.2.3 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.
Note:
• 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.
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
Antenna
a
height
Invalid Path loss matrices Sufficient Not necessary
Antenna
a
pattern
Invalid Path loss matrices Sufficient Not necessary
Downtilt
a
Invalid Path loss matrices Sufficient Not necessary
Azimuth
a
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
Receiver antenna
Valid because
Prediction study Sufficient Not necessary
Geographic layer order Invalid Path loss matrices
Insufficient
b
Necessary
Geographic file resolution Invalid Path loss matrices
Insufficient
b
Necessary
New DTM map Invalid Path loss matrices
Insufficient
b
Necessary
New clutter class edition Invalid Path loss matrices
Insufficient
b
Necessary
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
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.
L
ant
Rx
0 =
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.
© Forsk 2010 AT283_TRG_E1 77
Chapter 4: Calculations
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, S’1, S’’1, S’2 and S’’2, from the
DTM file (Centre of each DTM pixel).
1
st
step: Atoll draws a vertical line through S. This line respectively intersects (S’1,S’’1) and (S’2, S’’2) lines at S1 and S2.
2
nd
step: Atoll determines the S1 and S2 altitudes using a linear interpolation method.
3
rd
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.
Tip 2
Calculation area management
When performing prediction studies, it is recommended to follow this methodology to minimise recalculations:
1
st
step: Calculate without computation zone.
2
nd
step: Draw a computation zone and calculate.
3
rd
step: Decrease the calculation radius and calculate.
Figure 4.3Ground Altitude Determination - 1
Figure 4.4Ground Altitude Determination - 2
Figure 4.5Ground Altitude Determination - 3
Figure 4.6Ground Altitude Determination - 4
78 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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, S’1,
S’’1, S’2 and S’’2. Here, the nearest point to S is S”2; therefore Atoll takes the S”2 clutter height as clutter height of S.
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.
Figure 4.7Clutter Height
Figure 4.8Radial calculation method
© Forsk 2010 AT283_TRG_E1 79
Chapter 4: Calculations
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.
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 4.9Site-bin centre profile
Figure 4.10Radial calculation method
Transmitter
Geographic profiles
Receiver: it may be anywhere in point analysis or at the centre of each calculation bin in coverage studies
80 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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. 1
st
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. 2
nd
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.
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.
Explorer window Work space
DTM
•DTM 1 (25m)
•DTM 2 (40m)
Clutter
•Clutter (20m)
Notes:
• 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.
© Forsk 2010 AT283_TRG_E1 81
Chapter 4: Calculations
4.4 Coverage Predictions
4.4.1 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
multiplied by the area of one of its pixels gives the total area.
This area depends on the resolution of the coverage prediction. At the border of the focus (or computation) zone, pixels
are considered either inside or outside the zone. A pixel is inside 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 (or 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,
Atoll9955 automatically replaces it by 100%.
4.4.2 Filtering Coverage Prediction Exports
Filtering can directly be applied to any type of prediction export (raster or vector) in order to exclude holes and islands.
The principle is to set the colour of each pixel by extracting the dominant colour of the bounding box made of pixels around
the pixel to be filled using a dispersion factor: .
where is the distance from the pixel to be filled to each pixel within the bounding box and where is the value at that
pixel.
In other words, the pixel will be filled by the most representative value within this bounding box.
The user-defined filtering percentage gives the size of the bounding box: pixels in each direction. In other words,
the bounding box is increased by one pixel every 10% (since Y is defined as a percentage in the interface)
4.4.3 Smoothing Coverage Prediction Exports
Smoothing can be applied to any prediction export in vector format to simplify its contours.
The principle is to reduce the number of points defining the contour of the polygon. This is done using a vertex reduction
routine reducing successive vertices which are clustered too closely (vertex reduction within tolerance of prior vertex
cluster, Douglas-Peucher polyline simplification).
Two methods can be set up in order to define the degree of coverage smoothing.
4.4.3.1 Smoothing: Percentage Method
The user-defined smoothing percentage gives the approximation tolerance: , where is the user-defined
export resolution. The tolerance defines the interval within which the algorithm tries to reduce the number of points as
explained hereafter.
Figure 4.11Bounding box for prediction filtering
D
2
– X 2t ( ) ( ) exp
D X
Y Y 10
Z
2
2
------- R
Z
20
------ × × R
82 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
Let’s consider the case of 3 successive points, A
1
, A
2
, A
3
. The aim of smoothing is to reduce the number of points
according to the tolerance such that A
2
will be deleted if within this tolerance (and A
1
and A
3
will be directly linked) and A
2
will be conserved if outside this tolerance.
- If A
2
is outside this interval, the exported shape will be (in blue):
- If A2 is within this interval, the exported shape will be (in blue):
4.4.3.2 Smoothing: Number of points method
The second method consists in defining a maximum number of points to be deleted. This number of points helps the
algorithm to determine the optimised tolerance (See "Smoothing: Percentage Method" on page 81) such that, with this
obtained tolerance, the number of points to be deleted will be lower than this value.
Let’s consider the following initial coverage
Starting from the maximum possible tolerance, one can estimate the number of points to be filtered out (circled in red).
Figure 4.12Smoothing Tolerance Definition
© Forsk 2010 AT283_TRG_E1 83
Chapter 4: Calculations
If this number is greater than the maximum number of points defined in the interface, Atoll reduces the tolerance until
reaching the requested maximum number of points or less.
The first the number of points respecting the constraint is obtained, smoothing is applied by deleting these points and
linking the remaining closest points.
84 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
4.5 Propagation Models
Propagation models available in Atoll are listed in the table below along with their main characteristics.
C
O
S
T
-
H
a
t
a
O
k
u
m
u
r
a
-
H
a
t
a
1
5
0
-
2
0
0
0

M
H
z
L
(
d
,

f
,

H
R
x
)
(
p
e
r

e
n
v
i
r
o
n
m
e
n
t
)
D
i
f
f
r
a
c
t
i
o
n

l
o
s
s
D
e
y
g
o
u
t
(
1

o
b
s
t
a
c
l
e
)
D
T
M
R
a
d
i
a
l
M
a
c
r
o

c
e
l
l
M
i
n
i

c
e
l
l
S
t
r
e
e
t
M
o
b
i
l
e
G
S
M

9
0
0
,

G
S
M

1
8
0
0
,
U
M
T
S
,

C
D
M
A
2
0
0
0
,
L
T
E
E
r
c
e
g
-
G
r
e
e
n
s
t
e
i
n
(
S
U
I
)
1
9
0
0
-
6
0
0
0

M
H
z
L
(
d
,

f
,

H
T
x
,

H
R
x
)
(
p
e
r

e
n
v
i
r
o
n
m
e
n
t
)
D
i
f
f
r
a
c
t
i
o
n

l
o
s
s
D
e
y
g
o
u
t
(
1

o
b
s
t
a
c
l
e
)
D
T
M
R
a
d
i
a
l
M
a
c
r
o

c
e
l
l
M
i
n
i

c
e
l
l
S
t
r
e
e
t
F
i
x
e
d
U
r
b
a
n

a
n
d

s
u
b
u
r
b
a
n
a
r
e
a
s
1
0
0
m

<

d

<

8
k
m
F
i
x
e
d

W
i
M
A
X
I
T
U

5
2
9
-
3
3
0
0
-
1
5
0
0

M
H
z
L
(
d
,

f
,

H
R
x
)
(
p
e
r

e
n
v
i
r
o
n
m
e
n
t
)
D
i
f
f
r
a
c
t
i
o
n

l
o
s
s
D
e
y
g
o
u
t
(
1

o
b
s
t
a
c
l
e
)
D
T
M
R
a
d
i
a
l
M
a
c
r
o

c
e
l
l
M
i
n
i

c
e
l
l
S
t
r
e
e
t
M
o
b
i
l
e
1

<

d

<

1
0
0
k
m
G
S
M
,

C
D
M
A
2
0
0
0
,
L
T
E
S
t
a
n
d
a
r
d

P
r
o
p
a
g
a
t
i
o
n

M
o
d
e
l
1
5
0
-
3
5
0
0

M
H
z
L
(
d
,

H
T
x
e
f
f
,

H
R
x
e
f
f
,

D
i
f
f

l
o
s
s
,

c
l
u
t
t
e
r
)
D
e
y
g
o
u
t

(
3

o
b
s
t
a
c
l
e
s
)
E
p
s
t
e
i
n
-
P
e
t
e
r
s
o
n

(
3

o
b
s
t
a
c
l
e
s
)
D
e
y
g
o
u
t

c
o
r
r
e
c
t
e
d

(
3

o
b
s
t
a
c
l
e
s
)
M
i
l
l
i
n
g
t
o
n

(
1

o
b
s
t
a
c
l
e
)
D
T
M
C
l
u
t
t
e
r
R
a
d
i
a
l
S
y
s
t
e
m
a
t
i
c
M
a
c
r
o

c
e
l
l
M
i
n
i

c
e
l
l
S
t
r
e
e
t
R
o
o
f
t
o
p
M
o
b
i
l
e

a
n
d

F
i
x
e
d
1

<

d

<

2
0
k
m
G
S
M
,

U
M
T
S
,
C
D
M
A
2
0
0
0
,

W
i
M
A
X
,

L
T
E
W
L
L
3
0
-
1
0
0
0
0

M
H
z
F
r
e
e

s
p
a
c
e

l
o
s
s
D
i
f
f
r
a
c
t
i
o
n

l
o
s
s
D
e
y
g
o
u
t
(
3

o
b
s
t
a
c
l
e
s
)
D
T
M
C
l
u
t
t
e
r
R
a
d
i
a
l
-
S
t
r
e
e
t
R
o
o
f
t
o
p
F
i
x
e
d
F
i
x
e
d

r
e
c
e
i
v
e
r
s
W
L
L
,

M
i
c
r
o
w
a
v
e
l
i
n
k
s
,

W
i
M
A
X
I
T
U

5
2
6
-
5
3
0
-
1
0
0
0
0

M
H
z
F
r
e
e

s
p
a
c
e

l
o
s
s
D
i
f
f
r
a
c
t
i
o
n

l
o
s
s
D
e
y
g
o
u
t
(
3

o
b
s
t
a
c
l
e
s
)
D
e
y
g
o
u
t

c
o
r
r
e
c
t
e
d
(
3

o
b
s
t
a
c
l
e
s
)
D
T
M
R
a
d
i
a
l
M
a
c
r
o

c
e
l
l
S
t
r
e
e
t
F
i
x
e
d
F
i
x
e
d

r
e
c
e
i
v
e
r
s
W
L
L
I
T
U

1
5
4
6
3
0
-
3
0
0
0

M
H
z
F
r
e
e

s
p
a
c
e

l
o
s
s

+
C
o
r
r
e
c
t
i
o
n
s
- - -
M
a
c
r
o

c
e
l
l
R
o
o
f
t
o
p
M
o
b
i
l
e
1

<

d

<

1
0
0
0
k
m
L
a
n
d

a
n
d

m
a
r
i
t
i
m
e
m
o
b
i
l
e
,

b
r
o
a
d
c
a
s
t
I
T
U

3
7
0
-
7
(
V
i
e
n
n
a

9
3
)
1
0
0
-
4
0
0

M
H
z
F
r
e
e

s
p
a
c
e

l
o
s
s
C
o
r
r
e
c
t
e
d
s
t
a
n
d
a
r
d
l
o
s
s
- - -
M
a
c
r
o

c
e
l
l
R
o
o
f
t
o
p
F
i
x
e
d
d

>

1
0
k
m
L
o
w

f
r
e
q
u
e
n
c
i
e
s
B
r
o
a
d
c
a
s
t
P
r
o
p
a
g
a
t
i
o
n
m
o
d
e
l
F
r
e
q
u
e
n
c
y
b
a
n
d
P
h
y
s
i
c
a
l
p
h
e
n
o
m
e
n
a
D
i
f
f
r
a
c
t
i
o
n
c
a
l
c
u
l
a
t
i
o
n
m
e
t
h
o
d
P
r
o
f
i
l
e
b
a
s
e
d

o
n
P
r
o
f
i
l
e
e
x
t
r
a
c
t
i
o
n
m
o
d
e
C
e
l
l

s
i
z
e
R
e
c
e
i
v
e
r
l
o
c
a
t
i
o
n
R
e
c
e
i
v
e
r
U
s
e
© Forsk 2010 AT283_TRG_E1 85
Chapter 4: Calculations
4.5.1 Okumura-Hata and Cost-Hata Propagation Models
4.5.1.1 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:
f is the frequency (MHz).
h
Tx
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 A
1
, A
2
, A
3
, B
1
, B
2
, and B
3
can be user-defined. Default values are proposed in the table below:
4.5.1.2 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.
• For urban areas:
• For suburban areas:
• For quasi-open rural areas:
• For open rural areas:
a(h
Rx
) is a correction for a receiver antenna height different from 1.5m.
4.5.1.3 Calculations in Atoll
Hata models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the receiver.
1
st
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 .
2
nd
step: This step depends on whether the ‘Add diffraction loss’ option is checked.
• If the ‘Add diffraction loss’ option is unchecked, Atoll stops calculations.
Notes:
• In formulas described above, is stated in dB.
• Under Physical phenomena, L(...) expressions refer to formulas customisable in Atoll.
• SUI stands for Stanford University Interim models.
L
model
Lu A
1
A
2
f ( ) log A
3
h
Tx
( ) log B
1
B
2
h
Tx
( ) log B
3
h
Tx
+ + ( ) d log + + + =
Parameters
Okumura-Hata
f s1500 MHz
Cost-Hata
f > 1500 MHz
A
1
69.55 49.30
A
2
26.16 33.90
A
3
-13.82 -13.82
B
1
44.90 44.90
B
2
-6.55 -6.55
B
3
0 0
L
model 1
Lu a h
Rx
( ) – =
L
model 1
Lu a h
Rx
( ) – 2
f
28
------
\ .
| |
log
\ .
| |
2
– 5.4 – =
L
model 1
Lu a h
Rx
( ) – 4.78 f ( ) log ( )
2
– 18.33 f ( ) log 35.94 – + =
L
model 1
Lu a h
Rx
( ) – 4.78 f ( ) log ( )
2
– 18.33 f ( ) log 40.94 – + =
Environment a(h
Rx
)
Rural/Small city
Large city
1.1 f ( ) log 0.7 – ( )h
Rx
1.56 f ( ) log 0.8 – ( ) –
3.2 11.75h
Rx
( ) log ( )
2
4.97 –
Note:
• When receiver antenna height equals 1.5m, a(h
Rx
) is close to 0 dB regardless of
frequency.
L
model 1
86 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
• 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 .
4.5.2 ITU 529-3 Propagation Model
4.5.2.1 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:
where:
E is the field strength for 1 kW ERP
f is the frequency (MHz).
is the transmitter antenna height above ground (m) (H
b
notation is also used in Atoll)
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:
which gives the following path loss formula for the ITU 529-3 model:
4.5.2.2 Corrections to the ITU 529-3 Path Loss Formula
4.5.2.2.1 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.
for large city and urban environments
for suburban area
for rural area
4.5.2.2.2 Area Size Correction
In the formulas above, is the environment correction and is defined according to the area size
Note:
• Like for any Hata-based 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).
L
model
L
model 1
=
L
model 2
L
model
L
model 1
L
model 2
+ =
L
model
E 69.82 6.16 f log – 13.82 h
Tx
log 44.9 6.55 h
Tx
log – ( ) d log ( )
b
– + =
h
Tx
h
Rx
Lu 139.37 20 f log E – + =
Lu 69.55 26.16 f log 13.82 h
Tx
log – 44.9 6.55 h
Tx
log – ( ) d log ( )
b
+ + =
L
model 1
Lu a h
Rx
( ) – =
L
model 1
Lu a h
Rx
( ) 2
f
28
------
\ .
| |
log
\ .
| |
2
– 5.4 – – =
L
model 1
Lu a h
Rx
( ) – 4.78 f log ( )
2
– 18.33 f log 40.94 – + =
Environment a(Hr)
Rural/Small city
a h
Rx
( )
1.1 f log 0.7 – ( )h
Rx
1.56 f log 0.8 – ( ) –
© Forsk 2010 AT283_TRG_E1 87
Chapter 4: Calculations
4.5.2.2.3 Distance Correction
The distance correction refers to the term b above.
4.5.2.3 Calculations in Atoll
Hata-based models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the
receiver.
1
st
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 .
2
nd
step: This step depends on whether the ‘Add diffraction loss’ option is checked.
• If the ‘Add diffraction loss’ option is unchecked, Atoll stops calculations.
• 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 .
4.5.3 Standard Propagation Model (SPM)
4.5.3.1 SPM Path Loss Formula
SPM is based on the following formula:
with,
K
1
: constant offset (dB).
K
2
: multiplying factor for log(d).
d: distance between the receiver and the transmitter (m).
K
3
: multiplying factor for log(H
Txeff
).
HTxeff: effective height of the transmitter antenna (m).
K
4
: multiplying factor for diffraction calculation. K
4
has to be a positive number.
Diffraction loss: loss due to diffraction over an obstructed path (dB).
K
5
: multiplying factor for
K
6
: multiplying factor for .
K
7
: multiplying factor for .
: effective mobile antenna height (m).
Large city
3.2 11.75h
Rx
( ) log ( )
2
4.97 –
Distance b
d<20 km 1
d>20 km
where
b 1 0.14 1.87 10
4 –
f × 1.07 10
3 –
h'
Tx
× + + ( )
d
20
------ log
\ .
| |
0.8
× + =
h'
Tx
h
Tx
1 7 10
6 –
h
Tx
2
× +
-------------------------------------------- =
Note:
• Like for any Hata-based 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)
L
model 1
L
model
L
model 1
=
L
model 2
( )
L
model
L
model 1
L
model 2
+ =
L
model
L
model
K
1
K
2
d ( ) log K
3
H
Txeff
( ) log K
4
Di ffracti onLoss × K
5
d ( ) log H
Txeff
( ) log × + + + + +
K
6
H
Rxeff
( ) K
7
H
Rxeff
( ) log K
cl utter
f cl utter ( ) + +
=
d ( ) log H
Txeff
( ) log ×
H
Rxeff
H
Rxeff
( ) log
H
Rxeff
88 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
K
clutter
: multiplying factor for f(clutter).
f(clutter): average of weighted losses due to clutter.
4.5.3.2 Calculations in Atoll
4.5.3.2.1 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. The LOS is defined
by no obstruction along the direct ray between the transmitter and the receiver.
If the receiver is not in the transmitter line of sight, Atoll will use the set of values (K1,K2)NLOS.
4.5.3.2.2 Effective Transmitter Antenna Height
Effective transmitter antenna height (H
Txeff
) may be calculated with six different methods.
Height Above Ground
The transmitter antenna height is above the ground (H
Tx
in m).
H
Txeff
= H
Tx
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.
where,
is the ground height (ground elevation) above sea level at transmitter (m).
is the average ground height above sea level along the profile (m).
Slope at Receiver Between 0 and Minimum Distance
The transmitter antenna height is calculated using the ground slope at receiver.
where,
is the ground height (ground elevation) above sea level at receiver (m).
is the ground slope calculated over a user-defined distance (Distance min). In this case, Distance min is a distance from
receiver.
Spot H
t
If then,
If then,
Note:
• If the profile is not located between the transmitter and the receiver, H
Txeff
equals H
Tx
only.
Notes:
• If then, Atoll uses 20m in calculations.
• If then, Atoll takes 200m.
H
Txeff
H
Tx
H
0Tx
H
0
– ( ) + =
H
0Tx
H
0
H
Txeff
H
Tx
H
0Tx
+ ( ) H
0Rx
– K d × + =
H
0Rx
K
H
Txeff
20m <
H
Txeff
200m >
H
0Tx
H
0Rx
> H
Txeff
H
Tx
H
0Tx
H
0Rx
– ( ) + =
H
0Tx
H
0Rx
s H
Txeff
H
Tx
=
© Forsk 2010 AT283_TRG_E1 89
Chapter 4: Calculations
Absolute Spot H
t
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.
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:
1
st
step: Atoll determines line of sight between transmitter and receiver.
The LOS line equation is:
where,
is the receiver antenna height above the ground (m).
i is the point index.
Res is the profile resolution (distance between two points).
2
nd
step: Atoll extracts the transmitter-receiver terrain profile.
3
rd
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.
is the filtered height.
is the original height. Original terrain height is determined from extracted ground profile.
- Filter starting from transmitter
Let us assume that
For each point, we have three different cases:
1
st
case: If and ,
Then,
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).
Figure 4.13Enhanced Slope at Receiver
H
Txeff
H
Tx
H
0Tx
H
0Rx
– + =
Los i ( ) H
0Tx
H
Tx
+ ( )
H
0Tx
H
Tx
+ ( ) H
0Rx
H
Rx
+ ( ) – ( )
d
-------------------------------------------------------------------------------Res i ( ) – =
H
Rx
H
fi l t
H
ori g
H
fi l t Tx –
Tx ( ) H
ori g
Tx ( ) =
H
ori g
i ( ) H
ori g
i 1 – ( ) >
H
ori g
i ( ) H
ori g
i 1 – ( ) –
Res
------------------------------------------------------ 0.05 s
H
fi l t Tx –
i ( ) H
fi l t Tx –
i 1 – ( ) H
ori g
i ( ) H
ori g
i 1 – ( ) – ( ) + =
90 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
2
nd
case: If and
Then,
3
rd
case: If
Then,
If additionally
Then,
- Filter starting from receiver
Let us assume that
For each point, we have three different cases:
1
st
case: If and ,
Then,
2
nd
case: If and
Then,
3
rd
case: If
Then,
If additionally
Then,
Then, for every point of profile, Atoll compares the two filtered heights and chooses the higher one.
4
th
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 must contain at least three bins.
5
th
step: Atoll performs a linear regression on the filtered profile within R in order to determine a regression line.
The regression line equation is:
and
where,
i is the point index. Only points within R are taken into account.
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:
Notes:
• When several influence areas are possible, Atoll chooses the highest one.
• If d < 3000m, R = d.
H
ori g
i ( ) H
ori g
i 1 – ( ) >
H
ori g
i ( ) H
ori g
i 1 – ( ) –
Res
------------------------------------------------------ 0.05 >
H
fi l t Tx –
i ( ) H
fi l t Tx –
i 1 – ( ) =
H
ori g
i ( ) H
ori g
i 1 – ( ) s
H
fi l t Tx –
i ( ) H
fi l t Tx –
i 1 – ( ) =
H
fi l t
i ( ) H
ori g
i ( ) >
H
fi l t Tx –
i ( ) H
ori g
i ( ) =
H
fi l t
Rx ( ) H
ori g
Rx ( ) =
H
ori g
i ( ) H
ori g
i 1 + ( ) >
H
ori g
i ( ) H
ori g
i 1 + ( ) –
Res
------------------------------------------------------- 0.05 s
H
fi l t Rx –
i ( ) H
fi l t Rx –
i 1 + ( ) H
ori g
i ( ) H
ori g
i 1 + ( ) – ( ) + =
H
ori g
i ( ) H
ori g
i 1 + ( ) >
H
ori g
i ( ) H
ori g
i 1 + ( ) –
Res
------------------------------------------------------- 0.05 >
H
fi l t Rx –
i ( ) H
fi l t Rx –
i 1 + ( ) =
H
ori g
i ( ) H
ori g
i 1 + ( ) s
H
fi l t Rx –
i ( ) H
fi l t Rx –
i 1 + ( ) =
H
fi l t
i ( ) H
ori g
i ( ) >
H
fi l t Rx –
i ( ) H
ori g
i ( ) =
H
fi l t
i ( ) max H
fi l t Tx –
i ( ) H
fi l t Rx –
i ( ) , ( ) =
R 3000m >
R 0.01 d · >
y ax b + =
a
d i ( ) d
m
– ( ) H
fi l t
i ( ) H
m
– ( )
i
¿
d i ( ) d
m
– ( )
2
i
¿
------------------------------------------------------------------------ = b H
m
ad
m
– =
H
m
1
n
--- H
fi l t
i ( )
i
¿
=
d
m
d
R
2
---- – =
© Forsk 2010 AT283_TRG_E1 91
Chapter 4: Calculations
6
th
step: Then, Atoll calculates effective transmitter antenna height, (m).
If H
Txeff
is less than 20m, Atoll recalculates it with a new influence area, which begins at transmitter.
7
th
step: If is still less than 20m (even negative), Atoll evaluates path loss using and applies a
correction factor.
Therefore, if ,
where,
4.5.3.2.3 Effective Receiver Antenna Height
where,
is the receiver antenna height above the ground (m).
is the ground height (ground elevation) above sea level at the receiver (m).
is the ground height (ground elevation) above sea level at the transmitter (m).
4.5.3.2.4 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:
When the transmitter and the receiver are not in line of sight, the path loss formula is:
is determined in three steps. Influence area, R, and regression line are supposed available.
1
st
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%.
2
nd
step: Atoll evaluates the terrain roughness, Ah; it is the distance between the two lines.
3
rd
step: Atoll calculates .
We have
If ,
Else
If ,
Else
Notes:
• In case , 1000m will be used in calculations.
• If is still less than 20m, an additional correction is taken into account (7
th
step).
regr i ( ) a i Res · ( ) · b + =
H
Txeff
H
Txeff
H
0Tx
H
Tx
b – +
1 a
2
+
-------------------------------------- =
H
Txeff
1000m >
H
Txeff
H
Txeff
H
Txeff
20m =
H
Txeff
20m <
L
model
L
model
H
Txeff
20m = ( ) d f , , ( ) K
l owant
+ =
K
l owant
d
10
5
--------- 0.3 H
Txeff
20 – ( ) · ( ) –
20 1 H
Txeff
20 – ( ) – ( ) ·
9.63
d
1000
------------- +
\ .
| |
6.93
d
1000
------------- +
\ .
| |
·
------------------------------------------------------------------------------ – =
Note:
• The calculation of effective antenna heights ( and ) is based on extracted
DTM profiles. They are not properly performed if you have not imported heights (DTM file)
beforehand.
H
Rxeff
H
Rx
H
0Rx
+ ( ) H
0Tx
– =
H
Rx
H
0Rx
H
0Tx
H
Rxeff
H
Txeff
L
model
K
1 LOS ,
K
2 LOS ,
d ( ) log K
3
H
Txeff
( ) log K
5
H
Txeff
( ) d ( ) log log K
6
H
Rx
· K
cl utter
f cl utter ( ) K
hi l l LOS ,
+ + + + + + =
L
model
K
1 NLOS ,
K
2 NLOS ,
d ( ) log K
3
H
Txeff
( ) log K
4
Di ffracti on · K
5
H
Txeff
( ) d ( ) log log K
6
H
Rx
· K
cl utter
f cl utter ( ) + + + + + + =
K
hi l l LOS ,
K
hi l l LOS ,
K
hi l l LOS ,
K
h
K
hf
+ =
0 h A < 20m s K
h
0 =
K
h
7.73 h A ( ) log ( )
2
15.29 h A ( ) log – 6.746 + =
0 h A < 10m s K
hf
2 – 0.1924 H
0Rx
H
Rx
regr i
Rx
( ) – + ( ) · · =
K
hf
2 – 1.616 h A ( ) log ( )
2
– 14.75 h A ( ) log 11.21 – + ( )
H
0Rx
H
Rx
regr i
Rx
( ) – +
h A
------------------------------------------------------------ · · =
92 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
i
Rx
is the point index at receiver.
4.5.3.2.5 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:
• 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.
4.5.3.2.6 Losses due to Clutter
Atoll calculates f(clutter) over a maximum distance from receiver:
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:
• Uniform weighting function:
• Triangular weighting function:
• , where d’
i
is the distance between the receiver and the ith point and D is the maximum distance
defined.
• Logarithmic weighting function:
• Exponential weighting function:
The chart below shows the weight variation with the distance for each weighting function.
Figure 4.14Losses due to Clutter
f cl utter ( ) L
i
w
i
i 1 =
n
¿
=
w
i
1
n
--- =
w
i
d
i
d
j
j 1
n
¿
-------------- =
d
i
D d'
i
– =
w
i
d
i
D
---- 1 +
\ .
| |
log
d
j
D
---- 1 +
\ .
| |
log
j 1
n
¿
-------------------------------------- =
w
i
e
d
i
D
----
1 –
e
d
j
D
----
1 –
j 1 =
n
¿
-------------------------- =
© Forsk 2010 AT283_TRG_E1 93
Chapter 4: Calculations
4.5.3.2.7 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.
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 indoor loss as an additional
penetration loss.
4.5.3.3 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:
• K
1
, K
2
, K
3
, K
4
, K
5
, K
6
and K
7
Figure 4.15Tx-Rx profile
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).
• Like for any Hata-based 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)
• Even with no clearance, the clutter height (extracted either from clutter class or clutter
height folders) is never considered at the last profile point.
L
model
94 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
• Losses per clutter class (K
clutter
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.
4.5.3.3.1 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 x
0
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 x
0
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 x
0
. Atoll implementation is based on this method, which is explained in detail in [1].
4.5.3.3.2 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.
K
1
depends on the frequency and the technology. Here are some sample values:
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
References:
[1] Björck A. “Numerical Methods for Least Square Problems”, SIAM, 1996.
[2] Lawson C.L., Hanson R.J. “Solving Least Squares Problems”, SIAM, 1974.
Minimum Typical Maximum
K
1
Variable Variable Variable
K
2
20 44.9 70
K
3
-20 5.83 20
K
4
0 0.5 0.8
K
5
-10 -6.55 0
K
6
-1 0 0
K
7
-10 0 0
Project type Frequency (MHz) K
1
GSM 900 935 12.5
GSM 1800 1805 22
GSM 1900 1930 23
UMTS 2110 23.8
1xRTT 1900 23
WiMAX
2300 24.7
2500 25.4
2700 26.1
3300 27.8
3500 28.3
© Forsk 2010 AT283_TRG_E1 95
Chapter 4: Calculations
predictions.
All K paramaters can be defined by the automatic calibration wizard. Since K
clutter
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:
These values have to be entered only when considering statistical clutter class maps only.
4.5.3.4 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: {D5701837-B081-11D4-931D-00C04FA05664}
• Type: 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 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.
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 it’s 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.
Dense urban From 4 to 5
Woodland From 2 to 3
Urban 0
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
Note:
• 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.
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.
96 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
4.5.4 WLL Propagation Model
4.5.4.1 WLL Path Loss Formula
Where is the free space loss calculated using the formula entered in the model properties, is the diffraction loss
calculated using the 3-obstacle Deygout method, and is the diffraction multiplying factor defined in the model
properties.
4.5.4.2 Calculations in Atoll
4.5.4.2.1 Free Space Loss
Please refer to the Appendices for further details about free space loss calculation.
4.5.4.2.2 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. The final
diffraction losses are determined by multiplying the diffraction losses calculated using the Deygout method by the
Diffraction multiplying factor defined in the model properties.
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 L
model
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.
• In this case, receiver is considered in ‘line of sight’ and Atoll computes L
model
on each calculation bin using the
formula defined above.
• Otherwise, Atoll considers that L
model
tends to infinity.
4.5.5 ITU-R P.526-5 Propagation Model
4.5.5.1 ITU 526-5 Path Loss Formula
• 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.
• It’s highly recommended to use one method (Atoll.ini options) or the other one (new
identifier & signature) but not to combine both.
L
model
L
FS
F
Di ff
L
Di ff
× + =
L
FS
L
Di ff
F
Di ff
L
model
L
FS
L
Di ff
+ =
© Forsk 2010 AT283_TRG_E1 97
Chapter 4: Calculations
Where is the free space loss calculated using the formula entered in the model properties and is the diffraction
loss calculated using the 3-obstacle Deygout method.
4.5.5.2 Calculations in Atoll
4.5.5.2.1 Free Space Loss
Please refer to the Appendices for further details about free space loss calculation.
4.5.5.2.2 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.
4.5.6 ITU-R P.370-7 Propagation Model
4.5.6.1 ITU 370-7 Path Loss Formula
If d<1 km,
If d>1000 km,
If 1<d<1000 km,
d is the distance between the transmitter and the receiver (km).
4.5.6.2 Calculations in Atoll
4.5.6.2.1 Free Space Loss
Please refer to the Appendices for further details about free space loss calculation.
4.5.6.2.2 Corrected Standard Loss
This formula is given for a 60 dBm (1kW) transmitter power.
where,
C
n
is the field strength received in dBµV/m,
is a correction factor for effective receiver antenna height (dB),
A
cl
is the correction for terrain clearance angle (dB),
f is the frequency in MHz.
C
n
Calculation
The C
n
value is determined from charts C
n
=f(d, H
Txeff
).
In the following part, let us assume that C
n
=E
n
(d,H
Txeff
) (where E
n
(d,H
Txeff
) is the field received in dBµV/m) is read from
charts for a distance, d (in km), and an effective transmitter antenna height, H
Txeff
(in m).
First of all, Atoll evaluates the effective transmitter antenna height, , as follows:
If ,
If ,
If ,
where,
is the transmitter antenna height above the ground (m).
is the ground height (ground elevation) above sea level at the transmitter (m).
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).
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 H
Txeff
, Atoll determines C
n
using bilinear interpolation as follows.
L
FS
L
Di ff
L
model
L
FS
=
L
model
1000 =
L
model
max L
FS
CorrectedS dardLoss tan , ( ) =
CorrectedS dardLoss tan 60 C
n
– A
H
Rxeff
– A
cl
– 108.75 – 31.54 20 f log – + =
A
H
Rxef f
H
Txeff
0 d 3km < s H
Txeff
H
0Tx
H
Tx
H
0Rx
– + =
3 d 15km < s H
Txeff
H
0Tx
H
Tx
H
0
3 d ; ( ) – + =
15 d < H
Txeff
H
0Tx
H
Tx
H
0
3 15 ; ( ) – + =
H
Tx
H
0Tx
H
0
3 d ; ( )
H
0
3 15 ; ( )
98 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
If 37.5 H
Txeff
1200, C
n
= E
n
(d,H
Txeff
)
Otherwise, Atoll considers (d is stated in km)
Therefore,
If H
Txeff
< 37.5
If , we have
Else C
n
=E
n
(d, 37.5) – E
n
(d
horizon
, 37.5) + E
n
(25, 37.5)
If H
Txeff
> 1200
If , we have
Else C
n
=E
n
(d, 1200) – E
n
(d
horizon
, 1200) + E
n
(142, 1200)
A
HRxeff
Calculation
where,
H
Rx
is the user-defined receiver height,
c is the height gain factor.
A
cl
Calculation
If f 300 MHz,
Otherwise,
With
where,
is the clearance angle (in radians) determined according to the recommendation 370-7 (figure 19),
f is the frequency stated in MHz.
4.5.7 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:
Where . 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 d
0
is a fixed reference distance (100 m). a(H
BS
) is the
correction factor for base station antenna heights, H
BS
:
Where , 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:
Note:
• c values are provided in the recommendation 370-7; for example, c=4 in a rural case.
d
hori zon
4.1 H
Txeff
· =
d d
hori zon
> C
n
E
n
d 25 d
hori zon
– + 37.5 , ( ) =
d d
hori zon
> C
n
E
n
d 142 d
hori zon
– + 1200 , ( ) =
A
H
Rxef f
c
6
--- 20
H
Rx
10
----------
\ .
| |
log · · =
s A
cl
8.1 6.9 20 v 0.1 – ( )
2
1 + ( ) v 0.1 – ( ) + ( ) log + | | – =
A
cl
14.9 6.9 20 v 0.1 – ( )
2
1 + ( ) v 0.1 – ( ) + ( ) log + | | – =
v u – 4000
f
300
---------- · · =
u
PL A 10 a H
BS
( ) Log
10
d
d
0
------
\ .
| |
· · + =
A 20 Log
10
4td
0
ì
-------------
\ .
| |
· =
a H
BS
( ) a b H
BS
c
H
BS
---------- + · – =
10 m H
BS
80 m s s
PL A 10 a H
BS
( ) Log
10
d
d
0
------
\ .
| |
a f ( ) a H
R
( ) – + · · + =
© Forsk 2010 AT283_TRG_E1 99
Chapter 4: Calculations
Where a(f) is the correction factor for the operating frequency, , with f being the operating
frequency in MHz. a(H
R
) is the correction factor for the receiver antenna height, , where d
depends on the terrain type.
4.5.7.1 SUI Terrain Types
The SUI models are divided into three types of terrains
1
, 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.
4.5.7.2 Erceg-Greenstein (SUI) Path Loss Formula
The Erceg-Greenstein (SUI) propagation model formula can be simplified from the following equation:
(1)
to the equation below:
(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)
• H
BS
is the transmitter height in m
• H
R
is the receiver height in m
The above equation is divided into two parts in Atoll:
Where,
The above path loss formulas are valid for d > d
0
, 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:
instead of
Where a(f) and a(Hr) have the same definition as given above. Simplifying the above equation, we get,
, or
Note:
• a(H
R
) = 0 for H
R
= 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
a f ( ) 6 Log
10
f
2000
-------------
\ .
| |
· =
a H
R
( ) X Log
10
H
R
2
-------
\ .
| |
· =
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 description.
Model Parameter Terrain A Terrain B Terrain C
a 4.6 4.0 3.6
b (m
-1
)
0.0075 0.0065 0.005
c (m) 12.6 17.1 20
X 10.8 10.8 20
PL 20 Log
10
4td
0
ì
-------------
\ .
| |
· 10 a H
BS
( ) Log
10
d
d
0
------
\ .
| |
a f ( ) a H
R
( ) – + · · + =
PL 7.366 – 26 Log
10
f ( ) 10 a H
BS
( ) 1 Log
10
d ( ) + ( ) a H
R
( ) – · · + · + =
PL Lu a H
R
( ) – =
Lu 7.366 – 26 Log
10
f ( ) 10 a H
BS
( ) 1 Log
10
d ( ) + ( ) · · + · + =
PL 20 Log
10
4 t d · ·
ì
------------------
\ .
| |
a f ( ) a H
R
( ) – + · = PL 20 Log
10
4 t d · ·
ì
------------------
\ .
| |
· =
PL 12.634 26 Log
10
f ( ) 20 Log
10
d ( ) a H
R
( ) – · + · + = Lu 12.634 26 Log
10
f ( ) 20 Log
10
d ( ) · + · + =
100 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
The above equation is not user-modifiable in Atoll except for the coefficient of , i.e. 26. Atoll uses the same
coefficient as the one you enter for in Atoll for the case d > d
0
.
4.5.7.3 Calculations in Atoll
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.
1
st
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.
2
nd
step: This step depends on whether the ‘Add diffraction loss’ option is selected or not.
• If the ‘Add diffraction loss’ option is not selected, 1
st
step gives the final path loss result.
• If the ‘Add diffraction loss’ option is selected, Atoll proceeds as follows:
a. It extracts a geographic profile between the transmitter and the receiver using 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 . For more information on the Deygout method, see "3 Knife-Edge Deygout
Method" on page 107.
The final path loss is the sum of the path loss determined in 1
st
step and .
Shadow fading is computed in Atoll independent of the propagation model. For more information on the shadow fading
calculation, see "Shadowing Model" on page 115.
4.5.8 ITU-R P.1546-2 Propagation Model
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 , as a function of distance for:
• Nominal frequencies, : 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, : 10, 20, 37.5, 75, 150, 300, 600, and 1200 m
For any values of 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 below 10 m, the extrapolation to be applied is given
in Annex 5, § 4.2. It is possible for the value of to be negative, in which case the method is given in Annex 5,
§ 4.3.
• Time variability, : 1, 10, and 50 %
The propagation curves represent the field strength values exceeded for 1, 10 and 50 % of time.
• Receiver antenna height, : 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.
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.
Note:
• You can get the same resulting equation by setting a(hBS) = 2.
Log
10
f ( )
Log
10
f ( )
L
Di ffracti on
L
Di ffracti on
Notes:
• 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.
db µV m ( )
f
n
h
1
h
1
h
1
h
1
t
h
2
© Forsk 2010 AT283_TRG_E1 101
Chapter 4: Calculations
4.5.8.1 Calculations in 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, is the transmission frequency, is the transmitter-receiver distance, and 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.
4.5.8.1.1 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., .
Once and are known, along with the information about the percentage of time and the type of path (land or sea),
the sets of graphs which will be used for the calculation are also known.
4.5.8.1.2 Step 2: Calculation of Maximum Field Strength
A field strength must not exceed a maximum value, , which is given by:
for land paths, and
for sea paths.
Where is the free space field strength for 1 kW ERP, is an enhancement for sea graphs.
4.5.8.1.3 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
• Sea paths
Here, all antenna heights (i.e., , , and ) are in expressed in m. is the antenna height above ground and
is the effective height of the transmitter antenna, which is its height over the average level of the ground between distances
of and d km from the transmitter in the direction of the receiver.
4.5.8.1.4 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, , from the graphs for transmitter antenna height to determine . The second
interpolation/extrapolation is performed over the interpolated/extrapolated values of to determine . And, the thrid
and final interpolation/extrapolation is performed over the interpolated/extrapolated values of to determine .
Step 4.1: Interpolation/Extrapolation of Field Strength for Transmitter Antenna Height
If the value of 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
The field strength is interpolated or extrapolated from field strengths obtained from two curves using the following
equation:
Where if , otherwise is the nearest nominal effective height below ,
if , otherwise is the nearest nominal effective height above , is the field
strength value for at the required distance, and is the field strength value for at the required
distance.
• If
f d t
f
n1
f f
n2
< <
f
n1
f
n1
t
E
Max
E
Max
E
FS
106.9 20 Log d ( ) × – = =
E
Max
E
FS
E
SE
+ 106.9 20 Log d ( ) × – 2.38 1 d 8.94 – ( ) exp – { } Log 50 t ( ) × + = =
E
FS
E
SE
h
1
h
eff
=
h
1
Max 1 h
a
, ( ) =
h
1
h
eff
h
a
h
a
h
eff
0.2 d ×
E E
h1
E
h1
E
d
E
d
E
f
h
1
10 m h
1
3000 m < <
E
h1
E
Low
E
Up
E
Low
– ( )
Log h
1
h
Low
( )
Log h
Up
h
Low
( )
------------------------------------------ × + =
h
Low
600 m = h
1
1200 m > h
Low
h
1
h
Up
1200 m = h
1
1200 m > h
Up
h
1
E
Low
h
Low
E
Up
h
Up
0 m h
1
10 m < <
102 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
- For land path if the transmitter-receiver distance is less than the smooth-Earth horizon distance
, i.e., if ,
, or
because
- For land path if the transmitter-receiver distance is greater than or equal to the smooth-Earth horizon distance
, i.e., if ,
, or because
Where 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, even though , the field strength is
determined from linear extrapolation for Log (distance) of the graph given by:
Where is penultimate tabulation distance (km), is the final tabulation distance (km), is the
field strength value for , and is the field strength value for .
- For sea path, 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:
(km)
Where (km) with (frequency-dependent term),
and (asymptotic term defined by the horizon distance).
If the 0.6 Fresnel clearance distance for the sea path where the transmitter antenna height is 20 m is
also calculated as:
(km)
Once and are known, the field strength for the required distance is given by:
Where is the maximum field strength at the required distance as calculated in "Step 2: Calculation of
Maximum Field Strength" on page 101, is for ,
, ,
and is the field strength calculated as described for land paths. and are field strengths
interpolated for distance y and , respectively, and .
• If
A correction is applied to the field strength, , 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, .
Where with and ,
, and is 1.35 for 100 MHz, 3.31 for 600 MHz, 6.00 for 2000 MHz.
d
H
h
1
( ) 4.1 h
1
× = d 4.1 h
1
× <
E
h1
E
10
d
H
10 ( ) ( ) E
10
d ( ) E
10
d
H
h
1
( ) ( ) – + =
E
h1
E
10
12.9 km ( ) E
10
d ( ) E
10
d
H
h
1
( ) ( ) – + = d
H
10 ( ) 12.9 km =
d
H
h
1
( ) 4.1 h
1
× = d 4.1 h
1
× >
E
h1
E
10
d
H
10 ( ) d d
H
h
1
( ) – + ( ) = E
h1
E
10
12.9 km d d
H
h
1
( ) – + ( ) = d
H
10 ( ) 12.9 km =
E
x
y ( )
d
H
10 ( ) d d
H
h
1
( ) – + 1000 km > d 1000 km s
E
h1
E
Low
E
Up
E
Low
– ( )
Log d D
Low
( )
Log D
Up
D
Low
( )
-------------------------------------------- × + =
D
Low
D
Up
E
Low
D
Low
E
Up
D
Up
h
1
D
h1
D
0.6
f h
1
h
2
10 m = ( ) , , ( ) =
D
0.6
Max 0.001
D
f
D
h
×
D
f
D
h
+
------------------- ,
\ .
| |
= D
f
0.0000389 f h
1
h
2
× × × =
D
h
4.1 h
1
h
2
+ ( ) × =
d D
h1
>
D
20
D
0.6
f h
1
20 m = ( ) h
2
10 m = ( ) , , ( ) =
D
h1
D
20
E
h1
E
Max
for d D
h1
s
E
D
h1
E
D
20
E
D
h1
– ( )
Log d D
h1
( )
Log D
20
D
h1
( )
--------------------------------------- × + for D
h1
d D
20
< <
E' 1 F
S
– ( ) × E'' F
S
× + for d D
20
>
¹
¦
¦
´
¦
¦
¦
=
E
Max
E
D
h1
E
Max
d D
h1
=
E
D
20
E
10
D
20
( ) E
20
D
20
( ) E
10
D
20
( ) – ( )
Log h1 10 ( )
Log 20 10 ( )
---------------------------------- × + = E' E
10
d ( ) E
20
d ( ) E
10
d ( ) – ( )
Log h1 10 ( )
Log 20 10 ( )
---------------------------------- × + =
E'' E
10
y ( ) E
20
y ( )
h
1
10 m and 20 m = F
S
d D
20
– ( ) d =
h
1
0 m <
E
h1
C
h1
Max C
h1d
C
h1t
, ( ) =
C
h1d
6.03 J v ( ) – = J v ( ) 6.9 20 Log v 0.1 – ( )
2
1 + v 0.1 – + ( ) × + | | = v K
v
u
eff2
× =
u
eff2
arc
h
1

9000
-------------
\ .
| |
tan = K
v
© Forsk 2010 AT283_TRG_E1 103
Chapter 4: Calculations
with , (radius of the Earth), and is the
effective 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:
Where is the lower value of the nearest tabulated distance to , is the higher value of the nearest tabulated
distance to , is the field strength value for , and is the field strength value for .
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:
Where is the lower nominal frequency (100 MHz if f < 600 MHz, 600 MHz otherwise), is the higher nominal
frequency (600 MHz if f < 600 MHz, 2000 MHz otherwise), is the field strength value for , and is the field
strength value for .
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.
4.5.8.1.5 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, (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 is 10 m.
For land paths, the elevation angle of the arriving ray is taken into account by calculating a modified representative clutter
height , given by .
Note that for , .
The different correction factors are calculated as follows:
• For land path in urban and suburban zones
With and .
If , is reduced by .
• For land path other zones
C
h1t
30 Log
u
e
u
e
u
eff2
+
------------------------
\ .
| |
× = u
e
180 d ×
a t k × ×
---------------------- = a 6370 km = k 4 3 =
E
d
E
Low
E
Up
E
Low
– ( )
Log d d
Low
( )
Log d
Up
d
Low
( )
------------------------------------------ × + =
d
Low
d d
Up
d E
Low
d
Low
E
Up
d
Up
E
f
E
Low
E
Up
E
Low
– ( )
Log f f
Low
( )
Log f
Up
f
Low
( )
--------------------------------------- × + =
f
Low
f
Up
E
Low
f
Low
E
Up
f
Up
R
R
R' R' Max 1
1000 d R × × 15 h
1
× – ( )
1000 d × 15 –
--------------------------------------------------------------- ,
\ .
| |
=
h
1
6.5 d × R + < R' R ~
C
Recei ver
6.03 J v ( ) – for h
2
R' <
3.2 6.2 Log f ( ) × + ( ) Log
h
2
R'
------
\ .
| |
× for h
2
R' >
¹
¦
´
¦
¦
=
J v ( ) 6.9 20 Log v 0.1 – ( )
2
1 + v 0.1 – + ( ) × + | | = v 0.0108 f × R' h
2
– ( ) arc
R' h
2

27
-----------------
\ .
| |
tan × × =
R' 10 m < C
Recei ver
3.2 6.2 Log f ( ) × + ( ) Log
10
R'
------
\ .
| |
×
C
Recei ver
3.2 6.2 Log f ( ) × + ( ) Log
h
2
10
------
\ .
| |
× =
104 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
• For sea path
and 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 and variable , respectively. These distances are given by
and (km), respectively. Here
as explained earlier.
- If ,
- If and ,
- If and and ,
- If and and ,
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:
Where 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 and .
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:
Where , , and
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.
• : The reference angle, .
Where and are the heights of the transmitter and the receiver above sea level, respectively.
4.5.8.1.6 Step 6: Calculation of Path Loss
First, the final field strength is calculated from the interpolated/extrapolated field strength, , by applying the corrections
calculated earlier. The calculated field strength is given by:
The resulting field strength is given by , from which the path loss (basic transmission loss, ) is
calculated as follows:
4.5.9 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]:
d
10
d
h2
h
2
10 m = h
2
d
10
D
0.6
f h
1
h
2
10 m = ( ) , , ( ) = d
h2
D
0.6
f h
1
h
2
, , ( ) = D
0.6
Max 0.001
D
f
D
h
×
D
f
D
h
+
------------------- ,
\ .
| |
=
h
2
10 m > C
Recei ver
3.2 6.2 Log f ( ) × + ( ) Log
h
2
10
------
\ .
| |
× =
h
2
10 m < d d
10
> C
Recei ver
3.2 6.2 Log f ( ) × + ( ) Log
h
2
10
------
\ .
| |
× =
h
2
10 m < d d
10
< d d
h2
< C
Recei ver
0 =
h
2
10 m < d d
10
< d d
h2
> C
Recei ver
3.2 6.2 Log f ( ) × + ( ) Log
h
2
10
------
\ .
| |
×
Log d d
h2
( )
Log d
10
d
h2
( )
-------------------------------------
\ .
| |
× =
C
Bui l di ng
3.3 – Log f ( ) ( ) 1 0.85 Log d ( ) × – ( ) 1 0.46 Log 1 h
a
R – + ( ) × – ( ) =
h
a
d 15 km < h
1
R – 150 m <
C
Cl earance
J v' ( ) J v ( ) – =
J v ( ) 6.9 20 Log v 0.1 – ( )
2
1 + v 0.1 – + ( ) × + | | = v' 0.036 f × = v 0.065 u
Cl earance
f × × =
u
Cl earance
u
u
Ref
u
Ref
arc
h
1S
h
2S

1000 d ×
------------------------
\ .
| |
tan =
h
1S
h
2S
E
f
E
Cal c
E
f
C
Recei ver
C
Bui l di ng
C
Cl earance
+ + + =
E Mi n E
Cal c
E
Max
, ( ) = L
B
L
B
139 E – 20 Log f ( ) × + =
© Forsk 2010 AT283_TRG_E1 105
Chapter 4: Calculations
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
• h
s
is the height (in meters) of the buildings close to the receiver
• H
1
is the average height (in meters) of the buildings close to the receiver
• h
b
is the height (in meters) of the transmitter antenna with respect to the observer
• h
b0
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:
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:
Where a is a corrective factor with three components:
• W is the width (in meters) of the streets where the receiver is located
• H
0
(= h
s
= H
1
) is the height (in meters) of the buildings close to the receiver
• h
b
(= h
b0
) is the height (in meters) of the transmitter antenna with respect to the ground
• h
m
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:
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:
The extended Sakagami propagation model is valid for:
5 m < W < 50 m
0° < < 90°
5 m < h
s
< 80 m
5 m < H
1
< 50 m
20 m < h
b
< 100 m
0.5 km < d < 10 km
450 MHz < f < 2200 MHz
5 m < W < 50 m
10 m < H
0
< 30 m
10 m < h
b
< 100 m
0.1 km < d < 3 km
0.8 GHz < f < 8 GHz
1.5 m < h
m
< 5 m
10 m < h
b
< 100 m
0.1 km < d < 3 km
3 GHz < f < 8 GHz
1.5 m < h
m
< 5 m
L
Model
100 7.1 Log W ( ) × – 0.023 ¢ × 1.4 Log h
s
( ) × 6.1 Log H
1
( ) × 24.37 3.7
H
h
b0
--------
\ .
| |
2
× – Log h
b
( ) × – + + + +
43.2 3.1 Log h
b
( ) × – | | Log d ( ) × 20 Log f ( ) × e
13 Log f ( ) 3.23 – ( ) ×
+ +
=
¢
¢
h
b0
H >
L
Model
54 40 Log d ( ) × 30 Log h
b
( ) × – 21 Log f ( ) × a + + + =
a a H
0
( ) a W ( ) a h
m
( ) + + 11 Log
H
0
20
-------
\ .
| |
× 7.1 Log
W
20
------
\ .
| |
× – 5 Log
h
m
1.5
--------
\ .
| |
× – = =
L
Model
54 40 Log d ( ) × 30 Log h
b
( ) × – 21 Log f ( ) × 5 Log h
m
( ) × – + + =
106 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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:
For more information on the Standard Propagation Model, see "Standard Propagation Model (SPM)" on page 87.
4.5.10 Appendices
4.5.10.1 Free Space Loss
The calculation of free space loss is based on ITU 525 recommendations.

where,
f is the frequency in MHz,
d is the Tx-Rx distance in km,
Free space loss is stated in dB.
4.5.10.2 Diffraction Loss
The calculation of diffraction is based on ITU 526-5 recommendations. General method for one or more obstacles (knife-
edge 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).
4.5.10.2.1 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(v), depends on the obstruction parameter (v), which corresponds to the ratio of the
obstruction height (h) and the radius of the Fresnel zone (R).
K1 65.4 (calculated for 3.5 GHz)
K2 40
K3 -30
K4 0
K5 0
K6 0
K7 -5
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.
L
FS
32.4 20 f ( ) log 20 d ( ) log + + =
Figure 4.16Knife-Edge Diffraction
© Forsk 2010 AT283_TRG_E1 107
Chapter 4: Calculations
where,
n is the Fresnel zone index,
c
0
is the speed of light (2.99792 x10
8
ms
-1
),
f is the frequency in Hz
d
1
is the distance from the transmitter to obstacle in m,
d
2
is the distance from obstacle to receiver in m.
We have:
where,
h is the obstruction height (height from the obstacle top to the Tx-Rx axis).
Hence,
For 1 knife-edge method, if ,
Else,
4.5.10.2.2 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 (v) value.
1 Obstacle
A straight line between transmitter and receiver is drawn and the height of the obstacle above the Tx-Rx axis, h
i
, is
calculated. The obstruction position, d
i
, is also recorded. v
i
are evaluated from these data. The point with the highest v
value is termed the principal edge, p, and the corresponding loss is J(vp).
Therefore, we have
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-edge-
receiver section.
Note:
• In case of multiple-knife edge method, the minimum required to estimate diffraction loss
is -0.78.
R
c
0
n d
1
d
2
· · ·
f d
1
d
2
+ ( ) ·
---------------------------------- =
v
h
r
--- =
r
R
2
------- =
v 0.7 – > J v ( ) 6.9 20 v 0.1 – ( )
2
1 + v 0.1 – ( ) + ( ) log · + =
J v ( ) 0 =
v
Figure 4.17Deygout Construction – 1 Obstacle
Di ffracti onLoss J v
P
( ) =
108 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
The same procedure is repeated on each half profile to determine the edge with the higher v. The two obstacles found,
(points t and r), are called ‘secondary edges’. Losses induced by the secondary edges, J(v
t
) and J(v
r
), are then calculated.
Once the edge hierarchy is determined, the total loss is evaluated by adding all the intermediary losses obtained.
Therefore, if
we have
Otherwise, If ,
4.5.10.2.3 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, h
p
, is recalculated
according to the Epstein-Peterson construction. h
p
is the height above a straight line connecting t and r points. The main
edge position d
p
is recorded and v
p
and J(v
p
) are evaluated from these data.
Therefore, we have
4.5.10.2.4 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.
Figure 4.18Deygout Construction – 3 Obstacles
Note:
• In case of ITU 526-5 and WLL propagation models, Diffraction loss term is determined as
follows:
- If , we have
Where,
- Otherwise
v
P
0 >
Di ffracti onLoss J v
P
( ) J v
t
( ) J v
r
( ) + + =
v
P
0.7 – > Di ffracti onLoss J v
P
( ) =
v
P
0.78 – > Di ffracti onLoss J v
P
( ) J v
t
( ) J v
r
( ) + ( ) t · + =
t mi n
J v
P
( )
6
-------------- 1 ,
\ .
| |
=
Di ffracti onLoss 0 =
Figure 4.19Epstein-Peterson Construction
Di ffracti onLoss J v
P
( ) J v
t
( ) J v
r
( ) + + =
© Forsk 2010 AT283_TRG_E1 109
Chapter 4: Calculations
Therefore, If ,
we have
Otherwise
4.5.10.2.5 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, h
h
, is calculated. The position d
h
is recorded and
then, from these values, v
h
and J(v
h
) are evaluated using the same previous formulas.
Therefore, we have
4.6 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 tuned 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 60.
4.6.1 Transmitter Path Loss Tuning
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. Total 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 tuning is done to smooth the local corrections (step 2) of measured values and not the tuned bins.
2. Local 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 :
Note:
• In case of ITU 526-5 propagation model, Diffraction loss term is determined as follows:
- If , we have
Where,
(d: distance stated in km between the transmitter and the receiver).
- Otherwise
v
P
0 >
Di ffracti onLoss J v
P
( ) J v
t
( ) J v
r
( ) C + + + =
Di ffracti onLoss J v
P
( ) C + =
v
P
0.78 – > Di ffracti onLoss J v
P
( ) t J v
t
( ) J v
r
( ) C + + ( ) · + =
t mi n
J v
P
( )
6
-------------- 1 ,
\ .
| |
=
C 8.0 0.04d + =
Di ffracti onLoss 0 =
Figure 4.20Millington Construction
Di ffracti onLoss J v
h
( ) =
110 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
• The radius of the axis parallel to the Profile (A)
• The radius of the axis perpendicular to the Profile (B)
Let’s take M a measurement value and the path loss value at point i, before any tuning.
The squared elliptic distance between i and M is given by:
Where:
and are the X-coordinates of i and M respectively
and are the Y-coordinates of i and M respectively
The mean error for the first tuning is given by:
Where is the error between measurement and prediction at point i
Then, the path loss value is tuned using E:
Finally, a second tuning ( ) is applied where:
so
Where is (measurement gain - losses).
So, the final tuned path loss is:
so
When several ellipses overlap a pathloss bin, the final tuned path loss is given by:
Where is the number of overlapping ellipses
4.6.2 Repeater Path Loss Tuning
In the case of repeaters, Atoll provides only a composite measured value per pixel which is a combination of the
contribution of both a transmitter and one or several repeaters. In order to tune the path loss matrices of donor transmitters
and repaters, its is mandatory to split the contribution of each element in the measured value as starting point.
Let’s take M the measured value.
where :
represents the contribution of the donor transmitter in the measured value.
represents the contribution of the repeater in the measured value.
Note:
• M is limited by the minimum measurement threshold defined in the interface.
Note:
• E is limited by the maximum total correction defined in the interface.
Note:
• is limited by the maximum local correction defined in the interface.
P
i
D
i
X
i
X
M
– ( )
2
A
2
--------------------------
Y
i
Y
M
– ( )
2
B
2
-------------------------- + =
X
i
X
M
Y
i
Y
M
E
1
n
---
\ .
| |
e
i
i
¿
× =
e
i
P
i
new
P
i
ol d
E + =
R
i
R
i
1 D
i
– ( ) M g P
i

new
– ( ) × = R
i
1 D
i
– ( ) M g P
i
ol d
E + ( ) – – ( ) × =
g
R
i
P
i
tuned
P
i
new
R
i
+ = P
i
tuned
P
i
ol d
E R
i
+ + =
P
i
tuned
1 d
j
– ( )P
j
tuned
j
¿
\ .
|
| |
n d
j
j
¿

\ .
|
| |
---------------------------------------------------- =
n
Note:
• All the values are used in Watts
M M
d
M
r
+ =
M
d
M
r
© Forsk 2010 AT283_TRG_E1 111
Chapter 4: Calculations
If and represent respectively the filtered signal level from the donor transmitter and the repeater on a pixel, one
can define the contribution of each element as follows:
and .
Following the path loss tuning process described in "Transmitter Path Loss Tuning" on page 109, the donor transmitter
(resp. the repeater) is then tuned using (resp. ) values.
4.7 Antenna Attenuation Calculation
The modelling method used to evaluate transmitter antenna attenuation, , 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.
4.7.1 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).
a
Tx
and e
Tx
are respectively the transmitter (Tx) antenna azimuth and tilt in the coordinate system .
a
Rx
and e
Rx
are respectively the azimuth and tilt of the receiver (Rx) in the coordinate system .
d is the distance between the transmitter (Tx) and the receiver (Rx).
In the coordinate system , the receiver coordinates are:
(1)
Let az and el respectively be the azimuth and tilt of the receiver in the transmitter antenna coordinate system
. These angles describe the direction of the transmitter-receiver path in the transmitter antenna coordinate
system. Therefore, the receiver coordinates in are:
(2)
C
d
C
r
M
d
M
C
d
C
d
C
r
+
------------------- × = M
r
M
C
r
C
d
C
r
+
------------------- × =
M
d
M
r
L
antTx
Figure 4.21Azimuth and Tilt Computation
S
0
x y z , , ( )
S
0
x y z , , ( )
S
0
x y z , , ( )
x
Rx
y
Rx
z
Rx
e
Rx
( ) cos a
Rx
( ) sin d · ·
e
Rx
( ) cos a
Rx
( ) cos d · ·
e
Rx
( ) sin – d ·
=
S
Tx
x'' y'' z'' , , ( )
S
Tx
x'' y'' z'' , , ( )
x''
Rx
y''
Rx
z''
Rx
el ( ) cos az ( ) sin d · ·
el ( ) cos az ( ) cos d · ·
el ( ) sin – d ·
=
112 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
According to the figure above, we have the following relations:
(3)
and
(4)
Therefore, the relation between the system and the transmitter antenna system is:
(5)
We get,
(6)
Then, substituting the receiver coordinates in the system S
0
from Eq. (1) and the receiver coordinates in the system S
Tx
from Eq. (2) in Eq. (6) leads to a system where two solutions are possible:
1
st
solution: If , then and
2
nd
solution: If , then
and
If , then
4.7.2 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, .
x'
y'
z'
a
Tx
( ) cos a
Tx
( ) sin – 0
a
Tx
( ) sin a
Tx
( ) cos 0
0 0 1
x
y
z
- =
x''
y''
z''
1 0 0
0 e
Tx
( ) cos e
Tx
( ) sin –
0 e
Tx
( ) sin e
Tx
( ) cos
x'
y'
z'
- =
S
0
x y z , , ( ) S
Tx
x'' y'' z'' , , ( )
x''
y''
z''
1 0 0
0 e
Tx
( ) cos e
Tx
( ) sin –
0 e
Tx
( ) sin e
Tx
( ) cos
a
Tx
( ) cos a
Tx
( ) sin – 0
a
Tx
( ) sin a
Tx
( ) cos 0
0 0 1
-
x
y
z
- =
x''
y''
z''
a
Tx
( ) cos a
Tx
( ) sin – 0
e
Tx
( ) cos a
Tx
( ) sin · e
Tx
( ) cos a
Tx
( ) cos · e
Tx
( ) sin –
e
Tx
( ) sin a
Tx
( ) sin · e
Tx
( ) sin a
Tx
( ) cos · e
Tx
( ) cos
x
y
z
- =
a
Rx
a
Tx
= az 0 = el e
Rx
e
Tx
– =
a
Rx
a
Tx
=
az
1
e
Tx
( ) cos
a
Rx
a
Tx
– ( ) tan
--------------------------------------
e
Tx
( ) sin e
Rx
( ) tan ·
a
Rx
a
Tx
– ( ) sin
-------------------------------------------------- +
------------------------------------------------------------------------------------------------ atan =
el az ( ) sin
e
Tx
( ) sin –
a
Rx
a
Tx
– ( ) tan
--------------------------------------
e
Tx
( ) cos e
Rx
( ) tan ·
a
Rx
a
Tx
– ( ) sin
---------------------------------------------------- +
¹ )
´ `
¦ ¹
· atan =
az ( ) sin a
Rx
a
Tx
– ( ) sin · 0 < az az 180 + =
Notes:
• Atoll assumes 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(180)=V(180)
In case of an electrical tilt, o, the horizontal pattern is a conical section with a o degrees
elevation off the horizontal plane. Here, horizontal and vertical patterns must satisfy the
following:
H(0)=V(o) and H(180)=V(180-o)
If the constraints listed above are satisfied, this implies that:
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.
• The above interpolation is performed in dBs.
L
antTx
az el , ( )
L
antTx
az el , ( ) H az ( )
180 az –
180
------------------------- H 0 ( ) V el ( ) – ( ) ·
az
180
---------- H 180 ( ) V 180 el – ( ) – ( ) · + – =
© Forsk 2010 AT283_TRG_E1 113
Chapter 4: Calculations
4.7.3 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:
when
when
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, .
4.7.4 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 (A
Smoothing
) 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 (N
Smoothing
) and their corresponding angles
according to the defined Peak-to-Null Deviation (D
Peak-to-Null
). D
Peak-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
(A
Smoothing
) by applying the smoothing to a certain smoothing factor (F
Smoothing
) defined by the user.
Let’s take an example of an antenna pattern to be smoothed, as shown in Figure 4.23 on page 114. Let D
Peak-to-Null
be 10
dB, A
Smoothing
= 90 degrees, and F
Smoothing
= 0.5.
• Angle values in formulas are stated in degrees.
• The above interpolation is not used in case the transmitter antenna has a 3-D antenna
pattern.
Figure 4.22Vertical Pattern Transformation due to Electrical Downtilt
V x ( ) V x AEDT – ( ) = x 90 – 90 [ , ] e
V x ( ) V x AEDT + ( ) = x 90 270 [ , ] e
L
antTx
az el , ( )
114 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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.
Peaks (P) and Nulls (N):
Then, Atoll verifies whether the difference of attenuation at a given angle is D
Peak-to-Null
less than the before and after it.
This comparison determines the nulls to be smoothed (N
Smoothing
).
Nulls to be smoothed (N
Smoothing
):
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.
Figure 4.23Vertical Antenna Pattern
Figure 4.24Peaks and Nulls in the Antenna Pattern
Angle (°) Attenuation (dB)
1 0.1
15 33.5
21 13.2
30 37.6
38 16.9
49 32.2
67 15.6
Angle (°) Attenuation (dB)
15 33.5
30 37.6
49 32.2
© Forsk 2010 AT283_TRG_E1 115
Chapter 4: Calculations
4.7.4.1 Smoothing Algorithm
For all nulls surrounded by two peaks P
1
and P
2
at angles and ,
Where,
i is the angle in degrees from to incremented by 1 degree,
A
Angle
is the attenuation at any given angle which can be i, or , and
F
Smoothing
is the smoothing factor defined by the user.
4.8 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 transmitter-
receiver 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 4.25 on page 115, whose
standard deviation depends on the type of clutter.
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 4.25 on page 115.
The accuracy of this model depends upon:
• 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.
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.
n N
Smoothi ng
e o
1
o
2
A
i Smoothed ,
A
i
F
Smoothi ng
A
i
A
o
1
A
o
2
A
o
1

o
2
o
1

------------------------
\ .
|
| |
i o
1
– ( ) · +
¹ )
´ `
¦ ¹
– – =
o
1
o
2
o
1
o
2
Figure 4.25Log-normal Probability Density Function
o
116 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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.
GSM GPRS EGPRS Documents
The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 120, 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 ( ) is applied to the
received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage
probability, and depends on the model standard deviation ( 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 window’s Interference tab that require calculation of the received
signal level and interference received from other base stations.
In these calculations, ( calculations), the shadowing margin ( ) is applied 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 (
in dB) associated to the clutter class where the receiver is located.
UMTS HSPA and CDMA2000 1xRTT 1xEV-DO Documents
The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 120 and
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 121, and applied to signal level, E
c
/I
0
, or E
b
/N
t
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 ( ) is applied to the
received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage
probability, and depends on the model standard deviation ( in dB) associated to the clutter class where the
receiver is 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 interference and noise received from other base stations.
In these calculations, the shadowing margins ( , , or
) are applied to E
c
/I
0
or Eb/Nt. These shadowing margins are calculated for a given cell
edge coverage probability and depend on the E
c
/I
0
or Eb/Nt standard deviations ( , , or
, in dB) associated to the clutter class where the receiver is located.
• Macro-Diversity Gains
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. “Systèmes 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
M
Shadowi ng model –
o
model
C I M
Shadowi ng C I –
o
C I
M
Shadowi ng model –
o
model
M
Shadowi ng Ec Io –
M
Shadowi ng Eb Nt ( )
DL

M
Shadowi ng Eb Nt ( )
UL

o
Ec Io
o
Eb Nt ( )
DL
o
Eb Nt ( )
UL
© Forsk 2010 AT283_TRG_E1 117
Chapter 4: Calculations
Atoll calculates the uplink and downlink macro-diversity gains ( and ) 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 macro-
diversity gains, please refer to "Macro-Diversity Gains Calculation" on page 122.
• Monte-Carlo Simulations
Random values for shadowing margins are calculated for each transmitter-receiver link and applied to the
predicted signal level. 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 log-normal distribution as explained above.
TD-SCDMA Documents
The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 120 and
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 121, and applied to signal level or interference+noise
predictions as explained below.
• Signal Level-Based Predictions
Signal level-based predictions include coverage predictions (Best Server and RSCP P-CCPCH Coverages, P-
CCPCG 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.
In these calculations (signal level calculations), a shadowing margin ( ) is applied to the
received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage
probability, and depends on the model standard deviation ( in dB) associated to the clutter class where the
receiver is located.
• Interference+noise-Based Predictions
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 ( , , or
) are applied to Eb/Nt. These shadowing margins are calculated for a given cell edge
coverage probability and depend on the Eb/Nt standard deviations ( , , or ,
in dB) associated to the clutter class where the receiver is located.
• Monte-Carlo Simulations
Random values for shadowing margins are calculated for each transmitter-receiver link and applied to the
predicted signal level. 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 log-normal distribution as explained above.
WiMAX 802.16d and WiMAX 802.16e Documents
The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 120 and
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 121 , and applied to signal level or C/(I+N) 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 ( ) is applied to the
received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage
probability, and depends on the model standard deviation ( in dB) associated to the clutter class where the
receiver is located.
• Interference+noise-Based Predictions
Interference-based predictions include coverage predictions (Coverage by C/(I+N) Level, Coverage by Best
Bearer, Coverage by Throughput, etc.) that require calculation of the received signal level and interference.
In these calculations, (C/(I+N) calculations), in addition to the shadowing margin ( ) applied to
the received signal level calculated for each pixel, the ratio is applied to the
interfering signal levels (I). is calculated for a given cell edge coverage probability and depends
on the C/I standard deviation ( in dB) associated to the clutter class where the receiver is located.
G
macro di versi ty –
UL
G
macro di versi ty –
DL
M
Shadowi ng model –
o
model
M
Shadowi ng Eb Nt ( )
P CCPCH –

M
Shadowi ng Eb Nt ( )
DL

M
Shadowi ng Eb Nt ( )
UL

o
Eb Nt ( )
P CCPCH –
o
Eb Nt ( )
DL
o
Eb Nt ( )
UL
M
Shadowi ng model –
o
model
M
Shadowi ng model –
M
Shadowi ng model –
M
Shadowi ng C I –

M
Shadowi ng C I –
o
C I
118 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
The reason why the ratio is used can be understood from the following
derivation (linear, not it dB):
Inputs
- : The predicted received carrier power without any shadowing margin.
- : The predicted received interference power without any shadowing margin.
- : Shadowing margin based on the model standard deviation ( )
- : Shadowing margin based on the C/I standard deviation ( )
- : Thermal noise
Calculations
The effective received carrier power is given by:
The effective C/I is given by:
The above equations lead to:
Where corresponds to in dB.
Therefore, the effective C/(I+N) is given by:
• Monte-Carlo Simulations
Random values for shadowing margins are calculated for each transmitter-receiver link and applied to the
predicted signal level. 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 log-normal distribution as explained above.
LTE Documents
The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 120 and
"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 121 , and applied to signal level or C/(I+N) 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 ( ) is applied to the signal
level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage probability, and
depends on the model standard deviation ( in dB) associated to the clutter class where the receiver is
located.
• Interference+noise-Based Predictions
Interference-based predictions include coverage predictions (Coverage by C/(I+N) Level, Coverage by Best
Bearer, Coverage by Throughput, etc.) that require calculation of the received signal level and received
interference.
In these calculations, (C/(I+N) calculations), in addition to the shadowing margin ( ) applied to
the signal level calculated for each pixel, the ratio is applied to the interfering
signal levels (I). is calculated for a given cell edge coverage probability and depends on the C/I
standard deviation ( in dB) associated to the clutter class where the receiver is located.
M
Shadowi ng model –
M
Shadowi ng C I –

C
P
I
P
m
C
10
M
Shadowi ng model –
10
-----------------------------------------------------
m
C I
10
M
Shadowi ng C I –
10
----------------------------------------------
N
C m
C
C
P
× =
C
I
---- m
C I
C
P
I
P
------- × =
I
C
m
C I
C
P
I
P
------- ×
--------------------------
m
C
C
P
×
m
C I
C
P
I
P
------- ×
--------------------------
m
C
m
C I
------------ I
P
× = = =
m
C
m
C I
------------ M
Shadowi ng model –
M
Shadowi ng C I –

C
I N + ( )
-----------------
m
C
C
P
×
m
C
m
C I
------------ I
P
× N +
\ .
| |
---------------------------------------- =
M
Shadowi ng model –
o
model
M
Shadowi ng model –
M
Shadowi ng model –
M
Shadowi ng C I –

M
Shadowi ng C I –
o
C I
© Forsk 2010 AT283_TRG_E1 119
Chapter 4: Calculations
The reason why the ratio is used can be understood from the following
derivation (linear, not it dB):
Inputs
- : The predicted received carrier power without any shadowing margin.
- : The predicted received interference power without any shadowing margin.
- : Shadowing margin based on the model standard deviation ( )
- : Shadowing margin based on the C/I standard deviation ( )
- : Thermal noise
Calculations
The effective received carrier power is given by:
The effective C/I is given by:
The above equations lead to:
Where corresponds to in dB.
Therefore, the effective C/(I+N) is given by:
• Monte-Carlo Simulations
Random values for shadowing margins are calculated for each transmitter-receiver link and applied to the
predicted signal level. 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 log-normal distribution as explained above.
4.8.1 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:
M
Shadowi ng model –
M
Shadowi ng C I –

C
P
I
P
m
C
10
M
Shadowi ng model –
10
-----------------------------------------------------
m
C I
10
M
Shadowi ng C I –
10
----------------------------------------------
N
C m
C
C
P
× =
C
I
---- m
C I
C
P
I
P
------- × =
I
C
m
C I
C
P
I
P
------- ×
--------------------------
m
C
C
P
×
m
C I
C
P
I
P
------- ×
--------------------------
m
C
m
C I
------------ I
P
× = = =
m
C
m
C I
------------ M
Shadowi ng model –
M
Shadowi ng C I –

C
I N + ( )
-----------------
m
C
C
P
×
m
C
m
C I
------------ I
P
× N +
\ .
| |
---------------------------------------- =
Network Type
Standard
Deviation
M
Shadowing
Applied to
GSM GPRS EGPRS
C
C/I
UMTS HSPA
C
E
c
/I
0
E
b
/N
t
(DL)
E
b
/N
t
(UL)
CDMA2000 1xRTT 1xEV-DO
C
E
c
/I
0
E
b
/N
t
(DL)
E
b
/N
t
(UL)
o
model
M
Shadowi ng model –
o
C I
M
Shadowi ng C I –
o
model
M
Shadowi ng model –
o
Ec Io
M
Shadowi ng Ec Io –
o
Eb Nt ( )
DL
M
Shadowi ng Eb Nt ( )
DL

o
Eb Nt ( )
UL
M
Shadowi ng Eb Nt ( )
UL

o
model
M
Shadowi ng model –
o
Ec Io
M
Shadowi ng Ec Io –
o
Eb Nt ( )
DL
M
Shadowi ng Eb Nt ( )
DL

o
Eb Nt ( )
UL
M
Shadowi ng Eb Nt ( )
UL

120 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
4.8.1.1 Shadowing Margin Calculation in Predictions
Shadowing margins, M
Shadowing
, 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 applied to the path loss, L
path
.
Shadowing Error PDF (1 Signal)
The measured path loss in dB can be expressed as a Gaussian random variable:
where,
• L
path
is the predicted path loss,
• o
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:
The probability that the shadowing error exceeds z dB is
Normalising x by dividing it by o
dB
:
where Q is the complementary cumulative function.
To ensure a given cell edge coverage probability, , for the predicted value, a shadowing margin, , is added
to the link budget.
Confidence in the prediction can be expressed as:
where,
• is the signal level predicted at the receiver.

• EIRP is the effective isotropic radiated power of the transmitter.
• are receiver losses.
• is the receiver antenna gain.
The shadowing margin is calculated such that:
A lookup table is used for mapping the values of Q vs. a set of cell edge coverage probabilities.
TD-SCDMA
C
E
b
/N
t
P-CCPCH
E
b
/N
t
(DL)
E
b
/N
t
(UL)
WiMAX 802.16d
WiMAX 802.16e
C and C/(I+N)
C/(I+N)
LTE
C and C/(I+N)
C/(I+N)
o
model
M
Shadowi ng model –
o
Eb Nt ( )
P CCPCH –
M
Shadowi ng Eb Nt ( )
P CCPCH –

o
Eb Nt ( )
DL
M
Shadowi ng Eb Nt ( )
DL

o
Eb Nt ( )
UL
M
Shadowi ng Eb Nt ( )
UL

o
model
M
Shadowi ng model –
o
C I
M
Shadowi ng C I –
o
model
M
Shadowi ng model –
o
C I
M
Shadowi ng C I –
L L
path
o
dB
G 0 1 , ( ) × + =
p
L
x ( )
1
o
dB
2t
--------------------- e
x
2
2o
dB
2
-------------- –
× =
P
L
x z > ( ) p
L
x ( ) x d
z
·
}
1
o
dB
2t
--------------------- e
x
2
2o
dB
2
-------------- –
x d
z
·
}
× = =
P
L
x z > ( )
1
2t
----------- e
x
2
2
----- - –
x d
z
o
dB
----------
·
}
× Q
z
o
dB
---------
\ .
| |
= =
R
L
M
Shadowi ng
C
d
P'
Tx
L – P
rec
> L · P'
Tx
s P
rec
– G 0 1 , ( ) · o
dB
× M
Shadowi ng
s =
P
rec
P
rec
P'
Tx
L
path
– M
Shadowi ng
– =
P'
Tx
EIRP G
antRx
L
Rx
– + =
L
Rx
G
antRx
P C
d
P
rec
> ( ) R
L
M
Shadowi ng
( ) 1 P
L
x M
Shadowi ng
– 0 > ( ) – 1 Q
M
Shadowi ng
o
dB
------------------------------
\ .
| |
– = = =
© Forsk 2010 AT283_TRG_E1 121
Chapter 4: Calculations
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.
4.8.1.2 Shadowing Margin Calculation in Monte-Carlo Simulations
Shadowing margins, M
Shadowing
, 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, L
path
.
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 .
Here, is a zero mean gaussian random variable representing variation due to shadowing. It can be
expressed as the sum of two uncorrelated zero mean gaussian random variables, and . models the error related
to the receiver’s location (surrounding environment), and remains the same for all links between the receiver and the base
stations from which it is receiving signals. models the error related to the path between the transmitter and the receiver.
Therefore, in case of two links, we have:
for link 1
for link 2
Standard deviations of and can be calculated from , the model standard deviation , and the
correlation coefficient between and .
Assuming all have the same standard deviations, we have:
Therefore,
is set to 0.5 in Atoll, which gives:
and
Therefore, to model shadowing error common to all the signals received at a receiver ( ), values are
randomly generated for each receiver. These values have a zero-mean gaussian distribution with a standard deviation of
, where is the model standard deviation associated with the receiver’s clutter class.
Figure 4.26Normalised Margin M i n arg
M
Shadowi ng
o
dB
------------------------------ =
L L
path
ç + =
ç G 0 o
dB
, ( )
ç
L
ç
P
ç
L
ç
P
ç
1
ç
L
ç
P
1
+ =
ç
2
ç
L
ç
P
2
+ =
ç
L
o
L
( ) ç
P
i
o
P
( ) ç
i
o
model
( )
µ ( ) ç
1
ç
2
ç
P
o
model
2
o
L
2
o
P
2
+ =
µ
o
L
2
o
model
2
----------------- =
o
P
2
o
model
2
1 µ – ( ) × =
o
L
2
o
model
2
µ × =
µ
o
L
o
model
2
----------------- = o
P
o
model
2
----------------- =
E
Shadowi ng model –
Recei ver
o
model
2
-----------------
\ .
| |
o
model
( )
122 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
Next, Atoll generates another random value for each transmitter-receiver pair. This values represents the shadowing error
not related to the location of the receiver ( ). These values also have a zero-mean gaussian distribution
with a standard deviation .
So, we have:
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.
4.8.2 Macro-Diversity Gains Calculation
The following sections explain how uplink and downlink macro-diversity gains are calculated in UMTS HSPA and
CDMA2000 1xRTT 1xEV-DO documents for predictions and AS Analysis tab of the point analysis tool.
4.8.2.1 Uplink Macro-Diversity Gain Evaluation
In UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, 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.
4.8.2.1.1 Shadowing Error PDF (n Signals)
For each link, path loss (L) can be broken down as:
is a zero mean gaussian random variable representing variation due to shadowing. It can be expressed as
the sum of two uncorrelated zero mean gaussian random variables, and . models error related to the receiver
local environment; it is the same whichever the link. models error related to the path between transmitter and receiver.
Therefore, in case of two links, we have:
for the link 1
for the link 2
Knowing , the uplink Eb/Nt standard deviation and the correlation coefficient between and , we
can calculate standard deviations of and (assuming all have the same standard deviations).
We have:
Therefore,
2 Signals Without Recombination
In technologies supporting soft handoff (UMTS and CDMA2000), cell is interference limited. As for one link, to ensure a
required cell edge coverage probability for the prediction, we add to each link budget a shadowing margin,
.
Prediction reliability in order to have Eb/Nt higher or equal to Eb/Nt from the best server can be expressed as:
E
Shadowi ng model –
Path
o
model
2
-----------------
\ .
| |
E
Shadowi ng model –
E
Shadowi ng model –
Recei ver
E
Shadowi ng model –
Path
+ =
Note:
• The calculation and use of macro-diversity gains can be disabled through the Atoll.ini file.
For more information, see the Administrator Manual.
L L
path
ç + =
ç G 0 o
dB
, ( )
ç
L
ç
P
ç
L
ç
P
ç
1
ç
L
ç
P
1
+ =
ç
2
ç
L
ç
P
2
+ =
ç
i
o
Eb Nt ( )
UL
( ) µ ç
1
ç
2
ç
L
o
L
( ) ç
P
o
P
( ) ç
P
o
Eb Nt ( )
UL
2
o
L
2
o
P
2
+ =
µ
o
L
2
o
Eb Nt ( )
UL
2
-------------------------- =
o
P
2
o
Eb Nt ( )
UL
2
1 µ – ( ) × =
o
L
2
o
Eb Nt ( )
UL
2
µ × =
R
L
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
C
d
1
N
1
--------- P'
Tx1
L
1
– N
1
– CI
pred
1
> ç
1
P'
Tx1
L
path
1
– N
1
– CI
pred
1
– s · =
© Forsk 2010 AT283_TRG_E1 123
Chapter 4: Calculations
or
where
is the quality level (signal to noise ratio) predicted at the receiver for link i.
N
i
is the noise level for link i.
We note:
and
is the minimum needed margin on each link.
Therefore, the probability of having a quality at least equal to the best predicted one is:
We can express it using , and
Then, we have:
If we introduce user defined standard deviation and correlation coefficient , and consider that is a
Gaussian pdf:
C
d
2
N
2
--------- P'
Tx2
L
2
– N
2
– CI
pred
1
> ç
2
P'
Tx2
L
path
2
– N
2
– CI
pred
1
– s · =
CI
pred
i
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
P'
Txi
L
path
i
– N
i
– CI
pred
i
– =
A
1
2
CI
pred
1
CI
pred
2
– =
A
1
2
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
( ) 1 P
L1 L2 ,
C
d
1
N
1
--------- CI
pred
1
<
C
d
2
N
2
--------- CI
pred
1
< ,
\ .
|
| |
– =
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
( ) 1 P
ç
1
ç
2
,
ç
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
> ç
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– > , ( ) – =
ç
L
ç
P
1
ç
P
2
P
ç
1
ç
2
,
ç
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
> ç
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– > ç
L
A
L
= , ( )
P
ç
L
A
L
( ) P
ç
P
1
ç
P
2
,
× ç
P
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L
– > ç
P
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– A
L
– > , ( ) =
P
ç
1
ç
2
,
ç
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
> ç
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– > ç
L
A
L
= , ( )
P
ç
L
A
L
( ) P
ç
P
× A
P
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L
– > ( ) P
ç
P
A
P
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– A
L
– > ( ) × =
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
( )
1 P
ç
L
A
L
( ) P
ç
P
× A
P
1
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L
– > ( ) P
ç
P
A
P
2
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– A
L
– > ( ) × A
L
d
· –
·
}

\ .
|
|
| |
=
P
ç
P
A
P
i
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L
– > ( )
1
o
P
2t
------------------ e
x
2

2o
P
2
----------
x d
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L

\ .
| |
·
}
Q
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L

o
P
---------------------------------------------------------------------
\ .
|
| |
=
\ .
|
|
|
|
| |
=
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
( )
1 P
ç
L
A
L
( ) Q
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
L

o
P
---------------------------------------------------------------------
\ .
|
| |
× Q
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
A
1
2
– A
L

o
P
----------------------------------------------------------------------------------
\ .
|
| |
× A
L
d
· –
·
}

\ .
|
|
| |
=
o
Eb Nt ( )
UL
( ) µ ( ) P
ç
L
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
( )
1
1
2t
----------- e
x
L
2

2
---------
Q
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
x
L
o
Eb Nt ( )
UL
µ –
o
Eb Nt ( )
UL
1 µ –
--------------------------------------------------------------------------------------------------------
\ .
|
|
| |
× Q
M
Shadowi ng Eb Nt ( )
UL

2si gnal s
x
L
o
Eb Nt ( )
UL
µ – A
1
2

o
Eb Nt ( )
UL
1 µ –
--------------------------------------------------------------------------------------------------------------------
\ .
|
|
| |
× x
L
d
· –
·
}

\ .
|
|
| |
=
124 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
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):
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 and . 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]:
when
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 t (+/- t/3) and D1/D2 is close to 1.
In [1,5], when and .
In Atoll, is set to 0.5.
4.8.2.1.2 Uplink Macro-Diversity Gain
Atoll determines the uplink macro-diversity gain ( ) from the shadowing margins calculated in case of one
signal and n signals.
Therefore, we have:
Where n is the number of cell-mobile signals.
4.8.2.2 Downlink Macro-Diversity Gain Evaluation
In UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, 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.
4.8.2.2.1 Shadowing Error PDF (n Signals)
For each link, path loss (L) can be broken down as:
is a zero mean gaussian random variable representing variation due to shadowing. It can be expressed as
the sum of two uncorrelated zero mean gaussian random variables, and . models the error related to the receiver
local environment, which is the same for all links. models the error related to the path between the transmitter and the
receiver.
Therefore, in case of two links, we have:
for the link 1
for the link 2
R
L
noMRC
M
Shadowi ng Eb Nt ( )
UL

nsi gnal s
( )
1
1
2t
----------- e
x
L
2

2
---------
Q
M
Shadowi ng Eb Nt ( )
UL

nsi gnal s
x
L
o
Eb Nt ( )
UL
µ –
o
Eb Nt ( )
UL
1 µ –
--------------------------------------------------------------------------------------------------------
\ .
|
|
| |
× Q
M
Shadowi ng Eb Nt ( )
UL

nsi gnal s
x
L
o
Eb Nt ( )
UL
µ – A
1
2

o
Eb Nt ( )
UL
1 µ –
--------------------------------------------------------------------------------------------------------------------
\ .
|
|
| |
× x
L
d
· –
·
}

\ .
|
|
| |
=
µ ( ) ç
1
ç
2
µ
¢
T
¢
------
\ .
| |
¸
D1
D2
-------- = |
T
| t s s
¢
T
¢
µ 0.5 = ¸ 0.3 = ¢
T
t
10
------ =
µ
G
macro di versi ty –
UL
G
macro di versi ty –
UL
M
Shadowi ng Eb Nt ( )
UL

nsi gnal s
M
Shadowi ng Eb Nt ( )
UL

– =
L L
path
ç + =
ç G 0 o
dB
, ( )
ç
L
ç
P
ç
L
ç
P
ç
1
ç
L
ç
P
1
+ =
ç
2
ç
L
ç
P
2
+ =
© Forsk 2010 AT283_TRG_E1 125
Chapter 4: Calculations
Knowing , the Ec/Io standard deviation and the correlation coefficient between and , we can calculate
standard deviations of and (assuming all have the same standard deviations).
We have:
Therefore,
2 Available Signals
In technologies supporting soft handoff (UMTS and CDMA2000) cells are interference limited. As for one link, to ensure a
required cell edge coverage probability for the prediction, we add a shadowing margin, , to each
link budget.
Prediction reliability to have for the best server can be expressed as:
Or
We note:
is the minimum needed margin on each link.
Therefore, probability of having a quality at least equal to the best predicted one is:
We can express it by using , and
Then, we have:
ç
i
o
Ec I o
( ) µ ç
1
ç
2
ç
L
o
L
( ) ç
P
o
P
( ) ç
P
o
Ec I o
2
o
L
2
o
P
2
+ =
µ
o
L
2
o
Ec I o
2
---------------- =
o
P
2
o
Ec I o
2
1 µ – ( ) × =
o
L
2
o
Ec I o
2
µ × =
R
L
M
Shadowi ng Ec Io –
2si gnal s
Ec
Io
-------
Ec
Io
-------
\ .
| |
pred
>
Ec
1
Io
---------- P
pi l ot
1
L
1
– Io –
Ec
Io
-------
\ .
| |
pred
1
> = ç
1
P
pi l ot
1
L
m
1
– Io –
Ec
Io
-------
\ .
| |
pred
1
– s ·
Ec
2
Io
---------- P
pi l ot
2
L
2
– Io –
Ec
Io
-------
\ .
| |
pred
1
> = ç
2
P
pi l ot
2
L
m
2
– Io –
Ec
Io
-------
\ .
| |
pred
1
– s ·
M
Shadowi ng Ec Io –
2si gnal s
P
pi l ot
i
L
m
i
– Io –
Ec
Io
-------
\ .
| |
pred
1
– =
A
1
2 Ec
Io
-------
\ .
| |
pred
1
Ec
Io
-------
\ .
| |
pred
2
– =
A
1
2
R
L
noMRC
M
Shadowi ng Ec Io –
2si gnal s
( ) 1 P
L1 L2 ,
Ec
1
Io
----------
Ec
Io
-------
\ .
| |
pred
1
<
Ec
2
Io
----------
Ec
Io
-------
\ .
| |
pred
1
< ,
\ .
| |
– =
R
L
noMRC
M
Shadowi ng Ec Io –
2si gnal s
( ) 1 P
ç1 ç2 ,
ç
1
M
Shadowi ng Ec Io –
2si gnal s
> ç
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– > , ( ) – =
ç
L
ç
P
1
ç
P
2
P
ç1 ç2 ,
ç
1
M
Shadowi ng Ec Io –
2si gnal s
> ç
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– > ç
L
A
L
= , ( )
P
ç
L
A
L
( ) P
ç
P
1
ç
P
2
,
× ç
P
1
M
Shadowi ng Ec Io –
2si gnal s
A
L
– > ç
P
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– A
L
– > , ( ) =
P
ç1 ç2 ,
ç
1
M
Shadowi ng Ec Io –
2si gnal s
> ç
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– > ç
L
A
L
= , ( )
P
ç
L
A
L
( ) P
ç
P
× A
P
1
M
Shadowi ng Ec Io –
2si gnal s
A
L
– > ( ) P
ç
P
× A
P
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– A
L
– > ( ) =
R
L
noMRC
M
Shadowi ng Ec Io –
2si gnal s
( )
1 P
ç
L
A
L
( ) P
ç
P
× A
P
1
M
Shadowi ng Ec Io –
2si gnal s
A
L
– > ( ) P
ç
P
× A
P
2
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– A
L
– > ( ) A
L
d
· –
·
}
– =
P
ç
P
A
P
i
M
Shadowi ng Ec Io –
2si gnal s
A
L
– > ( )
1
o
P
2t
------------------ e
x
2

2o
P
2
----------
¸
SHO
A
L

·
}
dx Q
M
Shadowi ng Ec Io –
2si gnal s
A
L

o
P
-----------------------------------------------------------
\ .
|
| |
= =
126 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
If we introduce a user defined Ec/Io standard deviation and a correlation coefficient and consider that is a
Gaussian pdf:
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):
=1 dB
=5 dB
=10 dB
Figure 4.27Margin - Probability (Case of 2 Signals)
2 signals
=5 dB
=10 dB
Figure 4.28Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 1dB)
R
L
noMRC
M
Shadowi ng Ec Io –
2si gnal s
( ) 1 P
ç
L
A
L
( ) Q ×
M
Shadowi ng Ec Io –
2si gnal s
A
L

o
P
-----------------------------------------------------------
\ .
|
| |
Q ×
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– A
L

o
P
------------------------------------------------------------------------
\ .
|
| |
A
L
d
· –
·
}
– =
o ( ) µ ( ) P
ç
L
R
L
noMRC
M
Shadowi ng Ec Io –
2si gnal s
( )
1
1
2t
----------- e
x
L
2

2
---------
Q ×
M
Shadowi ng Ec Io –
2si gnal s
x
L
o
Ec I o
µ –
o
Ec I o
1 µ –
------------------------------------------------------------------------------------
\ .
|
| |
Q ×
M
Shadowi ng Ec Io –
2si gnal s
A
1
2
– x
L
o
Ec I o
µ –
o
Ec I o
1 µ –
------------------------------------------------------------------------------------------------
\ .
|
| |
x
L
d
· –
·
}
– =
R
L
noMRC
M
Shadowi ng Ec Io –
nsi gnal s
( )
1
1
2t
----------- e
x
L
2

2
---------
Q ×
M
Shadowi ng Ec Io –
nsi gnal s
x
L
o
Ec I o
µ –
o
Ec I o
1 µ –
------------------------------------------------------------------------------------
\ .
|
| |
x Q
M
Shadowi ng Ec Io –
nsi gnal s
A
1
i
– x
L
o
Ec I o
µ –
o
Ec I o
1 µ –
------------------------------------------------------------------------------------------------
\ .
|
| |
i 2 =
n
[
x
L
d
· –
·
}
– =
A
1
2
A
1
2
A
1
2
A
1
3
A
1
3
© Forsk 2010 AT283_TRG_E1 127
Chapter 4: Calculations
Correlation Coefficient Determination
For further information about determination of the correlation coefficient, please see "Correlation Coefficient
Determination" on page 127.
4.8.2.2.2 Downlink Macro-Diversity Gain
Atoll determines the downlink macro-diversity gain ( ) from the shadowing margins calculated in case of
one signal and n signals.
Therefore, we have:
Where n is the number of available signals.
4.9 Appendices
4.9.1 Transmitter Radio Equipment
Radio equipment such as TMA, feeder and BTS, are taken into account to evaluate:
• Total UL and DL losses of transmitter ( ) and transmitter noise figure in UMTS HSPA,
CDMA2000 1xRTT 1xEV-DO, TD-SCDMA, WiMAX 802.16d, WiMAX 802.16e, and LTE documents,
• Transmitter total losses 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.
2 signals
=5 dB
=10 dB
Figure 4.29Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 2dB)
A
1
3
A
1
3
G
macro di versi ty –
DL
G
macro di versi ty –
DL
M
Shadowi ng Ec Io –
nsi gnal s
M
Shadowi ng Ec Io –
– =
Figure 4.30Reference Point - Location of the Transmission/Reception parameters
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.
• Loss and gain inputs specified in .atl documents must be positive values.
L
total UL –
L
total DL –
, NF
Tx
( )
L
Total
( )
NF
Feeder
L
Feeder
UL
=
128 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
4.9.1.1 UMTS, CDMA2000, TD-SCDMA, WiMAX, and LTE Documents
As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we
have . Where is the BTS noise figure.
Uplink Total Losses:
Atoll9955 calculates total UL losses as follows:
Where,
• are the miscellaneous reception losses (Transmitter property).
• are the feeder reception losses ( , where , and
are respectively the feeder loss per metre (Feeder property), the reception feeder length in metre
(Transmitter property) and the connector reception losses.
• are the losses due to BTS configuration (BTS property).
• is the antenna diversity gain (Transmitter property). This gain does not exist in WiMAX and LTE
documents.
• is the noise rise at transmitter due to repeaters. This parameter is taken into account only if the
transmitter has active repeater(s). The noise rise at transmitter due to repeaters is calculated as follows:
For each active repeater ( ), Atoll9955 calculates a noise injection margin ( ). This is the difference
between the donor transmitter noise figure ( ) and the repeater noise figure received at the donor.
Where,
- is the repeater noise figure,
- is the repeater amplification gain (repeater property),
- are the losses between the donor transmitter and the repeater (repeater property).
- For each active repeater ( ), Atoll9955 converts the noise injection margin ( ) to Watt. Then, it uses
the values to calculate the noise rise at the donor transmitter due to active repeaters ( ).
• is the gain due to TMA, which is calculated as follows:
Where and 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.
And,
Where,
- is the feeder noise figure.
- is the TMA noise figure.
- is the BTS noise figure.
- is the TMA reception gain.
- is the feeder UL gain .
NF
TX
NF
BTS
= NF
BTS
L
Total UL –
L
Mi sc
UL
L
Feeder
UL
L
BTS Conf –
UL
NR
Repeaters
G
Ant di v –
UL
– G
TMA
– + + + =
L
Mi sc
UL
L
Feeder
UL
L
Feeder
UL
L
Feeder
I
Feeder
UL
L
Connector
UL
+ × = L
Feeder
I
Feeder
UL
L
Connector
UL
L
BTS Conf –
UL
G
Ant di v –
UL
NR
Repeaters
NR
Repeaters
10 Log 1
1
NIM
Rp
r
-------------------
r
¿
+
\ .
|
| |
× =
k NIM
Rp
k
NF
TX
NIM
Rp
r
NF
TX
NF
Rp
k
G
amp
Rp
k
L
TX Rp
k

– +
\ .
| |
– =
NF
Rp
k
G
amp
Rp
k
L
TX R – p
k
k NIM
Rp
k
NR
Repeaters
G
TMA
G
TMA
NF
Composi te
Wi thoutTMA
NF
Composi te
Wi thTMA
– =
NF
Composi te
Wi thTMA
NF
Composi te
Wi thoutTMA
NF
Composi te
Wi thTMA
10 Log 10
NF
TMA
10
--------------------
10
NF
Feeder
10
--------------------------
1 –
10
G
TMA
UL
10
----------------
------------------------------------
10
NF
BTS
10
-------------------
1 –
10
G
TMA
UL
10
----------------
10
G
Feeder
UL
10
----------------------
×
------------------------------------------------- + +
\ .
|
|
|
|
| |
× =
NF
Composi te
Wi thoutTMA
NF
BTS
NF
Feeder
+ =
NF
Feeder
NF
TMA
NF
BTS
G
TMA
UL
G
Feeder
UL
G
Feeder
UL
L
Feeder
UL
– =
© Forsk 2010 AT283_TRG_E1 129
Chapter 4: Calculations
- is the feeder reception loss ( , where , and
are respectively the feeder loss per metre, the reception feeder length in metre and the connector
reception loss).
Downlink Total Losses:
Atoll9955 calculates total DL losses as follows.
Where,
• is the TMA transmission loss.
• is the feeder transmission loss ( , where , and
are respectively the feeder loss per metre, the transmission feeder length in metre and the connector
transmission losses).
• are the miscellaneous transmission losses.
• are the losses due to BTS configuration (BTS property).
4.9.1.2 GSM Documents
Atoll calculates DL total losses as follows:
Where,
• is the TMA transmission loss.
• is the feeder transmission loss ( , where , and
are respectively the feeder loss per metre, the transmission feeder length in metre and the connector
transmission loss).
• are the miscellaneous transmission losses.
• are the losses due to BTS configuration (BTS property).
4.9.2 Secondary Antennas
When secondary antennas are installed on a transmitter, the signal level received from it is calculated as follows:
(not in dB
2
)
Where,
P
Tx
is the transmitter power (P
pilot
in UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, P
P-CCPCH
in TD-SCDMA, P
Preamble
in WiMAX 802.16d and WiMAX 802.16e, and P
DLRS
in LTE),
i is the secondary antenna index,
x
i
is the percentage of power dedicated to the secondary antenna, i,
is the gain of the main antenna installed on the transmitter,
L
Tx
are transmitter losses (L
Tx
=L
total-DL
),
is the gain of the secondary antenna, i, installed on the transmitter,
L
model
is the path loss calculated by the propagation model,
is the attenuation due to main antenna pattern,
is the attenuation due to pattern of the secondary antenna, i.
L
Feeder
UL
L
Feeder
UL
L
Feeder
I
Feeder
UL
L
Connector
UL
+ × = L
Feeder
I
Feeder
UL
L
Connector
UL
L
Total DL –
L
TMA
DL
L
Feeder
DL
L
Mi sc
DL
L
BTS Conf –
DL
+ + + =
L
TMA
DL
L
Feeder
DL
L
Feeder
DL
L
Feeder
I
Feeder
DL
L
Connector
DL
+ × = L
Feeder
I
Feeder
DL
L
Connector
DL
L
Mi sc
DL
L
BTS Conf –
DL
L
Total DL –
L
TMA
DL
L
Feeder
DL
L
Mi sc
DL
L
BTS Conf –
DL
+ + + =
L
TMA
DL
L
Feeder
DL
L
Feeder
DL
L
Feeder
I
Feeder
DL
L
Connector
DL
+ × = L
Feeder
I
Feeder
DL
L
Connector
DL
L
Mi sc
DL
L
BTS Conf –
DL
2. Formula cannot be directly calculated from components stated in dB and must be converted in linear values.
P
rec
P
Tx
1 X
i
i
¿

\ .
|
| |
G
ant m
Tx

L
Tx
------------------------ · ·
L
ant m
Tx

az
m
el
m
, ( )
-----------------------------------------------------------------------
P
Tx
X
i
G
ant i
Tx

L
Tx
--------------------- · ·
L
ant i
Tx

az
i
el
i
, ( )
----------------------------------------------
i
¿
+
\ .
|
|
|
|
|
| |
L
model
------------------------------------------------------------------------------------------------------------------------------------------ =
G
ant m
Tx

G
ant i
Tx

L
ant m
Tx

az
m
el
m
, ( )
L
ant i
Tx

az
i
el
i
, ( )
130 AT283_TRG_E1 © Forsk 2010
Technical Reference Guide
The definition of angles, az and el, depends on the used calculation method.
• Method 1 (must be indicated in an Atoll.ini file):
- az
m
: the difference between the receiver antenna azimuth and azimuth of the transmitter main antenna,
- el
m
: the difference between the receiver antenna tilt and tilt of the transmitter main antenna,
- az
i
: the difference between the receiver antenna azimuth and azimuth of the transmitter secondary antenna, i,
- el
i
: the difference between the receiver antenna tilt and tilt of the transmitter secondary antenna, i,
• Method 2 (default):
- az
m
: the receiver azimuth in the coordinate system of the transmitter main antenna,
- el
m
: the receiver tilt in the coordinate system of the transmitter main antenna,
- az
i
: the receiver azimuth in the coordinate system of the transmitter secondary antenna, i,
- el
i
: the receiver tilt in the coordinate system of the transmitter secondary antenna, i.
Chapter 5
GSM GPRS EDGE Networks
© Forsk 2010 AT283_TRG_E2 133
Chapter 5: GSM GPRS EDGE Networks
5 GSM GPRS EDGE Networks
This chapter describes all the calculations performed in Atoll GSM/GPRS/EDGE documents. The first four sections
describe the signal level, interference, GPRS/EDGE-specific, and CQI calculations, respectively. The following three
sections explain the traffic analysis, network dimensioning, and KPI calculation processes. The last section describes the
neighbour allocation process in GSM.
5.1 Signal Level Calculations
Three parameters can be studied in point analysis (Profile tab) and in signal level-based coverage predictions:
Where,
• EIRP is the effective isotropic radiated power of the transmitter,
• is the loss on the transmitter-receiver path (path loss) calculated by the propagation model,
• is the transmitter antenna attenuation (from antenna patterns),
• is the shadowing margin. This parameter is taken into account when the option “Shadowing
taken into account” is selected,
• are the indoor losses, taken into account when the option “Indoor coverage” is selected,
• are the receiver losses,
• is the receiver antenna gain,
• AP 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).
5.1.1 Point Analysis
5.1.1.1 Profile Tab
For a selected transmitter, it is possible to display the signal level received from a TRX type ( ), the path loss,
, or the total losses, . Path loss and total losses are the same for all TRX types.
If the power reduction values defined for all the subcells are the same, the received signal level from the selected
transmitter will be the same for all TRX types.
5.1.1.2 Reception Tab
Analysis provided in the Reception tab is based on path loss matrices. Therefore, it is possible to display the signal levels
received from TBC transmitters for which path loss matrices have been calculated over their calculation areas.
For each transmitter, Atoll can display the signal level received from a TRX type ( ), the path loss, , or the
total losses, . Path loss and total losses are the same for all TRX types.
If the power reduction values defined for all the subcells are the same, the received signal level from the selected
transmitter will be the same for all TRX types.
Reception level bars are displayed in the order of decreasing signal level. The number of displayed bars depends on the
signal level received from the best server. Bars are only displayed for transmitters whose signal level is within a 30 dB
margin from the best server signal level.
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.
• Logarithms used in this chapter (Log function) are base-10 unless stated otherwise.
Studied Parameter Formulas
Signal level ( )
Signal level received from a transmitter on a TRX type
Path loss ( )
Total losses ( )
P
rec
Txi
P
rec
Txi
tt ( ) EIRP tt ( ) AP tt ( ) – L
path
Txi
– M
Shadowi ng model –
– L
Indoor
– G
ant
Rx
L
Rx
– ( ) + =
L
path
Txi
L
path
Txi
L
model
L
ant
Tx
+ =
L
total
Txi
L
total
Txi
L
path
Txi
M
Shadowi ng model –
L
Indoor
L +
Tx
L
Rx
+ + + ( ) G
ant
Tx
G
ant
Rx
+ ( ) – =
L
model
L
ant
Tx
M
Shadowi ng model –
L
Indoor
L
Rx
G
ant
Rx
P
rec
Txi
tt ( )
L
path
Txi
L
total
Txi
P
rec
Txi
tt ( ) L
path
Txi
L
total
Txi
134 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.1.2 Signal Level-based Coverage Predictions
For each TBC transmitter, Txi, Atoll calculates the selected parameter on each pixel inside the Txi calculation area. In
other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.
Coverage prediction parameters to be set are:
• The coverage conditions in order to determine the service area of each TBC transmitter, and
• The display settings to select the displayed parameter and its shading levels.
5.1.2.1 Service Area Determination
Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas
where coverage will be displayed.
We can distinguish eight cases as below. Let us assume that:
• Each transmitter, Txi, belongs to a Hierarchical Cell Structure (HCS) layer, k, with a defined priority and a defined
reception threshold.
• No max range is set.
5.1.2.1.1 All Servers
The service area of Txi corresponds to the pixels where:
5.1.2.1.2 Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.1.2.1.3 Second Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
Note:
• 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.
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
P
rec
Txi
tt ( )
Best
j i =
P
rec
Txj
tt ( ) ( ) M – >
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
© Forsk 2010 AT283_TRG_E2 135
Chapter 5: GSM GPRS EDGE Networks
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.1.2.1.4 Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.1.2.1.5 Second Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.1.2.1.6 HCS Servers and a Margin
The service area of Txi corresponds to the pixels where:
And
And the received exceeds the reception threshold defined per HCS layer.
P
rec
Txi
tt ( )
2
nd
Best
j i =
P
rec
Txj
tt ( ) ( ) M – >
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
P
rec
Txi
BCCH ( )
2
nd
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
tt ( )
136 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.1.2.1.7 Highest Priority HCS Server and a Margin
The service area of Txi corresponds to the pixels where:
And
And Txi belongs to the HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest).
And the received exceeds the reception threshold defined per HCS layer.
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.1.2.1.8 Best Idle Mode Reselection Criterion (C2)
Such type of coverage is useful :
• To compare idle and dedicated mode best servers for voice traffic
• Display the GPRS/EDGE best server map (based on GSM idle mode)
The path loss criterion C1 used for cell selection and reselection is defined by:
The path loss criterion (GSM03.22) is satisfied if .
The reselection criterion C2 is used for cell reselection only and is defined by:
Where is the Cell Reselect Offset defined for the transmitter.
The service area of Txi corresponds to the pixels where:
And
The Best function considers the highest value from a list of values.
On each pixel, the transmitter with the highest C2 value is kept. It corresponds to the best server in idle mode. C2 is defined
as an integer in the 3GPP specifications, therefore, the C2 values in the above calculations are rounded down to the
nearest integer.
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
Note:
• 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.
Mi ni mumThreshol d P
rec
Txi
tt ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
tt ( )
Note:
• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with
or .
C1 P
rec
Txi
BCCH ( ) Mi ni mumThreshol d BCCH ( ) – =
C1 0 >
C2 C1 CELL_RESELECT_OFFSET + =
CELL_RESELECT_OFFSET
Mi ni mumThreshol d P
rec
Txi
BCCH ( ) s Maxi mumThreshol d <
P
rec
Txi
tt ( )
L
total
Txi
L
path
Txi
C2
Txi
BCCH ( )
Best
j
C2
Txj
BCCH ( ) ( ) =
© Forsk 2010 AT283_TRG_E2 137
Chapter 5: GSM GPRS EDGE Networks
5.1.2.2 Coverage Display
5.1.2.2.1 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, see "Path
Loss Calculations" on page 77 for more information).
5.1.2.2.2 Display Types
It is possible to display the coverage predictions with colours depending on any transmitter attribute or other criteria such
as:
Signal Level (in dBm, dBµV, dBµV/m)
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). 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, dBµV, dBµV/m)
Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. When other
serviceWhen 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
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 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
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 pixel 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
pixel of a service 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 minimum threshold.
Best Server Total Losses (dB)
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.
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.
>
>
>
>
>
>
>
138 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Cell Edge Coverage Probability (%)
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.
Best Cell Edge Coverage Probability (%)
On each pixel 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 pixel of each transmitter service area. When other service
areas overlap the studied one, Atoll chooses the highest value. A pixel of a service area is coloured if the C2 value
exceeds ( ) the defined thresholds (the pixel 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 Interference-based Calculations
Interference-based calculations include all the calculations that involve the calculation of interference received from
interfering transmitters in addition to the signal level received from the server.
5.2.1 Carrier-to-Interference Ratio Calculation
MSA (Mobile Station Allocation) Definition
A wide-ranging definition of an MSA, Mobile Station Allocation, can be that it is a list of channels and an associated MAIO.
More precisely, for different frequency hopping modes, this definition can be:
• Non-hopping (NH): An MSA is the channel assigned to a TRX used by a mobile.
• Baseband hopping (BBH): An MSA is the Mobile Allocation List (MAL) and the TRX index.
• Synthesised frequency hopping (SFH): An MSA is the Mobile Allocation List (MAL) and the Mobile Allocation
Index Offset (MAIO).
From the point of view of a mobile station, BBH and SFH work in the same way.
Notations and Assumptions
In the following description:
• v is a victim transmitter,
• MSAS(v) is the set of MSAs (Mobile Station Allocations) associated to v,
The number of MSAS(v) depends on TRX types to be analysed. You may study a given TRX type tt (there will be
as many MSA(v) as TRXs allocated to the subcell (v,tt)) or all the TRX types (the number of MSA(v) will correspond
to the number of TRXs allocated to v).
Several MSAs, m, are related to a transmitter. Therefore, Atoll calculates the C/I for each victim
transmitter v with MSA m (m e MSAS(v)).
>
TRX index Channel list MAIO MSA
1 53 - (53,-)
2 54 - (54,-)
TRX index Channel list MAIO MSA
1 53 * ([53,54,55],0)
2 54 * ([53,54,55],1)
3 55 * ([53,54,55],2)
TRX index Channel list MAIO MSA
1 53 54 55 56 2 ([53,54,55,56],2)
2 53 54 55 56 3 ([53,54,55,56],3)
C
v
m ( )
I
v
m ( )
-----------------
\ .
|
| |
© Forsk 2010 AT283_TRG_E2 139
Chapter 5: GSM GPRS EDGE Networks
Atoll considers the most interfered MSA, therefore, the displayed C/I or C/(I+N) are or
, respectively. If the Detailed Results check box is selected, the C/I values for all
MSAs are displayed.
• i is any potential interfering transmitter (TBC transmitters whose calculation areas intersect the 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,
• is the carrier power level received from v on m,
• corresponds to the interference received from interfering transmitters i on m,
• used in the C/I calculation is based on the C/I standard deviation.
Calculations
The carrier power level is the power received from the victim transmitter at the receiver.
If the interference conditions are based on C/(I+N), Atoll takes the total noise into account. The total noise is the sum
of the thermal noise (-121 dBm by default or user-defined), the noise figure NF, and the inter-technology
downlink noise rise .
Interference can be received from interfering transmitters i on co-channel and adjacent channels. Interference may also
be received from the transmitters of another technology.
Therefore,
Here, is the average power control gain defined for the interfering transmitter i.
Each interference component is explained below.
Co- and Adjacent Channel Interference:
is the interference received at v on m on co-channel, given by:
is the interference received at v on m on adjacent channels, given by:
Here, is the carrier power level received from i on n.
T
i
(n) is occupancy of the MSA n:
is the traffic load defined for the MSA n or i. It can be set to 100% in the coverage prediction properties.
is the activity factor defined for the MSA n of i. If the subcell (i,tt) supports DTX, the value specified in the
coverage prediction properties is used. Otherwise, the activity factor is 1.
Note:
• The C/I shadowing margin is applied on the carrier power level. The interference levels are
not changed.
Note:
• BCCH TRXs are always on. Therefore, DTX and traffic loads do not impact the interference
from BCCH. In other words, and for the BCCH TRXs of the
interferers.
C
I
----
\ .
| |
v
Mi n
k
C
v
m ( )
I
v
m ( )
-----------------
\ .
|
| |
=
C
I N
tot
+
------------------
\ .
| |
v
Mi n
k
C
v
m ( )
I
v
m ( ) N
tot
+
-------------------------------
\ .
|
| |
=
C
v
m ( )
I
v
m ( )
M
Shadowi ng
C
v
m ( ) P
rec
v
m ( ) =
N
tot
N
thermal
NR
i nter techno y log –
v DL ,
N
tot
N
thermal
NF NR
i nter techno y log –
v DL ,
+ + =
I
v
m ( ) I
co
v
m ( ) I
adj
v
m ( ) I
i nter techno y log –
DL
G
PC
i
– + + =
G
PC
i
I
co
v
m ( )
I
co
v
m ( ) p
m n ,
v i ,
P
rec
i
n ( ) T
i
n ( ) × ×
n MSAS i ( ) e
¿
\ .
|
| |
i INT v ( ) e
¿
co
=
I
adj
v
m ( )
I
adj
v
m ( ) p
m n ,
v i ,
P
rec
i
n ( )
F
------------------- T
i
n ( ) × ×
n MSAS i ( ) e
¿
\ .
|
| |
i INT v ( ) e
¿
adj
=
P
rec
i
n ( )
T
i
n ( ) L
traffi c
i
n ( ) f
act
i
n ( ) × =
L
traffi c
i
n ( )
f
act
i
n ( )
f
act
i
n ( ) 1 = L
traffi c
i
n ( ) 1 =
140 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
is the probability of having a co- or adjacent channel collision between MSAs n and m, depending on the
used frequency hopping mode.
- Collision Probability for Non Hopping Mode:
- Collision Probability for BBH and SFH Modes:
MSA m of v can be defined as the pair ([f
1
,f
2
,….f
n
], MAIO) and MSA n of i as the pair ([f’
1
,f’
2
,….f’
n
], MAIO’)
(where f and f’ are channels).
An occurence refers to the event when a channel f of m encounters a channel f’ of n during
hopping. A collision occurs when f and f’ are co- or adjacent channels:
such that
The probability of collision is the ratio of the number of collisions to the number of occurences:
The probibility of collision depends on the correlation between m and n. There can be two cases:
i. 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:
Here, the number of occurrences is 3, the number of co-channel collisions is 1, and the number of adjacent
channel collisions is 1. Therefore,
and
ii. MSAs m and n are not correlated
m and n do not have identical HSN and synchronisation. The probability of collision is the same for all the
channels.
Example:
Here, the number of occurrences is 9, the number of co-channel collisions is 1, and the number of adjacent
channel collisions is 3. Therefore,
and
Inter-technology Downlink Interference:
is the total inter-technology interference level on m due to transmitters in a linked Atoll document.
The interference from a transmitter Tx in a linked Atoll document is given as:
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
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
p
m n ,
v i ,
p
m n ,
v i ,
1 =
OCCUR f
m
v
f'
n
i
, ( )
Col l i si on OCCUR f
m
v
f'
n
i
, ( ) = f
m
v
f'
n
i
– 0 or 1 =
p
m n ,
v i ,
n
col l i si on
n
occurence
--------------------------- =
p
m n ,
v i ,
( )
co
1
3
--- = p
m n ,
v i ,
( )
adj
1
3
--- =
p
m n ,
v i ,
( )
co
1
9
--- = p
m n ,
v i ,
( )
adj
1
3
--- =
I
i nter techno y log –
DL
I
i nter techno y log –
DL
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
f ,
Tx
×
------------------------------------------
n
i
¿
=
© Forsk 2010 AT283_TRG_E2 141
Chapter 5: GSM GPRS EDGE Networks
Where is the frequency used by the transmitter Tx within its list of frequencies, is the total
transmitted Tx power on , are the total losses between the transmitter Tx and the receiver, and
is the inter-technology channel protection between the frequencies used by the transmitter Tx and the victim
transmitter v.
5.2.2 Point Analysis
Analysis provided in the Interference tab is based on path loss matrices. Therefore, it is possible to display the interference
levels received from TBC transmitters for which path loss matrices have been calculated over their calculation areas.
Atoll displays the following at the receiver:
• The carrier power level received from the victim transmitter v on the most interfered MAS m,
• Co-channel, adjacent channel, or both co- and adjacent channel interference received from interfering transmitters
i on MAS m (for further information about noise calculation, please refer to Signal to noise calculation: noise
calculation part),
Interferers are sorted in the order of descending carrier power levels.
5.2.3 Interference-based Coverage Predictions
Two interference-based coverage predictions are available:
• Coverage by C/I Level: Provides a global analysis of the network quality.
Atoll calculates the C/I on each pixel within the service area of studied transmitters, determines the pixels where
the calculated C/I exceeds the defined minimum threshold, and colours these pixels depending on C/I value.
• Interfered Zones: Shows the areas where a transmitter is interfered.
Atoll calculates the C/I on each pixel within the service area of studied transmitters, determines the pixels where
the calculated C/I is lower than the defined maximum threshold, and colours these pixels depending on colour of
the interfered transmitter.
For each TBC transmitter, Txi, Atoll calculates the selected parameter on each pixel inside the Txi calculation area. In
other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.
Coverage prediction parameters to be set are:
• The coverage conditions in order to determine the service area of each TBC transmitter,
• The interference conditions to meet for a pixel to be covered, and
• The display settings to select the displayed parameter and its shading levels.
The thermal noise (N = -121 dBm, by default) is used in the calculations if the coverage prediction is based on C/(I+N).
This value can be modified by the user.
5.2.3.1 Service Area Determination
Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas
where coverage will be displayed. Service areas are determined in the same manner as for signal level-based coverage
predictions. See "Service Area Determination" on page 134 for more information.
5.2.3.2 Coverage Area Determination
For each victim transmitter v, coverage area corresponds to pixels where or is between the lower and upper
thresholds defined in the coverage prediction properties.
The two options defining the thresholds are explained below.
5.2.3.2.1 Interference Condition Satisfied by At Least One TRX
In this case, the coverage area of a transmitter Txi corresponds to the pixels where:
Notes:
• In case of frequency hopping, the ICP value is weighted according to the fractional load.
• In the ICP, the frequency gap is based on the defined base frequency for each technology
(e.g., 935 MHz in GSM 900)
i c
i
i
th
P
Transmi tted
Tx
i c
i
( )
i c
i
L
total
Tx
ICP
i c
i
f ,
Tx
Notes:
• Neither DTX nor traffic load of TRXs are taken into account to evaluate interference levels.
Therefore, we have .
• The C/I shadowing margin is applied on the carrier power level. The interference levels are
not changed.
T
i
n ( ) L
traffi c
i
n ( ) f
act
i
n ( ) × 1 = =
C
I
----
\ .
| |
v
C
I N +
------------
\ .
| |
v
142 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
or
Where, TRX
j
is any TRX belonging to Txi.
5.2.3.2.2 Interference Condition Satisfied by The Worst TRX
In this case, the coverage area of a transmitter Txi corresponds to the pixels where:
or
Where, TRX
j
is the TRX (belonging to Txi) with the worst C/I or C/(I+N) at the pixel.
5.2.3.3 Coverage Display
5.2.3.3.1 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, see "Path
Loss Calculations" on page 77 for more information).
5.2.3.3.2 Display Types
It is possible to display the coverage predictions with colours depending on any transmitter attribute or other criteria such
as:
C/I Level
Each pixel of the transmitter coverage area is coloured if the calculated C/I (or C/(I+N)) level exceeds ( ) the specified
minimum thresholds (pixel 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.
Max C/I Level
Atoll compares calculated C/I (or C/(I+N)) levels received from transmitters on each pixel of each transmitter coverage
area where coverage areas overlap the studied one and chooses the highest value. A pixel of a coverage area is coloured
if the C/I (or C/(I+N)) level exceeds ( ) the specified thresholds (the pixel colour depends on the 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 defined thresholds. Each layer corresponds to an area where the highest received C/I level exceeds a defined
minimum threshold.
Min C/I Level
Atoll compares C/I (or C/(I+N)) levels received from transmitters on each pixel of each transmitter coverage area where
the coverage areas overlap the studied one and chooses the lowest value. A pixel of a coverage area is coloured if the C/
I (or C/(I+N)) level exceeds ( ) the specified thresholds (the pixel colour depends on the 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
defined thresholds. Each layer corresponds to an area where the lowest received C/I level exceeds a defined minimum
threshold.
5.3 GPRS/EDGE Calculations
GPRS/EDGE calculations include coding scheme selection and throughput calculation. Coding schemes may be selected
using ideal link adaptation or without it. Once coding schemes have been selected, throughputs corresponding to these
coding schemes are readily determined from the look-up tables.
The following sections describe the two categories of calculations, i.e., with and without ideal link adaptations. Ideal link
adaptation implies that the selected coding scheme corresponds to the highest available throughput under the given radio
conditions.
GPRS/EDGE calculations may be based on signal levels (C) alone, on C/I, or on C/(I+N). For calculating the noise, either
the noise figure defined for the calculations or that of the selected terminal type is used.
Different GPRS/EDGE configurations may be defined for transmitter and terminals. In this case, Atoll only selects the
coding schemes that are common in the two. If no terminal type is defined for the calculation, or if the terminal type does
not have any GPRS/EDGE configuration assigned to it, Atoll only uses the GPRS/EDGE configuration of the transmitter.
Similarly, if a transmitter does not have any GPRS/EDGE configuration assigned to it, Atoll only uses the GPRS/EDGE
configuration of the terminal type. If both the transmitter and the terminal type do not have any GPRS/EDGE configuration
assigned to them, no coding scheme selection and throughput calculation is carried out.
Minimum threshold
C
I
----
\ .
| |
v
TRX
j
s Maximum threshold < Minimum threshold
C
I N +
------------
\ .
| |
v
TRX
j
s Maximum threshold <
Minimum threshold
C
I
----
\ .
| |
v
TRX
j
s Maximum threshold < Minimum threshold
C
I N +
------------
\ .
| |
v
TRX
j
s Maximum threshold <
>
>
>
© Forsk 2010 AT283_TRG_E2 143
Chapter 5: GSM GPRS EDGE Networks
In the following calculations, we assume that:
• is the signal level received from the selected TRX type (tt) or on all the TRXs of Txi on each pixel of
the Txi coverage area,
• is the Power Backoff defined for the subcell for 8PSK, 16QAM, or 32QAM modulations,
• CS is the set of all available coding schemes,
• are the values of reception thresholds for the coding schemes available in the GPRS/
EDGE configuration,
• are the values of C/I thresholds for the coding schemes available in the GPRS/EDGE
configuration,
• are the values of C/(I+N) thresholds for the coding schemes available in the GPRS/EDGE
configuration,
• The priorities of the coding scheme lists are as follows: DBS > DAS > MCS > CS.
When the calculations are based on C/I and C/(I+N):
• Atoll calculates the carrier-to-interference ratio for all the GPRS/EDGE TBC transmitters but takes into account
all the TBC transmitters (GSM and GPRS/EDGE) to evaluate the interference.
• 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.
For more information on interference (I) calculation, see "Carrier-to-Interference Ratio Calculation" on page 138.
5.3.1 Coding Scheme Selection and Throughput Calculation
Without Ideal Link Adaptation
5.3.1.1 Calculations Based on C
Coding Scheme Selection
Atoll selects a coding scheme, cs, from among the coding schemes available in the GPRS/EDGE configuration, such that:
For each TRX type, tt,
The selected coding scheme, cs, is the coding scheme with the lowest coding scheme number from the lowest priority
coding scheme list.
Throughput Calculation
Once the coding scheme cs is selected, Atoll reads the corresponding throughput value for the received signal level from
the Throughput=f(C) graph associated with cs.
5.3.1.2 Calculations Based on C/I
Coding Scheme Selection
Atoll selects two coding schemes from among the coding schemes available in the GPRS/EDGE configuration, such that:
For each TRX type, tt,
And,
cs
C
is the coding scheme determined from the signal level, and cs
C/I
is the coding scheme determined from the C/I level.
Both coding schemes are the coding schemes with the lowest coding scheme number from the lowest priority coding
scheme list.
The selected coding scheme, cs, is the coding scheme with the lower coding scheme number among cs
C
and cs
C/I
:
.
Throughput Calculation Based on the Worst Case Between C and C/I
For the coding scheme cs
C
determined above, a throughput value, TP
C
, corresponding to the signal level is determined
from the TP = f(C) graph.
P
rec
Txi
TRX ( )
P
Backoff
Txi
TRX ( )
Recepti on Threshold ( )
CS
C
I
---- Threshold
\ .
| |
CS
C
I N +
------------ Threshold
\ .
| |
CS
cs Lowest CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) – Recepti on Threshold ( )
CS
>
\ .
| |
=
cs
C
Lowest CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) – Recepti on Threshold ( )
CS
>
\ .
| |
=
cs
C I
Lowest CS
P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) –
I
-----------------------------------------------------------------------------
C
I
---- Threshold
\ .
| |
CS
> \ .
|
|
| |
=
cs Mi n cs
C
cs
C I
, ( ) =
144 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
For the coding scheme cs
C/I
determined above, a throughput value, TP
C/I
, corresponding to the C/I is determined from the
TP = f(C/I) graph.
The resulting throughput TP is the lower of the two values, TP
C
and TP
C/I
: .
5.3.1.3 Calculations Based on C/(I+N)
Coding Scheme Selection
Atoll selects two coding schemes from among the coding schemes available in the GPRS/EDGE configuration, such that:
For each TRX type, tt,
And,
cs
C/N
is the coding scheme determined from the C/N, and cs
C/(I+N)
is the coding scheme determined from the C/(I+N) level.
Both coding schemes are the coding schemes with the lowest coding scheme numbers from the lowest priority coding
scheme list.
The selected coding scheme, cs, is the coding scheme with the higher coding scheme number among cs
C/N
and cs
C/(I+N)
:
.
Throughput Calculation Based on Interpolation Between C/N and C/(I+N)
For the coding scheme cs
C/N
determined above, the TP = f(C) graph is internally converted to TP = f(C/N) graph. A
throughput value, TP
C/N
, corresponding to the C/(I+N) is determined from the TP = f(C/N) graph.
For the coding scheme cs
C/(I+N)
determined above, the TP = f(C/I) graph is internally converted to TP = f(C/(I+N)) graph.
A throughput value, TP
C/(I+N)
, corresponding to the C/(I+N) is determined from the TP = f(C/(I+N)) graph.
The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The
throughput interpolation method consists in interpolating TP
C/N
and TP
C/(I+N)
according to the respective weights of I and
N values.
The resulting throughput TP is given by:
Where , pN is the thermal noise power (value in Watts), and p(I+N) is the interferences + thermal noise
power (value in Watts).
5.3.2 Coding Scheme Selection and Throughput Calculation With
Ideal Link Adaptation
5.3.2.1 Calculations Based on C
Throughput Calculation
For the received signal level, and coding schemes whose reception thresholds are lower than the received signal level,
Atoll determines the highest throughput from the graphs available in the GPRS/EDGE configuration.

Coding Scheme Selection
The selected coding scheme, cs, is the one corresponding to the highest throughput calculated above.
If there are more than one coding schemes providing the highest throughput at the pixel, the selected coding scheme, cs,
is the one with the lowest coding scheme number from the lowest priority coding scheme list.
5.3.2.2 Calculations Based on C/I
Throughput Calculation Based on Worst Case Between C and C/I
For the received signal level, and coding schemes whose reception thresholds are lower than the received signal level,
Atoll determines the highest throughput from the graphs available in the GPRS/EDGE configuration.
TP Mi n TP
C
TP
C I
, ( ) =
cs
C N
Lowest CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) –
N
-----------------------------------------------------------------------------
C
I N +
------------ Threshold
\ .
| |
CS
> \ .
|
|
| |
=
cs
C I N + ( )
Lowest CS
P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) –
I N +
-----------------------------------------------------------------------------
C
I N +
------------ Threshold
\ .
| |
CS
> \ .
|
|
| |
=
cs Max cs
C N
cs
C I N + ( )
, ( ) =
TP o TP
C N
× 1 o – ( ) TP
C I N + ( )
× + =
o
pN
p I N + ( )
--------------------- =
TP=f C ( )
TP
C
Hi ghest TP=f C P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) – = ( ) ( ) = CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) – Recepti on Threshold ( )
CS
>
\ .
| |
¬
TP=f C ( )
© Forsk 2010 AT283_TRG_E2 145
Chapter 5: GSM GPRS EDGE Networks

For the received C/I, and coding schemes whose C/I thresholds are lower than the received C/I, Atoll determines the
highest throughput from the graphs available in the GPRS/EDGE configuration.

The resulting throughput TP is the lower of the two values, TP
C
and TP
C/I
.
Coding Scheme Selection
The selected coding scheme, cs, is the one corresponding to the lower of the two highest throughputs calculated above.
If there are more than one coding schemes providing the highest throughputs at the pixel, the selected coding scheme,
cs, is the one with the lowest coding scheme number from the lowest priority coding scheme list.
5.3.2.3 Calculations Based on C/(I+N)
Throughput Calculation Based on Interpolation Between C/N and C/(I+N)
Atoll internally converts the TP = f(C) graphs into TP = f(C/N) graphs. For the received C/(I+N), and coding schemes
whose C/(I+N) thresholds are lower than the received C/(I+N), Atoll determines the highest throughput from the TP = f(C/
N) graphs available in the GPRS/EDGE configuration.

Atoll internally converts the TP = f(C/I) graphs into TP = f(C/(I+N)) graphs. For the received C/(I+N), and coding schemes
whose C/(I+N) thresholds are lower than the received C/(I+N), Atoll determines the highest throughput from the TP = f(C/
(I+N)) graphs available in the GPRS/EDGE configuration.

The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The
throughput interpolation method consists in interpolating TP
C/N
and TP
C/(I+N)
according to the respective weights of I and
N values.
The resulting throughput TP is given by:
Where , pN is the thermal noise power (value in Watts), and p(I+N) is the interferences + thermal noise
power (value in Watts).
Coding Scheme Selection
The selected coding scheme, cs, is the one corresponding to the higher of the two highest throughputs calculated above.
If there are more than one coding schemes providing the highest throughputs at the pixel, the selected coding scheme,
cs, is the one with the highest coding scheme number from the highest priority coding scheme list.
5.3.3 Application Throughput Calculation
Application throughput is calculated from the RLC/MAC throughput as follows:
Where is the RLC/MAC throughput, and and are the throughput offset (kbps) and the
throughput scaling factor (%) defined for the selected service.
TP
C
Hi ghest TP=f C P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) – = ( ) ( ) = CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) – Recepti on Threshold ( )
CS
>
\ .
| |
¬
TP=f C I ( )
TP
C I
Hi ghest TP=f C I
P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) –
I
--------------------------------------------------------------------------- =
\ .
|
| |
\ .
|
| |
= CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) –
I
-----------------------------------------------------------------------------
C
I
---- Threshold
\ .
| |
CS
> \ .
|
|
| |
¬
TP Mi n TP
C
TP
C I
, ( ) =
TP
C N
Hi ghest TP=f
C
N
----
P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) –
I N +
--------------------------------------------------------------------------- =
\ .
|
| |
\ .
|
| |
= CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) –
I N +
-----------------------------------------------------------------------------
C
I N +
------------ Threshold
\ .
| |
CS
> \ .
|
|
| |
¬
TP
C I N + ( )
Hi ghest TP=f
C
I N +
------------
P
rec
Txi
TRX ( ) P
Backoff
Txi
TRX ( ) –
I N +
--------------------------------------------------------------------------- =
\ .
|
| |
\ .
|
| |
= CS
P
rec
Txi
TRX ( ) P
Backof f
Txi
TRX ( ) –
I N +
-----------------------------------------------------------------------------
C
I N +
------------ Threshold
\ .
| |
CS
> \ .
|
|
| |
¬
TP o TP
C N
× 1 o – ( ) TP
C I N + ( )
× + =
o
pN
p I N + ( )
--------------------- =
TP
Appl i cati on
TP
RLC MAC
SF
100
---------- TP
Offset
– × =
TP
RLC MAC
TP
Offset
SF
146 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.3.4 BLER Calculation
Block error rate is calculated as follows:
Where TP is the throughput per timeslot calculated for a pixel and TP
MAX
is the maximum throughput per timeslot read
from the GPRS/EDGE configuration used for the calculations.
5.3.5 GPRS/EDGE Coverage Predictions
Two GPRS/EDGE coverage predictions are available:
• GPRS/EDGE Coding Schemes: Shows the areas where various coding schemes are available.
• Packet Throughput and Quality: Shows the throughputs corresponding to the coding schemes available.
For each TBC transmitter, Txi, Atoll calculates the selected parameter on each pixel inside the Txi calculation area. In
other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.
Coverage prediction parameters to be set are:
• The coverage conditions in order to determine the service area of each TBC transmitter,
• The interference conditions to meet for a pixel to be covered, and
• The display settings to select the displayed parameter and its shading levels.
The thermal noise (N = -121 dBm, by default) is used in the calculations if the coverage prediction is based on C/(I+N).
This value can be modified by the user.
5.3.5.1 Service Area Determination
Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas
where coverage will be displayed.
We can distinguish eight cases as below. Let us assume that:
• Each transmitter, Txi, belongs to a Hierarchical Cell Structure (HCS) layer, k, with a defined priority and a defined
reception threshold.
• Each transmitter, Txi, is GPRS/EDGE-capable.
• No max range is set.
5.3.5.1.1 All Servers
The service area of Txi corresponds to the pixels where:
5.3.5.1.2 Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.3.5.1.3 Second Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
BLER
TP
TP
MAX
------------------ If TP TP
MAX
s ( )
0 If TP TP
MAX
> ( )
¹
¦
´
¦
¦
=
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
tt ( )
Best
j i =
P
rec
Txj
tt ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
tt ( )
2
nd
Best
j i =
P
rec
Txj
tt ( ) ( ) M – >
© Forsk 2010 AT283_TRG_E2 147
Chapter 5: GSM GPRS EDGE Networks
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.3.5.1.4 Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.3.5.1.5 Second Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.3.5.1.6 HCS Servers and a Margin
The service area of Txi corresponds to the pixels where:
And
And the received exceeds the reception threshold defined per HCS layer.
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.3.5.1.7 Highest Priority HCS Server and a Margin
The service area of Txi corresponds to the pixels where:
And
And Txi belongs to the HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest).
And the received exceeds the reception threshold defined per HCS layer.
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
BCCH ( )
2
nd
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
tt ( )
SubcellReceptionThreshold P
rec
Txi
tt ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
tt ( )
148 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.3.5.1.8 Best Idle Mode Reselection Criterion (C2)
Such type of coverage is useful:
• To compare idle and dedicated mode best servers for voice traffic
• Display the GPRS/EDGE best server map (based on GSM idle mode)
The path loss criterion C1 used for cell selection and reselection is defined by:
The path loss criterion (GSM03.22) is satisfied if .
The reselection criterion C2 is used for cell reselection only and is defined by:
Where is the Cell Reselect Offset defined for the transmitter.
The service area of Txi corresponds to the pixels where:
And
The Best function considers the highest value from a list of values.
On each pixel, the transmitter with the highest C2 value is kept. It corresponds to the best server in idle mode. C2 is defined
as an integer in the 3GPP specifications, therefore, the C2 values in the above calculations are rounded down to the
nearest integer.
5.3.5.2 Coverage Display
5.3.5.2.1 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, see "Path
Loss Calculations" on page 77 for more information).
5.3.5.2.2 Display Types
It is possible to display the coverage predictions with colours depending on criteria such as:
GPRS/EDGE Coding Schemes: Coding Schemes
Only the pixels with a coding scheme assigned are coloured. The pixel 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.
GPRS/EDGE Coding Schemes: Best Coding Schemes
On each pixel, Atoll chooses the highest coding scheme available from the TRXs of different transmitters covering that
pixel. Only the pixels with a coding scheme assigned are coloured. The pixel 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.
Packet Throughput and Quality: RLC/MAC Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated RLC/MAC throughput per timeslot from any transmitter covering
that pixel exceeds the defined minimum threshold. The pixel colour depends on the RLC/MAC throughput per timeslot.
Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many
Note:
• 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.
C1 P
rec
Txi
BCCH ( ) Mi ni mumThreshol d BCCH ( ) – =
C1 0 >
C2 C1 CELL_RESELECT_OFFSET + =
CELL_RESELECT_OFFSET
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
C2
Txi
BCCH ( )
Best
j
C2
Txj
BCCH ( ) ( ) =
© Forsk 2010 AT283_TRG_E2 149
Chapter 5: GSM GPRS EDGE Networks
layers as transmitter coverage areas and throughput display thresholds. Each layer shows the RLC/MAC throughput that
a transmitter can provide on one timeslot.
Packet Throughput and Quality: Best RLC/MAC Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated highest RLC/MAC throughput per timeslot from any transmitter
covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the highest RLC/MAC throughput
per timeslot. Coverage consists of several independent layers whose visibility in the map window can be managed. There
are as many layers as throughput display thresholds. Each layer shows the best RLC/MAC throughput that any transmitter
can provide on one timeslot.
Packet Throughput and Quality: Average RLC/MAC Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated average RLC/MAC throughput per timeslot from all the
transmitters covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the average RLC/
MAC throughput per timeslot. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as throughput display thresholds. Each layer shows the average RLC/MAC
throughput that all the transmitters can provide on one timeslot.
Packet Throughput and Quality: Application Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated application throughput per timeslot from any transmitter covering
that pixel exceeds the defined minimum threshold. The pixel colour depends on the application throughput per timeslot.
Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many
layers as transmitter coverage areas and throughput display thresholds. Each layer shows the application throughput that
a transmitter can provide on one timeslot.
Packet Throughput and Quality: Best Application Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated highest application throughput per timeslot from any transmitter
covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the highest application
throughput per timeslot. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as throughput display thresholds. Each layer shows the best application throughput
that any transmitter can provide on one timeslot.
Packet Throughput and Quality: Average Application Throughput/Timeslot (kbps)
A pixel of the coverage area is coloured if the calculated average application throughput per timeslot from all the
transmitters covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the average
application throughput per timeslot. Coverage consists of several independent layers whose visibility in the map window
can be managed. There are as many layers as throughput display thresholds. Each layer shows the average application
throughput that all the transmitters can provide on one timeslot.
Packet Throughput and Quality: Max Application Throughput (kbps)
A pixel of the coverage area is coloured if the calculated application throughput from any transmitter covering that pixel
exceeds the defined minimum threshold. The pixel colour depends on the application throughput for all the timeslots
supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). Coverage consists
of several independent layers whose visibility in the map window can be managed. There are as many layers as transmitter
coverage areas and throughput display thresholds. Each layer shows the application throughput that a transmitter can
provide on all available timeslots in the terminal.
Packet Throughput and Quality: Best Max Application Throughput (kbps)
A pixel of the coverage area is coloured if the calculated highest application throughput from any transmitter covering that
pixel exceeds the defined minimum threshold. The pixel colour depends on the highest application throughput for all the
timeslots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). Coverage
consists of several independent layers whose visibility in the map window can be managed. There are as many layers as
throughput display thresholds. Each layer shows the highest application throughput that any transmitter can provide on all
available timeslots in the terminal.
Packet Throughput and Quality: Average Max Application Throughput (kbps)
A pixel of the coverage area is coloured if the calculated average application throughput from all the transmitters covering
that pixel exceeds the defined minimum threshold. The pixel colour depends on the average application throughput for all
the timeslots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots).
Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many
layers as throughput display thresholds. Each layer shows the average application throughput that all the transmitters can
provide on all available timeslots in the terminal.
Packet Throughput and Quality: User Throughput (kbps)
A pixel of the coverage area is coloured if the calculated user throughput from any transmitter covering that pixel exceeds
the defined minimum threshold. The pixel colour depends on the user throughput for all the timeslots supported by the
selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user throughput is calculated
150 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
by applying the throughput reduction factor, determined using the selected dimensioning model, to the application
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 and throughput display thresholds. Each layer shows the user
throughput that a transmitter can provide on all available timeslots in the terminal.
Packet Throughput and Quality: Max User Throughput (kbps)
A pixel of the coverage area is coloured if the calculated highest user throughput from any transmitter covering that pixel
exceeds the defined minimum threshold. The pixel colour depends on the highest user throughput for all the timeslots
supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user
throughput is calculated by applying the throughput reduction factor, determined using the selected dimensioning model,
to the application throughput. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as throughput display thresholds. Each layer shows the highest user throughput that
any transmitter can provide on all available timeslots in the terminal.
Packet Throughput and Quality: Average User Throughput (kbps)
A pixel of the coverage area is coloured if the calculated average user throughput from all the transmitters covering that
pixel exceeds the defined minimum threshold. The pixel colour depends on the average user throughput for all the
timeslots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user
throughput is calculated by applying the throughput reduction factor, determined using the selected dimensioning model,
to the application throughput. Coverage consists of several independent layers whose visibility in the map window can be
managed. There are as many layers as throughput display thresholds. Each layer shows the average user throughput that
all the transmitters can provide on all available timeslots in the terminal.
Packet Throughput and Quality: BLER (%)
A pixel of the coverage area is coloured if the calculated BLER from any transmitter exceeds the defined minimum
threshold. The pixel 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 and BLER display thresholds.
Each layer shows the BLERs that the covered pixels experience on one timeslot.
Packet Throughput and Quality: Max BLER (%)
A pixel of the coverage area is coloured if the calculated highest BLER from all the transmitters exceeds the defined
minimum threshold. The pixel 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 BLER display thresholds. Each layer shows
the BLER that the covered pixels experience on one timeslot.
5.4 Codec Mode Selection and CQI Calculations
Atoll supports FR, HR, EFR, and AMR codec modes. A codec configuration contains codec mode adaptation thresholds
and quality graphs for 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.
The default codec configurations in Atoll include default FER, BER, and MOS quality graphs with respect to the carrier to
interference ratio, and codec mode adaptation thresholds (calculated from the FER vs. C/I graphs for all codec modes at
5 % FER).
© Forsk 2010 AT283_TRG_E2 151
Chapter 5: GSM GPRS EDGE Networks
Figure 5.1FER vs. C/I Graphs
Figure 5.2BER vs. C/I Graphs
Figure 5.3MOS vs. C/I Graphs
152 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.4.1 Circuit Quality Indicator Calculations
Circuit quality indicator calculations include codec mode selection and CQI calculation. Codec modes may be selected
using ideal link adaptation or without it. Once codec modes have been selected, CQI corresponding to these codec modes
are determined from the look-up tables.
The following sections describe the two categories of calculations, i.e., with and without ideal link adaptations. Ideal link
adaptation implies that the selected codec mode corresponds to the best value of the reference CQI under the given radio
conditions. Without ideal link adaptation, the codec mode is selected based on the codec adaptation thresholds.
CQI calculations may be based on C/N or on C/(I+N). For calculating the noise, either the noise figure defined for the
calculations or that of the selected terminal type is used.
Different codec configurations may be defined for transmitter and terminals. In this case, Atoll only selects the codec
modes that are common in the two. If no terminal type is defined for the calculation, or if the terminal type does not have
any codec configuration assigned to it, Atoll only uses the codec configuration of the transmitter. Similarly, if a transmitter
does not have any codec configuration assigned to it, Atoll only uses the codec configuration of the terminal type. If both
the transmitter and the terminal type do not have any codec configuration assigned to them, no codec mode selection and
CQI calculation is carried out.
If more than one codec modes satisfy the C/N or C/I conditions, Atoll selects the higher priority codec mode.
In the following calculations, we assume that:
• is the signal level received from the selected TRX type (tt) or on all the TRXs of Txi on each pixel of
the Txi coverage area,
• CM is the set of all available codec modes,
• are the values of adaptation thresholds for the codec modes available in the codec
configuration,
The computed noise is compared to the codec configuration reference noise . If the values are the same,
the defined graphs are used as is, otherwise the graphs are downshifted by the difference .
When the calculations are based on C/(I+N):
• Atoll calculates the carrier-to-interference ratio for all the TBC transmitters with codec configurations assigned,
but takes into account all the TBC transmitters (with and without codec configurations) to evaluate the interference.
For more information on interference (I) calculation, see "Carrier-to-Interference Ratio Calculation" on page 138.
Ideal link adaptation for circuit quality indicator studies is defined at the codec configuration 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.
5.4.2 CQI Calculation Without Ideal Link Adaptation
5.4.2.1 Calculations Based on C/N
Atoll selects the highest priority codec mode, cm, from among the codec modes available in the codec configuration:
For each TRX type, tt,
For , Atoll determines the CQI from the CQI=f(C/N) graph associated to the selected codec mode, cm.
5.4.2.2 Calculations Based on C/(I+N)
Atoll selects the highest priority codec mode, cm, from among the codec modes available in the codec configuration:
References:
The 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)
P
rec
Txi
TRX ( )
Adaptati on Threshold ( )
CM
N N
Ref
N N
Ref

cm Highest Priority CM
P
rec
Txi
TRX ( )
N
------------------------------ Adaptati on Threshold ( )
CM
> \ .
|
|
| |
=
P
rec
Txi
TRX ( )
N
-----------------------------
© Forsk 2010 AT283_TRG_E2 153
Chapter 5: GSM GPRS EDGE Networks
For each TRX type, tt,
For , Atoll determines the CQI from the CQI=f(C/I) graph associated to the selected codec mode, cm.
5.4.3 CQI Calculation With Ideal Link Adaptation
5.4.3.1 Calculations Based on C/N
Ideal link adaptation is used by a codec configuration according to a defined reference CQI (MOS by default).
Atoll calculates signal level received from Txi on each pixel 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 configuration.
The selected codec mode among these filtered codec modes will be,
For each TRX type, tt, , for MOS
Or, , for BER and FER
Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI
gives the highest or the lowest value at the received C/N level, .
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
performed corresponding to for the selected codec mode.
5.4.3.2 Calculations Based on C/(I+N)
Ideal link adaptation is used by a codec configuration according to a defined reference CQI (MOS by default).
Atoll calculates the C/I level received from the transmitter on each pixel 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 configuration.
The selected codec mode among these filtered codec modes will be,
For each TRX type, tt, , for MOS
Or, , for BER and FER
Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI
gives the highest or the lowest value at the received C/(I+N) level, .
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
the CQI for which the study was performed corresponding to for the selected codec mode.
cm Highest Priority CM
P
rec
Txi
TRX ( )
I N +
------------------------------ Adaptati on Threshold ( )
CM
> \ .
|
|
| |
=
P
rec
Txi
TRX ( )
I N +
-----------------------------
cm Highest Priority CM
CQI
Ref
Hi ghest CQI=f
C
N
---- =
P
rec
Txi
TRX ( )
N
t ot
------------------------------
\ .
|
| |
\ .
|
| |
=
\ .
|
|
|
| |
=
cm Highest Priority CM
CQI
Ref
Lowest CQI=f
C
N
----=
P
rec
Txi
TRX ( )
N
t ot
------------------------------
\ .
|
| |
\ .
|
| |
=
\ .
|
|
|
| |
=
P
rec
Txi
TRX ( )
N
tot
-----------------------------
P
rec
Txi
TRX ( )
N
tot
-----------------------------
cm Highest Priority CM
CQI
Ref
Hi ghest CQI=f
C
I
---- =
P
rec
Txi
TRX ( )
I N
t ot
+
------------------------------
\ .
|
| |
\ .
|
| |
=
\ .
|
|
|
| |
=
cm Highest Priority CM
CQI
Ref
Lowest CQI=f
C
I
----=
P
rec
Txi
TRX ( )
I N
t ot
+
------------------------------
\ .
|
| |
\ .
|
| |
=
\ .
|
|
|
| |
=
P
rec
Txi
TRX ( )
I N +
tot
-----------------------------
P
rec
Txi
TRX ( )
I N +
tot
-----------------------------
154 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.4.4 Circuit Quality Indicators Coverage Predictions
The Circuit Quality Indicators coverage predictions show the areas BER, FER, and MOS values in the transmitter coverage
areas.
For each TBC transmitter, Txi, Atoll calculates the selected parameter on each pixel inside the Txi calculation area. In
other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.
Coverage prediction parameters to be set are:
• The coverage conditions in order to determine the service area of each TBC transmitter,
• The interference and quality indicator conditions to meet for a pixel to be covered, and
• The display settings to select the displayed parameter and its shading levels.
The thermal noise (N = -121 dBm, by default) is used in the calculations if the coverage prediction is based on C/(I+N).
This value can be modified by the user.
5.4.4.1 Service Area Determination
Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas
where coverage will be displayed.
We can distinguish seven cases as below. Let us assume that:
• Each transmitter, Txi, belongs to a Hierarchical Cell Structure (HCS) layer, k, with a defined priority and a defined
reception threshold.
• Each transmitter, Txi, has a codec configuration assigned.
• No max range is set.
5.4.4.1.1 All Servers
The service area of Txi corresponds to the pixels where:
5.4.4.1.2 Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.4.4.1.3 Second Best Signal Level and a Margin
The service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.4.4.1.4 Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
2
nd
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
© Forsk 2010 AT283_TRG_E2 155
Chapter 5: GSM GPRS EDGE Networks
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.4.4.1.5 Second Best Signal Level per HCS Layer and a Margin
For each HCS layer, k, the service area of Txi corresponds to the pixels where:
And
Where M is the specified margin (dB). The 2
nd
Best function considers the second highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within
a 2 dB margin from the second highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 3
rd
best servers.
5.4.4.1.6 HCS Servers and a Margin
The service area of Txi corresponds to the pixels where:
And
And the received exceeds the reception threshold defined per HCS layer.
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
5.4.4.1.7 Highest Priority HCS Server and a Margin
The service area of Txi corresponds to the pixels where:
And
And Txi belongs to the HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest).
And the received exceeds the reception threshold defined per HCS layer.
Where M is the specified margin (dB). The Best function considers the highest value from a list of values.
• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.
• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB
margin from the highest.
• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels
from transmitters which are 2
nd
best servers.
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
2
nd
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
BCCH ( )
Note:
• 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.
SubcellReceptionThreshold P
rec
Txi
BCCH ( ) s
P
rec
Txi
BCCH ( )
Best
j i =
P
rec
Txj
BCCH ( ) ( ) M – >
P
rec
Txi
BCCH ( )
156 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.4.4.2 Coverage Display
5.4.4.2.1 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, see "Path
Loss Calculations" on page 77 for more information).
5.4.4.2.2 Display Types
It is possible to display the coverage predictions with colours depending on criteria such as:
BER
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the BER 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 and BER display thresholds. Each layer shows the BER in the transmitter coverage area.
FER
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the FER 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 and FER display thresholds. Each layer shows the FER in the transmitter coverage area.
MOS
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the MOS 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 and MOS display thresholds. Each layer shows the MOS in the transmitter coverage area.
Max BER
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest BER value among the
BER values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in
the map window can be managed. There are as many layers as BER display thresholds. Each layer shows the BER value.
Max FER
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest FER value among the
FER values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in
the map window can be managed. There are as many layers as FER display thresholds. Each layer shows the FER value.
Max MOS
Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest MOS value among the
MOS values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in
the map window can be managed. There are as many layers as MOS display thresholds. Each layer shows the MOS
value.
5.5 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.
5.5.1 Traffic Distribution
5.5.1.1 Normal Cells (Nonconcentric, No HCS Layer)
5.5.1.1.1 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:
• The terminal, t, works on the frequency band used by the BCCH subcell,
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).
© Forsk 2010 AT283_TRG_E2 157
Chapter 5: GSM GPRS EDGE Networks
• The terminal, t, works on the frequency band used by the TCH subcell.
5.5.1.1.2 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:
• The transmitter is an GPRS/EDGE 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 BCCH subcell,
• The terminal, t, works on the frequency band used by the TCH subcell.
5.5.1.2 Concentric Cells
In case of concentric cells, TCH_INNER TRX type has the highest priority to carry traffic.
5.5.1.2.1 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 TCH_INNER,
BCCH and TCH subcells of a transmitter if:
• The terminal, t, works on the frequency band used by the BCCH subcell,
• The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells.
5.5.1.2.2 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 TCH_INNER,
BCCH and TCH subcells of a transmitter if:
• The transmitter is an GPRS/EDGE 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 BCCH subcell,
• The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells.
5.5.1.3 HCS Layers
For each HCS layer, k, you may specify the maximum mobile speed supported by the transmitters of the layer.
5.5.1.3.1 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 (and TCH_INNER in case of concentric cells) of a transmitter if:
• The terminal, t, works on the frequency band used by the BCCH subcell,
• The terminal, t, works on the frequency band(s) used by the TCH_INNER and TCH subcells,
• The user’s mobility, m, is less than the maximum speed supported by the layer, k.
5.5.1.3.2 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 (and TCH_INNER in case of concentric cells) of a transmitter if:
• The transmitter is an GPRS/EDGE 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 BCCH subcell,
• 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.
5.5.2 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.
5.5.2.1 User Profile Traffic Maps
5.5.2.1.1 Normal Cells (Nonconcentric, No HCS Layer)
Number of subscribers ( ) for each TCH subcell (Txi, TCH), per user profile up with a given mobility m, is inferred as:
Where S
up,m
is the TCH service area containing the user profile up with the mobility m and D is the user profile density.
X
up m ,
X
up m ,
Txi TCH , ( ) S
up m ,
Txi TCH , ( ) D × =
158 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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.
Circuit Switched Services
For a circuit switched service c, we have:
Where N
call
is the number of calls per hour and d is the average call duration (in seconds).
Then, Atoll evaluates the traffic demand, , in Erlangs for the subcell (Txi, TCH) service area.
Packet Switched Services (Max Rate)
For a max rate packet switched service p, we have:
Where N
call
is the number of calls per hour and V is the transmitted data volume per call (in Kbytes).
Then, Atoll evaluates the traffic demand, , in kbits/s for the subcell (Txi, TCH) service area.
Packet Switched Services (Constant Bit Rate)
For a constant bit packet switched service p, we have:
Where N
call
is the number of calls per hour and d is the average call duration (in seconds).
Then, Atoll evaluates the traffic demand, , in kbits/s for the subcell (Txi, TCH) service area.
5.5.2.1.2 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.
Number of subscribers ( ) for each TCH_INNER (Txi, TCH_INNER) and TCH (Txi, TCH) subcell, per user profile up
with a given mobility m, is inferred as:
and 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.
p
up c t , ( )
N
cal l
d ×
3600
---------------------- =
D
up c t , ( ) m ,
D
up c t , ( ) m ,
Txi TCH , ( ) X
up m ,
Txi TCH , ( ) p
up c t , ( )
× =
p
up p t , ( )
N
cal l
V × 8 ×
3600
-------------------------------- =
D
up p t , ( ) m ,
D
up p t , ( ) m ,
Txi TCH , ( ) X
up m ,
Txi TCH , ( ) p
up p t , ( )
× =
p
up p t , ( )
N
cal l
d ×
3600
---------------------- =
D
up p t , ( ) m ,
D
up p t , ( ) m ,
Txi TCH , ( ) X
up m ,
Txi TCH , ( ) p
up p t , ( )
× =
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.
Figure 5.4Representation of a Concentric Cell TXi
X
up m ,
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 ( ) S
up m ,
Txi,TCH ( )
© Forsk 2010 AT283_TRG_E2 159
Chapter 5: GSM GPRS EDGE Networks
Circuit Switched Services
For each user of the user profile up using a circuit switched service c with a terminal t, Atoll calculates the probability
( ) of the user being connected. Calculations are detailed in "Circuit Switched Services" on page 156.
Then, Atoll evaluates the traffic demand, , in Erlangs in the (Txi, TCH_INNER) and (Txi, TCH) subcell service
areas.
Where is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.
Packet Switched Services (Max Rate)
For each user of the user profile up using a max rate packet switched service p with a terminal t, probability of the user
being connected ( ) is calculated as explained in "Packet Switched Services" on page 157.
Atoll evaluates the traffic demand, , in kbits/s in the (Txi, TCH_INNER) and (Txi, TCH) subcell service areas.
Where is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.
Packet Switched Services (Constant Bit Rate)
For each user of the user profile up using a constant bit packet switched service p with a terminal t, probability of the user
being connected ( ) is calculated as explained in "Packet Switched Services" on page 157.
Atoll evaluates the traffic demand, , in kbits/s in the (Txi, TCH_INNER) and (Txi, TCH) subcell service areas.
Where is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.
5.5.2.1.3 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. The traffic contained in the input traffic map can be assigned to all the HCS layers.
Normal Cells
Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option “Best signal level per
HCS layer” meaning that there is an overlap between HCS layers service areas. Let 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.
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).
p
up c t , ( )
D
up c t , ( ) m ,
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 ( ) × + =
O
max
Txi,TCH_INNER ( )
p
up p t , ( )
D
up p t , ( ) m ,
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 ( ) × + =
O
max
Txi,TCH_INNER ( )
p
up p t , ( )
D
up p t , ( ) m ,
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 ( ) × + =
O
max
Txi,TCH_INNER ( )
Note:
• 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.
Figure 5.5Representation of Micro and Macro Layers
S
overl appi ng
macro
Txj TCH , ( )
160 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Number of subscribers ( ) for each TCH subcell (Txj, TCH) of the macro layer, per user profile up with the mobility
m, is inferred as:
Where 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 ( ) is calculated as explained in "Circuit Switched Services" on page 156.
Then, Atoll evaluates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service area.
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 ( ) is calculated as explained in "Packet Switched Services" on page 157.
Then, Atoll evaluates the traffic demand, , in kbits/s in the subcell (Txj, TCH) service area.
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro
layer) and 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.
The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area
overlapped by the TCH service area of the micro layer and another overlapped
by the TCH_INNER service area of the micro layer .
Let us consider three areas, S
1
, S
2
and S
3
.
Figure 5.6Concentric Cells
X
up m ,
macro
X
up m ,
macro
Txj TCH , ( ) S
up m ,
macro
Txj TCH , ( ) S
up m , overl appi ng –
macro
Txj TCH , ( ) – | | D × =
S
up m ,
macro
Txj TCH , ( )
p
up c t , ( )
D
up c t , ( ) m ,
macro
D
up c t , ( ) m ,
macro
Txj TCH , ( ) X
up m ,
macro
Txj TCH , ( ) p
up c t , ( )
× D
up c t , ( ) m ,
mi cro
Txi TCH , ( )
S
up m , overl appi ng –
macro
Txj TCH , ( )
S
up m ,
mi cro
Txi TCH , ( )
--------------------------------------------------------------------------- × O
max
Txi TCH , ( ) × + =
p
up p t , ( )
D
up p t , ( ) m ,
macro
D
up p t , ( ) m ,
macro
Txj TCH , ( ) X
up m ,
macro
Txj TCH , ( ) p
up p t , ( )
× D
up p t , ( ) m ,
mi cro
Txi TCH , ( )
S
up m , overl appi ng –
macro
Txj TCH , ( )
S
up m ,
mi cro
Txi TCH , ( )
--------------------------------------------------------------------------- × O
max
Txi TCH , ( ) × + =
O
max
Txi TCH , ( )
S
up m ,
mi cro
Txi TCH , ( )
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH_INNER ( )
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH_INNER ( )
S
1
S
up m ,
macro
Txj,TCH_INNER ( ) S
up m , overl appi ng – Txi TCH , ( ) –
macro
Txj,TCH_INNER ( ) – =
S
2
S
up m , overl appi ng – Txi,TCH_INNER ( ) –
macro
Txj,TCH_INNER ( ) =
S
3
S
up m , overl appi ng – Txi TCH , ( ) –
macro
Txj,TCH_INNER ( ) S
2
– =
© Forsk 2010 AT283_TRG_E2 161
Chapter 5: GSM GPRS EDGE Networks
Where 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.
On S
1
, the number of subscribers per user profile up with a given mobility m ( ) is inferred:
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 S
2
, the TCH subcell traffic coming from the
TCH_INNER subcell traffic overflow may overflow proportional to R
2
.
The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S
3
proportional to R
3
.
Where and 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 ( ) is calculated as explained in "Circuit Switched Services" on page 156. Then, Atoll evaluates
the traffic demand, , in Erlangs in the subcell (Txj, TCH_INNER) service area.
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 ( ) is calculated as explained in "Packet Switched Services" on page 157.
Then, Atoll evaluates the traffic demand, , stated in kbits/s in the subcell (Txj, TCH_INNER) service area.
Where and 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
the TCH service area of the micro layer and another one by the
TCH_INNER service area of the micro layer .
Let us consider three areas, S’
1
, S’
2
and S’
3
.
Where and 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.
On S’
1
, the number of subscribers per user profile up with a given mobility m ( ) is inferred:
Where D is the user profile density.
S
up m ,
macro
Txj,TCH_INNER ( )
X
up m ,
macro
X
up m ,
macro
Txj,TCH_INNER ( ) S
1
D × =
R
2
S
2
S
up m ,
mi cro
Txi,TCH_INNER ( )
----------------------------------------------------------------- =
R
3
S
3
S
up m ,
mi cro
Txi,TCH ( ) S
up m ,
mi cro
Txi,TCH_INNER ( ) –
------------------------------------------------------------------------------------------------------------------- =
S
up m ,
mi cro
Txi,TCH ( ) S
up m ,
mi cro
Txi,TCH_INNER ( )
p
up c t , ( )
D
up c t , ( ) m ,
macro
D
up c t , ( ) m ,
macro
Txj,TCH_INNER ( )
X
up m ,
macro
Txj,TCH_INNER ( ) p
up c t , ( )
× +
R
2
D
up c t , ( ) m ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi,TCH ( ) × +
R
3
X
up m ,
mi cro
Txi TCH , ( ) × p
up c t , ( )
× O
max
Txi TCH , ( ) ×
=
p
up p t , ( )
D
up p t , ( ) m ,
macro
D
up p t , ( ) m ,
macro
Txj,TCH_INNER ( )
X
up m ,
macro
Txj,TCH_INNER ( ) p
up p t , ( )
× +
R
2
D
up p t , ( ) m ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi,TCH ( ) × +
R
3
X
up m ,
mi cro
Txi TCH , ( ) × p
up p t , ( )
× O
max
Txi TCH , ( ) ×
=
O
max
Txi TCH , ( ) O
max
Txi,TCH_INNER ( )
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( )
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH -- TCH_INNER ( )
S'
1
S
up m ,
macro
Txj,TCH ( ) S
up m ,
macro
Txj,TCH_INNER ( ) S
up m , overl appi ng – Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( ) – – =
S'
2
S
up m , overl appi ng – Txi,TCH_INNER ( ) –
macro
Txj,TCH -- TCH_INNER ( ) =
S'
3
S
up m , overl appi ng – Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( ) S'
2
– =
S
up m ,
macro
Txj,TCH ( ) S
up m ,
macro
Txj,TCH_INNER ( )
X
up m ,
macro
X
up m ,
macro
Txj,TCH ( ) S'
1
D × =
162 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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 S’
2
, the TCH subcell traffic coming from
the TCH_INNER subcell traffic overflow may overflow proportionally to R’
2
.
The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S’
3
proportional to R’
3
.
Where and 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 ( ) is calculated as explained in "Circuit Switched Services" on page 156.
Then, Atoll evaluates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service area.
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 ( ) is calculated as explained in "Packet Switched Services" on page 157.
Then, Atoll evaluates the traffic demand, , in kbits/s in the subcell (Txj, TCH) service area.
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro
layer), the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txi (macro
layer), the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txj (macro
layer) and 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 157).
5.5.2.2 Sector Traffic Maps
We assume that the traffic map is built from a coverage by transmitter prediction 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 max
rate packet switched service and Erlangs for a circuit switched or constant bit rate packet switched service) and the global
distribution of terminals and mobility types.
Let denote the Erlangs for the circuit switched service, c, on the TCH subcell of Txi.
Let denote the throughput of the packet switched service (Max Bit Rate), p, on the TCH subcell of Txi.
Let denote the Erlangs for the packet switched service (Constant Bit Rate), p, on the TCH subcell of Txi.
We assume that 100% of users have the terminal, t, and the mobility type, m.
5.5.2.2.1 Normal Cells (Nonconcentric, No HCS Layer)
For each circuit switched service, c, Atoll evaluates the traffic demand, D
c,t,m
, in Erlangs in the subcell (Txi, TCH) service
area.
R'
2
S'
2
S
up m ,
mi cro
Txi,TCH_INNER ( )
----------------------------------------------------------------- =
R'
3
S'
3
S
up m ,
mi cro
Txi,TCH ( ) S
up m ,
mi cro
Txi,TCH_INNER ( ) –
------------------------------------------------------------------------------------------------------------------- =
S
up m ,
mi cro
Txi,TCH ( ) S
up m ,
mi cro
Txi,TCH_INNER ( )
p
up c t , ( )
D
up c t , ( ) m ,
macro
D
up c t , ( ) m ,
macro
Txj TCH , ( )
X
up m ,
macro
Txj TCH , ( ) p
up c t , ( )
× +
D
up c t , ( ) m ,
macro
Txj,TCH_INNER ( ) O
max
Txj,TCH_INNER ( ) × +
R'
2
D
up c t , ( ) m ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi,TCH ( ) × +
R'
3
X
up m ,
mi cro
Txi TCH , ( ) × p
up c t , ( ) m ,
× O
max
Txi TCH , ( ) ×
=
p
up p t , ( )
D
up p t , ( ) m ,
macro
D
up p t , ( ) m ,
macro
Txj TCH , ( )
X
up m ,
macro
Txj TCH , ( ) p
up p t , ( )
× +
D
up p t , ( ) m ,
macro
Txj,TCH_INNER ( ) O
max
Txj,TCH_INNER ( ) × +
R'
2
D
up p t , ( ) m ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi,TCH ( ) × +
R'
3
X
up m ,
mi cro
Txi TCH , ( ) × p
up p t , ( ) m ,
× O
max
Txi TCH , ( ) ×
=
O
max
Txi,TCH ( )
O
max
Txi,TCH_INNER ( )
O
max
Txj,TCH_INNER ( )
X
up m ,
mi cro
Txi TCH , ( )
E
c
Txi TCH , ( )
T
p
Txi TCH , ( )
E
p
Txi TCH , ( )
D
c t m , ,
Txi TCH , ( ) E
c
Txi TCH , ( ) =
© Forsk 2010 AT283_TRG_E2 163
Chapter 5: GSM GPRS EDGE Networks
For each packet switched service (Max Bit Rate), p, Atoll evaluates the traffic demand, D
p,t,m
, in kbits/s in the subcell (Txi,
TCH) service area.
For each packet switched service (Constant Bit Rate), p, Atoll evaluates the traffic demand, D
p,t,m
, in kbits/s in the subcell
(Txi, TCH) service area.
where is the guaranteed bit rate of the constant bit rate packet switched service p.
5.5.2.2.2 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.
For each circuit switched service, c, Atoll evaluates the traffic demand, D
c,t,m
, in Erlangs in the subcell, (Txi, TCH_INNER)
and (Txi, TCH), service areas.
and
For each packet switched service (Max Bit Rate), p, Atoll evaluates the traffic demand, D
p,t,m
, in kbits/s in the subcell,
(Txi, TCH_INNER) and (Txi, TCH), service areas.
and
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell,
and are the TCH and TCH_INNER service areas of Txi respectively.
For each packet switched service (Constant Bit Rate), p, Atoll evaluates the traffic demand, D
p,t,m
, in kbits/s in the subcell,
(Txi, TCH_INNER) and (Txi, TCH), service areas.
and
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell,
and are the TCH and TCH_INNER service areas of Txi respectively.
5.5.2.2.3 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. The traffic contained in the input traffic map can be assigned to all the HCS
layers.
Normal Cells
Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option “HCS Servers”. It means
that there is an overlapping area between HCS layers. Let 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
D
p t m , ,
Txi TCH , ( ) T
p
Txi TCH , ( ) =
D
p t m , ,
Txi TCH , ( ) E
p
Txi TCH , ( ) TP
p GBR ,
× =
TP
p GBR ,
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.
D
c t m , ,
Txi,TCH_INNER ( )
S Txi,TCH_INNER ( )
S Txi TCH , ( )
----------------------------------------------------- E
c
Txi TCH , ( ) × =
D
c t m , ,
Txi,TCH ( )
S Txi,TCH ( ) S Txi,TCH_INNER ( ) – ( )
S Txi TCH , ( )
----------------------------------------------------------------------------------------------- E
c
Txi TCH , ( ) × +
D
c t m , ,
Txi,TCH_INNER ( ) O
max
Txi,TCH_INNER ( ) ×
=
D
p t m , ,
Txi,TCH_INNER ( )
S Txi,TCH_INNER ( )
S Txi TCH , ( )
----------------------------------------------------- T
p
Txi TCH , ( ) × =
D
p t m , ,
Txi,TCH ( )
S Txi,TCH ( ) S Txi,TCH_INNER ( ) – ( )
S Txi TCH , ( )
----------------------------------------------------------------------------------------------- T
p
Txi TCH , ( ) × +
D
p t m , ,
Txi,TCH_INNER ( ) O
max
Txi,TCH_INNER ( ) ×
=
O
max
Txi,TCH_INNER ( )
S Txi,TCH ( ) S Txi,TCH_INNER ( )
D
p t m , ,
Txi,TCH_INNER ( )
S Txi,TCH_INNER ( )
S Txi TCH , ( )
----------------------------------------------------- E
p
Txi TCH , ( ) TP
p GBR ,
× × =
D
p t m , ,
Txi,TCH ( )
S Txi,TCH ( ) S Txi,TCH_INNER ( ) – ( )
S Txi TCH , ( )
----------------------------------------------------------------------------------------------- E
p
Txi TCH , ( ) TP
p GBR ,
× × +
D
p t m , ,
Txi,TCH_INNER ( ) O
max
Txi,TCH_INNER ( ) ×
=
O
max
Txi,TCH_INNER ( )
S Txi,TCH ( ) S Txi,TCH_INNER ( )
S
overl appi ng
macro
Txj TCH , ( )
164 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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.
Atoll starts evaluating the traffic demand on the micro layer (highest priority HCS layer).
For each circuit switched service, c, Atoll calculates the traffic demand, , in Erlangs in the subcell (Txi, TCH) service
area.
For each packet switched service (Max Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the subcell
(Txi, TCH) service area.
For each packet switched service (Constant Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the
subcell (Txi, TCH) service area.
Then, Atoll proceeds with the macro layer (lower priority HCS layer).
For each circuit switched service, c, Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service
area.
For each packet switched service (Max Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the subcell
(Txj, TCH) service area.
Where is the maximum rate of traffic overflow (in %) specified for the TCH subcell of Txi (micro cell) and
the TCH service area of Txi.
For each packet switched service (Constant Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the
subcell (Txj, TCH) service area.
Where is the maximum rate of traffic overflow (in %) specified for the TCH subcell of Txi (micro cell) and
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.
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.
Note:
• You can restrict the traffic assignement of each traffic map to a specific HCS layer in the
running options of the traffic capture. If you do so, no overflow occurs between HCS layers
and the only overflow which is considered occurs within concentric cells (See "Concentric
Cells" on page 157).
D
c t m , ,
mi cro
D
c t m , ,
mi cro
Txi TCH , ( ) E
c
Txi TCH , ( ) =
D
p t m , ,
mi cro
D
p t m , ,
mi cro
Txi TCH , ( ) T
p
Txi TCH , ( ) =
D
p t m , ,
mi cro
D
p t m , ,
mi cro
Txi TCH , ( ) E
p
Txi TCH , ( ) TP
p GBR ,
× =
D
c t m , ,
macro
D
c t m , ,
macro
Txj TCH , ( ) E
c
Txj TCH , ( ) D
c t m , ,
mi cro
Txi TCH , ( )
S
overl appi ng
macro
Txj TCH , ( )
S
mi cro
Txi TCH , ( )
------------------------------------------------------------- × O
max
Txi TCH , ( ) × + =
D
p t m , ,
macro
D
p t m , ,
macro
Txj TCH , ( ) T
p
Txj TCH , ( ) D
p t m , ,
mi cro
Txi TCH , ( )
S
overl appi ng
macro
Txj TCH , ( )
S
mi cro
Txi TCH , ( )
------------------------------------------------------------- × O
max
Txi TCH , ( ) × + =
O
max
Txi TCH , ( )
S
mi cro
Txi TCH , ( )
D
p t m , ,
macro
D
p t m , ,
macro
Txj TCH , ( ) E
p
Txi TCH , ( ) TP
p GBR ,
× D
p t m , ,
mi cro
Txi TCH , ( )
S
overl appi ng
macro
Txj TCH , ( )
S
mi cro
Txi TCH , ( )
------------------------------------------------------------- × O
max
Txi TCH , ( ) × + =
O
max
Txi TCH , ( )
S
mi cro
Txi TCH , ( )
© Forsk 2010 AT283_TRG_E2 165
Chapter 5: GSM GPRS EDGE Networks
The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area
overlapped by the TCH service area of the micro layer and another overlapped
by the TCH_INNER service area of the micro layer .
Let us consider three areas, S
1
, S
2
and S
3
.
Where is the TCH_INNER subcell service area of Txj.
The traffic specified for Txj in the map description ( ) is spread over S
1
proportionally to R
1
.
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 S
2
, the TCH subcell traffic coming from the
TCH_INNER subcell traffic overflow may overflow proportional to R
2
.
The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S
3
proportional to R
3
.
For each circuit switched service, c, Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj,
TCH_INNER) service area.
For each packet switched service (Max Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the subcell
(Txj, TCH_INNER) service area.
Figure 5.7Concentric Cells
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH_INNER ( )
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH_INNER ( )
S
1
S
macro
Txj,TCH_INNER ( ) S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH_INNER ( ) – =
S
2
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH_INNER ( ) =
S
3
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH_INNER ( ) S
2
– =
S
macro
Txj,TCH_INNER ( )
E
c
Txj TCH , ( )
R
1
S
1
S
map
Txj TCH , ( )
-------------------------------------------- =
S
map
Txj TCH , ( )
R
2
S
2
S
mi cro
Txi,TCH_INNER ( )
----------------------------------------------------------------- =
R
3
S
3
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( ) –
------------------------------------------------------------------------------------------------------------------- =
D
c t m , ,
macro
D
c t m , ,
macro
Txj,TCH_INNER ( )
R
1
E
c
Txj TCH , ( ) × +
R
2
D
c t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R
3
S
mi cro
Txi TCH , ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi TCH , ( )
--------------------------------------------------------------------------------------------------------------------------- × E
c
Txi TCH , ( ) × O
max
Txi TCH , ( ) ×
=
D
p t m , ,
macro
166 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi
and is the TCH subcell service area of Txi.
For each packet switched service (Constant Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the
subcell (Txj, TCH_INNER) service area.
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi
and 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
the TCH service area of the micro layer and another overlapped by the
TCH_INNER service area of the micro layer .
Let us consider three areas, S’
1
, S’
2
and S’
3
.
Where and are the TCH and TCH_INNER subcell service areas of Txj
respectively.
The traffic specified for Txj in the map description ( ) is spread over S’
1
proportional to R’
1
.
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 S’
2
, the TCH subcell traffic coming from
the TCH_INNER subcell traffic overflow may overflow proportional to R’
2
.
The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S’
3
proportional to R’
3
.
For each circuit switched service, c, Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj, TCH)
service area.
D
p t m , ,
macro
Txj,TCH_INNER ( )
R
1
T
p
Txj TCH , ( ) × +
R
2
D
p t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R
3
S
mi cro
Txi TCH , ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi TCH , ( )
--------------------------------------------------------------------------------------------------------------------------- × T
p
Txi TCH , ( ) × O
max
Txi TCH , ( ) ×
=
O
max
Txi TCH , ( )
O
max
Txi,TCH_INNER ( )
S
mi cro
Txi TCH , ( )
D
p t m , ,
macro
D
p t m , ,
macro
Txj,TCH_INNER ( )
R
1
E
p
Txi TCH , ( ) TP
p GBR ,
× × +
R
2
D
p t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R
3
S
mi cro
Txi TCH , ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi TCH , ( )
---------------------------------------------------------------------------------------------------------------------------
×
E
p
Txi TCH , ( ) TP
p GBR ,
× O
max
Txi TCH , ( ) ×
\ .
|
|
|
|
|
| |
×
=
O
max
Txi TCH , ( )
O
max
Txi,TCH_INNER ( )
S
mi cro
Txi TCH , ( )
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( )
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH -- TCH_INNER ( )
S'
1
S
macro
Txj TCH , ( ) S
macro
Txj,TCH_INNER ( ) – S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( ) – =
S'
2
S
overl appi ng Txi,TCH_INNER ( ) –
macro
Txj,TCH -- TCH_INNER ( ) =
S'
3
S
overl appi ng Txi TCH , ( ) –
macro
Txj,TCH -- TCH_INNER ( ) S'
2
– =
S
macro
Txj TCH , ( ) S
macro
Txj,TCH_INNER ( )
E
c
Txj TCH , ( )
R'
1
S'
1
S
map
Txj TCH , ( )
-------------------------------------------- =
S
map
Txj TCH , ( )
R'
2
S'
2
S
mi cro
Txi,TCH_INNER ( )
----------------------------------------------------------------- =
R'
3
S'
3
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( ) –
------------------------------------------------------------------------------------------------------------------- =
D
c t m , ,
macro
© Forsk 2010 AT283_TRG_E2 167
Chapter 5: GSM GPRS EDGE Networks
For each packet switched service (Max Bit Rate), p,Atoll calculates the traffic demand, , in kbits/s in the subcell
(Txj, TCH) service area.
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell
of Txj, is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi,
is the TCH subcell service area of Txi and is the TCH_INNER subcell service
area of Txi.
For each packet switched service (Constant Bit Rate), p,Atoll calculates the traffic demand, , in kbits/s in the
subcell (Txj, TCH) service area.
Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell
of Txj, is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,
is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi,
is the TCH subcell service area of Txi and is the TCH_INNER subcell service
area of Txi.
5.6 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 Atoll’s dimensioning engine.
5.6.1 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.
D
c t m , ,
macro
Txj TCH , ( )
R'
1
E
c
× Txj TCH , ( ) +
D
c t m , ,
macro
Txj,TCH_INNER ( ) O
max
Txj,TCH_INNER ( ) × +
R'
2
D
c t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R'
3
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi,TCH ( )
------------------------------------------------------------------------------------------------------------------------ × E
c
Txi TCH , ( ) × O
max
Txi TCH , ( ) ×
=
D
p t m , ,
macro
D
p t m , ,
macro
Txj TCH , ( )
R'
1
T
p
× Txj TCH , ( ) +
D
c t m , ,
macro
Txj,TCH_INNER ( ) O
max
Txj,TCH_INNER ( ) × +
R'
2
D
p t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R'
3
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi,TCH ( )
------------------------------------------------------------------------------------------------------------------------ × T
p
Txi TCH , ( ) × O
max
Txi TCH , ( ) ×
=
O
max
Txj,TCH_INNER ( )
O
max
Txi TCH , ( )
O
max
Txi,TCH_INNER ( )
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( )
D
p t m , ,
macro
D
p t m , ,
macro
Txj TCH , ( )
R'
1
E
p
Txi TCH , ( ) TP
p GBR ,
× × +
D
c t m , ,
macro
Txj,TCH_INNER ( ) O
max
Txj,TCH_INNER ( ) × +
R'
2
D
p t m , ,
mi cro
Txi,TCH_INNER ( ) × O
max
Txi,TCH_INNER ( ) × O
max
Txi TCH , ( ) × +
R'
3
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( ) – ( )
S
mi cro
Txi,TCH ( )
------------------------------------------------------------------------------------------------------------------------
×
E
p
Txi TCH , ( ) TP
p GBR ,
× O
max
Txi TCH , ( ) ×
\ .
|
|
|
|
|
| |
×
=
O
max
Txj,TCH_INNER ( )
O
max
Txi TCH , ( )
O
max
Txi,TCH_INNER ( )
S
mi cro
Txi,TCH ( ) S
mi cro
Txi,TCH_INNER ( )
168 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.6.1.1 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.
5.6.1.2 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.
5.6.1.2.1 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
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 network’s 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 (N
u
), number of timeslots available in the system (N
s
) and the
average system packet switched traffic load (L
p
) (utilization of resources in the system). Data Erlangs or data traffic is given
by:
© Forsk 2010 AT283_TRG_E2 169
Chapter 5: GSM GPRS EDGE Networks
More precisely, the reduction factor is a function of the ratio N
s
/N
u
(N
p
). N
p
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:
Where n is the instantaneous number of connections in the system. The throughput reduction factor is defined as:
Or,
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:
Hence the reduction factor can finally be written as:
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.
Figure 5.8Reduction of Throughput per Timeslot
Data Erlangs L
P
N
S
× =
X
0 if n = 0
1 if 0 < n N
P
s
N
P
n
------- if n > N
P
÷
RF X
P X n = ( )
P X 0 = ( )
----------------------- ·
n 0 =
·
¿
÷
RF X
P X n = ( )
P
i 0 =
·
¿
X i = ( )
------------------------------- ·
n 0 =
·
¿
=
P X n = ( )
L
P
N
P
· ( )
n
n!
--------------------------
L
P
N
P
· ( )
i
i!
------------------------
i 0 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
i N
P
– ( )
·
---------------------------------
i N
P
1 + =
·
¿
+
----------------------------------------------------------------------------------------------- = if 0 n N
P
s s
P X n = ( )
L
P
N
P
· ( )
n
N
P
! N
P
i N
P
– ( )
·
---------------------------------
L
P
N
P
· ( )
i
i!
------------------------
i 0 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
i N
P
– ( )
·
---------------------------------
i N
P
1 + =
·
¿
+
----------------------------------------------------------------------------------------------- = if n > N
P
RF
L
P
N
P
· ( )
i
i!
------------------------
i 1 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
i N
P
– ( )
·
---------------------------------
i N
P
1 + =
·
¿
N
P
i
-------
\ .
| |
· +
L
P
N
P
· ( )
i
i!
------------------------
i 1 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
i N
P
– ( )
·
---------------------------------
i N
P
1 + =
·
¿
+
----------------------------------------------------------------------------------------------------------------- =
170 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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 (N
p
) describing the reduction factor at different values of packet
switched traffic load (L
p
). 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.
Each curve in the above figure represents an equivalent number of packet switched timeslots, N
P
.
5.6.1.2.2 Delay
Delay is the time required for an LLC PDU to be completely transferred from the SGSN to the MS, or vice versa. 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.
5.6.1.2.3 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 * N
p
users in the system. This implies that if a new TBF is requested when there are already M * N
p
users active, it will be
blocked and placed in a queue. So the blocking probability is the probability of having M * N
p
+ 1 users in the system or
more, meaning,
as in this case n is always greater than N
p
, we have,
Figure 5.9Reduction Factor for Different Packet Switched Traffic Loads (L
p
, X-axis)
RF
L
P
N
P
· ( )
n
n!
--------------------------
n 1 =
N
P
¿
N
P
N
P
1 + ( )
N
P
!
--------------------- 1 L
P
– ( ) ln
L
P
n
n
------
n 1 =
N
P
¿
+
\ .
|
|
| |
· –
L
P
N
P
· ( )
n
n!
--------------------------
n 1 =
N
P
¿
L
P
N
P
· ( )
N
P
N
P
!
-----------------------------
L
P
1 L
P

--------------- · +
----------------------------------------------------------------------------------------------------------------------------------- =
P X n = ( ) for n M N
P
· ( ) 1 + =
© Forsk 2010 AT283_TRG_E2 171
Chapter 5: GSM GPRS EDGE Networks
So, the Blocking Probability can be given as:
Eliminating the summations to infinity, the blocking probability can be stated in a simpler form:
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, N
P
.
The curves depict the blocking probabilities for different number of available connections (N
p
) at different packet switched
traffic loads (L
p
) 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.
Figure 5.10Blocking Probability for Different Packet Switched Traffic Loads (L
p
, X-axis)
Reference:
T. Halonen, J. Romero, J. Melero; GSM, GPRS and EDGE performance – Evolution towards 3G/UMTS, John Wiley
and Sons Ltd.
P X n = ( )
L
P
N
P
· ( )
n
N
P
! N
P
i N
P
– ( )
·
---------------------------------
L
P
N
P
· ( )
i
i!
------------------------
i 0 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
i N
P
– ( )
·
---------------------------------
i N
P
1 + =
·
¿
+
----------------------------------------------------------------------------------------------- =
BP P X n = ( )
n M N · 1 + =
·
¿
L
P
N
P
· ( )
n
N
P
! N
P
1 N
P
– ( )
·
-----------------------------------
n M N
P
· 1 + =
·
¿
L
P
N
P
· ( )
i
i!
------------------------
i 0 =
N
P
¿
L
P
N
P
· ( )
i
N
P
! N
P
1 N
P
– ( )
·
-----------------------------------
i N
P
1 + =
·
¿
+
------------------------------------------------------------------------------------------------ = =
BP
L
P
N
P
· ( )
M N
P
·
N
P
! N
P
M N
P
· N
P
– ( )
·
---------------------------------------------
L
P
1 L
P

--------------- ·
L
P
N
P
· ( )
i
i!
------------------------
i 0 =
N
P
¿
L
P
N
P
· ( )
N
P
N
P
!
-----------------------------
L
P
1 L
P

--------------- · +
------------------------------------------------------------------------------------------- =
172 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.6.2 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.
5.6.2.1 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:
3. The throughput reduction factor is greater than the minimum throughput reduction factor,
4. Delay is less than the maximum permissible delay defined in the service properties, and
5. 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.
On the whole, following are the inputs and outputs of the network dimensioning process:
5.6.2.1.1 Inputs
• Circuit switched traffic demand
• Packet switched traffic demand
• 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.
5.6.2.1.2 Outputs
• 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
5.6.2.2 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.
5.6.2.2.1 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) either user-defined or 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, O
Target
, 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 TD
C
, Atoll determines the required number of
timeslots TS
req. C
for each subcell using formulas described below. In fact, Atoll searches for TS
req. C
value until the
defined grade of service is reached.
Figure 5.11Network Dimensioning Process
© Forsk 2010 AT283_TRG_E2 173
Chapter 5: GSM GPRS EDGE Networks
For Erlang B, we have:
For Erlang C, we have:
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 TS
req. 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 TS
eff.
that can carry this traffic mix is calculated by:
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.
5.6.2.2.2 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
carrying 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:
and
5.6.2.2.3 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, O
eff.
.
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 O
eff.
and the circuit switched traffic demand of TD
C
, the served circuit switched traffic ST
C
is computed as:
5.6.2.2.4 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 or is user-
defined in the subcells table. 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
configuration), Atoll calculates the average traffic demand in packet switched timeslots by:
GoS
TD
C
( )
TS
reqC
TS
reqC
( )!
--------------------------------
TD
C
( )
k
k!
-------------------
k 0 =
TS
reqC
¿
----------------------------------- =
GoS
TD
C
( )
TS
reqC
TD
C
( )
TS
reqC
TS
reqC
( )! 1
TD
C
TS
reqC
------------------- –
\ .
| |
TD
C
( )
k
k!
-------------------
k 0 =
TS
reqC
1 –
¿
· · +
------------------------------------------------------------------------------------------------------------------------------------------------------ =
TS
eff
TS
reqFR
1
HR
2
--------- –
\ .
| |
× =
TS
P
TS
S
TS
P dedi cated ,
+ = TS
C
TS
S
TS
C dedi cated ,
+ =
ST
C
TD
C
1 O
eff
– ( ) · =
174 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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,
The packet switched traffic load is calculated by the formula:
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. N
P
. 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, N
P
is calculated at this stage as:
Where, TS
Terminal
is the number of timeslots that a terminal will use in packet switched calls.
The number timeslots that a terminal can use in packet switched calls is the product of the number of available DL timeslots
for packet-switched services (on a frame) and the number of simultaneous carriers (in case of EDGE evolution).
The number of timeslots that a terminal will use in packet switched calls 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 (see above).
and
Here, the min(X,Y) function yields the lower value among X and Y as result.
Now, knowing the packet switched traffic load, L
P
, and the equivalent number of available timeslots, N
P
, 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
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, L
P
, and the equivalent number of available timeslots, N
P
. Then it recomputes the KPIs with these
new values of L
P
and N
P
. 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 user’s 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. If dimensioning is not based on a traffic analysis, the minimum throughput reduction factor is a
user-defined parameter.
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 configuration properties for each coding scheme, the total number of downlink timeslots defined in the properties
of the mobile terminal (See defintion above) 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 configuration.
The throughput per timeslot per pixel TP
TS, 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,
TD
P
Ti mesl ots
Traffic demand per pixel (kbps)
Throughput per pixel (kbps)
----------------------------------------------------------------------------------
pi xel
¿
=
TS
P
TS
S
TS
P dedi cated ,
+ =
TS
C
TS
S
TS
C dedi cated ,
+ =
L
P
ST
C
TS
C dedi cated ,
– TD
P
Ti mesl ot s
+ ( )
TS
P
----------------------------------------------------------------------------------------------- =
N
P
TS
P
TS
Termi nal
---------------------------- =
TS
Termi nal
mi n TS
Max Servi ce ,
TS
Max Termi nal Type ,
, ( ) =
TS
Max Termi nal Type ,
TS
DL Termi nal Type ,
Carri ers
DL Termi nal Type ,
× =
TS
Max Termi nal Type ,
© Forsk 2010 AT283_TRG_E2 175
Chapter 5: GSM GPRS EDGE Networks
Or
and
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:
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:
So for a reliability level of 90%, the corresponding RF
min
will be the one provided at least 90% of the pixels covered, i.e.
945 pixels. The corresponding value of the resulting RF
min
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 RF
min
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 RF
min
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.
5.6.2.2.5 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
RF
min
Number of pixels
0.3 189
0.36 57
0.5 20
0.6 200
0.72 473
0.9 23
0.98 87
Figure 5.12Minimum Throughput Reduction Factor
TP
TS Pi xel ,
f C ( ) =
TP
TS Pi xel ,
f C ( ) = TP
TS Pi xel ,
f
C
i
----
\ .
| |
=
RF
mi n Pi xel ,
TP
user mi n ,
TP
TS Pi xel ,
TS
Termi nal
×
-------------------------------------------------------------- =
176 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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
5.6.2.2.6 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:
Where,
• ST
C
is the served circuit switched traffic
• ST
P
is the served packet switched traffic
• TS
C, dedicated
is the number of dedicated circuit switched timeslots
• TS
P, dedicated
is the number of dedicated packet switched timeslots
• TS
S
is the number of shared timeslots
5.7 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,
• TS
C, dedicated
is the number of timeslots dedicated to the circuit switched traffic,
• TS
P, dedicated
is the number of timeslots dedicated to the packet switched traffic,
• TS
S
is the number of shared timeslots for a transmitter,
Then, the number of timeslots available for the circuit switched traffic, TS
C
, is defined as:
And the number of timeslots available for the packet switched traffic, TS
P
, is given by:
5.7.1 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, TD
C
, and the number of timeslots available for
circuit switched traffic, TS
C
, the blocking probability can be easily computed using the Erlang formulas or tables.
5.7.1.1 Erlang B
Under the current conditions of circuit switched traffic demand, TD
C
, and the number of timeslots available for the circuit
switched traffic, TS
C
, the percentage of blocked circuit switched traffic can be computed through:
In a network dimensioning based on Erlang B model, the circuit switched traffic overflow rate, O
C
, is the same as the
percentage of traffic blocked by the subcell calculated above.
5.7.1.2 Erlang C
Similarly, under the current conditions of circuit switched traffic demand, TD
C
, and the number of timeslots available for
the circuit switched traffic, TS
C
, the percentage of delayed circuit switched traffic can be computed through:
L
ST
C
ST
P
+
TS
C dedi cated ,
TS
P dedi cated ,
TS
S
+ +
------------------------------------------------------------------------------------------------ =
TS
C
TS
S
TS
C dedi cated ,
+ =
TS
P
TS
S
TS
P dedi cated ,
+ =
% of blocked traffic
TD
C
( )
TS
C
TS
C
( )!
--------------------------
TD
C
( )
k
k!
-------------------
k 0 =
TS
C
¿
------------------------------ =
© Forsk 2010 AT283_TRG_E2 177
Chapter 5: GSM GPRS EDGE Networks
If the circuit switched traffic demand, TD
C
, 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, O
C
, will be 0 if the circuit switched traffic demand, TD
C
, is less than the number
of timeslots available for the circuit switched traffic, TS
C
.
If, on the other hand, the circuit switched traffic demand, TD
C
, is higher than the number of timeslots available to carry the
circuit switched traffic, TS
C
, then there will be a certain percentage of circuit switched traffic that will overflow from the
subcell. This circuit switched traffic overflow rate, O
C
, is calculated as:
5.7.1.3 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, O
C
. The
served circuit switched traffic, ST
C
, is calculated as:
5.7.2 Packet Switched Traffic
Identifying the total traffic demand, TD
T
, (circuit switched traffic demand + packet switched traffic demand) as:
The following two cases can be considered.
5.7.2.1 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:
5.7.2.1.1 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.
5.7.2.1.2 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, O
P
. This overflow
rate is calculated as show below:
5.7.2.1.3 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.
5.7.2.1.4 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).
5.7.2.1.5 Blocking Probability
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%.
5.7.2.1.6 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, ST
P
, is calculated on the same principle as the served circuit switched traffic:
% of traffic delayed
TD
C
( )
TS
C
TD
C
( )
TS
C
TS
C
( )! 1
TD
C
TS
C
----------- –
\ .
| |
TD
C
( )
k
k!
-------------------
k 0 =
TS
C
1 –
¿
· · +
-------------------------------------------------------------------------------------------------------------------------- =
O
C
TD
C
TS
C

TD
C
----------------------------- =
ST
C
TD
C
1 O
C
– ( ) · =
TD
T
TD
C
TD
P
+ =
O
P
1
TS
C dedi cated ,
TS
P dedi cated ,
TS
S
+ + ( ) ST
C
– { }
TD
P
----------------------------------------------------------------------------------------------------------------------------- – 100 × =
ST
P
TD
P
1 O
P
– ( ) · =
178 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.7.2.2 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, L
P
, calculated as under:
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, N
P
, which is calculated in the same manner as in the
dimensioning process as well:
These parameters calculated, now Atoll can compute the required KPIs through their respective quality curves.
5.7.2.2.1 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:
5.7.2.2.2 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.
5.7.2.2.3 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, L
P
, and number of equivalent timeslots, N
P
.
5.7.2.2.4 Delay
The resulting delay the subcell is calculated through the delay quality curve for given packet switched traffic load, L
P
, and
number of equivalent timeslots, N
P
.
5.7.2.2.5 Blocking Probability
The resulting blocking probability for a normally loaded subcell is calculated through the blocking probability quality curve
for given packet switched traffic load, L
P
, and number of equivalent timeslots, N
P
.
5.7.2.2.6 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:
5.8 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.
L
P
ST
C
TS
C dedi cated ,
– TD
P
Ti mesl ot s
+ ( )
TS
P
----------------------------------------------------------------------------------------------- =
N
P
TS
P
TS
Termi nal
---------------------------- =
Traffic Load
TD
T
TS
C dedi cated ,
TS
P dedi cated ,
TS
S
+ +
------------------------------------------------------------------------------------------------ =
ST
P
TD
P
=
Note:
• If no focus zone exists in the .atl document, Atoll takes into account the computation zone.
© Forsk 2010 AT283_TRG_E2 179
Chapter 5: GSM GPRS EDGE Networks
5.8.1 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:
6. 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.
7. 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).
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.
8. There must be an overlapping zone ( ) with a given cell edge coverage probability where:
• S
A
is the area where the received signal level from the transmitter A is greater than a minimum signal level. S
A
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”)
• S
B
is the coverage area where the candidate transmitter B is the best server.
Notes:
• Adjacence criterion: Geographically adjacent transmitters are determined on the basis of
their Best Server coverages in 2G (GSM GPRS EDGE) 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.
• 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.
S
A
S
B
·
180 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Atoll calculates either the percentage of covered area ( ) if the option “Take into account Covered Area” is
selected, or the percentage of traffic covered on the overlapping area 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).
9. 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.
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%.
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
Figure 5.13Overlapping Zones
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
S
A
S
B
·
S
A
---------------------- 100 ×
S
A
S
B
·
© Forsk 2010 AT283_TRG_E2 181
Chapter 5: GSM GPRS EDGE Networks
The (IF) function is user-definable using the Min importance and Max importance fields.
The (IF) function is evaluated as follows:
Where Delta(x) = Max(x) - Min(x)
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 (km
2
) (or the
traffic covered on the area in Erlangs), the percentage of area meeting the adjacency conditions and the corresponding
surface area (km
2
). Finally, if cells have previous allocations in the list, neighbours are marked as existing.
Factor
Min
importance
Default value
Max
importance
Default value
Overlapping factor (O) Min(O) 1% Max(O) 30%
Adjacency factor (A) Min(A) 30% Max(A) 60%
Co-site factor (C) Min(C) 60% Max(C) 100%
Neighbourhood cause
(IF) function
(IF) function with default Min
and Max default values
Co-site Adjacent
no 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)}
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.
Notes:
• No coverage prediction 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.
• 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:
1
st
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.
2
nd
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.
• 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.
182 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.8.2 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 179.
5.9 AFP Appendices
5.9.1 The AFP Cost Function
The notations listed hereafter are used to describe the cost function:
• TRG: Group of TRXs
• TRGs: Set of all the TRGs
• ·: If and only if
• : Size of any group g
• ARFCN: Set of all the frequencies
• : Set of all the subsets of frequencies
• : The largest integer
• : Number of times a group is assigned to TRG
i
in the assignment A
For example:
- When i is NH, · g is a single member group containing one of the frequencies assigned at TRG
i
.
If |g| is not 1 or if g does not contain a frequency assigned at i, then .
- When i is BBH, can be either 0 or equal to the number of TRXs in TRG
i
.
= Number of TRXs in TRG
i
· g is the set of frequencies assigned to TRXs of TRG
i
. (|g| = number of
TRXs in TRG
i
).
When we talk about "TRXs of i using g", and in the case of BBH, then there are |g| such virtual TRXs, each
using the entire group g and having a virtual MAIO [0, |g| - 1].
- When i is SFH, must be less than or equal to the umber of TRXs in TRG
i
. · g is the set of
frequencies assigned to n TRXs of TRG
i
.
We assume all the groups assigned to TRG
i
to have the same length.
• TS
i
: Number of timeslots available for each TRX in TRG
i
• TL
i
: Traffic load of TRG
i
(calculated or user-defined)
of a single TRX in TRG
i
divided by TS
i
• TSU
i
: Downlink timeslot use ratio (due to DTX) at TRG
i
• CF
i
: Cost factor of TRG
i
(AFP Weight)
• QMIN
i
: Minimum required quality (in C/I) at TRG
i
• PMAX
i
: Percentage permitted to have quality lower than QMIN
i
at TRG
i
• REQ
i
: Required number of TRXs at TRG
i
A communication uses the group g in TRG
i
if its mobile allocation is g. The probability to be interfered is denoted by
(i’ is the TRX index). Different TRX indexes may have different MAIOs. is a function of the whole
frequency assignment. The precise definition of the term “to be interfered” is provided afterwards. The probability penalty
due to violating a separation constraint is . It is a function of the whole frequency assignment as well.
The term “Atom” will be used in the following context:
For two TRGs, i and k,
i and k are synchronised, have the same HSN, the same MAL length and the same hopping mode.
• 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.
g
2
ARFCN
x x s
A
i g ,
g 2
ARFCN
e
A
i g ,
1 =
A
i g ,
0 =
A
i g ,
A
i g ,
A
i g ,
A
i g ,
n =
TL
i
#Erlangs =
P
i i' g , ,
A ( ) P
i i' g , ,
A ( )
P
i i' g , ,
A ( )
ATOM i ( ) ATOM k ( ) ÷
© Forsk 2010 AT283_TRG_E2 183
Chapter 5: GSM GPRS EDGE Networks
NH TRGs or BBH TRGs are always in separate atoms. If two TRGs interfere but are not in the same atom, these can be
taken as unsynchronised. The quality of unsynchronised TRGs is a function of all possible frequency combinations. For
synchronised TRGs, pairs of frequencies emitted at the same time are known.
5.9.1.1 Cost Function
The Atoll AFP cost function is a TRX based cost and not an interference matrix entry based cost. It counts the impaired
traffic of the network TRXs in weighted Erlangs.
The cost function is reported to the user during the AFP progress with the help of its 5 components: , ,
, and .
= + + + +
where,
represents the missing TRX cost component
represents the separation component
represents the additional cost component (interference, cost of changing a TRX)
represents the corrupted TRX cost component
represents the out-of-domain frequency assignment cost component
In the above equations,
• i’ is the TRX index belonging to .
• is the number of missing TRXs for the subcell i.
• is the cost value for a missing TRX. This value can vary between 0 and 10. The default cost value is set to 1
and can be modified in the AFP module properties dialog.
• is the number of corrupted TRXs for the subcell i.
• is the cost value of a corrupted TRX. This value can vary between 0 and 10. The default cost value is set to 10
and can be modified in the AFP module properties dialog.
• is the number of TRXs, for the subcell i, having out-of-domain frequencies assigned.
• is the cost value of a TRX with out-of-domain frequencies assigned. This value can vary between 0 and 1. The
default cost value is set to 0.5 and can be modified in the AFP module properties dialog.
And, as mentioned earlier, a virtual TRX is considered in case of BBH.
If i’ is valid, the algorithm evaluates the cost of a valid TRX. This cost has two components, and
.
+ +
mi s
+
sep
+
comp
+
corr
+
dom
+ +
mi s
+
sep
+
comp
+
corr
+
dom
+
mi s
+
sep
+
comp
+
corr
+
dom
+
mi s
MIS_TRX
i
ì × ( ) TL
i
CF
i
TS
i
× × ×
i TRGs e
¿
=
+
corr
CORR_TRX
i
O × ( ) TL
i
CF
i
TS
i
× × ×
i TRGs e
¿
=
+
dom
DOM_TRX
i
e × ( ) TL
i
CF
i
TS
i
× × ×
i TRGs e
¿
=
+
sep
o'
i i' g , ,
A ( )
g 2
ARFCN
e
i' TRXs of i using g e
¿
\ .
|
|
|
|
|
|
| |
TL
i
CF
i
TS
i
× × ×
i TRGs e
¿
=
+
comp
o''
i i' g , ,
A ( )
g 2
ARFCN
e
i' TRXs of i using g e
¿
\ .
|
|
|
|
|
|
| |
TL
i
CF
i
TS
i
× × ×
i TRGs e
¿
=
0 1 ... A
i g ,
1 – , , , { }
MIS_TRX
i
MIS_TRX
i
MAX 0 REQ
i
A
i g ,
g 2
ARFCN
e
¿
– , =
ì
CORR_TRX
i
O
DOM_TRX
i
e
o'
i i' g , ,
A ( )
o''
i i' g , ,
A ( )
184 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• is the separation violation probability penalty.
• is complementary probability penalty due to interference and the cost of modifying a TRX.
If the option “Take into account the cost of all the TRXs” available in the AFP module properties dialog is selected,
then,
and
Or if the option “Do not include the cost of TRXs having reached their quality target” available in the AFP module
properties dialog is selected, the algorithm compares with the quality target specified for i,
:
If ,
Then and .
Otherwise,
Both and will be equal 0.
is the same as (separation violation probability penalty) and the same as
(complementary probability penalty due to interference and the cost of modifying a TRX) in most cases. These are
explained in detail in the next sections.
5.9.1.2 Cost Components
Separation violation and interference cost components are described hereafter. Parameters considered in the cost
function components can be fully controlled by the user. Some of these parameters are part of the general data model
(quality requirements, percentage of interference allowed per subcell), while others (such as separation costs and diversity
gains) can be managed through the properties dialog of the Atoll AFP module.
5.9.1.2.1 Separation Violation Cost Component
The separation violation cost component is evaluated for each TRX. Estimation is based on costs specified for the required
separations.
Let denote the required separation constraint between TRG
i
and TRG
k
. Let denote the user
defined separation penalty for a required separation “s” and actual separation “z”. is used instead of
as abbreviation.
is considered to be the effect of a separation violation on the th TRX of TRG
i
assigned the group g, caused by
the th TRX of TRG
k
assigned the group .
denotes the overall weight of the separation violation cost component. This value can be between 0 and 1, set to 1 by
default. It can be modified in the AFP module properties dialog.
represents the weight of the specific separation constraint between i and k. This specific weight depends on the type
of separation violation and follows the following priority rule:
1. Exceptional pairs
2. Co-transmitters
3. Co-site
4. Neighbours
For example, if a pair of subcells are co-site and neighbours at the same time, they will be considered as co-site because
higher priority. Hence, of these subcells will be the weight of co-site relations. If only a neighbour relation exists
between two subcells, then will be further weighted by the neighbour relation importance. The value of remains
between 0 and 1. The default weights of each type of separation are available in the Separation cost tab.
If
Then , which is same for all values of k.
If
o'
i i' g , ,
A ( )
o''
i i' g , ,
A ( )
o'
i i' g , ,
A ( ) P'
i i' g , ,
A ( ) = o''
i i' g , ,
A ( ) P''
i i' g , ,
A ( ) =
P'
i i' g , ,
A ( ) P''
i i' g , ,
A ( ) +
P
MAX
P'
i i' g , ,
A ( ) P''
i i' g , ,
A ( ) + P
MAX
>
o'
i i' g , ,
A ( ) P'
i i' g , ,
A ( ) = o''
i i' g , ,
A ( ) P''
i i' g , ,
A ( ) =
o'
i i' g , ,
A ( ) o''
i i' g , ,
A ( )
P'
i i' g , ,
A ( ) o'
i i' g , ,
A ( ) P''
i i' g , ,
A ( ) o''
i i' g , ,
A ( )
Note:
• The AFP module properties dialog takes probability percentages as inputs while this
document deals in probability values.
SEP_CONSTR
i k ,
Cost
s z ,
SEP
i k v , ,
Cost
SEP_CONSTR
i k ,
z ,
ç
i i' kgg' k'
i'
k' g'
¸
¸i k
¸i k
¸i k ¸i k
ATOM i ( ) ATOM k ( ) =
ç
i i' kgg' k'
¸ ¸i k
SEP
i k f f' – , ,
f g e
f' g' e
¿
g g' ×
---------------------------------------------- × × =
ATOM i ( ) ATOM k ( ) =
© Forsk 2010 AT283_TRG_E2 185
Chapter 5: GSM GPRS EDGE Networks
Then
In the above equations, is the number of frames in the MAL g. .
Let denote the instantaneous frame number from 0 to .
While modulo and is the frequency in g,
And modulo and is the frequency in g’.
In addition, frequencies belonging to a MAL with a low fractional load, and breaking a separation constraint, should not be
weighted equally as in a non-hopping separation breaking case. Therefore, the cost is weighted by an interferer diversity
gain.
The separation gain, denoted by is basically a function of the MAL length (and, of course, of the
hopping mode). With frequency hopping, the effects of DTX and traffic load become more significant (due to the
consideration of the average case instead of the worst case). For this reason, it is possible to consider these effects in
through the relevant option available in the Advanced tab of the AFP module properties dialog.
Without this option, the is:
is the user defined interferer diversity gain (dB) for a given MAL length. It is used in definition as well.
On the other hand, if this option is selected, the becomes,
Where ,
And
More than one separation violations may exist for a TRX. Many “small” and have to be combined to form
one cost element, the . This is done through iterating over all violating assignments and by summing up an
equivalent to the probability of not being violated while considering each separation violation as an independent probability
event. This sum is naturally limited to 100% of the TRX traffic, and is given by,
In the above formula, if , then , so that interference with itself is not taken into account.
5.9.1.2.2 Interference Cost Component
The interference cost component is evaluated for each TRX. Its estimation is based on interference histograms calculated
for pairs of subcells. In addition, it takes into account frequency and interferer diversity gains and models frequency
hopping and gain due to DTX.
When estimating , the following problems are encountered:
Note:
• Since , we shortly denote the two as .
ç
i i' kgg' k'
¸ ¸i k
SEP
i k g
u
g'
t
– , ,
f_n 0 1 ... F_N 1 – , , , { } e
¿
F_N
------------------------------------------------------------------------------------- × × =
F_N g ( ) F_N g ( ) g =
F_N g ( ) F_N g' ( ) = F_N
f_n F_N
u f_n MAIO
A
i g i' , ,
+ ( ) = F_N g
u
u
th
t f_n MAIO
A
k g' k' , ,
+ ( ) = F_N g'
t
t
th
G
ˆ
i k g g' , , ,
1
10
0.1 SEP_GAIN i k g g' , , , ( ) × ( )
-------------------------------------------------------------------- =
SEP_GAIN i k g g' , , , ( )
SEP_GAIN i k g g' , , , ( )
SEP_GAIN i k g g' , , , ( )
SEP_GAIN i k g g' , , , ( ) I_DIV g ( ) =
I_DIV g ( ) P
i i' g , ,
A ( )
SEP_GAIN i k g g' , , , ( )
SEP_GAIN i k g g' , , , ( ) I_DIV g ( ) 0.5 TSU_GAIN k ( ) × mi n 10 4 2 I_DIV g ( ) + ( )
2 ASYN_GAIN i k g' , , ( ) + ( )
4
----------------------------------------------------------------------- ×
\ .
| |
+ ,
\ .
| |
×
\ .
| |
+ =
TSU_GAIN k ( ) log
10
1
TL
k
TSU
k
×
-------------------------------
\ .
| |
=
ASYN_GAIN i k g' , , ( )
0 if ATOM(i) = ATOM(k)
I_DIV( g' ) Otherwise
=
G
ˆ
i k g g' , , ,
ç'
i i' kgg'
P'
i i' g , ,
A ( )
P'
i i' g , ,
A ( )
1 1 ç
i i' kgg' k'
G
ˆ
i k g g' , , ,
× – ( )
k TRGs e
g' 2
ARFCN
e
k' TRXs of k using g' e
[

\ .
|
|
|
|
|
|
|
| |
=
k i = ( ) k' i' = ( )
Note:
• Interference histograms are described in User Manual (GSM GPRS EGPRS project
management, GSM GPRS EGPRS network optimisation, GSM GPRS EGPRS generic
AFP management). Interference histograms can also be exported to files. For further
description, refer to "Interferences" on page 190.
P''
i i' g , ,
A ( )
186 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• The QMINi C/I quality indicator corresponds to the accumulated interference level of all interferers while the C/I
interference histograms correspond to pair-wise interferences.
• Both QMINi and the histograms correspond to a single frequency. In case of a MAL containing more than one
frequencies, interferences on several different frequencies of a MAL must be combined.
This estimation, presented below, is the simplest possible as it solves the first problem by linear summation and
truncation at the value of 1 and it solves the second problem by averaging and adding the two diversity gains:
• , the frequency diversity gain, and
• , the interferer diversity gain.
Hereafter, denotes the global weight of interference cost component. This value can vary between 0 and 1 and is set
to 0.35 by default, which can be modified in the AFP module properties dialog.
Let be the number of frames in the MAL g. .
Let denote the instantaneous frame number from 0 to .
Let be the j’th MAIO of , where j is one of the TRXs.
The value of is one of
If TRG
k
is NH, then .
If TRG
k
is BBH, then .
As said earlier, in case of BBH, we consider virtual TRXs, the jth TRX has the MAIO j.
Let be the ith frequency in the group g.
Similar to the definition of , is defined as an interference event. is the effect interference on the
th TRX of TRG
i
assigned the group g, caused by the th TRX of TRG
k
assigned the group .
If
Then
Where
If
Then,
Since , these are both represented by .
Where,
,
,
modulo ,
modulo ,
Therefore, we have,
In the above formula, if , then , so that interference with itself is not taken into account.
The sum is limited to 100% of the TRX traffic. is quite similar to . The
only difference is the frequency diversity gain, , added to .
F_DIV g ( )
I_DIV g ( )
o
F_N g ( ) F_N g ( ) g =
f_n F_N
MAIO
A
k g' j , ,
A
k g' ,
0 1 ... A
k g' ,
1 – , , , { }
MAIO
A
k g' j , ,
0 1 ... g' , , , { }
MAIO
A
k g' j , ,
0 =
MAIO
A
k g' j , ,
j =
g'
g
i
ç
i i' kgg' k'
ç'
i i' kgg' k'
ç'
i i' kgg' k'
i' k' g'
ATOM i ( ) ATOM k ( ) =
ç'
i i' kgg' k'
Probabi l i ty
C
I
i k
----- Q_UB
i k f f' , , ,
<
\ .
| |
g g' ×
--------------------------------------------------------------------------------
f g e f' g' e ,
¿
=
Q_UB
i k f f' , , ,
QMIN
i
f f' – ADJ_SUP INTERF_GAIN i k g g' , , , ( ) + × – =
ATOM i ( ) ATOM k ( ) =
F_N g ( ) F_N g' ( ) = F_N
ç
i i' kgg' k'
Probabi l i ty
C
I
i k
----- Q_UB
i k f f' , , ,
<
\ .
| |
F_N
--------------------------------------------------------------------------------
¹ )
¦ ¦
´ `
¦ ¦
¦ ¹
f_n 0 1 ... F_N 1 – , , , { } e
¿
=
f g
u
=
f' g'
t
=
u f_n MAIO
A
i g i' , ,
+ ( ) = F_N
t f_n MAIO
A
k g' k' , ,
+ ( ) = F_N
Q_UB
i k f f' , , ,
QMIN
i
f f' – ADJ_SUP INTERF_GAIN i k g g' , , , ( ) + × – =
P''
i i' g , ,
A ( ) 1 1 P'
i i' g , ,
A ( ) – ( ) 1 ç
i i' kgg' k'
– ( )
k TRGs e
g' 2
ARFCN
e
k' TRXs of k using g' e
[
×
¹ )
¦ ¦
¦ ¦
¦ ¦
´ `
¦ ¦
¦ ¦
¦ ¦
¦ ¹
– P'
i i' g , ,
A ( ) – =
i k = ( ) k' i' = ( )
INTERF_GAIN i k g g' , , , ( ) SEP_GAIN i k g g' , , , ( )
F_DIV g ( ) SEP_GAIN i k g g' , , , ( )
© Forsk 2010 AT283_TRG_E2 187
Chapter 5: GSM GPRS EDGE Networks
5.9.1.2.3 I_DIV, F_DIV and Other Advanced Cost Parameters
When combining interference effects (or separation violation effects) on different frequencies belonging to a MAL, the
following considerations should be taken into account:
1. Non-linearity of Frame Error Rate (FER) with respect to average C/I conditions and MAL length.
2. Interference Diversity Gain. This factor represents that the effect of average negative effects over user geographic
location are directly proportional to the MAL length.
3. Frequency Diversity Gain. This factor models the gain due to diversity of multi-path effects and should be applied
to the interference cost component only.
4. The fact that long MALs with synthesized hopping permit discarding the worst case estimation and include a gain
due to DTX and low traffic load at the interferer end.
The Advanced properties tab shown in the figure below facilitates modelling these effects.
The Interference Diversity Gain table lists the values of I_DIV provided as a functions of MAL length. This gain is applied
to the interference cost component and to the separation constraint violation cost component. Therefore, it provides a
means to model the non-linear FER effects and interference diversity both. The default values in this table correspond to
the curve . This equation generates values somewhat lower than empirical best-found values (this is
because we prefer a slightly pessimistic cost function to be on the safe side).
The other table contains the F_DIV values, which are the same as the I_DIV values by default.
5.9.2 The AFP Blocked Traffic Cost
This section provides additional information on the AFP cost components used for the optimisation of the number of TRXs.
This optimisation is performed for each traffic pool in the network. In most cases, the traffic pool is equivalent to a
transmitter and corresponds to the BCCH and TCH subcells. In more complex cases, a traffic pool may include additional
subcells, and more than one traffic pools may exist per transmitter.
The cost component described below, and the recalculation of traffic loads, is only used when the AFP performs the
oprimisation of the number of TRXs.
The notations listed hereafter are used for the description.
• {BCCH, TCH(1), TCH(2), …, TCH(n)}: Subcells of a traffic pool.
For concentric cells, at least two traffic pools exist per transmitter. The BCCH subcell may not always be part of
the pool’s TRX types.
• {d(0), d(1), d(2), …, d(n)}: Number of required TRXs of
each TRX type in the pool
• {ts(0), ts(1), ts(2), …, ts(n)}: Numbers of traffic timeslots
Figure 5.14The Advanced tab of the AFP module Properties dialogue
y 2 log
10
x ( ) × =
188 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• {L(0), L(1), L(2), …, L(n)}: Traffic loads
• {CF(0), CF(1), CF(2), …, CF(n)}: AFP cost factors
• CS (Erlangs): Overall circuit-switched traffic
demand of the traffic pool (Subcells table or traffic analysis results)
• PS (Data Timeslots): Overall packet-switched traffic
demand of the traffic pool (Subcells table or traffic analysis results)
If CS or PS is less than 1, its value is set to 1 in order to avoid working with transmitters carrying no traffic.
• {nb(0), nb(1), nb(2), …, nb(n)}: Number of TRXs in the
frequency plan
• {HR(0), HR(1), HR(2), …, HR(n)}: TCH HR use ratios
5.9.2.1 Calculation of New Traffic Loads Including Blocked Traffic Loads
During the optimisation of the number of TRXs, traffic loads are calculated in order to determine the blocked traffic loads
. The blocked traffic load is then multiplied by the AFP cost weight and the number of timeslots to calculate the
blocked traffic cost.
Without the optimisation of the number of required TRXs, the network’s weighted Erlangs are calculated as follows:
With the optimisation of the number of TRXs, the network’s weighted Erlangs are calculated as follows:
and represent the load estimation and the blocked load estimation of the AFP. They are calculated at
traffic pool level for the vector {nb(0), nb(1), nb(2), …, nb(n)} as follows:
Where
is determined from the above equation once is known. is obtained from the Erlang B equation
applied to the traffic pool demand and the total number of timeslots (TTS):
The Max() function above gives 1 timeslot when there is no TRX.
The above equations give the number of served circuit-switched timeslots (SCS):
The number of served packet-switched timeslots (SPS) is obtained as follows:
is given by:
is given by:
BL nb ( )
WE d i ( ) ts i ( ) × L i ( ) × CF i ( ) ×
i 0 =
n
¿
=
WE nb i ( ) ts i ( ) × BL nb ( ) L nb ( ) + { } × CF i ( ) ×
i 0 =
n
¿
=
BL nb ( ) L nb ( )
BL nb ( ) L nb ( ) +
PS CS 1
HR
2
--------------- –
\ .
| |
×
¹ )
´ `
¦ ¹
+
Max 1 nb i ( ) ts i ( ) ×
i 0 =
n
¿
,
\ .
|
|
| |
------------------------------------------------------------------- =
HR Max
i 0 =
n
HR i ( ) ( ) =
BL nb ( ) L nb ( ) L nb ( )
TTS Max 1
nb i ( ) ts i ( ) ×
1
HR
2
--------------- –
\ .
| |
-------------------------------
i 0 =
n
¿
,
\ .
|
|
|
| |
=
P
Bl ocki ng
Erl angB CS TTS , ( ) =
SCS 1
HR
2
--------------- –
\ .
| |
CS 1 P
Bl ocki ng
– ( ) × × =
SPS Mi n PS Max 1 nb i ( ) ts i ( ) ×
i 0 =
n
¿
,
\ .
|
|
| |
SCS – ,
¹ )
¦ ¦
´ `
¦ ¦
¦ ¹
=
L nb ( )
L nb ( )
SCS SPS +
Max 1 nb i ( ) ts i ( ) ×
i 0 =
n
¿
,
\ .
|
|
| |
------------------------------------------------------------------ =
BL nb ( )
© Forsk 2010 AT283_TRG_E2 189
Chapter 5: GSM GPRS EDGE Networks
Once and are known, replaces TLi in the cost function (See "The AFP Cost Function" on
page 182), and is used to generate a new cost component, the blocked Erlangs of the pool:
5.9.2.2 Recalculation of CS and PS From Traffic Loads
In earlier versions, the detailed traffic demand information is not available. In order to guide the AFP to generate it from
the loads, the following two equations with three variables must be solved. The equations are solvable due to the
monotone nature of the Erlang B function.
Inputs for a given traffic pool:
• {d(0), d(1), d(2), …, d(n)}: Number of required TRXs of each TRX type in the pool
• L: Traffic load
• :
• MB: Maximum blocking rate (between 0 and 1).
The ratio of packet-switched demand is given by:
Here, we assume that a traffic load of 1 is generated by a demand of (1+MB)*TTS’ which generates a blocking rate of MB.
In other words, the ratio is calculated so that the worst case blocking rate is BM, giving a load of 1.
The following equations are solved to find PS’, CS’, and R’, which are calculated for a traffic load of 1.
When the traffic load of a pool is not 1, PS is different from PS’ and CS is different from CS’. Here, however, we assume
that R’ = R. This assumption implies that R is more or less the same as MB for big traffic pools and considerably larger
than MB for smaller pools.
The following equations are solved to find PS, CS, and R, which are calculated for the actual traffic loads.
The above five equations are solved to get the values of the five variables PS, PC, , SCS, SPS, and calculate
the cost.
BL nb ( )
PS CS 1
HR
2
--------------- –
\ .
| |
× +
Max 1 nb i ( ) ts i ( ) ×
i 0 =
n
¿
,
\ .
|
|
| |
------------------------------------------------------------------ L nb ( ) – =
L nb ( ) BL nb ( ) L nb ( )
BL nb ( )
nb i ( ) ts i ( ) × BL nb ( ) × CF i ( ) ×
i 0 =
n
¿
TTS' TTS' Max 1
d i ( ) ts i ( ) ×
1
HR
2
--------------- –
\ .
| |
------------------------------
i 0 =
n
¿
,
\ .
|
|
|
| |
=
R
PS
PS CS 1
HR
2
--------------- –
\ .
| |
× +
------------------------------------------------------------ =
MB Erl angB CS' TTS' , ( ) =
R'
PS'
PS' CS' 1
HR
2
--------------- –
\ .
| |
× +
-------------------------------------------------------------- =
1 MB + ( ) TTS' ×
PS'
1
HR
2
--------------- –
\ .
| |
------------------------------ CS' + =
R
PS
PS CS 1
HR
2
--------------- –
\ .
| |
× +
------------------------------------------------------------ =
P
Bl ocki ng
Erl angB CS TTS' , ( ) =
SCS 1
HR
2
--------------- –
\ .
| |
CS 1 P
Bl ocki ng
– ( ) × × =
SPS Mi n PS Max 1 d i ( ) ts i ( ) ×
i 0 =
n
¿
,
\ .
|
|
| |
SCS – ,
¹ )
¦ ¦
´ `
¦ ¦
¦ ¹
=
SCS SPS + d i ( ) ts i ( ) × L i ( ) ×
i 0 =
n
¿
=
P
Bl ocki ng
190 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
5.9.2.3 Testing the Blocked Cost Using Traffic Analysis
As long as the conditions below hold truw, the blocked cost calculation in the AFP and the effective overflow calculation
in the KPI calculation and dimensioning use the same algorithm. The conditions are:
• The AFP cost factors are 1,
• The HR ratios are the same within the subcells of a traffic pool,
• The dimensioning model is based on Erlang B,
• The timeslot configurations are the default ones,
• There exists at least one TRX in the traffic pool (and at least one Erlang of traffic),
• All transmitters belong to the same HCS Layer.
Effective Overflow rate =
Output: New values for CS and PS.
5.9.3 Interferences
This appendix provides a high-level overview of interferences taken into account by the AFP.
5.9.3.1 Using Interferences
If interferences are to be taken into account by the AFP, they must be calculated or imported beforehand. In order to do
this, the user should previously decide to take interferences into account (enabling the loading of all the potential
interferers). Otherwise, Atoll does not allow performing their computation by disabling the histogram part in the
corresponding dialog.
5.9.3.2 Cumulative Density Function of C/I Levels
For each [interfered subcell, interfering subcell] pair, Atoll calculates a C/I value on each pixel of the interfered subcell
service area (as if the two subcells share the same channel). Then, Atoll integrates these C/I values to determine a C/I
distribution and transforms this distribution function into a cumulative density function in the normal way.
In Atoll, both the IM
co
and IM
adj
are represented by this Cumulative Density function This implies that each query for the
probability to have C/I conditions worse than X dB requires a single memory access: the co-channel interference
probability at X dB. In order to deduce the adjacent interference probability value, Atoll looks up the cumulative density
function at the value corresponding to X - Y dB, Y dB being the adjacency suppression value. The following example may
be helpful in further clarifying this concept:
Example: Let [TX1, BCCH] and [TX2, BCCH] be the interfered and interfering subcells respectively. The service
areas for both have been defined by Best Server with 0 dB margin. The interference probability is stated in
percentage of interfered area.
In this case, we observe that the probability for C/I (BCCH of TX2 effecting the BCCH of TX1) being greater than
0 is 100% (which is normal because TX1 is the Best Server). The probability of having a C/I value at least equal
to 31 dB is 31.1%. For a required C/I level of 12 dB on the BCCH of TX1, the interference probability is 6.5% (as
this requirement is fulfilled with a probability of 93.5%).
1
L nb ( )
L nb ( ) BL nb ( ) +
------------------------------------------ –
Figure 5.15The cumulative density of C/I levels between [TX1, BCCH] and [TX2, BCCH]
Note:
• The subcell power offset does not enter the calculation results in the .clc file. It is added
later by the AFP interface. On the other hand, its influence on the subcell service zone is
taken into account in the .clc file.
© Forsk 2010 AT283_TRG_E2 191
Chapter 5: GSM GPRS EDGE Networks
5.9.3.3 Precise Definition
is defined to be the probability of a communication (call) occupying a timeslot in subcell v (victim) to have
C/I conditions of C_I with respect to a co-channel interference from the BCCH TRX of cell n (neighbour). We assume C_I
values to be discrete and in dB. CDF(Pci) is the cumulative density function of Pci:

5.9.3.4 Precise Interference Distribution Strategy
Why does Atoll calculate and maintain precise interference distributions, while the most common solution (used by most
other tools) is rather to compress the information into two values: the co-channel and adjacent-channel interference
probabilities?
The reason is simply that it,
• improves the AFP result,
• introduces very little (or no) overhead, and
• creates more generic interference information.
5.9.3.4.1 Direct Availability of Precise Interference Distribution to the AFP
In the presence of frequency hopping, and when one or more frequencies are common (or adjacent) in two interfering MAL
sequences, the hopping gain depends on following factors:
• the MAL length,
• the traffic load on the interferer TRX,
• DTX level, and
• the number of common (and adjacent) frequencies in the two MALs.
All these factors cannot be pre-calculated since it is the AFP that determines the MAL length and the MAL frequencies.
5.9.3.4.2 Efficient Calculation and Storage of Interference Distribution
In the innermost loop of the calculation process Atoll increments a counter each time a C/I level has a certain value. In
the case of a two-entry IM, there are only two counters for each [interfered, interferer] pair. In the case of precise
distribution information, there are about 40 counters per pair. In both cases, the number of operations is the same: one
increment of an integer value. Once Atoll finishes the counting for an [interfered, interferer] pair, it compresses the
information from the counters to a Cumulative Density Function (CDF) representation. In this way, access to interference
probability at a certain level is instantaneous. Thus, the only overheads are the read / write times to the files and the
memory occupation at running time. These two overheads are negligible and do not affect the calculations, the heaviest
part of the task.
5.9.3.4.3 Robustness of the IM
By having precise C/I distributions calculated and exported, the user is free to change the following settings without the
need for recalculating their interference distributions:
1. Quality requirements of network elements (required C/I, % Probability Max, …),
2. C/I weighting (the interference levels above and below the C/I target),
3. Separation requirements and/or neighbour relations,
4. Hopping gain values, DTX activities, traffic load levels, HSNs, synchronisation information,
5. Any frequency assignment setting (MAL length directives, frequency domains, assignment strategies, number of
required TRXs, cost function parameters, …), or
6. Remove equipment
By not mixing any of the elements above, the interference information keeps its original probability units and is easier to
check and validate. Therefore, the user spends less time on interference recalculations than in the case of a two-entry
matrix (where “everything” is included).
5.9.3.5 Traffic Load and Interference Information Discrimination
Atoll maintains the traffic load separate from the interference information. The reasons for implementing this strategy are
explained here.
Let us look at the possible alternatives to this strategy:
1. The mixed option: The interference information contains the traffic information as well. In this way, each IM entry
will contain the quantity of traffic interfered if a co-channel / adjacent channel reuse exists.
2. The separated option: The AFP has separate access to traffic load information and to interference probabilities
(As in Atoll).
Knowing the difference between the two alternative solutions explains why the second strategy has been opted for for
Atoll. However, in detail, this has been done because:
• Option 2 is a superset that contains option 1. But option 1, being a subset, does not contain option 2 (i.e. once the
information are mixed they cannot be separated).
Pci v n C_I , , ( )
CDF Pci v n C_I , , ( ) ( ) Pci v n x , , ( )
x C_I >
¿
=
192 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• It does not create any overhead (the size of the additional information is negligible compared to the size of the IM).
• It helps keeping the unit definitions simpler.
• It is facilitates merging IMs with different traffic units.
• The traffic information can be used for weighting the separation violation component.
• The traffic load can be used in deciding whether a TRX can be left uncreated.
For example, if there are too many TRXs at a site and the user wishes that the AFP remove one of them, in order
to be able to not violate site constraints, the AFP must know the traffic loads in order to choose a low load TRX to
be removed.
• The gain introduced by the traffic load of the interferer depends on the hopping mode and the MAL size.
Incorporating this gain in the IM (as a result of the mixed option) means that the IMs become hopping-mode and
MAL-size dependent. This is a bad idea since the AFP should be able to change the MAL. And the user should
be able to change the hopping mode without recalculating the IM. In addition, an IM calculated externally to Atoll,
with a non-hopping BCCH can be used for the hopping TCH.
A third option also exists. Though, this option is so practically useless due to its inefficiency. It consists in mixing IM and
traffic but still keeping the traffic in its isolated form. This is again a bad idea because of the unit definition and the variety
of IM sources. It involves less benefits than the option chosen in Atoll.
Chapter 6
UMTS HSPA Networks
© Forsk 2010 AT283_TRG_E2 195
Chapter 6: UMTS HSPA Networks
6 UMTS HSPA Networks
6.1 General Prediction Studies
6.1.1 Calculation Criteria
Three criteria can be studied in point analysis (Profile tab) and in common coverage studies. Study criteria are detailed in
the table below:
where,
EIRP is the effective isotropic radiated power of the transmitter,
ic is a carrier number,
is the loss on the transmitter-receiver path (path loss) calculated by the propagation model,
is the transmitter antenna attenuation (from antenna patterns),
is the shadowing margin. This parameter is taken into account when the option “Shadowing taken into
account” is selected,
are the indoor losses, taken into account when the option “Indoor coverage” is selected,
are the receiver losses,
is the receiver antenna gain,
is the transmitter antenna gain,
is the transmitter loss ( ). For information on calculating transmitter loss, see "UMTS, CDMA2000,
TD-SCDMA, WiMAX, and LTE Documents" on page 128.
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 ( ), or the highest signal
level received from the selected transmitter on all the carriers.
6.1.2.2 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.
Study criteria Formulas
Signal level ( ) in dBm
Signal level received from a transmitter on a carrier (cell)
Path loss ( ) in dBm
Total losses ( ) in dBm
Notes:
• ( 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 and equal zero.
P
rec
P
rec
i c ( ) EIRP i c ( ) L
path
– M
Shadowi ng model –
– L
Indoor
– G
term
L
term
– + =
L
path
L
path
L
model
L
ant
Tx
+ =
L
total
L
total
L
path
L
Tx
L
term
L
i ndoor
M
Shadowi ng model –
+ + + + ( ) G
Tx
G
term
+ ( ) – =
L
model
L
ant
Tx
M
Shadowi ng model –
L
Indoor
L
term
G
term
G
Tx
L
Tx
L
Tx
L
total DL –
=
EIRP i c ( ) P
pi l ot
i c ( ) G
Tx
L
Tx
– + = P
pi l ot
i c ( )
G
term
L
term
Note:
• For a selected transmitter, it is also possible to study the path loss, , or the total
losses, . Path loss and total losses are the same on any carrier.
P
rec
i c ( )
L
path
L
total
196 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
For each transmitter, Atoll displays either the signal level received on a carrier, ( ), or the highest signal level
received on all the carriers.
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.
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.
Coverage study parameters to be set are:
• The study conditions in order to determine the service area of each TBC transmitter,
• The display settings to select how to colour service areas.
6.1.3.1 Service Area Determination
Atoll uses parameters entered in the Condition tab of the coverage study property dialogue to predetermine areas where
it will display coverage.
We can distinguish three cases:
6.1.3.1.1 All Servers
The service area of Txi corresponds to the bins where:
6.1.3.1.2 Best Signal Level and a Margin
The service area of Txi corresponds to the bins where:
And
M is the specified margin (dB).
Best function: considers the highest value.
6.1.3.1.3 Second Best Signal Level and a Margin
The service area of Txi corresponds to the bins where:
And
M is the specified margin (dB).
2
nd
Best function: considers the second highest value.
Note:
• For a selected transmitter, it is also possible to study the path loss, , or the total
losses, . Path loss and total losses are the same on any carrier.
• 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.
P
rec
i c ( )
L
path
L
total
Mi ni mumThreshol d P
rec
Txi
i c ( ) or L
total
Txi
or L
path
Txi
( ) s Maxi mumThreshol d <
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 2
nd
best servers.
Mi ni mumThreshol d P
rec
Txi
i c ( ) or L
total
Txi
or L
path
Txi
( ) s Maxi mumThreshol d <
P
rec
Txi
i c ( )
Best
j i =
P
rec
Txj
i c ( ) ( ) M – >
Mi ni mumThreshol d P
rec
Txi
i c ( ) or L
total
Txi
or L
path
Txi
( ) s Maxi mumThreshol d <
P
rec
Txi
i c ( )
2
nd
Best
j i =
P
rec
Txj
i c ( ) ( ) M – >
© Forsk 2010 AT283_TRG_E2 197
Chapter 6: UMTS HSPA Networks
6.1.3.2 Coverage Display
6.1.3.2.1 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).
6.1.3.2.2 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, dBµV, dBµV/m)
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). 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, dBµV, dBµV/m)
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
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 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
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 pixel of each transmitter service area. Where other
service areas overlap the studied one, Atoll determines the best transmitter and evaluates path loss from the best
transmitter. A pixel of a service 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 minimum threshold.
Best Server Total Losses (dB)
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.
Notes:
• If the margin equals 0 dB, Atoll will consider bins where the signal level received from Txi
is the second highest.
• If the margin is set to 2 dB, Atoll will consider bins where the signal level received from Txi
is either the second highest or 2dB lower than the second 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 3
rd
best servers.
>
>
>
>
>
>
198 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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.
Cell Edge Coverage Probability (%)
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.
Best Cell Edge Coverage Probability (%)
On each pixel 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.
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
Clutter parameter None Orthogonality factor
Site equipment parameter None MUD factor
Terminal parameter - HSDPA properties None MUD factor
Frequency band parameter None Carrier number
Cell parameter None
Threshold for macro diversity
specified for a transmitter on a given
carrier ic
Mobility parameter None
Ec/I0 target on downlink for the best
server
Global parameter None
Pilot RSCP threshold for compressed
mode activation
Global parameter None
Ec/I0 threshold for compressed mode
activation
(Reception equipment, R99 bearer, Mobility) parameter None Eb/Nt target on downlink
Global parameter None
Downlink Eb/Nt target increase due to
compressed mode activation
(Reception equipment, R99 bearer, Mobility) parameter None Eb/Nt target on uplink
Global parameter None
Uplink Eb/Nt target increase due to
compressed mode activation
Site parameter None
Number of channel elements available
for a site on uplink
Site parameter None
Number of channel elements available
for a site on downlink
Simulation result None
Number of channel elements of a site
consumed by users on uplink
Simulation result None
Number of channel elements of a site
consumed by users on downlink
F
ortho
F
MUD
Tx
F
MUD
Term
i c
AS_Th Txi i c , ( )
Q
pi l ot
req
E
c
I
0
------
\ .
| |
threshol d
RSCP
pi l ot
CM acti vati on –
Q
pi l ot
CM acti vati on –
Q
req
DL
E
b
N
t
------
\ .
| |
req
DL
Q
req
DL
A
Q
req
UL
E
b
N
t
------
\ .
| |
req
UL
Q
req
UL
A
N
max
CE UL –
N
I
( )
N
max
CE DL –
N
I
( )
N
CE UL –
N
I
( )
N
CE DL –
N
I
( )
© Forsk 2010 AT283_TRG_E2 199
Chapter 6: UMTS HSPA Networks
Site equipment parameter - UL overhead resources for common
channels/cell
None
Number of channel elements used by
the cell for common channels on
uplink
Site equipment parameter - DL overhead resources for common
channels/cell
None
Number of channel elements used by
the cell for common channels on
downlink
(R99 bearer, site equipment) parameter None
Number of channel elements used for
R99 traffic channels on uplink
(R99 bearer, site equipment) parameter None
Number of channel elements used for
R99 traffic channels on downlink
(HSUPA bearer, site equipment) parameter None
Number of channel elements
consumed by the HSUPA bearer on
uplink
Site parameter kbps
Maximum Iub backhaul throughput for
a site in the uplink
Site parameter kbps
Maximum Iub backhaul throughput for
a site in the downlink
Simulation result kbps
Iub backhaul throughput for a site in
the uplink
Simulation result kbps
Iub backhaul throughput for a site in
the downlink
Site equipment parameter kbps
Iub throughput required by the cell for
common channels in the downlink
Site equipment parameter % HSDPA Iub backhaul overhead
Site equipment parameter kbps
Throughput carried by an E1/T1/
Ethernet link
(R99 bearer, site equipment) parameter kbps
Iub backhaul throughput consumed by
the R99 bearer in the uplink
(R99 bearer, site equipment) parameter kbps
Iub backhaul throughput consumed by
the R99 bearer in the downlink
(HSUPA bearer, site equipment) parameter kbps
Iub backhaul throughput consumed by
the HSUPA bearer in the uplink
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
Site equipment parameter - DL overhead resources for common
channels/cell
None
Number of 256 bit-length OVSF codes
used by the cell for common channels
Cell parameter (for HSDPA only) None
Maximum number of 16 bit-length
OVSF codes available per cell for HS-
PDSCH
Cell parameter (for HSDPA only) None
Minimum number of 16 bit-length
OVSF codes available per cell for HS-
PDSCH
Terminal parameter None Terminal Noise Figure
Transmitter parameter (user-defined or calculated from transmitter
equipment characteristics)
None Transmitter Noise Figure
K
1.38 10
-23
J/K Boltzman constant
T 293 K Ambient temperature
W 3.84 MHz Hz Spreading Bandwidth
Cell parameter None Inter-technology downlink noise rise
Cell parameter
Only used in uplink interference-based calculations of the Monte-Carlo
simulation
None Inter-technology uplink noise rise
Network parameter
If not defined, it is assumed that there is no inter-carrier interference
None
Interference reduction factor between
two adjacent carriers and
N
Overhead C – E UL –
N
Overhead C – E DL –
N
R99 T – CH C – E UL –
N
R99 T – CH C – E DL –
N
HSUPA C – E
T
Iub m – ax
UL
N
I
( )
T
Iub m – ax
DL
N
I
( )
T
Iub
UL
N
I
( )
T
Iub
DL
N
I
( )
T
Iub
Overhead DL –
Overhead
Iub
HSDPA
T
E1 T1 Ethernet
T
Iub
R99 T – CH UL –
T
Iub
R99 T – CH DL –
T
Iub
HSUPA
N
max
Codes
Txi i c , ( )
N
Codes
Txi i c , ( )
N
Overhead C – odes
N
max
Codes HS – PDSCH
Txi i c , ( )
N
mi n
Codes HS – PDSCH
Txi i c , ( )
NF
term
NF
Tx
NR
i nter techno y log –
Tx DL ,
NR
i nter techno y log –
Tx UL ,
RF i c i c
adj
, ( )
i c i c
adj
200 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Network parameter
If not defined, it is assumed that there is no inter-technology downlink
interferences due to external transmitters
None
Inter-technology Channel Protection
between the signal transmitted by Tx
and received by m assuming the
frequency gap between (external
network) and
Cell parameter (user-defined or simulation result) % Total uplink load factor of the cell
Simulation result %
Uplink cell load contribution due to
R99 traffic
Cell parameter %
Uplink cell load contribution due to
HSUPA traffic
Simulation constraint (global parameter or cell parameter) % Maximum uplink load factor of the cell
Simulation constraint (global parameter or cell parameter) % Maximum percentage of used power
W Thermal noise at transmitter
W Thermal noise at terminal
W
bps Chip rate
Site equipment parameter None Uplink rake receiver efficiency factor
Terminal parameter None
Downlink rake receiver efficiency
factor
R99 bearer parameter kbps R99 bearer downlink nominal bit rate
R99 bearer parameter None
Downlink spreading factor for active
users
R99 bearer parameter None
Downlink spreading factor for inactive
users
R99 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
Cell parameter kbps
Maximum connection rate per user on
downlink
R99 bearer parameter kbps R99 bearer uplink nominal bit rate
Service parameter kbps
Uplink activity factor on E-DPCCH
channels
Service parameter kbps
Downlink Activity factor on A-DPCH
channel
Service parameter kbps
Minimum required bit rate that the
service should have in order to be
available in the uplink
Service parameter kbps
Minimum required bit rate that the
service should have in order to be
available in the downlink
R99 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
Cell parameter kbps
Maximum connection rate per user on
uplink
None Service downlink processing gain
None Service uplink processing gain
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
X
UL
X
R99
UL
X
HSUPA
UL
X
max
UL
%Power
max
DL
N
0
Tx
NF
Tx
K T × W × × NR
i nter techno y log –
Tx UL ,
×
N
0
Term
NF
Term
K T × W NR
i nter techno y log –
Tx DL ,
× × ×
R
c W 10
3 –
·
f
rake efficiency
UL
f
rake efficiency
DL
R
nomi nal
DL
F
spreading
DL
Active user ( )
F
spreading
DL
Inactive user ( )
r
c
DL
R
max
DL
R
nomi nal
UL
f
act ADPCH –
UL
f
act ADPCH –
DL
R
Guaranteed
UL
R
Guaranteed
DL
r
c
UL
R
max
UL
G
p
DL
W
R
nomi nal
DL
----------------------
G
p
UL
W
R
nomi nal
UL
----------------------
© Forsk 2010 AT283_TRG_E2 201
Chapter 6: UMTS HSPA Networks
HSDPA bearer parameter None
Index of the HSDPA bearer obtained
by the user in the cell (Txi,ic)
HSDPA bearer parameter kbps
RLC peak rate supported by the
HSDPA bearer
Without MIMO:
With MIMO (transmit diversity):
With MIMO (spatial multiplexing):
kbps
RLC peak rate provided to the user in
the cell (Txi,ic) in the downlink
HSDPA study result
For single-carrier HSDPA users
For dual-cell HSDPA users
kbps
RLC peak rate provided to the user in
the downlink
START HERE %
HSDPA bearer consumption for a
packet (HSPA - Constant Bit Rate)
service user
HSDPA study result kbps
RLC peak throughput supported by
the HSDPA bearer
HSDPA study result kbps
Average RLC throughput supported
by the HSDPA bearer
HSDPA study result kbps
MAC rate supported by the HSDPA
bearer
HSDPA study result kbps
MAC throughput supported by the
HSDPA bearer
HSDPA study result kbps
User application throughput on
downlink
HSUPA study result kbps User application throughput on uplink
HSUPA study result kbps
User average application throughput
on uplink
HSUPA Bearer parameter None
Index of the HSUPA bearer obtained
in the cell (Txi,ic)
HSUPA bearer selection parameter kbps
Maximum number of retransmissions
a HARQ process will perform for a
block of data before moving on to a
new block of data, for the HSUPA
bearer index
HSUPA bearer parameter kbps
RLC peak rate supported by the
HSUPA bearer
HSUPA study result
kbps
RLC peak rate provided to the user in
the cell (Txi,ic) in the uplink
%
HSUPA bearer consumption for a
packet (HSPA - Constant Bit Rate)
service user
HSUPA study result kbps
Minimum RLC throughput supported
by the HSUPA bearer
HSUPA study result kbps
Average RLC throughput supported
by the HSUPA bearer
HSUPA study result kbps
MAC rate supported by the HSUPA
bearer
Service parameter (for HSDPA only) kbps Throughput offset
Service parameter (for HSDPA only) % Scaling factor
I
HSDPABearer
R
RLC peak –
DL
I
HSDPABearer
( )
R
RLC peak –
DL
Tx i c , ( )
R
RLC peak –
DL
I
HSDPABearer
( )
R
RLC peak –
DL
I
HSDPABearer
( )
R
RLC peak –
DL
I
HSDPABearer
( ) 1 f
SM Gai n –
G
SM
Max
1 – ( ) × + ( ) ) ×
R
RLC peak –
DL
R
RLC peak –
DL
Tx i c , ( )
R
RLC peak –
DL
Tx i c , ( )
i c Tx e
¿
C
HSDPABearer
R
Guaranteed
DL
R
RLC peak –
DL
I
HSDPABearer
( )
---------------------------------------------------------------------
T
RLC peak –
DL
T
RLC Av –
DL
R
MAC
DL
T
MAC
DL
T
appl i cati on
DL
T
appl i cati on
UL
T
appl i cati on Av –
UL
I
HSUPABearer
N
Rtx
I
HSUPABearer
( )
R
RLC peak –
UL
I
HSUPABearer
( )
R
RLC peak –
UL
R
RLC peak –
UL
I
HSUPABearer
( )
C
HSUPABearer
R
Guaranteed
UL
R
RLC peak –
UL
I
HSUPABearer
( )
---------------------------------------------------------------------
T
RLC Mi n –
UL
T
RLC Av –
UL
R
MAC
UL
AR
SF
Rate
202 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Transmitter parameter W
Maximum shared power
Available only if the inter-carrier power
sharing option is activated
Cell parameter W Cell synchronisation channel power
Cell parameter W
Cell other common channels (except
CPICH and SCH) power
Cell parameter W Cell pilot power
Cell parameter (user-defined or simulation result) (for HSDPA only)
W
Available cell HSDPA power
HSDPA: High Speed Downlink Packet
Access
Simulation result (for HSDPA only) W
Cell HS-PDSCH power
HS-PDSCH: High Speed Physical
Downlink Shared Channel
Cell parameter (for HSDPA only) W
Cell HS-SCCH power
HS-SCCH: High Speed Shared
Control Channel
Cell parameter (for HSDPA only)
number of HS-SCCH channels
managed by the cell
Cell parameter (for HSDPA only) W Cell headroom power
Cell parameter W Maximum Cell power
Simulation result W
R99 traffic channel power transmitted
on carrier ic
R99 bearer parameter W
Minimum power allowed on R99 traffic
data channel
R99 bearer parameter W
Maximum power allowed on R99
traffic data channel
Cell parameter W
Cell HSUPA power
HSUPA: High Speed Uplink Packet
Access
Simulation result W
Transmitter HSDPA power transmitted
on carrier ic
Simulation result
W
Transmitter R99 power transmitted on
carrier ic
Cell parameter (user-defined or simulation result)
W
Transmitter total power transmitted on
carrier ic
Simulation result W
Terminal power transmitted to obtain
the R99 radio bearer
Simulation result W
Terminal power transmitted to obtain
the HSUPA radio bearer
Simulation result
for HSPA users
for R99 users
W
Total power transmitted by the
terminal
Terminal parameter W Minimum terminal power allowed
Terminal parameter W Maximum terminal power allowed
BTS parameter %
Percentage of BTS signal correctly
transmitted
Terminal parameter %
Percentage of terminal signal correctly
transmitted
Clutter parameter %
Percentage of pilot finger - percentage
of signal received by the terminal pilot
finger
Antenna parameter None Transmitter antenna gain
P
max
Txi ( )
P
SCH
Txi i c , ( )
P
OtherCCH
Txi i c , ( )
P
pi l ot
Txi i c , ( )
P
HSDPA
Txi i c , ( )
P
HS PDSCH –
Txi i c , ( ) n
HS SCCH –
P ×
HS SCCH –
Txi i c , ( ) +
P
HS PDSCH –
Txi i c , ( )
P
HS SCCH –
Txi i c , ( )
n
HS SCCH –
P
Headroom
Txi i c , ( )
P
max
Txi i c , ( )
P
tch
Txi i c , ( )
P
tch
mi n
P
tch
max
P
HSUPA
Txi i c , ( )
P
tx H – SDPA
Txi i c , ( )
P
tx R99 –
Txi i c , ( )
P
pi l ot
Txi i c , ( ) P
SCH
Txi i c , ( ) P
OtherCCH
Txi i c , ( ) + + +
P
tch
Txi i c , ( )
tch(ic) used for
R99 users
¿
P
tch
Txi i c , ( ) f
act ADPCH –
DL
×
tch(ic) used for
HSUPA users
¿
+
P
tx
Txi i c , ( )
P
tx R99 –
Txi i c , ( ) P
tx H – SDPA
Txi i c , ( ) P
HSUPA
Txi i c , ( ) + +
P
term R99 –
P
term HSUPA –
P
term
P
term R99 –
f
act ADPCH –
UL
× P
term HSUPA –
+
P
term R99 –
P
term
mi n
P
term
max
µ
BTS
µ
term
o
G
Tx
© Forsk 2010 AT283_TRG_E2 203
Chapter 6: UMTS HSPA Networks
Terminal parameter None Terminal gain
R99 bearer parameter - Depends on the transmitter Tx diversity None Gain due to transmit diversity
R99 bearer parameter - Depends on the transmitter Rx diversity None Gain due to receive diversity
MIMO configuration parameter dB
Maximum spatial multiplexing gain for
a given number of transmission and
reception antennas
MIMO configuration parameter dB
Downlink Transmit Diversity gain for a
given number of transmission and
reception antenna ports
Clutter parameter None Spatial multiplexing gain factor
Clutter parameter dB Additional diversity gain in downlink
Transmitter parameter (user-defined or calculated from transmitter
equipment characteristics)
None
Transmitter loss
a
Service parameter None Body loss
Terminal parameter None Terminal loss
Clutter (and, optionally, frequency band) parameter Indoor loss
Propagation model result None Path loss
Result calculated from cell edge coverage probability and model
standard deviation
None
Model Shadowing margin
Only used in prediction studies
Result calculated from cell edge coverage probability and Ec/I0
standard deviation
None
Ec/I0 Shadowing margin
Only used in prediction studies
n=2 or 3
None
DL gain due to availability of several
pilot signals at the mobile
b
.
Result calculated from cell edge coverage probability and DL Eb/Nt
standard deviation
None
DL Eb/Nt Shadowing margin
Only used in prediction studies
Result calculated from cell edge coverage probability and UL Eb/Nt
standard deviation
None
UL Eb/Nt Shadowing margin
Only used in prediction studies
n=2 or 3
Global parameter (default value)
None
UL quality gain due to signal diversity
in soft handoff
c
.
Simulation result None
Random shadowing error drawn
during Monte-Carlo simulation
Only used in simulations
In prediction studies
d
For Ec/I0 calculation
For DL Eb/Nt calculation
For UL Eb/Nt calculation
In simulations
None Transmitter-terminal total loss
W Chip power received at terminal
W
Bit power received at terminal on
carrier ic
W
Total power received at terminal from
a transmitter on carrier ic
G
Term
G
Di v
DL
G
Di v
UL
G
SM
Max
G
TD
DL
f
SM Gai n –
AG
TD
DL
L
Tx
L
body
L
Term
L
i ndoor
L
path
M
Shadowi ng model –
M
Shadowi ng Ec Io –
G
macro di versi ty –
DL G
macro di versi ty –
DL
M
Shadowi ng Ec Io –
npaths
M
Shadowi ng Ec Io –
– =
M
Shadowi ng Eb Nt ( )
DL

M
Shadowi ng Eb Nt ( )
UL

G
macro di versi ty –
UL
G
macro di versi ty –
UL
M
Shadowi ng Eb Nt ( )
UL

npaths
M
Shadowi ng Eb Nt ( )
UL

– =
E
Shadowi ng
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
---------------------------------------------------------------------------------------------------------------------------------------------------
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Eb Nt ( )
DL

× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------------------------------------
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Eb Nt ( )
UL

× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------------------------------------
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------
P
c
Txi i c , ( )
P
pi l ot
Txi i c , ( )
L
T
-----------------------------------
P
b
DL
Txi i c , ( )
P
tch
Txi i c , ( )
L
T
--------------------------------
P
tot
DL
Txi i c , ( )
P
tx
Txi i c , ( )
L
T
------------------------------
204 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
6.2.2 Ec/I0 Calculation
This table details the pilot quality ( or ) calculations.
6.2.3 DL Eb/Nt Calculation
This table details calculations of downlink traffic channel quality ( or ).
W
Total power received at terminal from
traffic channels of a transmitter on
carrier ic
W
Bit power received at transmitter on
carrier ic used by terminal
W
Bit power received at transmitter on
carrier ic used by terminal
W
Bit power received at transmitter on
DPDCH from a terminal on carrier ic
a. on uplink and on downlink. For information on calculating transmitter
losses on uplink and downlink, see "UMTS, CDMA2000, TD-SCDMA, WiMAX, and LTE Documents" on page 128.
b. corresponds to the shadowing margin evaluated from the shadowing error probability
density function (n paths) in case of downlink Ec/I0 modelling.
c. corresponds to the shadowing margin evaluated from the shadowing error probability
density function (n paths) in case of uplink soft handoff modelling.
d. In uplink prediction studies, only carrier power level is downgraded by the shadowing margin
( ). In downlink prediction studies, carrier power level and intra-cell interference are downgraded by
the shadowing model ( or ) while extra-cell interference level is not. Therefore,
or is set to 1 in downlink extra-cell interference calculation.
P
traf
DL
Txi i c , ( )
P
tch
Txi i c , ( )
L
T
--------------------------------
tch i c ( )
¿
P
b
UL
i c ( )
P
term
L
T
--------------
P
b R99 –
UL
i c ( )
P
term R99 –
L
T
---------------------------
P
b DPDCH –
UL
i c ( ) P
b R99 –
UL
i c ( ) 1 r
c
UL
– ( ) ×
L
Tx
L
total UL –
= L
Tx
L
total DL –
=
M
Shadowi ng Ec Io –
npaths
M
Shadowi ng Eb Nt ( )
UL

npaths
M
Shadowi ng Eb Nt ( )
UL

M
Shadowi ng Eb Nt ( )
DL

M
Shadowi ng Ec Io –
M
Shadowi ng Eb Nt ( )
DL

M
Shadowi ng Ec Io –
Q
pi l ot
Ec Io
Name Value Unit Description
W
Downlink intra-cell interference at
terminal on carrier
W
Downlink extra-cell interference at
terminal on carrier
W
Downlink inter-carrier interference at
terminal on carrier
W
Downlink inter-technology interference
at terminal on carrier ic
a
Without Pilot:
Total noise:
W
Total received noise at terminal on
carrier
None
Quality level at terminal on pilot for
carrier
a. In the case of an interfering GSM external network in frequency hopping, the ICP value is weighted according to the
fractional load.
I
i ntra
DL
txi i c , ( )
P
tot
DL
txi i c , ( )
µ
BTS
o × –
P
tot
DL
txi i c , ( )
P
SCH
txi i c , ( )
L
T
---------------------------------- –
\ .
| |
×
i c
I
extra
DL
i c ( )
P
tot
DL
txj i c , ( )
txj j i = ,
¿
i c
I
i nter carri er –
DL
i c ( )
P
tot
DL
txj i c
adj
, ( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
-------------------------------------------------
i c
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
I
0
DL
i c ( )
I
i ntra
DL
txi i c , ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) + + +
N
0
Term
1 o – ( ) µ
BTS
P
c
txi i c , ( ) × × – +
P
tot
DL
txi i c , ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) + +
I
i nter techno y log –
DL
i c ( ) N
0
Term
+ +
i c
Q
pi l ot
txi i c , ( )
E
c
I
0
------
\ .
| |
·
µ
BTS
o × P
c
txi i c , ( ) ×
I
0
DL
i c ( )
-------------------------------------------------------
i c
Q
tch
DL Eb
Nt
-------
\ .
| |
DL
Name Value Unit Description
© Forsk 2010 AT283_TRG_E2 205
Chapter 6: UMTS HSPA Networks
6.2.4 UL Eb/Nt Calculation
This table details calculations of uplink traffic channel quality ( or ).
W
Downlink intra-cell interference at
terminal on carrier
W
Downlink extra-cell interference at
terminal on carrier
W
Downlink inter-carrier interference at
terminal on carrier
W
Downlink inter-technology interference
at terminal on carrier ic
a
W
Total received noise at terminal on
carrier ic
Without useful signal:
Total Noise:
None
Quality level at terminal on a traffic
channel from one transmitter on
carrier
b
None
Quality level at terminal using carrier
due to combination of all
transmitters of the active set (Macro-
diversity conditions).
None Soft handover gain on downlink
W
Required transmitter traffic channel
power to achieve Eb/Nt target at
terminal on carrier
a. In the case of an interfering GSM external network in frequency hopping, the ICP value is weighted according to the
fractional load.
b. 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.
I
i ntra
DL
txi i c , ( )
P
tot
DL
txi i c , ( )
µ
BTS
F
ortho
× –
P
tot
DL
txi i c , ( )
P
SCH
txi i c , ( )
L
T
---------------------------------- –
\ .
| |
×
i c
I
extra
DL
i c ( )
P
tot
DL
txj i c , ( )
txj j i = ,
¿
i c
I
i nter carri er –
DL
i c ( )
P
tot
DL
txj i c
adj
, ( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
-------------------------------------------------
i c
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
Term
+ +
Q
tch
DL
txi i c , ( )
E
b
N
t
------
\ .
| |
DL
·
µ
BTS
P
b
DL
txi i c , ( ) ×
N
tot
DL
i c ( ) 1 F
ortho
– ( ) µ
BTS
P
b
DL
txi i c , ( ) × × –
---------------------------------------------------------------------------------------------------------------- G
Di v
DL
G
p
DL
× ×
µ
BTS
P
b
DL
txi i c , ( ) ×
N
tot
DL
i c ( )
------------------------------------------------- G
Di v
DL
× G
p
DL
×
i c
Q
DL
i c ( )
f
rake efficiency
DL
Q
tch
DL
tx
k
i c , ( )
tx
k
Acti veSet e
¿
×
i c
G
SHO
DL
Q
DL
i c ( )
Q
tch
DL
BestServer i c , ( )
----------------------------------------------------------
P
tch
req
txi i c , ( )
Q
req
DL
Q
DL
i c ( )
--------------------- P
tch
txi i c , ( ) ×
i c
Q
tch
UL Eb
Nt
-------
\ .
| |
UL
Name Value Unit Description
W
Total power received at transmitter
from intra-cell terminals using carrier
W
Total power received at transmitter
from extra-cell terminals using carrier
W
Uplink inter-carrier interference at
terminal on carrier
W
Total received interference at
transmitter on carrier
W
Total noise at transmitter on carrier
(Uplink interference)
I
tot
UL
i ntra
txi i c , ( )
P
b
UL
i c ( )
term
txi
¿
i c
I
tot
UL
ext ra
txi i c , ( )
P
b
UL
i c ( )
term
txj j i = ,
¿
i c
I
i nter carri er –
UL
txi i c , ( )
P
b
UL
i c
adj
( )
term
txj j ¬ ,
¿
RF i c i c
adj
, ( )
---------------------------------------
i c
I
tot
UL
txi i c , ( )
I
tot
UL
extra
txi i c , ( ) 1 F
MUD
Tx
µ
term
× – ( ) + I
tot
UL
i ntra
txi i c , ( ) × I
i nter carri er –
UL
txi i c , ( ) +
i c
N
tot
UL
txi i c , ( ) I
tot
UL
txi i c , ( ) N
0
tx
+
i c
206 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
6.3 Active Set Management
The mobile’s 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 mobile’s active set must use a frequency band
supported by the terminal. Finally, the quality of the pilot (Ec⁄I0) is what determines whether or not a transmitter can belong
to the active set. The active set management is detailed hereafter. Cells entering a mobile’s 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/I0 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.
Without useful signal:
Total noise:
None
Quality level at transmitter on a traffic
channel for carrier
a
No HO:
Softer HO:
Soft, softer/soft HO (No MRC):
Softer/soft HO (MRC):
None
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
the quality due to macro-diversity
(macro-diversity gain).
In simulations .
None Soft handover gain on uplink
W
Required terminal power to achieve
Eb/Nt target at transmitter on carrier
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.
Q
tch
UL
txi i c , ( )
E
b
N
t
------
\ .
| |
UL
·
µ
term
P
b DPDCH –
UL
i c ( ) ×
N
tot
UL
txi i c , ( ) 1 F
MUD
Tx
– ( ) µ
term
× P
b
UL
i c ( ) × –
---------------------------------------------------------------------------------------------------------------- G
Di v
UL
× G
p
UL
×
µ
term
P
b DPDCH –
UL
i c ( ) ×
N
tot
UL
txi i c , ( )
---------------------------------------------------------- G
Di v
UL
× G
p
UL
×
i c
Q
UL
i c ( )
Q
tch
UL
txi i c , ( )
f
rake efficiency
UL
Q
tch
UL
tx
k
i c , ( )
tx
k
Acti veSet e
samesi te ( )
¿
×
Max
tx
k
Act i veSet e
Q
tch
UL
tx
k
i c , ( ) ( ) G
macro di versi ty –
UL
×
Max


t x
k
t x
l
, Acti veSet e
tx
k
samesi t e e
tx
l
ot hersi te e
f
rake efficiency
UL
Q
tch
UL
tx
k
i c , ( )
tx
k
¿
× Q
tch
UL
tx
l
i c , ( ) ,
\ .
|
|
|
| |
G
macro di versi ty –
UL
×
G
macro di versi ty –
UL
1 =
G
SHO
UL
Q
UL
i c ( )
Q
tch
UL
BestServer i c , ( )
----------------------------------------------------------
P
term
req
i c ( )
Q
req
UL
Q
UL
i c ( )
--------------------- P
term
×
i c
© Forsk 2010 AT283_TRG_E2 207
Chapter 6: UMTS HSPA Networks
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.
6.4.1.1 Simulations Based on User Profile Traffic Maps
User profile environment based traffic maps: 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 km²).
User profile traffic maps: 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.
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 ,
• The average duration of a call (seconds) .
The number of users and their distribution per activity status is determined as follows:
10. Calculation of the service usage duration per hour ( : probability of a connection):
11. Calculation of the number of users trying to access the service i ( ):
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
Notes:
• When user profile traffic maps are 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:
• The number of subscribers (X) is an input when a user profile traffic map is composed of
points.
X S D × =
X L D × =
N
cal l
d
p
0
p
o
N
cal l
d ×
3600
---------------------- =
n
i
n
i
X p
0
× =
208 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Next, we can take into account activity periods during the connection in order to determine the activity status of each user.
12. Calculation of activity probabilities:
Probability of being inactive on UL and DL:
Probability of being active on UL only:
Probability of being active on DL only:
Probability of being active both on UL and DL:
Where, and are respectively the UL and DL activity factors defined for the circuit switched service i.
13. Calculation of number of users per activity status:
Number of inactive users on UL and DL:
Number of users active on UL and inactive on DL:
Number of users active on DL and inactive on UL:
Number of users active on UL and DL both:
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 ,
• The volume (in kbytes) which is transferred on the downlink and the uplink 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:
: Average number of packet calls on the uplink during a session,
: Average number of packet calls on the downlink during a session,
: Average time (millisecond) between two packets calls on the uplink ,
: Average time (millisecond) between two packets calls on the downlink ,
: Average time (millisecond) between two packets on the uplink ,
: Average time (millisecond) between two packets on the downlink ,
: Packet size (Bytes) on uplink,
: Packet size (Bytes) on downlink.
p
i nacti ve
1 f
act
UL
– ( ) 1 f
act
DL
– ( ) × =
p
UL
f
act
UL
1 f
act
DL
– ( ) × =
p
DL
f
act
DL
1 f
act
UL
– ( ) × =
p
UL DL +
f
act
UL
f
act
DL
× =
f
act
UL
f
act
DL
n
i
i nacti ve
n
i
p
i nacti ve
× =
n
i
UL ( ) n
i
p
UL
× =
n
i
DL ( ) n
i
p
DL
× =
n
i
UL DL + ( ) n
i
p
UL DL +
× =
Figure 6.1Description of a Packet Session
N
sess
V
DL
V
UL
N
packet c – al l
UL
N
packet c – al l
DL
AT
packet cal l –
UL
AT
packet cal l –
DL
AT
packet
UL
AT
packet
DL
S
packet
UL
S
packet
DL
© Forsk 2010 AT283_TRG_E2 209
Chapter 6: UMTS HSPA Networks
The number of users and their distribution per activity status is determined as follows:
14. Calculation of the average packet call size (kBytes):
and
Where and are the UL and DL efficiency factors defined for the packet switched service j.
15. Calculation of the average number of packets per packet call:
and
16. Calculation of the average duration of inactivity within a packet call (s):
and
17. Calculation of the average duration of inactivity in a session (s):
and
18. Calculation of the average duration of activity in a session (s):
and
Where and are the uplink and downlink average requested rates defined for the service j.
Therefore, the average duration of a connection (in s) is:
and
19. Calculation of the service usage duration per hour (probability of a connection):
and
20. Calculation of the probability of being connected:
Therefore, the number of users who want to get the service j is:
As you can see on the picture above, we have to consider three possible cases when a user is connected:
• 1
st
case: At a given time, packets are downloaded and uploaded.
In this case, the probability of being connected is:
• 2
nd
case: At a given time, packet are uploaded (no packet is downloaded).
Here, the probability of being connected is:
• 3
rd
case: At a given time, packet are downloaded (no packet is uploaded).
In this case, the probability of being connected is:
Note:
• For packet (HSDPA) and packet (HSPA) services, and are set to 1.
Note:
• 1kBytes = 1024Bytes.
S
packet c – al l
UL
V
UL
N
packet c – al l
UL
f
eff
UL
×
-------------------------------------------- = S
packet c – al l
DL
V
DL
N
packet c – al l
DL
f
eff
DL
×
-------------------------------------------- =
f
eff
UL
f
eff
DL
f
eff
UL
f
eff
DL
N
packet
UL
i nt
S
packet c – al l
UL
S
packet
UL
1024
------------------------------------
\ .
|
| |
1 + = N
packet
DL
i nt
S
packet c – al l
DL
S
packet
DL
1024
------------------------------------
\ .
|
| |
1 + =
D
Inacti vi ty
UL
( )
packet cal l –
N
packet
UL
1 – ( ) AT
packet
UL
×
1000
--------------------------------------------------------------- = D
Inacti vi ty
DL
( )
packet cal l –
N
packet
DL
1 – ( ) AT
packet
DL
×
1000
--------------------------------------------------------------- =
D
Inacti vi ty
UL
( )
sessi on
N
packet c – al l
UL
D
Inacti vi ty
UL
( )
packet cal l –
× = D
Inacti vi ty
DL
( )
sessi on
N
packet c – al l
DL
D
Inacti vi ty
DL
( )
packet cal l –
× =
D
Acti vi ty
UL
( )
sessi on
N
packet c – al l
UL
N
packet
UL
S
packet
UL
8 × ×
R
average
UL
1000 ×
------------------------------------------------------ × = D
Acti vi ty
DL
( )
sessi on
N
packet c – al l
DL
N
packet
DL
S
packet
DL
8 × ×
R
average
DL
1000 ×
------------------------------------------------------ × =
R
average
UL
R
average
DL
D
Connecti on
UL
D
Acti vi ty
UL
( )
sessi on
D
Inacti vi ty
UL
( )
sessi on
+ = D
Connecti on
DL
D
Acti vi ty
DL
( )
sessi on
D
Inacti vi ty
DL
( )
sessi on
+ =
p
Connecti on
UL
N
sess
3600
-------------- D
Connecti on
UL
× = p
Connecti on
DL
N
sess
3600
-------------- D
Connecti on
DL
× =
p
Connected
1 1 p
Connecti on
UL
– ( ) 1 p
Connecti on
DL
– ( ) × – =
n
j
X p
Connected
× =
p
Connected
UL DL +
p
Connecti on
UL
p
Connecti on
DL
×
p
Connected
----------------------------------------------------------------- =
p
Connected
UL
p
Connecti on
UL
1 p
Connecti on
DL
– ( ) ×
p
Connected
------------------------------------------------------------------------------- =
p
Connected
DL
p
Connecti on
DL
1 p
Connecti on
UL
– ( ) ×
p
Connected
------------------------------------------------------------------------------- =
210 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Now, we have to take into account activity periods during the connection in order to determine the activity status of each
user.
21. Calculation of the probability of being active:
and
Therefore, we have:
• 1
st
case: At a given time, packets are downloaded and uploaded.
The user can be active on UL and inactive on DL; this probability is:
The user can be active on DL and inactive on UL; this probability is:
The user can be active on both links; this probability is:
The user can be inactive on both links; this probability is:
• 2
nd
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:
The user can be inactive on both links; this probability is:
• 3
rd
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:
The user can be inactive on both links; this probability is:
22. Calculation of number of users per activity status
Number of inactive users on UL and DL:
Number of users active on UL and inactive on DL:
Number of users active on DL and inactive on UL:
Number of users active on UL and 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.
6.4.1.2 Simulations Based on Sector Traffic Maps
Sector traffic maps can be based on live traffic data from OMC (Operation and Maintenance Centre). Traffic is spread over
the best server coverage area of each transmitter and each coverage area is assigned either the throughputs in the uplink
and in the downlink or the number of users per activity status or the total number of users (including all activity statuses).
Note:
• 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.
f
UL
D
Acti vi ty
UL
( )
sessi on
D
Inacti vi ty
UL
( )
sessi on
D
Acti vi ty
UL
( )
sessi on
+ ( )
------------------------------------------------------------------------------------------------------ = f
DL
D
Acti vi ty
DL
( )
sessi on
D
Inacti vi ty
DL
( )
sessi on
D
Acti vi ty
DL
( )
sessi on
+ ( )
------------------------------------------------------------------------------------------------------ =
p
UL
1
f
UL
1 f
DL
– ( ) × p
Connected
UL DL +
× =
p
DL
1
f
DL
1 f
UL
– ( ) × p
Connected
UL DL +
× =
p
UL DL +
1
f
UL
f
DL
× p
Connected
UL DL +
× =
p
i nacti ve
1
1 f –
UL
( ) 1 f
DL
– ( ) × p
Connected
UL DL +
× =
p
UL
2
f
UL
p
Connected
UL
× =
p
i nacti ve
2
1 f –
UL
( ) p
Connected
UL
× =
p
DL
3
f
DL
p
Connected
DL
× =
p
i nacti ve
3
1 f –
DL
( ) p
Connected
DL
× =
n
j
i nacti ve
n
j
p
i nacti ve
1
p
i nacti ve
2
p
i nacti ve
3
+ + ( ) × =
n
j
UL ( ) n
j
p
UL
1
p
UL
2
+ ( ) × =
n
j
DL ( ) n
j
p
DL
1
p
DL
3
+ ( ) × =
n
j
UL DL + ( ) n
j
p
UL DL +
1
× =
© Forsk 2010 AT283_TRG_E2 211
Chapter 6: UMTS HSPA Networks
6.4.1.2.1 Throughputs in Uplink and Downlink
When selecting Throughputs in Uplink and Downlink, you can input the throughput demands in the uplink and downlink
for each sector and for each listed service.
Atoll calculates the number of users active in uplink and in downlink in the Txi cell using the service (N
UL
and N
DL
) as
follows:
and
is the kbits per second transmitted in UL in the Txi cell to supply the service.
is the kbits per second transmitted in DL in the Txi cell to supply the service.
is the downlink average requested rate defined for the service,
is the uplink average requested rate defined for the service.
N
UL
and N
DL
values include:
• Users active in uplink and inactive in downlink (n
i
(UL)),
• Users active in downlink and inactive in uplink (n
i
(DL)),
• And users active in both links (n
i
(UL+DL)).
Atoll takes into account activity periods during the connection in order to determine the activity status of each user.
Activity probabilities are calculated as follows:
Probability of being inactive in UL and DL:
Probability of being active in UL only:
Probability of being active in DL only:
Probability of being active both in UL and DL:
Where, and are respectively the UL and DL activity factors defined for the service i.
Then, Atoll calculates the number of users per activity status:
We have:
Therefore, we have:
Number of users active in UL and DL both:
Number of users active in UL and inactive in DL:
Number of users active in DL and inactive in UL:
Number of inactive users in UL and DL:
Therefore, a connected user can have four different activity status: either active in both links, or inactive in both links, or
active in UL only, or active in DL only.
6.4.1.2.2 Total Number of Users (All Activity Statuses)
When selecting Total Number of Users (All Activity Statuses), you can input the number of connected users for each
sector and for each listed service ( ).
Atoll takes into account activity periods during the connection in order to determine the activity status of each user.
Activity probabilities are calculated as follows:
Probability of being inactive in UL and DL:
Note:
• For packet (HSDPA) and packet (HSPA) services, and are set to 1.
N
UL
R
t
UL
R
average
UL
---------------------- = N
DL
R
t
DL
R
average
DL
---------------------- =
R
t
UL
R
t
DL
R
average
DL
R
average
UL
p
i nacti ve
1 f
act
UL
– ( ) 1 f
act
DL
– ( ) × =
p
UL
f
act
UL
1 f
act
DL
– ( ) × =
p
DL
f
act
DL
1 f
act
UL
– ( ) × =
p
UL DL +
f
act
UL
f
act
DL
× =
f
act
UL
f
act
DL
f
act
UL
f
act
DL
p
UL
p
UL DL +
+ ( ) n
j
UL ( ) n
j
DL ( ) n
j
UL DL + ( ) + + ( ) × N
UL
=
p
DL
p
UL DL +
+ ( ) n
j
UL ( ) n
j
DL ( ) n
j
UL DL + ( ) + + ( ) × N
DL
=
n
i
UL DL + ( ) mi n
N
UL
p
UL DL +
×
p
UL
p
UL DL +
+
-------------------------------------
N
DL
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 + ( ) – =
n
i
i nacti ve
n
j
UL ( ) n
j
DL ( ) n
j
UL DL + ( ) + + ( )
1 p
i nacti ve

-------------------------------------------------------------------------------------- p
i nacti ve
× =
n
i
p
i nacti ve
1 f
act
UL
– ( ) 1 f
act
DL
– ( ) × =
212 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Probability of being active in UL only:
Probability of being active in DL only:
Probability of being active both in UL and DL:
Where, and are respectively the UL and DL activity factors defined for the service i.
Then, Atoll calculates the number of users per activity status:
Number of inactive users in UL and DL:
Number of users active in UL and inactive in DL:
Number of users active in DL and inactive in UL:
Number of users active in UL and DL both:
Therefore, a connected user can have four different activity status: either active in both links, or inactive in both links, or
active in UL only, or active in DL only.
6.4.1.2.3 Number of Users per Activity Status
When selecting Number of Users per Activity Status, you can directly input the number of inactive users ( ), the
number of users active in the uplink ( ), in the downlink ( ) and in the uplink and downlink ( ), for
each sector and for each service.
6.4.2 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 (i.e., Packet (HSDPA), Packet (HSPA) and Packet (HSPA - Constant Bit Rate) service 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 (i.e., Packet
(HSPA) and Packet (HSPA - Constant Bit Rate) service 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 (i.e., Circuit (R99), Packet
(R99), Packet (HSDPA), Packet (HSPA) and Packet (HSPA - Constant Bit Rate) service users) selected during the user
distribution generation (1
st
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.
Note:
• For packet (HSDPA) and packet (HSPA) services, and are set to 1.
p
UL
f
act
UL
1 f
act
DL
– ( ) × =
p
DL
f
act
DL
1 f
act
UL
– ( ) × =
p
UL DL +
f
act
UL
f
act
DL
× =
f
act
UL
f
act
DL
f
act
UL
f
act
DL
n
i
i nacti ve
n
i
p
i nacti ve
× =
n
i
UL ( ) n
i
p
UL
× =
n
i
DL ( ) n
i
p
DL
× =
n
i
UL DL + ( ) n
i
p
UL DL +
× =
Note:
• 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.
n
i
i nacti ve
n
i
UL ( ) n
i
DL ( ) n
i
UL DL + ( )
© Forsk 2010 AT283_TRG_E2 213
Chapter 6: UMTS HSPA Networks
As shown in Figure 6.2 on page 213, the simulation algorithm is divided in three parts. All users are evaluated by the R99
part of the algorithm. HSDPA and HSUPA bearer users, unless they have been rejected during the R99 part of the
algorithm, are then evaluated by the HSDPA part of the algorithm. Finally, HSUPA bearer users, unless they have been
rejected during the R99 or HSDPA parts of the algorithm, are then evaluated by the HSUPA part of the 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 , , is initialised to
.
Uplink powers received by the base station txi on carrier , , and
are initialised to 0 W (i.e. no connected mobile).
Therefore, we have:
Figure 6.2UMTS HSPA Power Control Algorithm
R99 part
Initialisation
Mi Best Server Determination
Mi Active Set Determination
UL and DL Interference Update
For each R99, HSDPA
and HSUPA mobile, Mi
DL Power Control
UL Power Control
Congestion and Radio Resource Control
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/A-
DPCH bearer (R99 bearer)
HSDPA part
Fast Link Adaptation
Mobile Scheduling
Radio Resource Control
For each HSDPA and
HSUPA mobile, Mi
HSUPA part
Admission Control
For each HSUPA
mobile, Mi
Radio Resource Control
Convergence Study
Noise Rise Scheduling
i c
m
P
Tx
txi i c
m
, ( )
P
pi l ot
txi i c
m
, ( ) P
SCH
txi i c
m
, ( ) P
otherCCH
txi i c
m
, ( ) P
HSDPA
txi i c
m
, ( ) P
HSUPA
txi i c , ( ) + + + +
i c
m
I
tot
UL
i ntra
txi i c
m
, ( ) I
tot
UL
ext ra
txi i c
m
, ( )
I
i nter carri er –
UL
txi i c
m
, ( )
X
R99
UL
txi i c
m
, ( ) ( )
k
I
tot
UL
txi i c
m
, ( )
N
tot
UL
txi i c
m
, ( )
---------------------------------- 0 = =
214 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
6.4.2.2 R99 Part of the Algorithm
The algorithm is detailed for any iteration k. X
k
is the value of the X (variable) at the iteration k. In the algorithm, all
and thresholds depend on the user mobility type and are defined in the R99 bearer property dialogue. All variables
are described in Definitions and formulas part. The bearer downgrading is not dealt with.
The algorithm applies to single frequency band networks and to dual-band networks. 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).
For each mobile M
b
Determination of M
b
’s Best Server
For each transmitter txi containing M
b
in its calculation area and working on the main frequency band supported by the
M
b
’s 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).
Calculation of
If user selects “without Pilot”
Determination of the best transmitter, tx
BS
, for each carrier ic.
For each carrier ic, selection of the transmitter with the highest , .
Analysis of candidate cells, (tx
BS
,ic).
For each pair (tx
BS
,ic), calculation of the uplink load factor:

corresponds to the load rise due to the mobile. For information on how this parameter is calculated, see "Admission
Control in the R99 Part" on page 252.
Rejection of bad candidate cells if the pilot is not received or if the uplink load factor is exceeded during the admission load
control (if simulation respects a loading factor constraint and M
b
was not connected in previous iteration)
If then (tx
BS
,ic) is rejected by M
b
If , then (tx
BS
,ic) is rejected by M
b
Else
Keep (tx
BS
,ic) as good candidate cell
For dual band terminals with the configuration 1 or terminals working on one frequency band only, if no good candidate
cell has been selected, M
b
has failed to be connected to the network and is rejected.
For dual band terminals with the configuration 2, if no good candidate cell has been selected, try to connect M
b
to
transmitters txi containing M
b
in their calculation area and working on the secondary frequency band supported by the M
b
’s
terminal (i.e. f2). If no good candidate cell has been selected, M
b
has failed to be connected to the network and is rejected.
For each NodeB having candidate cells, determination of the best carrier, ic
BS
, within the set of candidate cells
of the NodeB.
For DC-HSDPA users, this carrier is referred to as the "anchor" carrier.
If a given carrier is specified for the service requested by M
b
is the carrier specified for the service
Else the carrier selection mode defined for the site equipment is considered.
If carrier selection mode is “Min. UL Load Factor”
Q
req
UL
Q
req
DL
Q
pi l ot
k
txi i c Mb , , ( )
o µ
BTS
× P ×
c
txi M
b
i c , , ( )
P
tot
DL
txi i c , ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
Term
+ +
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- =
Q
pi l ot
k
txi i c Mb , , ( )
o µ
BTS
× P ×
c
txi M
b
i c , , ( )
I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) + + +
N
0
Term
1 o – ( ) µ
BTS
× P
c
txi M
b
i c , , ( ) × – + \ .
|
|
| |
----------------------------------------------------------------------------------------------------------------------------------------------------------------------- =
Q
pi l ot
k
txi M
b
i c , , ( ) tx
BS
M
b
( )
X
R99
UL
tx
BS
i c , ( ) ( )
k
I
tot
UL
tx
BS
i c , ( )
N
tot
UL
tx
BS
i c , ( )
----------------------------------- AX
UL
+ =
AX
UL
Q
pi l ot
k
tx
BS
M
b
i c , , ( ) Q
req
pi l ot
Mobi l i ty M
b
( ) ( ) <
X
R99
UL
tx
BS
i c , ( ) ( )
k
X
max
UL
>
i c
BS
M
b
( )
© Forsk 2010 AT283_TRG_E2 215
Chapter 6: UMTS HSPA Networks
is the carrier where we obtain the lowest
Else if carrier selection mode is “Min. DL Total Power”
is the carrier where we obtain the lowest
Else if carrier selection mode is “Random”
is randomly selected
Else if carrier selection mode is "Sequential"
is the first carrier where
Endif
is the best serving cell ( ) and its pilot quality is
In the following lines, we will consider as the carrier used by the best serving cell
Selection of the second serving cell for DC-HSDPA (Dual-cell HSDPA) users
If M
b
is a DC-HSDPA user and if tx
BS
supports DC-HSDPA and has several carriers, selection of the second carrier, ic
2
.
For each carrier other than the best serving carrier, ic
p
,

calculation of
Selection of the carrier, ic
2
, with the highest
If then (tx
BS
,ic
2
) is rejected by M
b
Else
Keep (tx
BS
,ic
2
) as second serving cell
Active Set Determination
For each station txi containing M
b
in its calculation area, using , and, if neighbours are used, neighbour of
Calculation of
If user selects “without Pilot”
Rejection of txi from the active set if difference with the best server is too high
If then txi is rejected
Else txi is included in the M
b
active set
Rejection of a station if the mobile active set is full
Station with the lowest in the active set is rejected
EndFor
Uplink Power Control
Calculation of the terminal power required by M
b
to obtain the R99 radio bearer:
For each cell (txi,ic) of the M
b
active set
Calculation of quality level on M
b
traffic channel at (txi,ic), with the minimum power allowed on traffic channel for the M
b
service
i c
BS
M
b
( ) X
R99
UL
tx
BS
i c , ( ) ( )
k
i c
BS
M
b
( ) P
tx
tx
BS
i c , ( )
k
i c
BS
M
b
( )
i c
BS
M
b
( ) X
R99
UL
tx
BS
i c , ( ) ( )
k
X
max
UL
s
tx
BS
i c
BS
( , )
k
M
b
( ) BestCel l
k
M
b
( ) Q
pi l ot
k
max
M
b
( )
i c
Q
pi l ot
k
tx
BS
i c
p
M
b
, , ( )
Q
pi l ot
k
tx
BS
i c
p
M
b
, , ( )
Q
pi l ot
k
tx
BS
i c
2
M
b
, , ( ) Q
req
pi l ot
Mobi l i ty M
b
( ) ( ) <
i c
BestCel l
k
M
b
( )
Q
pi l ot
k
txi M
b
i c , , ( )
o µ
BTS
× P ×
c
txi M
b
i c , , ( )
P
tot
DL
txi i c , ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N
0
Term
+ + + +
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- =
Q
pi l ot
k
txi M
b
i c , , ( )
o µ
BTS
× P ×
c
txi M
b
i c , , ( )
I
i ntra
DL
txi i c , ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) + + +
N
0
Term
1 o – ( ) µ
BTS
× P
c
txi M
b
i c , , ( ) × – + \ .
|
|
| |
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- =
Q
pi l ot
k
max
M
b
( ) Q
pi l ot
k
txi M
b
i c , , ( ) – AS_Th BestCel l
k
M
b
( ) ( ) >
Q
pi l ot
k
P
term
R99 req –
M
b
i c , ( )
k
P
b R99 –
UL
txi M
b
i c , , ( )
P
term R99 –
req
M
b
i c , ( )
k 1 –
L
T
txi M
b
, ( )
--------------------------------------------------------- =
P
b DPDCH –
UL
txi M
b
i c , , ( ) P
b R99 –
UL
txi M
b
i c , , ( ) 1 r
c
UL
– ( ) × =
P
b DPCCH –
UL
txi M
b
i c , , ( ) P
b R99 –
UL
txi M
b
i c , , ( ) r
c
UL
× =
216 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
if the user is active,
if the user is inactive,
If user selects "Total noise",
End For
If (M
b
is in not in handoff)
Else if (M
i
is in softer handoff)
Else if (M
b
is in soft, or softer/soft without MRC)
Else if (M
b
is in soft/soft)
Else if (M
b
is in softer/soft with MRC)
End If
If compressed mode is operated,
If then
If then M
b
cannot select any cell and its active set is cleared
If then M
b
cannot be connected
Endif
Downlink Power Control
If (mobile does not use a packet switched service that is inactive on the downlink)
Note:
• Compressed mode is operated if:
- M
i
and S
j
support compressed mode,
And
- Either if the Ec/I0 Active option is selected,
- Or if the RSCP Active option is selected.
P
b R99 –
UL
txi M
b
i c , , ( ) P
b DPCCH –
UL
txi M
b
i c , , ( ) P
b DPDCH –
UL
txi M
b
i c , , ( ) + =
P
b R99 –
UL
txi M
b
i c , , ( ) P
b DPCCH –
UL
txi M
b
i c , , ( ) =
Q
tch
UL
txi M
b
i c , , ( )
k
µ
term
P ×
b DPDCH –
UL
txi M
b
i c , , ( )
k
N
tot
UL
txi i c , ( ) 1 F
MUD
Tx
– ( ) µ
term
× P
b R99 –
UL
txi M
b
i c , , ( )
k 1 –
× –
-------------------------------------------------------------------------------------------------------------------------------------------------------- G
p
UL
Servi ce M
b
( ) ( ) G
di v
UL
× × =
Q
tch
UL
txi M
b
i c , , ( )
k
µ
term
P ×
b DPDCH –
UL
txi M
b
i c , , ( )
k
N
tot
UL
txi i c , ( )
--------------------------------------------------------------------------------- G
p
UL
Servi ce M
b
( ) ( ) G
di v
UL
× × =
Q
k
UL
M
b
( ) Q
tch
UL
txi M
b
i c , , ( )
k
=
Q
k
UL
M
b
( ) f
rake efficiency
UL
Q
tch
UL
txi M
b
i c , , ( )
k
txi Acti veSet e
¿
× =
Q
k
UL
M
b
( )
Max


txi Acti veSet e
Q
tch
UL
txi M
b
i c , , ( )
k
( ) G
macro di versi ty –
UL
( )
2 links
× =
Q
k
UL
M
b
( )
Max


txi Acti veSet e
Q
tch
UL
txi M
b
i c , , ( )
k
( ) G
macro di versi ty –
UL
( )
3 links
× =
Q
k
UL
M
b
( ) Max f
rake efficiency
UL
Q
tch
UL
i c ( )
txi Acti veSet e
samesi te ( )
¿
× Q
other site
tch
UL
i c ( ) ,
\ .
|
|
|
| |
G
macro di versi ty –
UL
( )
2 links
× =
P
term R99 –
req
M
b
i c , ( )
k
Q
req
UL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( )
Q
k
UL
M
b
( )
-------------------------------------------------------------------------------------------- P
term R99 –
req
M
b
i c , ( )
k 1 –
× =
Q
pi l ot
k
Resul ti ng
txi M
b
i c , , ( ) Q
pi l ot
CM acti vati on –
s
P
c
txi M
b
i c , , ( ) RSCP
pi l ot
CM acti vati on –
s
P
term R99 –
req
M
b
i c , ( )
k
Q
req
UL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( ) AQ
req
UL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( ) ( ) ×
Q
k
UL
M
b
( )
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- P
term R99 –
req
M
b
i c , ( )
k 1 –
× =
P
term R99 –
req
M
b
i c , ( )
k
P
term
mi n
M
b
( ) < P
term R99 –
req
M
b
i c , ( )
k
P
term
mi n
txi M ,
b
( ) =
P
term R99 –
req
M
b
i c , ( )
k
P
term
max
M
b
( ) >
R
nomi nal
UL
M
b
( ) R
max
UL
txi i c , ( ) >
© Forsk 2010 AT283_TRG_E2 217
Chapter 6: UMTS HSPA Networks
For each cell (tx
i
,ic) in M
b
active set
Calculation of quality level on (txi,ic) traffic channel at M
b
with the minimum power allowed on traffic channel for the M
b
service
If the user selects the option "Total noise"
End For
Do
For each cell (txi,ic) in M
b
active set
Calculation of the required power for DL traffic channel between (txi,ic) and M
b
:
If compressed mode is operated.
If then is set to
Recalculation of a decreased (a part of the required quality is managed by the cells set to )
If the user is inactive, then his contribution to interference in the calculation of is .
EndFor
While and M
b
active set is not empty
If then M
b
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)
Note:
• Compressed mode is operated if:
- M
i
and S
j
support compressed mode,
And
- Either if the Ec/I0 Active option is selected,
- Or if the RSCP Active option is selected.
P
b
DL
txi M
b
i c , , ( )
P
tch
mi n
Servi ce M
b
( ) ( )
L
T
txi M
b
, ( )
----------------------------------------------------- =
Q
tch
DL
txi M
b
i c , , ( )
k
µ
BTS
P ×
b
DL
txi M
b
i c , , ( )
k
N
tot
DL
i c ( ) 1 F
ortho
– ( ) µ
BTS
× P
b
DL
txi M
b
i c , , ( )
k 1 –
× –
------------------------------------------------------------------------------------------------------------------------------------ G
p
DL
Servi ce M
b
( ) ( ) × G
di v
DL
× =
Q
tch
DL
txi M
b
i c , , ( )
k
µ
BTS
P ×
b
DL
txi M
b
i c , , ( )
k
N
tot
DL
i c ( )
-------------------------------------------------------------- G
p
DL
Servi ce M
b
( ) ( ) G
di v
DL
× × =
Q
k
DL
M
b
( ) f
rake efficiency
DL
Q
tch
DL
txi M
b
i c , , ( )
k
txi Acti veSet e
¿
× =
P
tch
req
txi M
b
i c , , ( )
k
Q
req
DL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( )
Q
k
DL
M
b
( )
-------------------------------------------------------------------------------------------- P
tch
mi n
Servi ce M
b
( ) ( ) × =
P
tch
req
txi M
b
i c , , ( )
k
Q
req
DL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( ) AQ
req
DL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( ) ( ) ×
Q
k
DL
M
b
( )
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- P
tch
mi n
Servi ce M
b
( ) ( ) × =
Q
pi l ot
k
Resul ti ng
txi M
b
i c , , ( ) Q
pi l ot
CM acti vati on –
s
P
c
txi M
b
i c , , ( ) RSCP
pi l ot
CM acti vati on –
s
P
tch
req
txi M
b
i c , , ( )
k
P
tch
max
Servi ce M
b
( ) ( ) > txi i c , ( ) P
tch
max
Q
req
DL
P
tch
max
P
b
DL
txi M
b
i c , , ( )
P
tch
req
Servi ce M
b
( ) ( )
L
T
txi M
b
, ( )
---------------------------------------------------- =
Q
tch
DL
txi M
b
i c , , ( )
k
µ
BTS
P ×
b
DL
txi M
b
i c , , ( )
N
tot
DL
i c ( ) 1 F
ortho
– ( ) µ
BTS
× P
b
DL
txi M
b
i c , , ( ) × –
--------------------------------------------------------------------------------------------------------------------------- G
p
DL
Servi ce M
b
( ) ( ) × G
di v
DL
× =
N
tot
DL
i c ( ) P
b
DL
txi M
b
i c , , ( ) r
c
DL
×
Q
k
DL
M
b
( ) f
rake efficiency
DL
Q
tch
DL
txi M
b
i c , , ( )
k
txi Acti veSet e
¿
× =
Q
k
DL
M
b
( ) Q
req
DL
Servi ce M
b
( ) Mobi l i ty M
b
( ) , ( ) <
R
nomi nal
DL
M
b
( ) R
max
DL
txi i c , ( ) >
218 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Update of
EndFor
For each mobile M
i
Update of
EndFor
EndFor
Control of Radio Resource Limits (OVSF Codes, Cell Power, Channel Elements, Iub Backhaul
Throughput)
For each cell (txi,ic)
While
Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
For each cell (txi,ic)
While
Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
For each NodeB, Ni
While
Rejection of the mobile with the lowest service priority starting from the last admitted
While
Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
For each NodeB, Ni
While
Rejection of the mobile with the lowest service priority starting from the last admitted
While
Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
Uplink Load Factor Control
For each cell (txi,ic) with
Rejection of the mobile with the lowest service priority starting from the last admitted
EndFor
While at least one cell with exists.
6.4.2.3 HSDPA Part of the Algorithm
Packet (HSDPA) and packet (HSPA) service users active on DL as well as all packet (HSPA - Constant Bit Rate) service
users (i.e., active and incative), 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 ( ) is the sum of the transmitted R99 power, the HSUPA power and the
transmitted HSDPA power.
• In case of a static HSDPA power allocation strategy, Atoll checks in the simulation that:
N
tot
UL
txi i c , ( )
N
tot
DL
i c ( )
P
tx
txi i c , ( )
k
P
max
------------------------------ %Power
max
DL
>
N
Codes
txi i c , ( )k N
max
Codes
txi i c , ( ) >
N
CE DL –
N
i
( )
k
N
max
CE DL –
N
i
( ) >
N
CE UL –
N
i
( )
k
N
max
CE UL –
N
i
( ) >
T
Iub
DL
N
I
( )
k
T
Iub m – ax
DL
N
I
( ) >
T
Iub
UL
N
I
( )
k
T
Iub m – ax
UL
N
I
( ) >
X
R99
UL
txi i c , ( ) X
max
UL
>
X
R99
UL
txi i c , ( ) X
max
UL
>
P
tx
i c ( )
P
tx
i c ( ) P
tx R99 –
i c ( ) P
tx H – SDPA
i c ( ) P
HSUPA
i c ( ) + + =
P
tx
i c ( ) P
max
i c ( ) %Power
max
DL
× s
© Forsk 2010 AT283_TRG_E2 219
Chapter 6: UMTS HSPA Networks
where:
is the maximum DL load allowed.
Therefore, if the maximum DL load is set to 100%, we have:
• In case of dynamic HSDPA power allocation strategy, Atoll checks in the simulation that:
And it calculates the available HSDPA power as follows:
6.4.2.3.2 Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users
The number of HS-SCCH channels ( ) is the maximum number of HS-SCCH channels that the cell can manage.
This parameter is used to manage the number of packet (HSDPA) and packet (HSPA) service users simultaneously
connected to an HSDPA bearer. This parameter is not taken into account for packet (HSPA - Constant Bit Rate) service
users as HS-SCCH-less operation (i.e., HS-DSCH transmissions without any accompanying HS-SCCH) is performed.
Each packet (HSDPA) and packet (HSPA) service user consumes one HS-SCCH channel. Therefore, at a time (over a
transmission time interval), the number of these users connected to an HSDPA bearer cannot exceed the number of HS-
SCCH channels per cell.
The maximum number of HSDPA users ( ) corresponds to the maximum number of HSDPA bearer users that the cell
can support. Here, all HSDPA bearer users, i.e., packet (HSDPA) service users, packet (HSPA) service users and packet
(HSPA - Constant Bit Rate) service users, are taken into consideration.
Let us assume there are 30 HSDPA bearer users in the cell:
• 10 packet (HSPA - Constant Bit Rate) service users with any activity status.
• 20 packet (HSDPA) and packet (HSPA) service users active on DL.
All users 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 25.
The scheduler manages the maximum number of users within each cell. Packet (HSPA - Constant Bit Rate) service users
have the highest priority and are processed first, in the order established during the generation of the user distribution.
After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler ranks the remaining HSDPA bearer
users (i.e., packet (HSDPA) and packet (HSPA) service users) according to the selected scheduling technique. Users are
treated as described in the figure below.
• All packet (HSPA - Constant Bit Rate) service users may be served if there are enough HSDPA power, Iub
backhaul throughput and OVSF codes available in order for them to obtain the lowest HSDPA bearer that provides
a RLC peak rate higher or equal to the guaranted bit rate defined for the service. In this case, they will be
connected. Else, they will be rejected.
• Then, among the packet (HSDPA) and packet (HSPA) service users:
- The first four users may be simultaneously served if there are enough HSDPA power, Iub backhaul throughput
and OVSF codes available in order for them to obtain an 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 "HS-SCCH Channels Saturation".
- Finally, the last five users will be rejected beacuse the maximum number of HSDPA user has been fixed to
25. Their connection status will be "HSDPA Scheduler Saturation".
6.4.2.3.3 HSDPA Bearer Allocation Process
The HSDPA bearer allocation process depends on the type of service requested by the user. As explained before, packet
(HSPA - Constant Bit Rate) service users have the highest priority and are processed first, in the order established during
the generation of the user distribution. After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler
ranks the remaining HSDPA bearer users (i.e., packet (HSDPA) and packet (HSPA) service users) and shares the cell
radio resources between them.
Packet (HSPA - Constant Bit Rate) Service Users
Let us focus on the ten packet (HSPA - Constant Bit Rate) service users mentionned in the example of the previous
paragraph "Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users" on page 219. Fast link
%Power
max
DL
P
tx
i c ( ) P
max
i c ( ) s
P
tx R99 –
i c ( ) P
HSUPA
i c ( ) + P
max
i c ( ) %Power
max
DL
× s
P
HSDPA
i c ( ) P
max
i c ( ) P
Headroom
i c ( ) – P
tx R99 –
i c ( ) – P
HSUPA
i c ( ) – =
Figure 6.3Connection status of HSDPA bearer users
n
HS SCCH –
n
max
220 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
adaptation is carried out on these users in order to determine if they can obtain an HSDPA bearer that provides a RLC
peak rate higher or equal to the service guaranteed bit rate. As HS-SCCH less operation is performed, only HSDPA
bearers using the QPSK modulation and two HS-PDSCH channels at the maximum can be selected and allocated to the
users. The users are processed in the order established during the generation of the user distribution and the cell’s
available HSDPA power is shared between them as explained below. Several Packet (HSPA - Constant Bit Rate) service
users can share the same HSDPA bearer. Then, Atoll calculates the HSDPA bearer consumption ( in %) for each user
and takes into account this parameter when it determines the resources consumed by the user (i.e., the HSDPA power
used, the number of OVSF codes and the Iub backhaul throughput).
In the bearer allocation process shown below, the 10 packet (HSPA - Constant Bit Rate) service users are represented by
M
j
, with j = 1 to 10. And, the initial values of their respective HSDPA powers is 0, i.e. P
HSDPA
(B(M
X
)) = 0, where X = 0 to
10. 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.
Packet (HSDPA) and Packet (HSPA) Service Users
After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler share the cell’s remaining resources
between packet (HSDPA) and packet (HSPA) service users. Let us focus on the packet (HSDPA) and packet (HSPA)
service users, especially on the first four users mentionned in the example of the previous paragraph, "Number of HS-
SCCH Channels and Maximum Number of HSDPA Bearer Users" on page 219. A new fast link adaptation is carried out
on these users in order to determine if they can obtain an HSDPA bearer. They are processed in the order defined by the
For the user, M
j
, with j varying from 1 to 10:
Figure 6.4HSDPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users
C
© Forsk 2010 AT283_TRG_E2 221
Chapter 6: UMTS HSPA Networks
scheduler and the cell’s HSDPA power available after all Packet (HSPA - Constant Bit Rate) service users have been
served is shared between them as explained below.
In the bearer allocation process shown below, the 4 packet (HSDPA) and packet (HSPA) service users are represented
by M
j
, with j = 1 to 4. And, the initial values of their respective HSDPA powers is 0, i.e. P
HSDPA
(B(M
X
)) = 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.
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, 16QAM, 64QAM), 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
For the user, M
j
, with j varying from 1 to 4:
Figure 6.5HSDPA Bearer Allocation Process for Packet (HSDPA) and Packet (HSPA) Service Users
222 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
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
Let us assume the following notation: corresponds to the CPICH quality.
Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total
noise.
Therefore, we have:
for the total noise option,
And
for the without useful signal option.
With
ic
adj
is a carrier adjacent to ic.
is the interference reduction factor, defined between ic and ic
adj
and set to a value different from 0.
is the inter-technology interference at the receiver on ic.
is the interfering carrier of an external transmitter
is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming
the frequency gap between (external network) and .
(
3
)
, and are defined in "Inputs" on page 198.
3. In the HSDPA coverage prediction, is calculated as follows:
)
Ec
Nt
------- i c ( )
\ .
| |
pi l ot
Ec
Nt
------- i c ( )
\ .
| |
pi l ot
µ
BTS
o × P
c
i
i c ( ) ×
N
tot
DL
i c ( )
---------------------------------------------- =
Ec
Nt
------- i c ( )
\ .
| |
pi l ot
µ
BTS
o × P
c
i
i c ( ) ×
N
tot
DL
i c ( ) 1 o – ( ) µ
BTS
P
c
i
i c ( ) × × –
--------------------------------------------------------------------------------------- =
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
term
+ + =
I
i ntra
DL
i c ( )
P
tot
DL
i c ( )
txi
µ
BTS
1 F –
MUD
term
( ) 1 o – ( ) × × +
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
µ
BTS
– ×
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
× =
I
extra
DL
i c ( ) P
tot
DL
i c ( )
txj j i = ,
¿
=
I
i nter carri er –
DL
i c ( )
P
tot
DL
i c
adj
( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
--------------------------------------- =
RF i c i c
adj
, ( )
I
i nter techno y log –
DL
i c ( )
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
=
i c
i
i
th
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
P
c
i
i c ( )
P
pi l ot
i c ( )
L
T
i
----------------------- =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
BTS
o N
0
term
© Forsk 2010 AT283_TRG_E2 223
Chapter 6: UMTS HSPA Networks
2. CPICH CQI Determination
Let us assume the following notation: corresponds to the CPICH CQI. is read in the table
. This table is defined for the terminal reception equipment and the selected mobility.
3. HS-PDSCH Quality Calculation
Atoll proceeds as follows:
1
st
step: Atoll calculates the HS-SCCH power ( ).
is the HS-SCCH power on carrier ic. It is either fixed by the user (when the option “HS-SCCH Power
Dynamic Allocation”in the cell property dialogue is unchecked) or dynamically calculated (when the option “HS-SCCH
Power Dynamic Allocation” is selected).
In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ). It
is specified in mobility properties.
We have:
for the total noise option,
And
for the without useful signal option.
With
ic
adj
is a carrier adjacent to ic.
is the interference reduction factor, defined between ic and ic
adj
and set to a value different from 0.
is the inter-technology interference at the receiver on ic.
is the interfering carrier of an external transmitter
is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming
the frequency gap between (external network) and .
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:
I
i ntra
DL
i c ( )
P
max
i c ( )
L
T
----------------------- µ
BTS
1 F –
MUD
term
( ) 1 o – ( ) × × +
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
µ
BTS
– ×
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
× =
CQI ( )
pi l ot
CQI ( )
pi l ot
CQI ( )
pi l ot
f
Ec
Nt
------- i c ( )
\ .
| |
pi l ot
\ .
| |
=
P
HS SCCH –
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( )
------------------------------------ =
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( ) 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) × µ
BTS
P
c
i
i c ( ) × × –
--------------------------------------------------------------------------------------------------------------------------------------- =
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N
0
term
+ + + + =
I
i ntra
DL
i c ( )
P
tot
DL
i c ( )
txi
µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
µ
BTS
– ×
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
× =
I
extra
DL
i c ( ) P
tot
DL
i c ( )
txj j i = ,
¿
=
I
i nter carri er –
DL
i c ( )
P
tot
DL
i c
adj
( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
--------------------------------------- =
RF i c i c
adj
, ( )
I
i nter techno y log –
DL
i c ( )
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
=
i c
i
i
th
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
224 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
and
(
4
)
, , and are defined in "Inputs" on page 198.
Therefore,
for the total noise option,
And
for the without useful signal option.
2
nd
step: Atoll calculates the HS-PDSCH power ( ).
is the power available for HSDPA on the carrier ic. This parameter is either a simulation output, or a user-
defined cell input.
Therefore, we have:
is the number of HS-SCCH channels.
3
rd
step: Then, Atoll evaluates the HS-PDSCH quality
Let us assume the following notation: corresponds to the HS-PDSCH quality.
We have:
for the total noise option,
And
for the without useful signal option.
Here, Atoll works on the assumption that five HS-PDSCH channels are used (n=5).
With
4. In the HSDPA coverage prediction, is calculated as follows:
)
P
c
i
i c ( )
P
HS SCCH –
i c ( )
L
T
i
--------------------------------------- =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
BTS
F
ortho
F
MUD
term
N
0
term
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
N
tot
DL
i c ( ) ×
µ
BTS
---------------------------------------------------------------------------
\ .
|
|
|
| |
L
T
i
× =
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
N
tot
DL
i c ( ) ×
µ
BTS
1 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) ×
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
× +
\ .
| |
×
----------------------------------------------------------------------------------------------------------------------------------------------------------
\ .
|
|
|
| |
L
T
i
× =
P
HS PDSCH –
P
HSDPA
i c ( )
P
HSDPA
i c ( ) P
HS PDSCH –
i c ( ) n
HS SCCH –
P ×
HS SCCH –
i c ( ) + =
P
HS PDSCH –
i c ( ) P
HSDPA
i c ( ) n
HS SCCH –
P ×
HS SCCH –
i c ( ) – =
n
HS SCCH –
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( )
------------------------------------ =
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( ) 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) × µ
BTS
P
c
i
i c ( )
n
----------------- × × –
--------------------------------------------------------------------------------------------------------------------------------------- =
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
term
+ + =
I
i ntra
DL
i c ( )
P
tot
DL
i c ( )
txi
µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
µ
BTS
– ×
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
× =
I
extra
DL
i c) ( ) P
tot
DL
i c ( )
txj j i = ,
¿
=
I
i nter carri er –
DL
i c ( )
P
tot
DL
i c
adj
( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
--------------------------------------- =
© Forsk 2010 AT283_TRG_E2 225
Chapter 6: UMTS HSPA Networks
ic
adj
is a carrier adjacent to ic.
is the interference reduction factor, defined between ic and ic
adj
and set to a value different from 0.
is the inter-technology interference at the receiver on ic.
is the interfering carrier of an external transmitter
is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming
the frequency gap between (external network) and .
And
(
5
)
, , and are defined in "Inputs" on page 198.
4. HS-PDSCH CQI Determination
The best bearer that can be used depends on the HS-PDSCH CQI. Let us assume the following notation:
corresponds to the HS-PDSCH CQI. Atoll calculates as follows:
5. HSDPA Bearer Selection
Atoll selects the HSDPA bearer associated to this CQI (in the table Best Bearer=f(HS-PDSCH CQI) defined for the
terminal reception equipment and the user mobility) and compatible with the user equipment and cell capabilities.
HSDPA bearers can be classified into two categories:
• HSDPA bearers using QPSK and 16QAM modulations: They can be selected for all users connected to HSPA and
HSPA+ capable cells. The number of HS-PDSCH channels required by the bearer must not exceed the maximum
number of HS-PDSCH codes available for the cell.
For packet (HSPA - Constant Bit Rate) service users, HS-SCCH-less operation (i.e., HS-DSCH transmissions
without any accompanying HS-SCCH) is performed. In this case, the UE is not informed about the transmission
format and has to revert to blind decoding of the transport format used on the HS-DSCH. Complexity of blind
detections in the UE is decreased by limiting the transmission formats that can be used (i.e., the HSDPA bearers
available). Therefore, only HSDPA bearers using the QPSK modulation and two HS-PDSCH channels at the
maximum can be selected and allocated to these users. Additionally, the selected HSDPA bearer must provide a
RLC peak rate higher or equal to the guaranted bit rate defined for the service.
• HSDPA bearers using 64QAM modulation (improvement introduced by the release 7 of the 3GPP UTRA
specifications, referred to as HSPA+): These HSDPA bearers can be allocated to packet (HSDPA) and packet
(HSPA) users connected to cells with HSPA+ capabilities only. The number of HS-PDSCH channels required by
the bearer must not exceed the maximum number of HS-PDSCH codes available for the cell. These HSDPA
bearers cannot be allocated to packet (HSPA - Constant Bit Rate) service users.
Atoll considers an HSDPA bearer as compatible with the user equipment if:
• The transport block size does not exceed the maximum transport block size supported by the user equipment.
5. In the HSDPA coverage prediction, is calculated as follows:
)
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:
RF i c i c
adj
, ( )
I
i nter techno y log –
DL
i c ( )
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
=
i c
i
i
th
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
P
c
i
i c ( )
P
HS PDSCH –
i c ( )
L
T
i
------------------------------------------ =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
BTS
F
ortho
F
MUD
term
N
0
term
I
i ntra
DL
i c ( )
P
max
i c ( )
L
T
----------------------- µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
µ
BTS
– ×
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
× =
CQI ( )
HS PDSCH –
CQI ( )
HS PDSCH –
CQI ( )
HS PDSCH –
CQI ( )
pi l ot
P
pi l ot
– P
HS PDSCH –
+ =
226 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
• The number of HS-PDSCH channels required by the bearer does not exceed the maximum number of HS-PDSCH
channels that the terminal can use.
• The modulation is supported by the user equipment.
When there are several HSDPA bearers compatible, Atoll selects the HSDPA bearer that provides the highest RLC peak
rate. When several HSDPA bearers can supply the same RLC peak rate, Atoll chooses the HSDPA bearer with the
highest modulation scheme. Finally, if no HSDPA bearer is compatible, Atoll allocates a lower HSDPA bearer compatible
with the user equipment and cell capabilities which needs fewer resources.
Let’s consider the following examples.
Example1: One packet (HSDPA) user with category 13 user equipment and a 50km/h mobility.
The user equipment capabilities are:
• Maximum transport block size: 35280 bits
• Maximum number of HS-PDSCH channels: 15
• Highest modulation supported: 64QAM
• MIMO Support: No
The cell to which the user is connected supports HSPA+ functionalities (i.e. 64QAM modulation in the DL and MIMO
systems) and the maximum number of HS-PDSCH channels is 15.
1
st
case: The CQI experienced by the user equals 26. Therefore, Atoll can choose between two HSDPA bearers, the
bearer indexes 26 and 31.
Characteristics of the bearer index 26 are:
• Transport block size: 17237 bits
• Number of HS-PDSCH channels used: 12
• 16QAM modulation is used
• RLC Peak Rate: 8.32 Mb/s
Characteristics of the bearer index 31 are:
• Transport block size: 15776 bits
• Number of HS-PDSCH channels used: 10
• 64QAM modulation is used
• RLC Peak Rate: 7.36 Mb/s
Both HSDPA bearers are compatible with the user equipment and cell capabilities. Atoll selects the HSDPA bearer that
provides the highest RLC peak rate, i.e. the bearer index 26.
Figure 6.6HSDPA UE Categories Table
© Forsk 2010 AT283_TRG_E2 227
Chapter 6: UMTS HSPA Networks
2
nd
case: The CQI experienced by the user equals 27. Therefore, Atoll can choose between two HSDPA bearers, the
bearer indexes 27 and 32.
Characteristics of the bearer index 27 are:
• Transport block size: 21754 bits
• Number of HS-PDSCH channels used: 15
• 16QAM modulation is used
• RLC Peak Rate: 10.24 Mb/s
Characteristics of the bearer index 32 are:
• Transport block size: 21768 bits
• Number of HS-PDSCH channels used: 12
• 64QAM modulation is used
• RLC Peak Rate: 10.24 Mb/s
Both HSDPA bearers are compatible with the user equipment and cell capabilities and the RLC peak rate they provide is
the same. Atoll selects the HSDPA bearer using the highest modulation scheme, i.e. the bearer index 32.
Example 2: One packet (HSDPA) user experiencing a CQI of 26.
Therefore, Atoll can choose between two HSDPA bearers, the bearer indexes 26 and 31.
Characteristics of the bearer index 26 are:
• Transport block size: 17237 bits
• Number of HS-PDSCH channels used: 12
• 16QAM modulation is used
• RLC Peak Rate: 8.32 Mb/s
Characteristics of the bearer index 31 are:
• Transport block size: 15776 bits
• Number of HS-PDSCH channels used: 10
• 64QAM modulation is used
• RLC Peak Rate: 7.36 Mb/s
1
st
case: The user equipment category is 9. The cell to which the user is connected supports HSPA+ functionalities (i.e.
64QAM modulation in the DL and MIMO systems) and the maximum number of HS-PDSCH channels is 15.
The user equipment characteristics are the following:
• Maximum transport block size: 20251 bits
• Maximum number of HS-PDSCH channels: 15
• Highest modulation supported: 16QAM
• MIMO Support: No
The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the terminal. Only the
bearer index 26 is compatible with the user equipment capabilities. Atoll selects it.
Figure 6.7HSDPA Radio Bearers Table
228 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
2
nd
case: The user equipment category is 8. The cell to which the user is connected supports HSPA+ functionalities (i.e.
64QAM modulation in the DL and MIMO systems) and the maximum number of HS-PDSCH channels is 15.
The user equipment characteristics are the following:
• Maximum transport block size: 14411 bits
• Maximum number of HS-PDSCH channels: 10
• Highest modulation supported: 16QAM
• MIMO Support: No
Here, none of HSDPA bearers are compatible with the user equipment capabilities.
The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the terminal. With the
bearer index 26, the number of HS-PDSCH channels (12) exceeds the maximum number of HS-PDSCH channels the
terminal can use (10), and the transport block size (17237 bits) exceeds the maximum transport block size (14411 bits)
the terminal can carried.
In the HSDPA Radio Bearer table, Atoll selects a lower HSDPA bearer compatible with cell and UE category capabilities.
It selects the bearer index 25.
• The number of HS-PDSCH channels (10) does not exceed the maximum number of HS-PDSCH channels the
terminal can use (10) and the maximum number of HS-PDSCH channels available at the cell level (15),
• The transport block size (14411 bits) does not exceed the maximum transport block size (14411 bits) the terminal
can carried.
• 16QAM modulation is supported by the terminal and the cell.
3
rd
case: The user equipment category is 13. The cell to which the user is connected supports HSPA functionalities and
the maximum number of HS-PDSCH channels is 15.
The user equipment capabilities are:
• Maximum transport block size: 35280 bits
• Maximum number of HS-PDSCH channels:15
• Highest modulation supported: 64QAM
• MIMO Support: No
The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the cell. On the other
hand, the bearer index 26 is compatible with cell and UE category capabilities. Therefore, it is allocated.
6. 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:
1
st
step: Atoll calculates the HS-SCCH power ( ).
is the HS-SCCH power on carrier ic. It is either fixed by the user (when the option “HS-SCCH Power
Dynamic Allocation”in the cell property dialogue is unchecked) or dynamically calculated (when the option “HS-SCCH
Power Dynamic Allocation” is selected).
In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ). It
is specified in mobility properties.
We have:
for the total noise option,
And
for the without useful signal option.
With
P
HS SCCH –
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( )
------------------------------------ =
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( ) 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) × µ
BTS
P
c
i
i c ( ) × × –
--------------------------------------------------------------------------------------------------------------------------------------- =
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
term
+ + =
I
i ntra
DL
i c ( )
P
tot
DL
i c ( )
txi
µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
µ
BTS
– ×
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
× =
© Forsk 2010 AT283_TRG_E2 229
Chapter 6: UMTS HSPA Networks
ic
adj
is a carrier adjacent to ic.
is the interference reduction factor, defined between ic and ic
adj
and set to a value different from 0.
is the inter-technology interference at the receiver on ic.
is the interfering carrier of an external transmitter
is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming
the frequency gap between (external network) and .
And
(
6
)
, , and are defined in "Inputs" on page 198.
Therefore,
for the total noise option,
And
for the without useful signal option.
2
nd
step: Atoll calculates the HS-PDSCH power ( )
is the power available for HSDPA on the carrier ic. This parameter is either a simulation output, or a user-
defined cell input.
Therefore, we have:
is the number of HS-SCCH channels.
3
rd
step: Then, Atoll evaluates the HS-PDSCH quality
Let us assume the following notation: corresponds to the HS-PDSCH quality.
Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total
noise.
We have:
6. In the HSDPA coverage prediction, is calculated as follows:
)
I
extra
DL
i c ( ) P
tot
DL
i c ( )
txj j i = ,
¿
=
I
i nter carri er –
DL
i c ( )
P
tot
DL
i c
adj
( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
--------------------------------------- =
RF i c i c
adj
, ( )
I
i nter techno y log –
DL
i c ( )
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
=
i c
i
i
th
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
P
c
i
i c ( )
P
HS SCCH –
i c ( )
L
T
i
--------------------------------------- =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
BTS
F
ortho
F
MUD
term
N
0
term
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
N
tot
DL
i c ( ) ×
µ
BTS
---------------------------------------------------------------------------
\ .
|
|
|
| |
L
T
i
× =
P
HS SCCH –
i c ( )
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
N
tot
DL
i c ( ) ×
µ
BTS
1 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) ×
Ec
Nt
------- i c ( )
\ .
| |
HS SCCH –
req
× +
\ .
| |
×
----------------------------------------------------------------------------------------------------------------------------------------------------------
\ .
|
|
|
| |
L
T
i
× =
P
HS PDSCH –
P
HSDPA
i c ( )
P
HSDPA
i c ( ) P
HS PDSCH –
i c ( ) n
HS SCCH –
P ×
HS SCCH –
i c ( ) + =
P
HS PDSCH –
i c ( ) P
HSDPA
i c ( ) n
HS SCCH –
P ×
HS SCCH –
i c ( ) – =
n
HS SCCH –
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
230 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
for the total noise option,
And
for the without useful signal option.
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
ic
adj
is a carrier adjacent to ic.
is the interference reduction factor, defined between ic and ic
adj
and set to a value different from 0.
is the inter-technology interference at the receiver on ic.
is the interfering carrier of an external transmitter
is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming
the frequency gap between (external network) and .
And
(
7
)
, , and are defined in "Inputs" on page 198.
2. HS-PDSCH CQI Determination
7. In the HSDPA coverage prediction, is calculated as follows:
)
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:
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( )
------------------------------------ =
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
µ
BTS
P
c
i
i c ( ) ×
N
tot
DL
i c ( ) 1 F
ortho
– ( ) 1 F
MUD
term
– ( ) × µ
BTS
P
c
i
i c ( )
n
----------------- × × –
--------------------------------------------------------------------------------------------------------------------------------------- =
N
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) N + +
0
term
+ + =
I
i ntra
DL
i c ( )
P
tot
DL
i c ( )
txi
µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
µ
BTS
– ×
P
tot
DL
i c ( )
txi
P
SCH
i c ( )
L
T
------------------------ –
\ .
|
|
| |
× =
I
extra
DL
i c) ( ) P
tot
DL
i c ( )
txj j i = ,
¿
=
I
i nter carri er –
DL
i c ( )
P
tot
DL
i c
adj
( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
--------------------------------------- =
RF i c i c
adj
, ( )
I
i nter techno y log –
DL
i c ( )
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
i c
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
=
i c
i
i
th
ICP
i c
i
i c ,
Tx m ,
i c
i
i c
P
c
i
i c ( )
P
HS PDSCH –
i c ( )
L
T
i
------------------------------------------ =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Ec Io –
× × × ×
G
Tx
G
term
×
--------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
BTS
F
ortho
F
MUD
term
N
0
term
I
i ntra
DL
i c ( )
P
max
i c ( )
L
T
----------------------- µ
BTS
1 F –
MUD
term
( ) 1 F
ortho
– ( ) × × +
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
µ
BTS
– ×
P
max
i c ( ) P –
SCH
i c ( )
L
T
------------------------------------------------------
\ .
| |
× =
© Forsk 2010 AT283_TRG_E2 231
Chapter 6: UMTS HSPA Networks
Let us assume the following notation: corresponds to the HS-PDSCH CQI. is read in
the table . This table is defined for the terminal reception equipment and the
specified mobility.
3. HSDPA Bearer Selection
The bearer is selected as described in "HSDPA Bearer Selection" on page 225.
6.4.2.3.5 MIMO Modelling
MIMO - Transmit Diversity
If the user is connected to a cell that supports HSPA+ with transmit diversity and if he has a MIMO-capable terminal (i.e.,
a terminal with an HSDPA UE category supporting MIMO), he will benefit from downlink diversity gain on the HS-PDSCH
Ec/Nt.
in dB
Where
is the downlink transmit diversity gain (in dB) corresponding to the numbers of transmission and reception antenna
ports (respectively defined in the transmitter and terminal properties).
is the additional diversity gain in downlink (in dB). It is defined for the clutter class of the user.
MIMO - Spatial Multiplexing
If the user is connected to a cell that supports HSPA+ with spatial multiplexing and if he has a MIMO-capable terminal (i.e.,
a terminal with an HSDPA UE category supporting MIMO), he will benefit from the spatial multiplexing gain in its RLC peak
rate.
In this case, the RLC peak rate obtained by the user is the following:
Where
is the RLC peak rate that the selected HSDPA bearer ( ) can provide
in the cell (Txi, ic). It is read in the HSDPA Radio Bearer table.
is the maximum spatial multiplexing gain (in dB) for a given number of transmission and reception antennas
(respectively defined in the transmitter and terminal properties).
is the spatial multiplexing gain factor defined for the clutter
6.4.2.3.6 Scheduling Algorithms
The scheduler manages the maximum number of users within each cell. Packet (HSPA - Constant Bit Rate) service users
have the highest priority and are processed first, in the order established during the generation of the user distribution.
After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler ranks the remaining HSDPA bearer
users (i.e., packet (HSDPA) and packet (HSPA) service users) according to the selected scheduling technique.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 us consider a cell with 16 packet (HSDPA) and packet (HSPA) service users. All of them are active on DL and
connected to the A-DCH R99 bearer. There is no packet (HSPA - Constant Bit Rate) service user in the cell and the
number of HS-SCCH channels and the maximum number of HSDPA users have been respectively set to 4 and 15.
Max C/I
15 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.
CQI ( )
HS PDSCH –
CQI ( )
HS PDSCH –
CQI ( )
HS PDSCH –
f
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
\ .
| |
=
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
Ec
Nt
------- i c ( )
\ .
| |
HS PDSCH –
G
TD
DL
AG
TD
DL
+ + =
G
TD
DL
AG
TD
DL
R
RLC peak –
DL
R
RLC peak –
DL
Index
HSDPABearer
( ) 1 f
SM Gai n –
G
SM
Max
1 – ( ) × + ( ) × =
R
RLC peak –
DL
Index
HSDPABearer
( ) Index
HSDPABearer
G
SM
Max
f
SM Gai n –
Mobiles Simulation Rank
Best Bearer
(kbps)
DL Obtained Rate
(kbps)
Connection
Status
M1 2 2400 2400+3.4 Connected
M2 15 2400 1440+3.4 Connected
M3 8 2080 160+3.4 Connected
M4 9 2080 3.4 Delayed
M5 10 2080 3.4 Delayed
232 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Round Robin
Users are taken into account in the same order than the one in the simulation (random order).
Proportional Fair
15 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 is calculated as follows:
Where,
is the user rank in the simulation.
is the user rank according to the CQI.
M6 12 2080 3.4 Delayed
M7 13 2080 3.4 Delayed
M8 14 2080 3.4 Delayed
M9 7 1920 3.4 Delayed
M10 1 1600 3.4 Delayed
M11 3 1600 3.4 Delayed
M12 4 1600 3.4 Delayed
M13 5 1600 3.4 Delayed
M14 6 1600 3.4 Delayed
M15 11 1440 3.4 Delayed
M16 16 2080 0 Scheduler Saturation
Mobiles Simulation Rank
Best Bearer
(kbps)
DL Obtained Rate
(kbps)
Connection
Status
M1 1 1600 1600+3.4 Connected
M2 2 2400 960+3.4 Connected
M3 3 1600 3.4 Delayed
M4 4 1600 3.4 Delayed
M5 5 1600 3.4 Delayed
M6 6 1600 3.4 Delayed
M7 7 1920 3.4 Delayed
M8 8 2080 3.4 Delayed
M9 9 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 0 Scheduler Saturation
Note:
• You can change the default weights by editing the atoll.ini file. For more information, see
the Administrator Manual.
Mobiles
Simulation
Rank
CQI Rank RP
Best Bearer
(kbps)
DL Obtained
Rate (kbps)
Connection
Status
M1 2 1 150 2400 2400 Connected
RP
i
RP
i
50 R
i
Si mu
× 50 R
i
CQI
× + =
R
i
Si mu
R
i
CQI
© Forsk 2010 AT283_TRG_E2 233
Chapter 6: UMTS HSPA Networks
6.4.2.3.7 Dual-Cell HSDPA
For transmitters that support dual-cell HSDPA mode, the sheduler manages a single queue of users at the Node B. All
users belonging to the transmitter, i.e., dual-cell HSDPA and single-carrier HSDPA users, are ranked together in a unique
list. Dual-cell HSDPA users are considered twice in the list as they may be assigned two different HSDPA bearers in the
two cells.
Packet (HSPA - Constant Bit Rate) service users have the highest priority and are processed first, in the order established
during the generation of the user distribution. After processing the packet (HSPA - Constant Bit Rate) service users, the
scheduler ranks the remaining HSDPA bearer users (i.e., packet (HSDPA) and packet (HSPA) service users) according
to the selected scheduling technique (Max C/I, Round Robin and Proportional Fair). After the HSDPA users have been
ranked, the scheduler allocates HSDPA resources to each user following the calculated order as long as there are
resources available. Even if there is a unique list of users at the transmitter level, the resources of each cell are not shared
and each carrier has its own pool of resources (number of HS-SCCH channels, maximum number of HSDPA users,
HSDPA power, number of OVSF codes). Only site-level resources (such as the Iub throughput and the channel elements)
are shared between the users of the two cells.
Let us consider a dual-cell HSDPA transmitter with 16 packet (HSDPA) and packet (HSPA) service users. There is no
packet (HSPA - Constant Bit Rate) service user. All users are active in DL and connected to the A-DCH R99 bearer.
Among the users, there are 6 dual-cell HSDPA users (i.e., terminal with UE categories 21 to 24).
In each cell, the number of HS-SCCH channels and the maximum number of HSDPA bearer users have been respectively
set to 4 and 7.
The scheduling algorithms defined for the two cells are the same as the one selected for the transmitter.
M2 1 10 550 1600 960 Connected
M3 8 3 550 2080 160 Connected
M4 9 4 650 2080 3.4 Delayed
M5 3 11 700 1600 3.4 Delayed
M6 10 5 750 2080 3.4 Delayed
M7 4 12 800 1600 3.4 Delayed
M8 7 9 800 1920 3.4 Delayed
M9 15 2 850 2400 3.4 Delayed
M10 5 13 900 1600 3.4 Delayed
M11 12 6 900 2080 3.4 Delayed
M12 6 14 1000 1600 3.4 Delayed
M13 13 7 1000 2080 3.4 Delayed
M14 14 8 1100 2080 3.4 Delayed
M15 11 15 1300 1440 3.4 Delayed
M16 16 - - 2080 0
Scheduler
Saturation
Simulation Rank Dual-cell HSDPA Support Carriers Comments
1 Yes 1 and 2 Anchor carrier: 2
2 No 2
3 No 1
4 Yes 1 and 2 Anchor carrier: 2
5 No 1
6 No 2
7 No 1
8 No 2
9 Yes 1 and 2 Anchor carrier: 1
10 No 1
11 No 2
12 Yes 1 and 2 Anchor carrier: 1
13 No 2
14 Yes 1 and 2 Anchor carrier: 1
15 No 1
16 Yes 1 and 2 Anchor carrier: 2
234 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Each dual-cell HSDPA user is counted twice, once in each cell, as he may be assigned two different HSDPA bearers in
the two cells. Therefore, the scheduler manages the users ranked 1
st
to 11
th
(i.e. 4 single-carrier users connected to the
first carrier, 4 single-carrier users connected to the second carrier and 3 dual-cell users). Users ranked 12
th
to 16
th
are
rejected because the maximum number of HSDPA bearer users that the scheduler can manage in a cell is exceeded.
Impact the scheduling algorithms have on the simulation results is described in the tables below.
Max C/I
7 users from each cell (where 7 corresponds to the maximum number of HSDPA users defined for each cell), i.e., a total
of 14 users enter the scheduler in the same order as in the simulation. Then, they are sorted in the order of decreasing
channel quality indicator (CQI), i.e. in a best bearer descending order.
The scheduled dual-cell HSDPA users have the following status:
• The user ranked 4
th
(here M2) is connected to an HSDPA bearer in each cell. He obtains a total DL data rate of
4323.4 kbps (2403.4+1920).
• The user ranked 9
th
(here M4) is connected to an HSDPA bearer in each cell. He obtains a total DL data rate of
2083.4 kbps (963.4+1120).
• The first user (here M9) is delayed in the two cells. He obtains a total DL data rate of 3.4 kbps.
Round Robin
7 users from each cell (where 7 corresponds to the maximum number of HSDPA users defined for each cell), i.e., a total
of 14 users enter the scheduler in the same order as in the simulation.
Mobiles Carrier
Simulation
Rank
CQI
Best Bearer
(kbps)
DL Obtained
Rate (kbps)
Connection
Status
M1 1 5 21 3040 3040+3.4 Connected
M2
(DC-HSDPA)
2 4 19 2400 2400+3.4 Connected
M3 2 8 18 2080 1440+3.4 Connected
M2
(DC-HSDPA)
1 4 17 1920 1920 Connected
M4
(DC-HSDPA)
1 9 17 1920 960+3.4 Connected
M5 1 3 16 1600 3.4 Delayed
M4
(DC-HSDPA)
2 9 16 1600 1120 Connected
M6 2 2 15 1440 3.4 Delayed
M7 1 7 14 1120 3.4 Delayed
M8 1 10 14 1120 3.4 Delayed
M9
(DC-HSDPA)
2 1 13 960 3.4 Delayed
M10 2 6 13 960 3.4 Delayed
M9
(DC-HSDPA)
1 1 12 800 0 Delayed
M11 2 11 12 800 3.4 Delayed
M12
(DC-HSDPA)
1
2
12
14
15
1120
1440
0 Scheduler Saturation
M13 2 13 17 1920 0 Scheduler Saturation
M14
(DC-HSDPA)
1
2
14
13
15
960
1440
0 Scheduler Saturation
M15 1 15 17 1920 0 Scheduler Saturation
M16
(DC-HSDPA)
1
2
16
12
14
800
1120
0 Scheduler Saturation
Mobiles Carrier
Simulation
Rank
CQI
Best Bearer
(kbps)
DL Obtained
Rate (kbps)
Connection
Status
© Forsk 2010 AT283_TRG_E2 235
Chapter 6: UMTS HSPA Networks
The scheduled dual-cell HSDPA users have the following status:
• The first user (here M1) is connected to an HSDPA bearer in each cell. He obtains a total DL data rate of 1763.4
kbps (800+963.4).
• The user ranked 4
th
(here M4) is connected to an HSDPA bearer in each cell. He obtains a total DL data rate of
2563.4 kbps (1603.4+960).
• The user ranked 9
th
(here M9) is delayed in the two cells. He obtains a total DL data rate of 3.4 kbps.
Proportional Fair
7 users from each cell (where 7 corresponds to the maximum number of HSDPA users defined for each cell), i.e., a total
of 14 users enter 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 is calculated as follows:
Where,
is the user rank in the simulation.
is the user rank according to the CQI.
M1
(DC-HSDPA)
1 1 12 800 800 Connected
M1
(DC-HSDPA)
2 1 13 960 960+3.4 Connected
M2 2 2 15 1440 1440+3.4 Connected
M3 1 3 16 1600 1600+3.4 Connected
M4
(DC-HSDPA)
2 4 19 2400 1600+3.4 Connected
M4
(DC-HSDPA)
1 4 17 1920 960 Connected
M5 1 5 21 3040 480+3.4 Connected
M6 2 6 13 960 160+3.4 Connected
M7 1 7 14 1120 3.4 Delayed
M8 2 8 18 2080 3.4 Delayed
M9
(DC-HSDPA)
2 9 16 1600 0 Delayed
M9
(DC-HSDPA)
1 9 17 1920 3.4 Delayed
M10 1 10 14 1120 3.4 Delayed
M11 2 11 12 800 3.4 Delayed
M12
(DC-HSDPA)
1
2
12
14
15
1120
1440
0 Scheduler Saturation
M13 2 13 17 1920 0 Scheduler Saturation
M14
(DC-HSDPA)
1
2
14
13
15
960
1440
0 Scheduler Saturation
M15 1 15 17 1920 0 Scheduler Saturation
M16
(DC-HSDPA)
1
2
16
12
14
800
1120
0 Scheduler Saturation
Note:
• You can change the default weights by editing the atoll.ini file. For more information, see
the Administrator Manual.
RP
i
RP
i
50 R
i
Si mu
× 50 R
i
CQI
× + =
R
i
Si mu
R
i
CQI
236 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
The scheduled dual-cell HSDPA users have the following status:
• The user ranked 4
th
(here M1) is connected to an HSDPA bearer in each cell. He obtains a total DL data rate of
3843.4 kbps (2403.4+1440).
• The first user (here M6) is connected to an HSDPA bearer in his anchor cell and delayed in the other cell. He
obtains a total DL data rate of 483.4 kbps (483.4+0).
• The user ranked 9
th
(here M7) is delayed in the two cells. He obtains a total DL data rate of 3.4 kbps.
6.4.2.4 HSUPA Part of the Algorithm
Packet (HSPA) service users active in the UL as well as all packet (HSPA - Constant Bit Rate) service users (i.e., active
and incative), unless they have been rejected during the R99 or HSDPA parts of the algorithm, are then evaluated by the
HSUPA part of the algorithm. Atoll manages the maximum number of users within each cell. Packet (HSPA - Constant
Bit Rate) service users have the highest priority and are processed first, in the order established during the generation of
the user distribution. Then, Atoll considers packet (HSPA) service users in the order established during the generation of
the user distribution.
Let us assume there are 12 HSUPA bearer users in the cell:
• 3 packet (HSPA - Constant Bit Rate) service users with any activity status. All of them have been connected to an
HSDPA bearer.
• 9 packet (HSPA) service users active on UL. The first four packet (HSPA) have been connected to an HSDPA
bearer, the last one has been rejected and the remaining four have been delayed in the HSDPA part.
Finally, the maximum number of HSUPA users equals 10.
In this case, Atoll will consider the first ten HSUPA bearer users only and will reject the last two users in order not to
exceed the maximum number of HSUPA users allowed in the cell (their connection status is "HSUPA scheduler
saturation").
Mobiles Carrier
Simulation
Rank
CQI
CQI
Rank
RP
Best
Bearer
(kbps)
DL Obtained
Rate (kbps)
Connection
Status
M1
DC-HSDPA
2 4 19 2 300 2400 2400+3.4 Connected
M2 1 5 21 1 300 3040 3040+3.4 Connected
M1
DC-HSDPA
1 4 17 4 400 1920 1440 Connected
M3 1 3 16 6 450 1600 800+3.4 Connected
M4 2 2 15 8 500 1440 1120+3.4 Connected
M5 2 8 18 3 550 2080 800+3.4 Connected
M6
DC-HSDPA
2 1 13 11 600 960 480+3.4 Connected
M6
DC-HSDPA
1 1 12 13 700 800 0 Delayed
M7
DC-HSDPA
1 9 17 5 700 1920 3.4 Delayed
M8 1 7 14 9 800 1120 3.4 Delayed
M7
DC-HSDPA
2 9 16 7 800 1600 0 Delayed
M9 2 6 13 12 900 960 3.4 Delayed
M10 1 10 14 10 1000 1120 3.4 Delayed
M11 2 11 12 14 1250 800 3.4 Delayed
M12
(DC-HSDPA)
1
2
12
14
15
1120
1440
0
Scheduler
Saturation
0
Scheduler
Saturation
M13 2 13 17 1920 0
Scheduler
Saturation
0
Scheduler
Saturation
M14
(DC-HSDPA)
1
2
14
13
15
960
1440
0
Scheduler
Saturation
0
Scheduler
Saturation
M15 1 15 17 1920 0
Scheduler
Saturation
0
Scheduler
Saturation
M16
(DC-HSDPA)
1
2
16
12
14
800
1120
0
Scheduler
Saturation
0
Scheduler
Saturation
© Forsk 2010 AT283_TRG_E2 237
Chapter 6: UMTS HSPA Networks
6.4.2.4.1 Admission Control
During admission control, Atoll selects a list of HSUPA bearers for each user. The selected HSUPA bearers have to be
compatible with the user equipment and capabilities of each HSUPA cell of the active set. For packet (HSPA - Constant
Bit Rate) service users, the list is restricted to HSUPA bearers that provide a RLC peak rate higher than the guaranteed
bit rate.
Let us focus on one packet (HSPA) service user with category 3 user equipment and a 50km/h mobility. This user is
connected to one cell only. The cell supports HSPA+ functionalities, i.e the cell supports QPSK and 16QAM modulations
in the UL.
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
• Highest Modulation Supported: QPSK
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.
• Modulation: the modulation used (QPSK or 16QAM)
• RLC Peak Rate (bps): The RLC peak rate represents the peak rate without coding (redundancy, overhead,
addressing, etc.).
Mobiles Service
Simulation
Rank
HSDPA
Connection
Status
Evaluation by
the HSUPA
part of the
algorithm
M1 Packet (HSPA - Constant Bit Rate) 4 Connected Yes
M2 Packet (HSPA - Constant Bit Rate) 7 Connected Yes
M3 Packet (HSPA - Constant Bit Rate) 9 Connected Yes
M4 Packet (HSPA) 1 Connected Yes
M5 Packet (HSPA) 2 Connected Yes
M6 Packet (HSPA) 3 Connected Yes
M7 Packet (HSPA) 5 Connected Yes
M8 Packet (HSPA) 6 Delayed Yes
M9 Packet (HSPA) 8 Delayed Yes
M10 Packet (HSPA) 10 Delayed Yes
M11 Packet (HSPA) 11 Delayed No
M12 Packet (HSPA) 12 Rejected No
Figure 6.8HSUPA UE Categories Table
238 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
HSUPA bearers can be classified into two categories:
• HSUPA bearers using QPSK modulation: They can be selected for users connected to HSPA and HSPA+ capable
cells.
• HSUPA bearers using 16QAM modulation (improvement introduced by the release 7 of the 3GPP UTRA
specifications, referred to as HSPA+). These HSUPA bearers can be allocated to users connected to cells with
HSPA+ capabilities only.
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 modulation required by the bearer is supported by the terminal.
The HSUPA bearers compatible with category 3 user equiment are framed in red:
Then, during admission control, Atoll checks that the lowest compatible bearer in terms of the required E-DPDCH Ec⁄Nt
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.
Then, from the required Ec/Nt threshold, , Atoll calculates the required terminal power, .
With
(
8
)
, , , , and are defined in "Inputs" on page 198.
Figure 6.9HSUPA Radio Bearers Table
8. In the HSUPA coverage prediction, is calculated as follows:
)
Ec
Nt
-------
\ .
| |
E DPDCH –
req
P
term HSUPA –
req
P
term HSUPA –
req Ec
Nt
-------
\ .
| |
=
E DPDCH –
req
L
T
N
tot
UL
× ×
N
tot
UL
i c ( ) 1 F –
MUD
tx
µ
term
× ( ) I ×
tot
UL
i nt ra
i c ( ) I
tot
UL
ext ra
i c ( ) I
i nter carri er –
UL
i c ( ) N +
0
tx
+ + =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Eb Nt ( )
UL

× × × ×
G
Tx
G
term
×
------------------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
term
F
MUD
tx
I
tot
UL
i ntra
I
tot
UL
ext ra
I
i nter carri er –
UL
N
0
tx
© Forsk 2010 AT283_TRG_E2 239
Chapter 6: UMTS HSPA Networks
Atoll rejects the user if the terminal power required to obtain the lowest compatible HSUPA bearer ( )
exceeds the maximum terminal power (his connection status is "HSUPA Admission Rejection").
At the end of this step, the number of non-rejected HSUPA bearer users is . All of them will be connected to an
HSUPA bearer at the end.
6.4.2.4.2 HSUPA Bearer Allocation Process
The HSUPA bearer allocation process depends on the type of service requested by the user. As explained before, packet
(HSPA - Constant Bit Rate) service users have the highest priority and are processed first, in the order established during
the generation of the user distribution. After the admission control on packet (HSPA - Constant Bit Rate) service users,
Atoll performs a noise rise scheduling, followed by a radio resource control. Then, it repeats the same steps on packet
(HSPA) service users.
Packet (HSPA - Constant Bit Rate) Service Users
Let us focus on the three packet (HSPA - Constant Bit Rate) service users mentionned in the example of the previous
paragraph "HSUPA Part of the Algorithm" on page 236. We assume that all of them have been admitted. Noise rise
scheduling and radio resource control are carried out on each user in order to determine the best HSUPA bearer that the
user can obtain. Several Packet (HSPA - Constant Bit Rate) service users can share the same HSUPA bearer. Then, Atoll
calculates the HSUPA bearer consumption ( in %) for each user and takes into account this parameter when it
determines the resources consumed by the user (i.e., the terminal power used, the number of channel elements and the
Iub backhaul throughput).
In the bearer allocation process shown below, the 3 packet (HSPA - Constant Bit Rate) service users are represented by
M
j
, with j = 1 to 3.
Figure 6.10HSUPA Bearer SelectionTable
P
term HSUPA –
req
n
HSUPA
C
240 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
Packet (HSPA) Service Users
Let us focus on the seven packet (HSPA) service users mentionned in the example of the previous paragraph "HSUPA
Part of the Algorithm" on page 236. We assume that all of them have been admitted. Noise rise scheduling and radio
resource control are carried out on each user in order to determine the best HSUPA bearer that the user can obtain.
In the bearer allocation process shown below, the 7 packet (HSPA) service users are represented by M
j
, with j = 1 to 7.
6.4.2.4.3 Noise Rise Scheduling
Determination of the Obtained HSUPA Bearer
The obtained HSUPA radio bearer is the bearer that the user obtains after noise rise scheduling and radio resource control.
Packet (HSPA - Constant Bit Rate) service users have the highest priority and are processed first. Therefore, after the
admission control, the noise rise scheduling algorithm attempts to evenly share the remaining cell load between the packet
For the user, M
j
, with j varying from 1 to 3:
Figure 6.11HSUPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users
For the user, M
j
, with j varying from 1 to 7:
Figure 6.12HSUPA Bearer Allocation Process for Packet (HSPA) Service Users
© Forsk 2010 AT283_TRG_E2 241
Chapter 6: UMTS HSPA Networks
(HSPA - Constant Bit Rate) service users admitted in admission control; in terms of HSUPA, each user is allocated a right
to produce interference. The remaining cell load factor on uplink ( ) 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:
Then, Atoll evenly shares the remaining cell load factor between the packet (HSPA - Constant Bit Rate) service users
admitted during the previous step ( ).
From this value, Atoll calculates the maximum E-DPDCH Ec⁄Nt allowed ( ) for each packet (HSPA -
Constant Bit Rate) service user. For further information on the calculation, see "Uplink Load Factor Due to One User" on
page 257.
for the Without useful signal option
for the Total noise option
Then, it selects an HSUPA bearer. The allocation depends on the maximum E-DPDCH Ec⁄Nt allowed and on UE and cell
capabilities. Atoll selects the best HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer
( ) with the highest potential throughput ( ) where:

• And
When several HSUPA bearers are available, Atoll selects the one with the lowest .
After the noise rise scheduling, Atoll carries out radio resource control, verifying if enough channel elements and Iub
backhaul throughput are available for the HSUPA bearer assigned to the user. For information on radio resource control,
see "Radio Resource Control" on page 244.
After processing all packet (HSPA - Constant bit rate) service users, Atoll carries out noise rise scheduling and radio
resource control on packet (HSPA) service users. During the noise rise scheduling, Atoll distributes the remaining cell
load factor available after all packet (HSPA - Constant Bit Rate) service users have been served. It can be expressed as
follows:
The remaining cell load factor is shared equally between the admitted packet (HSPA) service users ( ).
From this value, Atoll calculates the maximum E-DPDCH Ec⁄Nt allowed ( ) as explained above and selects
an HSUPA bearer for each packet (HSPA) service user. After the noise rise scheduling, Atoll carries out radio resource
control on packet (HSPA) service users. For information on radio resource control, see "Radio Resource Control" on
page 244.
Example: We have a cell with six packet (HSPA) service users and no packet (HSPA - Constant Bit Rate) user. All packet
(HSPA) service users have been admitted.
The remaining cell load factor equal to 0.6 is shared between the packet (HSPA) service users. Therefore, the UL load
factor alloted to each user is 0.1. Let’s take the cell UL reuse factor equal to 1.5. Atoll calculates the maximum E-DPDCH
Ec⁄Nt allowed (the Without useful signal option is selected).
We have:
Here, the obtained HSUPA bearer is the index 5 HSUPA bearer. It provides a potential throughput of 128 kbps and
requires E-DPDCH Ec⁄Nt of -13 dB (lower than -11.5 dB) and a terminal power lower than the maximum terminal power
allowed.
AX
HSPA CBR –
UL
txi i c , ( )
AX
HSPA CBR –
UL
txi i c , ( ) X
max
UL
txi i c , ( ) X
R99
UL
txi i c , ( ) – =
n
HSPA CBR –
AX
user
UL
txi i c , ( )
AX
HSPA CBR –
UL
txi i c , ( )
n
HSPA CBR –
-------------------------------------------------------- =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Ec
Nt
-------
\ .
| |
E DPDCH –
max
1
F
UL
txi i c , ( )
AX
user
UL
txi i c , ( )
-------------------------------------- 1 –
----------------------------------------------- =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
AX
user
UL
F
UL
------------------ =
Index
HSUPABearer
R
RLC peak –
UL
Index
HSUPABearer
( )
N
Rtx
Index
HSUPABearer
( )
------------------------------------------------------------------------------------
Ec
Nt
-------
\ .
| |
E DPDCH –
req
Ec
Nt
-------
\ .
| |
E DPDCH –
max
s
P
term HSUPA –
req
P s
term
max
Ec
Nt
-------
\ .
| |
E DPDCH –
req
AX
HSPA
UL
txi i c , ( ) X
max
UL
txi i c , ( ) X
R99
UL
txi i c , ( ) – X
HSPA CBR –
UL
txi i c , ( ) – =
n
HSPA
AX
user
UL
txi i c , ( )
AX
HSPA
UL
txi i c , ( )
n
HSPA
----------------------------------------- =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Ec
Nt
-------
\ .
| |
E DPDCH –
max
-11.5 dB =
242 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
.
Noise Rise Scheduling in Soft Handover
With HSUPA, uplink soft handover impacts the scheduling operation. While HSDPA sends data from one cell only, with
HSUPA all cells in the active set receive the transmission from the terminal. Therefore, all the cells are impacted by the
transmission in terms of noise rise.
For each HSUPA capable cell of the active set , Atoll calculates the maximum E-DPDCH Ec⁄Nt allowed
( ) as explained in "HSUPA Bearer Allocation Process" on page 239.
For each cell of the active set , Atoll calculates the maximum terminal power allowed to obtain an HSUPA radio
bearer ( ).
With
(
9
)
, , , , and are defined in "Inputs" on page 198.
As HSUPA bearer users in soft handover use the lowest granted noise rise, Atoll chooses the lowest of maximum terminal
power allowed for each cell of the active set .
Once Atoll knows the selected maximum terminal power ( ), it recalculates the maximum E-DPDCH Ec⁄Nt
allowed ( ) for each HSUPA capable cell of the active set.
Then, Atoll calculates the maximum E-DPDCH Ec⁄Nt allowed ( ) after signal recombination of all HSUPA
capable cells of the active set
10
.
For softer (1/2) and softer-softer (1/3) handovers, we have:
HSUPA Bearers
Index
Required Ec/Nt
Threshold (dB)
Nb of
Retransmissions
RLC Peak Rate
(kbps)
Potential
Throughput
(kbps)
1 -21.7 2 32 16
2 -19 2 64 32
3 -16.1 2 128 64
4 -13.9 2 192 96
5 -13 2 256 128
6 -10.1 2 512 256
7 -8 2 768 384
8 -7 2 1024 512
9. In the HSUPA coverage prediction, is calculated as follows:
)
tx
k
i c , ( )
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
tx
k
i c , ( )
P
term HSUPA –
max
tx
k
i c , ( )
P
term HSUPA –
max
tx
k
i c , ( ) mi n
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( ) L
T
N
tot
UL
× ×
\ .
| |
P
term
max
,
\ .
| |
=
N
tot
UL
i c ( ) 1 F –
MUD
tx
µ
term
× ( ) I ×
tot
UL
i nt ra
i c ( ) I
tot
UL
ext ra
i c ( ) I
i nter carri er –
UL
i c ( ) N +
0
tx
+ + =
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
E
Shadowi ng
× × × ×
G
Tx
G
term
×
-------------------------------------------------------------------------------------------------------------------------------- =
L
T
L
T
L
path
L
Tx
L
term
× L
body
L
i ndoor
M
Shadowi ng Eb Nt ( )
UL

× × × ×
G
Tx
G
term
×
------------------------------------------------------------------------------------------------------------------------------------------------------------- =
µ
term
F
MUD
tx
I
tot
UL
i ntra
I
tot
UL
ext ra
I
i nter carri er –
UL
N
0
tx
tx
k
i c , ( )
P
term HSUPA –
max
mi n
tx
k
AS e
P
term HSUPA –
max
tx
k
i c , ( ) ( ) =
P
term HSUPA –
max
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
P
term HSUPA –
max
L
T
N
tot
UL
×
----------------------------------- =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
© Forsk 2010 AT283_TRG_E2 243
Chapter 6: UMTS HSPA Networks

For soft (2/2) and soft-soft (3/3) handovers, we have:
For softer-soft handover (2/3), it depends on if the MRC option is selected (option available in Global parameters). If se-
lected, we have:
Else, we have:
Then, Atoll selects an HSUPA bearer as previously explained in "HSUPA Bearer Allocation Process" on page 239. The
allocation depends on the maximum E-DPDCH Ec⁄Nt allowed and on UE and cell capabilities. Atoll selects the best
HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer ( ) with the highest po-
tential throughput ( ) where:

10. In HSUPA coverage predictions, we have the following:
For softer (1/2) and softer-softer (1/3) handovers:
For soft handover (2/2):
For soft-soft handover (3/3):
For softer-soft handover (2/3), it depends on if the MRC option is selected (option available in Global parameters). If se-
lected, we have:
Else, we have:
Ec
Nt
-------
\ .
| |
E DPDCH –
max
f
rake efficiency
UL Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
tx
k
Acti veSet e
samesi te ( )
¿
× =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
Act i veSet e
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
\ .
| |
G
macro di versi ty –
UL
( )
2l i nks
× =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
Act i veSet e
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
\ .
| |
G
macro di versi ty –
UL
( )
3l i nks
× =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
tx
l
, Act i veSet e
t x
k
samesi te e
tx
l
othersi te e
f
rake efficiency
UL Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
tx
k
¿
×
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
l
i c , ( ) ,
\ .
|
|
|
| |
G
macro di versi ty –
UL
( )
2l i nks
×
=
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
Act i veSet e
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
\ .
| |
G
macro di versi ty –
UL
( )
2l i nks
× =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
f
rake efficiency
UL Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
tx
k
Acti veSet e
samesi te ( )
¿
× =
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
Act i veSet e
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
\ .
| |
=
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
tx
l
, Act i veSet e
t x
k
samesi te e
tx
l
othersi te e
f
rake efficiency
UL Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
tx
k
¿
×
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
l
i c , ( ) ,
\ .
|
|
|
| |
=
Ec
Nt
-------
\ .
| |
E DPDCH –
max
Max


tx
k
Act i veSet e
Ec
Nt
-------
\ .
| |
E DPDCH –
max
tx
k
i c , ( )
\ .
| |
=
Index
HSUPABearer
R
RLC peak –
UL
Index
HSUPABearer
( )
N
Rtx
Index
HSUPABearer
( )
------------------------------------------------------------------------------------
Ec
Nt
-------
\ .
| |
E DPDCH –
req
Ec
Nt
-------
\ .
| |
E DPDCH –
max
s
244 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
When several HSUPA bearers are available, Atoll selects the one with the lowest .
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 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 Ec⁄Nt
allowed is equal to -1.8 dB and the requested HSUPA bearer is the index 7 HSUPA bearer. It requires E-DPDCH Ec⁄Nt of
-8 dB (lower than -1.8 dB) and a terminal power lower than the maximum terminal power allowed.
6.4.2.4.4 Radio Resource Control
Atoll checks to see if enough channel elements are available and if the Iub backhaul throughput is sufficient for the HSUPA
bearer assigned to the user (taking into account the maximum number of channel elements defined for the site and the
maximum Iub backhaul throughput allowed on the site in the uplink). If not, Atoll allocates a lower HSUPA bearer
("downgrading") which needs fewer channel elements and consumes lower Iub backhaul throughput. If no channel
elements are available, the user is rejected. On the same hand, if the maximum Iub backhaul throughput allowed on the
site in the uplink is still exceeded even by using the lowest HSDPA bearer, the user is rejected.
6.4.2.5 Convergence Criteria
The convergence criteria are evaluated for each iteration, and can be written as follow:
Atoll stops the algorithm if:
1
st
case: Between two successive iterations, and 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 and between the 4
th
and the 5
th
iteration, Atoll stops the algorithm after the 5
th
iteration. Convergence
has been reached.
2
nd
case: After 30 iterations, and/or are still higher than their respective thresholds and from the 30
th
iteration,
and/or 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 30
th
iteration, and/or equal 100 and do not decrease during the next 15 successive iterations:
Atoll stops the algorithm at the 46
th
iteration. Convergence has not been reached.
2. After the 30
th
iteration, and/or equal 80, they start decreasing slowly until the 40
th
iteration (without going
under the thresholds) and then, do not change during 15 successive iterations: Atoll stops the algorithm at the
56
th
iteration without reaching convergence.
3
rd
case: After the last iteration.
If and/or are still strictly higher than their respective thresholds, the simulation has not reached convergence
(specific divergence symbol).
If and 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
dialogue.
Ec
Nt
-------
\ .
| |
E DPDCH –
req
A
DL
max i nt
max
Stati ons
P
tx
i c ( )
k
P
tx
i c ( )
k 1 –

P
tx
i c ( )
k
---------------------------------------------------------------------------------------- 100 ×
\ .
|
| |
i nt
max
Stat i ons
N
user
DL
i c ( )
k
N
user
DL
i c ( )
k 1 –

N
user
DL
i c ( )
k
---------------------------------------------------------------------------------------------------- 100 ×
\ .
|
|
| |
,
\ .
|
|
| |
=
A
UL
max i nt
max
Stati ons
I
tot
UL
i c ( )
k
I
tot
UL
i c ( )
k 1 –

I
tot
UL
i c ( )
k
---------------------------------------------------------------------------------------- 100 ×
\ .
|
|
| |
i nt
max
St ati ons
N
user
UL
i c ( )
k
N
user
UL
i c ( )
k 1 –

N
user
UL
i c ( )
k
---------------------------------------------------------------------------------------------------- 100 ×
\ .
|
|
| |
,
\ .
|
|
| |
=
A
UL
A
DL
s
A
UL
5 s A
DL
5 s
A
UL
A
DL
A
UL
A
DL
A
UL
A
DL
A
UL
A
DL
A
UL
A
DL
A
UL
A
DL
© Forsk 2010 AT283_TRG_E2 245
Chapter 6: UMTS HSPA Networks
Name Value Unit Description
None
Number of E1/T1/Ethernet links
required by the site
None
Downlink intra-cell interference at
terminal on carrier
W
Downlink extra-cell interference at
terminal on carrier
W
Downlink inter-carrier interference at
terminal on carrier
W
Downlink inter-technology interference
at terminal on carrier ic
a
W
Total effective interference at terminal
on carrier (after unscrambling)
W
Total received noise at terminal on
carrier
W
Total power received at transmitter
from intra-cell terminals using carrier
W
Total power received at transmitter
from extra-cell terminals using carrier
W
Uplink inter-carrier interference at
terminal on carrier
W
Total received interference at
transmitter on carrier
W
Total noise at transmitter on carrier
(Uplink interference)
None Cell uplink load factor on carrier
None Cell uplink reuse factor on carrier
None
Cell uplink reuse efficiency factor on
carrier
Simulation result available per cell
with
Simulation result available per mobile
None Downlink load factor on carrier
Nb
E1 T1 Ethernet RoundUp Max T
Iub
DL
N
I
( ) T
E1 T1 Ethernet
T
Iub
UL
N
I
( ) T
E1 T1 Ethernet
, ( ) ( )
I
i ntra
DL
txi i c , ( )
P
tot
DL
txi i c , ( )
F
ortho
µ
BTS
× –
P
tot
DL
txi i c , ( )
txi
P
SCH
txi i c , ( )
L
T
---------------------------------- –
\ .
|
|
| |
×
1 F
ortho
– ( ) µ
BTS
× – P
b
DL
txi i c , ( ) ×
i c
I
extra
DL
i c ( )
P
tot
DL
txj i c , ( )
txj j i = ,
¿
i c
I
i nter carri er –
DL
i c ( )
P
tot
DL
txj i c
adj
, ( )
txj j ¬ ,
¿
RF i c i c
adj
, ( )
-------------------------------------------------
i c
I
i nter techno y log –
DL
i c ( )
P
Transmi tted
Tx
i c
i
( )
L
total
Tx
ICP
n
i
i c ,
Tx m ,
×
------------------------------------------
n
i
¿
I
tot
DL
i c ( ) I
i ntra
DL
i c ( ) I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) I
i nter techno y log –
DL
i c ( ) + + +
i c
N
tot
DL
i c ( ) I
tot
DL
i c ( ) N
0
Term
+
i c
I
tot
UL
i ntra
txi i c , ( )
P
b
UL
i c ( )
term
txi
¿
i c
I
tot
UL
ext ra
txi i c , ( )
P
b
UL
i c ( )
term
txj j i = ,
¿
i c
I
i nter carri er –
UL
txi i c , ( )
P
b
UL
i c
adj
( )
term
txj j ¬ ,
¿
RF i c i c
adj
, ( )
---------------------------------------
i c
I
tot
UL
txi i c , ( )
I
tot
UL
extra
txi i c , ( ) 1 F
MUD
Tx
µ
term
× – ( ) + I
tot
UL
i ntra
txi i c , ( ) × I
i nter carri er –
UL
txi i c , ( ) +
i c
N
tot
UL
txi i c , ( ) I
tot
UL
txi i c , ( ) N
0
tx
+
i c
X
UL
txi i c , ( )
I
tot
UL
txi i c , ( )
N
tot
UL
txi i c , ( )
------------------------------ i c
F
UL
txi i c , ( )
I
tot
UL
txi i c , ( )
I
tot
UL
i ntra
txi i c , ( ) 1 F
MUD
Tx
– µ
term
× ( ) ×
-------------------------------------------------------------------------------------------
i c
E
UL
txi i c , ( )
1
F
UL
txi i c , ( )
------------------------------
i c
X
DL
txi i c , ( )
I
extra
DL
i c ( ) I
i nter carri er –
DL
i c ( ) + ( ) L
T
×
P
Tx
DL
txi i c , ( )
---------------------------------------------------------------------------------------- 1 F
ortho
µ
BTS
× – +
1
CI
req
DL
------------- 1 F
ortho
µ
BTS
× – ( ) +
-------------------------------------------------------------------------------------------------------------------------------------------
tch
¿
CI
req
DL
Q
req
DL
G
p
DL
------------ =
I
tot
DL
i c ( )
N
tot
DL
i c ( )
--------------------
i c
246 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
6.4.3.2 HSPA Related Results
At the end of the R99 part, packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users can be:
• Either connected and in this case, they obtain the requested R99 bearer,
• Or rejected exactly for the same reasons as R99 users.
Only connected packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users are considered in
the HSDPA part. At the end of the HSDPA part, packet (HSDPA) and packet (HSPA) service users 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).
Packet (HSPA - Constant Bit Rate) service users can be:
• Either connected if they obtain an HSDPA bearer,
• Or rejected for the following reasons: the maximum number of HSDPA users per cell is exceeded, the lowest
HSDPA bearer the user can obtain does not provide a RLC peak rate higher than the guaranted bit rate, the HS-
SCCH signal quality is not sufficient, there are no more OVSF codes available, the maximum Iub backhaul
throughput allowed on the site in the downlink is exceeded.
In the HSUPA part, Atoll processes packet (HSPA) service users and packet (HSPA - Constant Bit Rate) service 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 for the following reasons: the maximum number of HSUPA users per cell is exceeded, the terminal
power required to obtain the lowest compatible HSUPA bearer exceeds the maximum terminal power, there are
no more channel elements available, the maximum Iub backhaul throughput allowed on the site in the uplink is
exceeded, the lowest compatible HSUPA bearer they can obtain does not provide a RLC peak rate higher than
the guaranted bit rate (only for packet (HSPA - Constant Bit Rate) service users).
In the following parts, a dual-cell HSDPA user refers to a user which has a dual-cell HSDPA-capable terminal and which
is simultaneously connected to two HSDPA cells of a transmitter supporting dual-cell HSDPA mode.
6.4.3.2.1 Statistics Tab
In the Statistics tab, Atoll displays as results:
• The number of rejected users.
• The number of 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 bearer users since all of them request an R99 bearer.
- The number of R99 bearer users per frequency band.
- The number of R99 bearer users per activity status.
- The downlink and uplink rates ( and ) generated by their connection to R99 bearers. Only active
users are considered.
and
is the downlink nominal rate of the user R99 radio bearer and 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. Packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users are considered
since they all request an HSDPA bearer. On the other hand, only active users are taken into consideration in the
downlink rate calculation ( ).
is the RLC peak rate provided in the downlink.
None Downlink reuse factor on a carrier
dB Noise rise on downlink
dB Noise rise on uplink
a. In the case of an interfering GSM external network in frequency hopping, the ICP value is weighted according to the
fractional load.
F
DL
txi i c , ( )
I
tot
DL
i c ( )
I
i ntra
DL
txi i c , ( )
-------------------------------- i c
NR
DL
txi i c , ( ) 10 1 X
DL
txi i c , ( ) – ( ) log –
NR
UL
txi i c , ( ) 10 1 X
UL
txi i c , ( ) – ( ) log –
R
R99
DL
R
R99
UL
R
R99
DL
R
nomi nal
DL
R99 Bearer ( )
Acti ve
users
¿
= R
R99
UL
R
nomi nal
UL
R99 Bearer ( )
Acti ve
users
¿
=
R
nomi nal
DL
R99 Bearer ( ) R
nomi nal
UL
R99 Bearer ( )
R
HSDPA
DL
R
HSDPA
DL
R
RLC peak –
DL
Acti ve
users
¿
=
R
RLC peak –
DL
© Forsk 2010 AT283_TRG_E2 247
Chapter 6: UMTS HSPA Networks
• The number of connected HSUPA bearer users (result of the HSUPA part). Only packet (HSPA) and packet
(HSPA - Constant Bit Rate) service users are considered.
In addition, Atoll indicates the uplink data rate generated by active users connected with an HSUPA bearer
( ):
is the RLC peak rate provided in the uplink.
6.4.3.2.2 Mobiles Tab
In the Mobiles tab, Atoll indicates for each user:
• The uplink and downlink total requested rates in kbps (respectively, and )
For circuit and packet (R99) service users, the DL and UL total requested rates correspond to the DL and UL nominal rates
of the R99 bearer associated to the service.
For packet (HSDPA) service users, 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(s) that the
selected HSDPA radio bearer(s) can provide. Here, the user is treated as if he is the only user in the cell and then, Atoll
determines the HSDPA bearer the user would obtain by considering the entire HSDPA power available of the cell.
for single-carrier users
for dual-carrier users
For HSUPA bearer users (i.e., packet (HSPA) and packet (HSPA - Constant Bit Rate) service users), 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 bearers compatible with the user
equipment. Here, the 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(s) that the requested HSDPA radio bearer(s) can
provide. The requested HSDPA radio bearer is determined as explained in the previous paragraph.
for single-carrier users
for dual-carrier
users
• The uplink and downlink total obtained rates in kbps (respectively, and )
For circuit and packet (R99) service users, 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.
In the downlink, HSDPA bearer users can be connected to a single cell or to two cells of the same transmitter when the
user has a dual-cell HSDPA-capable terminal and when the transmitter supports the dual-cell HSDPA mode.
For a single-carrier packet (HSDPA) service user connected to an HSDPA bearer, 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 user is delayed (he is only connected to
an R99 radio bearer), downlink obtained rate corresponds to the downlink nominal rate of the ADPCH radio bearer. Finally,
if the user is rejected either in the R99 part or in the HSDPA part (i.e., because the HSDPA scheduler is saturated), the
downlink obtained rate is zero.
For a dual-carrier packet (HSDPA) service user connected to two HSDPA bearers, the downlink obtained rate corresponds
to the instantaneous rate; this is the sum of the nominal rate provided by the A-DPCH radio bearer in the anchor cell and
the RLC peak rates provided by the selected HSDPA radio bearers after scheduling and radio resource control. If the user
is connected to one cell and delayed in the other cell, the downlink obtained rate is the sum of the nominal rate provided
by the A-DPCH radio bearer in the anchor cell and the RLC peak rate provided by the selected HSDPA radio bearer after
scheduling and radio resource control. If the user is delayed in the two cells (he is only connected to an R99 radio bearer
R
HSUPA
UL
R
HSUPA
UL
R
RLC peak –
UL
Acti ve
users
¿
=
R
RLC peak –
UL
R
requested
UL
M
b
( ) R
requested
DL
M
b
( )
R
requested
DL
M
b
( ) R
nomi nal
DL
R99 Bearer ( ) =
R
requested
UL
M
b
( ) R
nomi nal
UL
R99 Bearer ( ) =
R
requested
DL
M
b
( ) R
nomi nal
DL
ADPCH R99 Bearer ( ) R
RLC peak –
DL
+ =
R
requested
DL
M
b
( ) R
nomi nal
DL
ADPCH R99 Bearer ( )
AnchorCel l
R
RLC peak –
DL
c ( )
c Serving Cells e
¿
+ =
R
requested
UL
M
b
( ) R
nomi nal
UL
ADPCH R99 Bearer ( ) =
R
requested
DL
M
b
( ) R
nomi nal
DL
ADPCH EDPCCH R99 Bearer – ( ) R
RLC peak –
DL
+ =
R
requested
DL
M
b
( ) R
nomi nal
DL
ADPCH EDPCCH R99 Bearer – ( )
AnchorCel l
R
RLC peak –
DL
c ( )
c Serving cells e
¿
+ =
R
requested
UL
M
b
( ) R
nomi nal
UL
ADPCH EDPCCH R99 Bearer – ( ) R
RLC peak –
UL
+ =
R
obtai ned
UL
M
b
( ) R
obtai ned
DL
M
b
( )
248 AT283_TRG_E2 © Forsk 2010
Technical Reference Guide
in the anchor cell), the downlink obtained rate corresponds to the downlink nominal rate of the ADPCH radio bearer in the
anchor cell. Finally, if the user is rejected either in the R99 part or in the HSDPA part (i.e., because the HSDPA scheduler
is saturated), the downlink obtained rate is zero.
In the uplink, packet (HSDPA) service users can only have a single-carrier connection. When the user is either connected
or delayed, the uplink obtained rate corresponds to the uplink nominal rate of the ADPCH radio bearer. If the user is
rejected either in the R99 part or in the HSDPA part (i.e., because the HSDPA scheduler is saturated), the uplink obtained
rate is zero.
For a single-carrier packet (HSPA) service user, on downlink, if the user is connected to an HSDPA bearer, the downlink
obtained 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 downlink obtained rate corresponds to the downlink nominal rate of ADPCH-
EDPCCH radio bearer. If the user is rejected, the downlink obtained rate is "0."
For a dual-carrier packet (HSPA) service user connected to two HSDPA bearers, the downlink obtained rate corresponds
to the instantaneous rate; this is the sum of the nominal rate provided by the ADPCH-EDPCCH radio bearer in the anchor
cell and the RLC peak rates provided by the selected HSDPA radio bearers after scheduling and radio resource control.
If the user is connected to one cell and delayed in the other cell, the downlink obtained rate is the sum of the nominal rate
provided by the ADPCH-EDPCCH radio bearer in the anchor cell and the RLC peak rate provided by the selected HSDPA
radio bearer after scheduling and radio resource control. If the user is delayed in the two cells (he is only connected to an
R99 radio bearer in the anchor cell), the downlink obtained rate corresponds to the downlink nominal rate of the ADPCH-
EDPCCH radio bearer in the anchor cell. Finally, if the user is rejected, the downlink obtained rate is zero.
In uplink, packet (HSPA) service users can only have a single-carrier connection. When the user is connected to an
HSUPA bearer, the uplink obtained 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. If the user is rejected, the uplink obtained
rate is zero.
For a connected packet (HSPA - Constant Bit Rate) service user, the uplink and downlink total obtained rates are the sum
of the ADPCH-EDPCCH radio bearer nominal rate and the guaranteed bit rate defined for the service. If the user is
rejected, the uplink and downlink total obtained rates are "0".
• The mobile total power ( )
for packet (HSPA) service users
for packet (HSPA - Constant Bit Rate) service users
And
for circuit and packet (R99) service users and packet (HSDPA) service users
• The HSDPA application throughput in kbps ( )
This is the net HSDPA throughput without coding (redundancy, overhead, addressing, etc.).
Where:
is the RLC peak rate provided to the user by the selected HSDPA radio bearer after scheduling and radio
resource control.
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.
and represent the scaling factor between the application throughput and the RLC (Radio Link Control)
throughput and the throughput offset respectively. 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.
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 user.
• The required HSDPA power in dBm ( )
Note:
• For packet (HSPA - Constant Bit Rate) service users, .
P
term
P
term
P
term R99 –
f
act EDPCCH –
UL
× = P
term HSUPA –
+
P
term
P
term R99 –
f
act EDPCCH –
UL
× = P
term HSUPA –
C
HSDPABearer
× +
f
act EDPCCH –
UL
0.1 =
P
term
P
term R99 –
=
T
appl i cati on
DL
M
b
( )
T
appl i cati on
DL
M
b
( )
R
RLC peak –
DL
c ( )
c Serving cells e
¿
1 BLER
HSDPA
– ( ) ×
ATTI
------------------------------------------------------------------------------------------------------------------------------- SF
Rate
× AR – =
R
RLC peak –
DL
BLER
HSDPA
SF
Rate
AR
ATTI
P
HSDPA
( )
requi red
© Forsk 2010 AT283_TRG_E2 249
Chapter 6: UMTS HSPA Networks
It corresponds to the HSDPA power required to provide the HSDPA bearer user with the downlink requested rate. The
downlink requested rate is the data rate the user would obtain if he was the only user in the cell. In this case, Atoll
determines the HSDPA bearer the user would obtain by considering the entire HSDPA power available of the cell.
is the HS-PDSCH power required to obtain the selected HSDPA bearer (in dBm). If the HSDPA bearer
allocated to the user is the best one, corresponds to the available HS-PDSCH power of the cell. On
the other hand, if the HSDPA bearer has been downgraded in order to be compliant with cell and UE capabilities or for
another reason, will be lower than the available HS-PDSCH power of the cell.
• The served HSDPA power in dBm ( )
This is the HSDPA power required to provide the HSDPA bearer user with the downlink obtained rate. The downlink
obtained rateis the data rate experienced by the user after scheduling and radio resource control.
for packet (HSDPA) and packet (HSPA) service users
And
for packet (HSPA - Constant Bit Rate) service users
Where
is the HS-PDSCH power required to obtain the selected HSDPA bearer.
• The No. of HSUPA Retransmissions (Required)
The maximum number of retransmissions in order to have the requested HSUPA radio bearer with a given BLER.
• The No. of HSUPA Retransmissions (Obtained)
The maximum number of retransmissions in order to have the obtained HSUPA radio bearer with a given BLER.
• The HSUPA application throughput in kbps ( )
This is the net HSUPA throughput without coding (redundancy, overhead, addressing, etc.).
Where:
is the RLC peak rate provided by the selected HSUPA radio bearer after noise rise scheduling.
is the residual BLER after 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.
and 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.
is the maximum number of retransmissions for the obtained HSUPA bearer. This figure is read in the HSUPA Bearer
Se