UMTS/HSPA RRM Algorithms and Parameters Technical Training Part-II

UMTS/HSPA RRM Algorithms
and parameters
Technical Training
Part- II
Table of Contents
1 Overview of Radio Resource Management (RRM)
2 Idle Mode Procedures and Parameters
3 Random Access & Paging Procedures and Parameters
4 Power Control Algorithm and Parameters
5 Admission Control Algorithm and Parameters
6 Load Control Algorithm and Parameters
7 Channel Switching Algorithm and Parameters
8 Handover Algorithm and Parameters
9 HSDPA RRM Algorithm and Parameters
10 HSUPA RRM Algorithm and Parameters

Page 2
Load Control (LC)
LC (Load Control) consist of LDR (Load Reshuffling) and OLC (Over Load Control).
In basic congestion state, LDR will be used to optimize resource distribution, to minimize the impact on the on-going users perceptions.
In overload state, OLC will be used to reduce cell load quickly, keep system stability and the services of high priority users.
Page 3
LDR (Load Reshuffling )
When the usage of cell resource exceeds the basic congestion triggering threshold, the cell enters the basic congestion state.
In this case, Load Reshuffling (LDR) is required to reduce the cell load and increase the access success rate.
The basic congestion of a cell can be caused by power resource, code resource, Iub resource, or NodeB credit resource.
For power resource, the RNC performs periodic measurement and checks whether the cells are congested.
For code, Iub, and NodeB credit resources, the RNC checks whether the cells are congested when resource usage changes.
LDR algorithm consists of three steps as following,
Basic Congestion Triggering

LDR Procedure
LDR Actions
If the congestion of all resources is triggered in a cell, the congestion is relieved in order of resource priority for load reshuffling as
configured through the SET LDCALGOPARA command.
Assume that the parameters are set as follows:

The first priority for load reshuffling (LdrFirstPri) is set to IUBLDR.
The second priority for load reshuffling (LdrSecondPri) is set to CREDITLDR.
The third priority for load reshuffling (LdrThirdPri) is set to CODELDR.
The fourth priority for load reshuffling (LdrFourthPri) is set to UULDR.
Then, the basic congestion is relieved in the following sequence:

LDR based on Iub resource
LDR based on credit resource
LDR based on code resource
LDR based on power resource Page 4
LDR (Load Reshuffling ) – Step(1)
STEP (1) Basic Congestion Triggering
Power Resource
Congestion control based on power resource can be enabled through
the DL_UU_LDR and UL_UU_LDR subparameters of the
NBMLdcAlgoSwitch parameter.
If the parameter NBMUlCacAlgoSelSwitch /

NBMDlCacAlgoSelSwitch is set to ALGORITHM_SECOND , the
load reffuffling algorithm will trigger basic congestion based on
Equivalent Number of Users (ENU).
For an R99 cell:
If the current UL/DL load of the R99 cell is higher than or equal to
the UL/DL LDR trigger threshold (UlLdrTrigThd or
DlLdrTrigThd) for 1,000 ms, the cell is in the basic congestion
state, and the related load reshuffling actions are taken.
If the current UL/DL load of the R99 cell is lower than the UL/DL
LDR relief threshold (UlLdrRelThd or DlLdrRelThd) for 1,000 ms,
the cell enters the normal state again.
Page 5
For an HSPA cell:
In the uplink,
If the uncontrollable load of the cell is higher than or equal to the UL LDR trigger threshold (UlLdrTrigThd) for 600 ms, the cell is in the
basic congestion state, and the related load reshuffling actions.
If the uncontrollable load of the cell is lower than the UL LDR relief threshold (UlLdrRelThd) for 600 ms, the cell enters the normal state
again.
In the downlink, when the cell does not support DC-HSDPA,

If the load of non-HSPA power and GBP in a cell is higher than or equal to the DL LDR trigger threshold (DlLdrTrigThd) for a hysteresis
time that is specified by the parameter DlLdTrnsHysTime, the cell is in the basic congestion state, and the related load reshuffling actions.
If the load of non-HSPA power and GBP in a cell is lower than the DL LDR relief threshold (DlLdrRelThd) for a hysteresis time that is
specified by the parameter DlLdTrnsHysTime, the cell enters the normal state again.
In the downlink, when the cell support DC-HSDPA,

If both cells are in the basic congestion state, the cell group is in the basic congestion state. In such a case, LDR needs to be performed in
each cell respectively.
If both cells are in the normal state, the cell group is in the normal state.
Page 6
Code Resource
Congestion control based on code resource can be enabled through the CELL_CODE_LDR subparameter of the NBMLdcAlgoSwitch
parameter.
If the SF corresponding to the current remaining code of the cell is larger than the value of CellLdrSfResThd, code congestion is
triggered and the related load reshuffling actions are taken.
Iub Resource
Congestion control based on Iub resource can be enabled through the IUB_LDR subparameter of the NodeBLdcAlgoSwitch
parameter in the ADD NODEBALGOPARA or MOD NODEBALGOPARA command.
Iub congestion control in both the uplink and downlink is NodeB-oriented.For the basic congestion caused by Iub resource, all UEs under
the NodeB are the objects of related LDR actions.Iub congestion detection is controlled by TRM: Transmission Resource Management.
For a path, the following congestion-related parameters are applicable:
Congestion detection parameters: FWDCONGBW , BWDCONGBW

The default values of the two parameters are 0, which indicates that no congestion detection will be performed. If the parameters are set to values
other than 0, TRM performs congestion detection according to the settings.
Congestion resolving parameters:FWDCONGCLRBW , BWDCONGCLRBW

These two parameters are used to determine whether the congestion is resolved. Assume that the forward parameters of a port for congestion
detection are defined as follows:
•Configured bandwidth: AVE
•Forward congestion threshold: CON
•Forward congestion resolving threshold: CLEAR (Note that CLEAR is greater than CON.)
•Used bandwidth: USED
Then, the mechanism of congestion detection for the port is as follows:

•Congestion occurs on the port when CON + USED ≥ AVE.
•Congestion disappears from the port when CLEAR + USED < AVE. Page 7
NodeB Credit Resource

The basic congestion caused by NodeB credit resource is of the following types:
Type A: Basic congestion at local cell level

If the cell UL/DL current remaining SF (mapped to credit resource) is higher than UlLdrCreditSfResThd or DlLdrCreditSfResThd
(set through the ADD CELLLDR command), credit congestion at cell level is triggered and related load reshuffling actions are taken in
the current cell.
Type B: Basic congestion at local cell group level (if any)

Type C: Basic congestion at NodeB level
If the cell group or NodeB UL/DL current remaining SF (mapped to credit resource ) is higher than UlLdrCreditSfResThd or
DlLdrCreditSfResThd (set through the ADD NODEBLDR command), credit congestion at cell group or NodeB level is triggered and
related load reshuffling actions are taken. The range of LDR actions is the same as the first type, but the range of UEs to be sorted by
priority is different. All the UEs in the normal cells that belong to the cell group or NodeB will be sorted.
Load Control Algorithm Switch LDC Algorithm Switch

Algorithms
(LdcSwitch) (NBMLdcAlgoSwitch)
Type A LC_CREDIT_LDR_SWITCH CELL_CREDIT_LDR
Type B LCG_CREDIT_LDR_SWITCH LCG_CREDIT_LDR
Type C NODEB_CREDIT_LDR_SWITCH NODEB_CREDIT_LDR
Page 8
LDR (Load Reshuffling ) Parameters
GUI Value Actual Value

MO Parameter ID Parameter Name Parameter Description Unit Baseline Value
Range Range
NODEB_CREDIT_LDR_SWITCH: NodeB credit congestion control algorithm.

When NodeB credit load is heavy, the load can be reshuffled through BE service
rate reduction, renegotiation of uncontrollable real-time service QoS, and CS/PS
inter-RAT handover.
LCG_CREDIT_LDR_SWITCH: Local cell group credit congestion control

load control algorithm. When the local cell group credit load is heavy, the load can be
ULDCALGOPARA LdcSwitch ON,OFF None OFF None
algorithm switch reshuffled through BE service rate reduction, renegotiation of uncontrollable real-
time service QoS, and CS/PS inter-RAT handover.
LC_CREDIT_LDR_SWITCH: Local cell credit congestion control algorithm.

When the local cell credit load is heavy, the load can be reshuffled through BE
service rate reduction, renegotiation of uncontrollable real-time service QoS, and
CS/PS inter-RAT handover.
UL_UU_LDR and DL_UU_LDR: UL /DL UU load reshuffling algorithm. When

the cell is heavily loaded in UL/DL, this algorithm reduces the cell load in UL /DL
by using inter-frequency load handover, BE service rate reduction, uncontrollable
real-time service QoS renegotiation, CS should be inter-RAT, PS should be inter-
UL_UU_LDR:ON
RAT handover, CS should not be inter-RATand, PS should not be inter-RAT
handover and AMR service rate reduction.
DL_UU_LDR :ON
NBMLdcAlgoSw Cell LDC algorithm
UCELLALGOSWITCH CELL_CODE_LDR: Code reshuffling algorithm. When the cell CODE is heavily ON,OFF None OFF
itch switch CELL_CODE_LDR: ON
loaded, this algorithm reduces the cell CODE load by using BE service rate
reduction and code tree reshuffling.
CELL_CREDIT_LDR :
ON
CELL_CREDIT_LDR:Credit reshuffling algorithm. When the cell credit is heavily
loaded, this algorithm reduces the credit load of the cell by using BE service rate
reduction, uncontrollable real-time service QoS renegotiation, CS should be inter-
RAT, PS should be inter-RAT handover, CS should not be inter-RATand and PS
should not be inter-RAT handover.
Page 9
GUI Value Actual Value Baseline

MO Parameter ID Parameter Name Parameter Description Unit
Range Range Value
IUB_LDR (Iub congestion control algorithm): When the NodeB Iub load is
heavy, users are assembled in priority order among all the NodeBs and some
users are selected for LDR action (such as BE service rate reduction) in order to
reduce the NodeB Iub load.
NODEB_CREDIT_LDR (NodeB level credit congestion control algorithm): When
NodeB LDC algorithm the NodeB level credit load is heavy, users are assembled in priority order
UNODEBALGOPARA NodeBLdcAlgoSwitch ON,OFF None OFF None
switch among all the NodeBs and some users are selected for LDR action in order to
reduce the NodeB level credit load.
LCG_CREDIT_LDR (Cell group level credit congestion control algorithm): When
the cell group level credit load is heavy, users are assembled in priority order
among all the NodeBs and some users are selected for LDR action in order to
reduce the cell group level credit load.
first priority for load
ULDCALGOPARA LdrFirstPri If congestion is triggered by multiple resources such as credit and code at the IUBLDR
reshuffling
same time, the congestion of resources specified in this parameter is processed
second priority for load IUBLDR, IUBLDR,
ULDCALGOPARA LdrSecondPri with the first priority. CREDITLDR
reshuffling CODELDR, CODELDR,
IUBLDR refers to processing of LDR action trigged by Iub bandwidth. None
third priority for load UULDR, UULDR,
ULDCALGOPARA LdrThirdPri CREDITLDR refers to processing of LDR action trigged by credit. CODELDR
reshuffling CREDITLDR CREDITLDR
CODELDR refers to processing of LDR action trigged by code.
fourth priority for load
ULDCALGOPARA LdrFourthPri UULDR refers to processing of LDR action trigged by Uu. UULDR
reshuffling
Forward congestion If the available forward bandwidth is less than or equal to this value, the forward
TRMLOADTH FWDCONGBW 0~200000 0~200000 kbit/s 0
remain bandwidth congestion alarm is emitted and forward congestion control is triggered.
If the available backward bandwidth is less than or equal to this value, the
Backward congestion
TRMLOADTH BWDCONGBW backward congestion alarm is emitted and backward congestion control is 0~200000 0~200000 kbit/s 0
remain bandwidth
triggered.
Forward congestion clear If the available forward bandwidth is greater than this value, the forward
TRMLOADTH FWDCONGCLRBW 0~200000 0~200000 kbit/s 0
remain bandwidth congestion alarm is cleared and forward congestion control is stopped.
Backward congestion If the available backward bandwidth is greater than this value, the backward
TRMLOADTH BWDCONGCLRBW 0~200000 0~200000 kbit/s 0
clear remain bandwidth congestion alarm is cleared and backward congestion control is stopped.
Page 10
Range Range Value
If the ratio of UL/DL load of the cell to the uplink capacity is not lower than this threshold, the
UL LDR trigger
UlLdrTrigThd UL/DL load reshuffling function of the cell is triggered.
threshold
UCELLLDM DlLdrTrigThd After the basic congestion state of the cell load is released, the system no longer implements 0~100 0~1 % 55/70
DL LDR trigger
the LDR action. Because the load fluctuates, the difference between the LDR release
threshold
threshold and trigger threshold should be higher than 10% to avoid ping-pong effect.
If the ratio of UL /DL load of the cell to the uplink capacity is lower than this threshold, the
UL LDR release
UlLdrRelThd UL/DL load reshuffling function of the cell is stopped.
threshold
UCELLLDM DlLdrRelThd After the basic congestion state of the cell load is released, the system no longer implements 0~100 0~1 % 45/60
DL LDR release
the LDR action. Because the load fluctuates, the difference between the LDR release
threshold
threshold and trigger threshold should be higher than 10% to avoid ping-pong effect.
DL State Trans If the DL load state of the cell is lasted longer than this threshold, the DL load state of the cell
UCELLLDM DlLdTrnsHysTime 10~600000 10~600000 ms 1000
Hysteresis threshold transfers.
SF4, SF8,
This parameter specifies the Cell SF reserved threshold used for judging whether the code SF4, SF8, SF16,
Cell LDR SF reserved SF16, SF32,
UCELLLDR CellLdrSfResThd load reshuffling (LDR) is allowed. The code load reshuffling could be triggered only when the SF32, SF64, None SF8
threshold SF64, SF128,
minimum available SF of a cell is higher than this threshold. SF128, SF256
SF256
SF4, SF8,
Reserved SF threshold in uplink credit LDR. The uplink credit LDR could be triggered only SF4, SF8, SF16,
Ul LDR credit SF SF16, SF32,
UCELLLDR UlLdrCreditSfResThd when the SF factor corresponding to the uplink reserved credit is higher than the uplink or SF32, SF64, None SF8
reserved threshold SF64, SF128,
downlink credit SF reserved threshold. SF128, SF256
SF256
SF4, SF8,
Reserved SF threshold in downlink credit LDR. The downlink credit LDR could be triggered SF4, SF8, SF16,
Dl LDR credit SF SF16, SF32,
UCELLLDR DlLdrCreditSfResThd only when the SF factor corresponding to the downlink reserved credit is higher than the SF32, SF64, None SF8
uplink or downlink credit SF reserved threshold. SF128, SF256
SF256
Threshold of SF reserved in uplink credit LDR. The uplink credit LDR is triggered when the SF
factor corresponding to the uplink reserved credit is higher than the uplink or downlink credit
SF4, SF8,
SF reserved threshold. The lower the parameter value is, the easier the credit enters the SF4, SF8, SF16,
Ul LDR Credit SF SF16, SF32,
UNODEBLDR UlLdrCreditSfResThd congestion status, the easier the LDR action is triggered, and the easier the user experience SF32, SF64, None SF8
is affected. A lower code (SF)resource LDR trigger threshold, however, causes a higher SF128, SF256
SF256
admission success rate because the resource is reserved. The parameter should be set
based on the operator's requirement.
Threshold of SF reserved in downlink credit LDR. The downlink credit LDR is triggered when
the SF factor corresponding to the downlink reserved credit is higher than the uplink or
SF4, SF8,
downlink credit SF reserved threshold. The lower the parameter value is, the easier the credit SF4, SF8, SF16,
Dl LDR Credit SF SF16, SF32,
UNODEBLDR DlLdrCreditSfResThd enters the congestion status, the easier the LDR action is triggered, and the easier the user SF32, SF64, None SF8
experience is affected. A lower code (SF) resource LDR trigger threshold, however, causes a SF128, SF256
higher admission success rate because the resource is reserved. The parameter should be
Page 11
SF256
set based on the operator's requirement.
STEP (2) LDR Procedure
The RNC periodically takes actions if the basic congestion is detected.
The following procedures apply to HSPA cells and R99 cells. For R99 cells, only DCH UEs are selected by LDR actions.
Whether the users of gold priority are selected by LDR actions is specified by the GoldUserLoadControlSwitch parameter.
When the cell is in the basic congestion state, the RNC takes one of the following actions in each period (specified by the
LdrPeriodTimerLen parameter) until the congestion is relieved:
LDR Actions:
Inter-frequency load handover

Code reshuffling
BE service rate reduction
AMR rate reduction
Inter-RAT load handover in the CS domain, which involves the following actions:
•Inter-RAT Should Be Load Handover in the CS Domain
•Inter-RAT Should Not Be Load Handover in the CS Domain
Inter-RAT load handover in the PS domain, which involves the following actions:
Inter-RAT Should Be Load Handover in the PS Domain
Inter-RAT Should Not Be Load Handover in the PS Domain
Iu QoS renegotiation

MBMS power reduction
The sequence of LDR actions can be changed through the ADD UCELLLDR command. Page 12
Turn on the LDR algorithm switch
Mark "current LDR state = uncongested"
Start sending the LDM congestion indication LDR Procedure

Example, the sequence of LDR actions is fixed to inter-frequency
Mark "current action = first LDR action" load handover, code reshuffling, BE rate reduction, inter-RAT
handover in CS domain, inter-RAT handover in PS domain, AMR
Clear "selected" mark of all UE LDR actions
Congestion rate reduction, QOS renegotiation on Iu interface, and MBMS power
state indication Wait for congestion indication reduction.
Trigger the LDR
period timer Yes No when the system is congested, the inter-frequency load handover is
Current LDR state = congested?
initiated first.
Inter-frequency
load handover
Successful?
Yes
If the handover succeeds, the algorithm continues to check
No whether the system is congested. If the system is still
Code
Successful?
Yes congested, the inter-frequency load handover is initiated
reshuffling Wait for the
No
expiration of the timer
again.
BE rate Yes
Successful?
reduction
Keep the
No If the handover fails, code reshuffling is performed: If the
Inter-RAT
action
handover Successful?
Yes
Mark
code reshuffling succeeds, the algorithm continues to check
sequence
unchanged
in CS domain
No "current action whether the system is congested. If the system is still
Inter-RAT = successful
and take the
handover Successful?
Yes action" congested, the code reshuffling is initiated again.
current action
in PS domain
firstly No
AMR rate
Successful?
Yes If the code reshuffling fails, the next action, that is, BE rate
reduction
No reduction, is taken.
QoS
Yes
renegotiation Successful?
on Iu interface No The rest may be deduced by analogy.
MBMS power Yes
Successful?
reduction
No
No LDR action is taken Mark Page 13

or all actions fail "current action = first LDR action"
LDR Actions Intended for Different Resources
When the basic congestion is triggered by different resources, the congestion can be relieved in a order
set by running the SET ULDCALGOPARA command.
LDR Actions
Renegotiation
MBMS Power
Handover in
Handover in
Reshuffling
CS Domain
PS Domain
Frequency
Reduction
Reduction
Reduction
AMR Rate
Handover
Inter-RAT
Inter-RAT
BE Rate
Iu QoS
Code
Inter-
Load
Resource UL/DL Channel
DCH √ √ √ √ √ √
UL
HSUPA √ √ √
DCH √ √ √ √ √* √
Power HSDPA √ √ √
DL DC-HSDPA √ √
FACH (MBMS) √*
DCH √ √ √
UL
HSUPA √ √
Iub DCH √ √ √
DL HSDPA √ √
FACH (MBMS)
– –
DCH √ √ √
Code HSDPA
DL
FACH (MBMS)
DCH √ √ √ √
UL
HSUPA √ √ √
DCH √ √ √ √
Credit
HSDPA
DL Page 14
FACH (MBMS)

Range Range Value
Indicates whether gold users involve in the switch of congestion control. According to
GoldUserLoadControl Gold user load OFF(OFF),
UCELLLDR the policy set for gold users by operators, if service quality of gold users should be OFF, ON None OFF
Switch control switch ON(ON)
guaranteed even in resource congestion, the switch should be disabled.
Identifying the period of the LDR execution. When basic congestion occurs, execution
of LDR can dynamically reduce the cell load. The lower the parameter value is, the
more frequently the LDR action is executed, which decreases the load quickly. If the
parameter value is excessively low, an LDR action may overlap the previous one
before the previous result is displayed in LDM.
LDR period timer
ULDCPERIOD LdrPeriodTimerLen The LDR algorithm aims to slowly reduce the cell load and control the load belowthe 1~86400 1~86400 s 10
length
admission threshold, each LDR action takes a period (for example the inter-RAT load
handover needs a delay of about 5 s if the compressed mode is needed),
and there is a delay for the LDM module responds to the load decreasing (the delay is
about3 s when the L3 filter coefficient is set to 6), so the parameter value should be
higher than 8s.
Page 15
MO Parameter ID Parameter Name Parameter Description GUI Value Range Actual Value Range Unit Baseline Value
UCELLLDR DlLdrFirstAction DL LDR first action CodeAdj
UCELLLDR DlLdrSecondAction DL LDR second action InterFreqLDHO
UCELLLDR DlLdrThirdAction DL LDR third action This parameter is used to set the sequence of BERateRed
LDR actions.
UCELLLDR DlLdrFourthAction DL LDR fourth action NoAct, InterFreqLDHO, NoAct, InterFreqLDHO, NoAct
NOACT: No load reshuffling action is taken.
BERateRed, QoSRenego, BERateRed, QoSRenego,
INTERFREQLDHO: The inter-frequency load
UCELLLDR DlLdrFifthAction DL LDR fifth action CSInterRatShouldBeLDHO, CSInterRatShouldBeLDHO, NoAct
handover is performed.
PSInterRatShouldBeLDHO, PSInterRatShouldBeLDHO,
BERATERED: Channels are reconfigured for
AMRRateRed, AMRRateRed, MBMSDecPower,
UCELLLDR DlLdrSixthAction DL LDR sixth action the BE service. NoAct
MBMSDecPower, CodeAdj, CodeAdj,
QOSRENEGO: The renegotiation on the QoS of
CSInterRatShouldNotLDHO, CSInterRatShouldNotLDHO,
the uncontrollable real-time service is performed.
UCELLLDR DlLdrSeventhAction DL LDR seventh action PSInterRatShouldNotLDHO PSInterRatShouldNotLDHO NoAct
CSINTERRATSHOULDBELDHO: The inter-
RAT SHOULDBE load handover of the CS
UCELLLDR DlLdrEighthAction DL LDR eighth action domain is performed. NoAct
PSINTERRATSHOULDBELDHO: The inter-RAT
UCELLLDR DlLdrNinthAction DL LDR ninth action SHOULDBE load handover of the PS domain is NoAct
performed.
None
AMRRATERED (AMR service rate decreasing):
UCELLLDR DlLdrTenthAction DL LDR tenth action The setting of the TFC subset and the NoAct
negotiation of the service rate can be performed
UCELLLDR UlLdrFirstAction UL LDR first action for the AMR voice service. InterFreqLDHO
MBMSDECPOWER (MBMS power limiting):
UCELLLDR UlLdrSecondAction UL LDR second action The MBMS service is configured with the BERateRed
minimum power.
CODEADJ (code tree reshuffling): The
UCELLLDR UlLdrThirdAction UL LDR third action fragments of the downlink code tree are NoAct
NoAct, InterFreqLDHO, NoAct, InterFreqLDHO,
arranged. BERateRed, QoSRenego, BERateRed, QoSRenego,
UCELLLDR UlLdrFourthAction UL LDR fourth action CSINTERRATSHOULDNOTBELDHO: The CSInterRatShouldBeLDHO, CSInterRatShouldBeLDHO, NoAct
inter-RAT SHOULDNOTBE load handover of the PSInterRatShouldBeLDHO, PSInterRatShouldBeLDHO,
CS domain is performed. AMRRateRed, AMRRateRed,
UCELLLDR UlLdrFifthAction UL LDR fifth action NoAct
PSINTERRATSHOULDNOTLDHO: The inter- CSInterRatShouldNotLDHO, CSInterRatShouldNotLDHO,
RAT SHOULDNOTBE load handover of the PS PSInterRatShouldNotLDHO PSInterRatShouldNotLDHO
UCELLLDR UlLdrSixthAction UL LDR sixth action domain is performed. NoAct
UCELLLDR UlLdrSeventhAction UL LDR seventh action NoAct

Page 16
UCELLLDR UlLdrEighthAction UL LDR eighth action NoAct
STEP (3) LDR Actions
LDR Action (1) Inter-Frequency Load Handover Based on Blind HO
The inter-frequency load handover algorithm is restricted by the inter frequency hard handover algorithm switch.
If the UE is in the soft handover state, inter-frequency load handover can be performed only when the HO_ALGO_LDR_ALLOW_SHO_SWITCH
subparameter of the HoSwitch parameter is set to 1.
The CodeCongSelInterFreqHoInd parameter can be set to allow inter-frequency handover in order to reduce congestion on code resource.
The inter-frequency load handover can be performed based on blind handover or measurement, which can be decided by the parameter
InterFreqLDHOMethodSelection.
If the InterFreqLDHOMethodSelection parameter is set to BLINDHO, the inter-frequency load handover based on blind handover performs the
following steps:
1. The algorithm checks whether cells for inter-frequency blind handover are available. If available, the algorithm goes to the next step. Otherwise,
the action fails, and the algorithm takes the next action.
Whether the neighboring cells support blind handover is specified by the parameter BlindHoFlag.
2. The algorithm selects the target cell according to the type of resource that causes the basic congestion:
− If the basic congestion is caused by power resource:
If the candidate cell does not support DC-HSDPA, the algorithm checks whether the load margin of the target cell is higher than both
UlInterFreqHoCellLoadSpaceThd and DlInterFreqHoCellLoadSpaceThd and whether the load of the target cell is normal.
If the candidate cell supports DC-HSDPA, the concerned cell group and the candidate cell must have sufficient power margin.
The load margin refers to the difference between the load of the target cell and the basic congestion triggering threshold of the target cell.
If the margin is not higher than the threshold, the action fails, and the algorithm takes the next action.
If there is more than one cell meeting the requirements, the first cell in the list of the neighbor cells is selected as the blind handover target cell.
− If the basic congestion is caused by code resource:
Whether there are blind handover target cells meeting the requirements is decided by the following conditions:
a. The minimum SF of the target cell is not greater than that of the current cell.
b. The difference of code usage between the current cell and the target cell is greater than LdrCodeUsedSpaceThd.
c. The state of target cell is normal.
If there is no such cell, this action fails and the algorithm takes the next action.
Page 17
If there is more than one cell meeting the requirements, the first cell in the list of the neighbor cells is selected as the blind handover target cell.
3. The algorithm selects the UEs to be handed over according to the setting of InterFreqLdHoForbidenTC and
NbmLdcUeSelSwitch:
 If NbmLdcUeSelSwitch is set to NBM_LDC_MATCH_UE_ONLY, the algorithm performs the following steps:
a. Selects the UEs that meet the following conditions as candidate UEs.
 The service types of UEs are not forbidden for LDR handover by parameter InterFreqLdHoForbidenTC.
 The service types of UEs are supported by the target cell.
b. Sorts the candidate UEs whose rates are no higher than the handover bandwidth thresholds, based on the integrated priority.
c. Selects the UE with the lowest integrated priority for handover.
If NbmLdcUeSelSwitch is set to NBM_LDC_MATCH_UE_FIRST, the algorithm performs the following steps:
 The service types of UEs are not restricted for LDR handover by parameter InterFreqLdHoForbidenTC.
 The service types of UEs are supported by the target cell.
b. Sorts the candidate UEs whose rates are not higher than the handover bandwidth thresholds, based on the integrated priority.
If the rates of all the candidate UEs are higher than the handover bandwidth thresholds, the algorithm performs the following steps:
The service types of UEs are not restricted for LDR handover by parameter InterFreqLdHoForbidenTC.
The service types of UEs are not supported by the target cell.
b. Sorts the UEs whose rates are not higher than the handover bandwidth threshold, based on the integrated priority.
If NbmLdcUeSelSwitch is set to NBM_LDC_ALL_UE, the algorithm performs the following steps:
a. From the current cell, selects the UEs whose service types are not forbidden for LDR handover by parameter
InterFreqLdHoForbidenTC.
b. Sorts the UEs whose rates are no higher than the handover bandwidth thresholds, based on the integrated priority.
If multiple UEs have the same lowest integrated priority, the algorithm selects the one with the highest rate for handover.
The UL and DL handover bandwidth thresholds are specified by UlInterFreqHoBWThd and DlInterFreqHoBWThd respectively.
Both the thresholds are considered in the selection of the target UE.
4. After selecting the target cell and the UE, the RNC makes blind handover decision. Page 18
LDR Action (1)Inter-Frequency Load Handover Based on Measurement
The inter-frequency load handover based on measurement can be performed only if the basic congestion is caused by power resource.
If the InterFreqLDHOMethodSelection parameter is set to MEASUREHO, the inter-frequency load handover is performed based on
measurement. The LDR algorithm is implemented by performing the following steps:
1. The RNC selects the UE whose service types are not forbidden for LDR handover by parameter InterFreqLdHoForbidenTC, and
then sorts the selected UEs according to their integrated priority and performs inter-frequency load handover based on
measurement on the UE with the lowest integrated priority.
2. The RNC selects the candidate cells that meet the following conditions:
 The cell must be an inter-frequency neighboring cell of the current cell. The cell must not be a DRNC inter-frequency neighboring cell.
 The frequency of the cell is within the band supported by the UE.
 The cell must meet the following conditions on load margin:the algorithm checks whether the load margin of the target cell is higher than
both UlInterFreqHoCellLoadSpaceThd and DlInterFreqHoCellLoadSpaceThd and whether the load of the target cell is normal.
 The DrdOrLdrFlag parameter of the cell is set to True, indicating that the cell can be measured.
 If the NbmLdcUeSelSwitch parameter is set to NBM_LDC_MATCH_UE_ONLY, the cell must support the service requested by the UE.
 If such candidate target cells do not exist, the inter-frequency load handover action fails and the algorithm takes the next action.
 If such candidate cells exist, the following step is performed.
3. The RNC issues a measurement control message to the UE, requesting the UE to measure the signal quality of all candidate cells.
4. The UE measures the RSCP and Ec/No of the candidate cells and periodically reports the measurement results to the RNC.
The reporting period is specified by the PrdReportInterval parameter.
5. Based on the received measurement results, the RNC selects the candidate target cells. The candidate target cells must meet the
following conditions:
 The cell is not in the basic congestion state.
 The measured RSCP is higher than the RSCP threshold that is specified by the TargetFreqThdRscp parameter.
 The measured Ec/No is higher than the Ec/No threshold that is specified by the TargetFreqThdEcN0 parameter.
 If such candidate target cells do not exist, the inter-frequency load handover action fails and the algorithm takes the next action.
 If such candidate target cells exist, the following step is performed.
6. The RNC selects the cell with the highest priority from the candidate target cells to perform inter-frequency hard handover.
 If the handover succeeds, the LDR action is complete.
 If the handover fails, the RNC tries accessing the cell with the second highest priority to perform inter-frequency hard handover until the
handover succeeds or it has tried accessing all the candidate target cells. Page 19

Range Range Value
HO_ALGO_LDR_ALLOW_SH When the switch is on, the LDR inter-frequency

UCORRMALGOSWITCH HO_ALGO_LDR_ALLOW_SHO_SWITCH OFF,ON OFF,ON None ON
O_SWITCH handover is allowed during soft handover.
This switch is valid only when the inter-frequency

handover switch is enabled. TRUE means that inter-
Code congestion select inter- frequency handover is selected in code resource FALSE,
UCELLLDR CodeCongSelInterFreqHoInd FALSE, TRUE None FALSE
freq indication congestion. FALSE means that inter-frequency TRUE
handover is not selected in code resource
congestion.
This parameter specifies load handover

InterFreq Load Handover method.When network is composed of same BLINDHO, BLINDHO,
UCELLLDR InterFreqLDHOMethodSelection None BLINDHO
Method Selection frequency band,Blind Handover method is suggested MEASUREHO MEASUREHO
.Otherwise,Measure handover is suggested.
Whether to perform blind handover.

FALSE,
UINTERFREQNCELL BlindHoFlag Blind Handover Flag The value FALSE indicates that the cell is not FALSE, TRUE None FALSE
TRUE
considered as a candidate cell for blind handover.
The inter-frequency neighboring cell could be

UCELLLDR UlInterFreqHoCellLoadSpaceThd UL HO load space threshold selected as the destination of load handover only 0~100 0~1 % 20
when its load remaining space is larger than this
threshold. The lower the parameter is, the easier it is
to find a qualified target cell for the blind handover.
UCELLLDR DlInterFreqHoCellLoadSpaceThd DL HO load space threshold The higher the parameter is, the more difficult it is for 0~100 0~1 % 20
the inter-frequency blind handover occurs.
Code resource usage difference threshold.

Inter-frequency handover is triggered when the
InterFreq HO code used ratio
UCELLLDR LdrCodeUsedSpaceThd difference of the resource usage of the current cell 0~100 0~1 % 13
space threshold
and that of the target cell is greater than this
threshold.
Page 20
R99_CONVERSATIONAL, R99_CONVERSATIONAL,
R99_STREAMING, R99_STREAMING,
R99_BE, R99_BE,
This parameter specifies the forbidden traffic
HSDPA_CONVERSATION HSDPA_CONVERSATIONA
Forbidden Traffic Class classes when perform inter-frequency
UCELLLDR InterFreqLdHoForbidenTC AL, HSDPA_STREAMING, L, HSDPA_STREAMING, None None
for inter-freq HO handover, in order to prevent disarranging of
HSDPA_BE, HSDPA_BE,
the layers.
HSPA_CONVERSATIONA HSPA_CONVERSATIONAL,
L, HSPA_STREAMING, HSPA_STREAMING,
HSPA_BE HSPA_BE
The UE can be selected to process load

UL HO maximum
UCELLLDR UlInterFreqHoBWThd handover only when its bandwidth is less than 0~400000 0~400000 bit/s 200000
bandwidth
this threshold.
The UE can be selected to process load

DL HO maximum
UCELLLDR DlInterFreqHoBWThd handover only when its bandwidth is less than 0~400000 0~400000 bit/s 200000
bandwidth
this threshold.
The algorithms with the above values represent

are as follow:
NBM_LDC_ALL_UE: When inter-freq handover
select user occurs, no need to consider NBM_LDC_ALL_UE, NBM_LDC_ALL_UE,
whether target cell support Ue. NBM_LDC_MATCH_UE_ NBM_LDC_MATCH_UE_O
Inter-freq Handover Select NBM_LDC_MAT
UCELLALGOSWITCH NbmLdcUeSelSwitch NBM_LDC_MATCH_UE_ONLY: When inter- ONLY, NLY, None
User algorithm switch CH_UE_ONLY
freq handover select user occurs, only consider NBM_LDC_MATCH_UE_F NBM_LDC_MATCH_UE_FI
Ues supported by target cell. IRST RST
NBM_LDC_MATCH_UE_FIRST: When inter-
freq handover select user occurs, first consider
Ues supported by target cell.
Page 21
Specify the flags of the cells that the DRD measurement or

LDR measurement is performed.
UINTERFREQN The value "TRUE" indicates that the cell can be considered FALSE(Do not send),
DrdOrLdrFlag DrdOrLdrFlag FALSE, TRUE None FALSE
CELL as the measurement object in the DRD measurement TRUE(Send)
algorithm or LDR measurement algorithm. The value
"FALSE" indicates that the cell is invalid.
D250, D500, D1000,

Interval between sending of periodic measurement reports.
D2000, D3000, D4000, 250, 500, 1000, 2000,
Inter-Frequency This parameter has impact on the Uu signaling flow. If this
D6000, D8000, 3000, 4000, 6000, 8000,
Periodic parameter is set to a small value, the RNC may have high
UCELLMCDRD PrdReportInterval D12000, D16000, 12000, 16000, 20000, ms D3000
Measurement load when processing signaling. If this parameter is set to a
D20000, D24000, 24000, 28000, 32000,
Report Interval great value, the network cannot detect the signal changes
D28000, D32000, 64000
in time. This may delay the inter-frequency handover.
D64000
RSCP Threshold for the target cell. This parameter is used

to estimate the signal quality of the periodic reports.The
DRD is triggered only when the signal quality of the target
cell is higher than this parameter. If this parameter is set to
Target Frequency
a greater value, it is difficult for subscribers to re-access
UCELLMCDRD TargetFreqThdRscp RSCP Trigger -115~-25 -115~-25 dBm -92
another cell with a higher priority; however, the re-attempt
Threshold
success rate is high. If this parameter is set to a lower
value, it is easy for subscribers to re-access another cell
with a higher priority; however, the re-attempt success rate
however is low.
Ec/No Threshold for the target cell. This parameter is used

to estimate the signal quality of the periodic reports. The
DRD is triggered only when the signal quality of the target
cell is higher than this parameter. If this parameter is set to
a greater value, it is difficult for subscribers to re-access
Target Frequency
another cell with a higher priority; however, the re-attempt
UCELLMCDRD TargetFreqThdEcN0 EcNo Trigger -24~0 -24~0 dB -12
success rate is high. If this parameter is set to a lower
Threshold
value, it is easy for subscribers to re-access another cell
with a higher priority; however, the re-attempt success rate
however is low. Note:In order to increase the successful
rate of handover, inner protection mechanism keep Ec/No of
target cell larger than -16. Page 22
LDR Action (2) BE Rate Reduction

When admission control of Power/NodeB Credit is disabled, it is not recommended that the BE Rate Reduction to be configured as an
LDR action in order to avoid ping-pong effect.
BE rate reduction can only be performed when the DRA_DCCC_SWITCH subparameter of the DraSwitch parameter is set to 1.
Different from the TF restriction to the OLC function, the BE rate reduction is implemented by bandwidth reconfiguration. The bandwidth
reconfiguration requires signaling interaction on the Uu interface. This procedure is relatively long.
In the same environment, different rates have different downlink transmit powers. The higher the rate, the greater the downlink transmit
power. Therefore, the load can be reduced by bandwidth reconfiguration.
For HSUPA services, the consumption of CEs is based on the bit rate. The higher the rate, the more the consumption of CEs. Therefore,
the consumption of CEs can be reduced by bandwidth reconfiguration.
The LDR algorithm operates as follows:

1. Based on the integrated priority, the algorithm sorts the BE RABs in descending order.
2. The algorithm selects the BE RABs that meet the following condition:
− The current rate of the BE RAB is higher than the GBR specified by running the SET USERGBR command.
− The BE RAB has the lower integrated priorities.
The number of selected RABs is specified by the UlLdrBERateReductionRabNum or DlLdrBERateReductionRabNum parameter.

If the integrated priorities of some RABs are identical, the RAB with the highest rate is selected.
3. If services can be selected, the action is successful. If services cannot be selected, the action fails and go to next actions.
4. The bandwidth of the selected services is reduced to the specified rate based on the DCCC (Dynamic Channel
Configuration Control).
5. The reconfiguration is complete as indicated by the RADIO BEARER RECONFIGURATION message on the Uu interface and
through the synchronized radio link reconfiguration procedure on the Iub interface.
Page 23
When the cell enters the basic congestion state, the RNC decreases BE rates through the DCCC function to relieve the congestion.
The following table shows the basic principle of DCCC based on LDR.
Direction Event Event description DCCC
Cell load (uplink or downlink power) is Decrease rate for UEs selected by LDR
DCH UL/DL or E-DCH UL Basic congestion
greater than LDR trigger threshold until congestion is relieved.
Cell leaves basic congestion state The previously downgraded bearers may
DCH UL/DL or E-DCH UL Congestion relieved + 4a
and receives 4a have their rate increased.
Rate Adjustment Based on LDR

When LDR triggers BE service rate decrease, the selected RAB rate is decreased to the target rate. The target rates are different for DCH
RAB and HSUPA RAB.
For a DCH RAB, the rate adjustment level is fixed to 3_Rates. If the rate decrease level (UlRateDnAdjLevel or DlRateDnAdjLevel) is
set to 2_Rates, the middle rate is calculated by the RNC and the middle rate is approximately equal to the MBR divided by two.
− If the current rate is equal to the MBR, it is decreased to the UL/DL middle rate threshold (UlMidRateThd/DlMidRateThd).
− If the current rate is higher than the GBR but lower than the MBR, it is decreased to the GBR.
For an E-DCH RAB, a rate is selected from the HSUPA UL rate adjustment set (EdchRateAdjustSet) as the target rate for rate decrease.
− The target rate must be smaller than but the closest to the current rate.
− The target rate must be greater than or equal to the GBR.
Page 24
The figure shows an example of UL BE rate decrease and increase in the case
that the RAB is a DCH RAB.
In this example, a rate decrease is triggered by UL basic congestion, and a rate

increase is triggered when basic congestion is relieved.
In addition, the current UL rate can be the initial rate, middle rate, or MBR.
When the current UL rate is the MBR, it is decreased to the middle rate if a UL
basic congestion report is received, and then decreased to the GBR if a second
UL basic congestion report is received.
The figure shows an example of DL BE rate decrease and increase in

the case that the RAB is a DCH RAB.
In this example, a rate decrease is triggered by DL basic congestion, and a rate

increase is triggered when basic congestion is relieved. In addition, the current DL
rate can be the initial rate, middle rate, or MBR.
When the current DL rate is the MBR, it is decreased to the middle rate if a DL
basic congestion report is received, and then decreased to the GBR if a second
DL basic congestion report is received.
Page 25

Range Range Value
UCORRMALGO When the switch is on, the dynamic channel

DRA_DCCC_SWITCH DRA_DCCC_SWITCH OFF,ON OFF,ON None ON
SWITCH reconfiguration control algorithm is used for the RNC.
This parameter determines whether the UL rate is raised

Uplink Rate Increase
UDCCC UlRateUpAdjLevel to the maximum rate in one step or two steps for BE 2_Rates, 3_Rates 2_Rates, 3_Rates None 3_Rates
Adjust Level
service set up on DCH in uplink.
This parameter determines whether the UL rate is

Uplink Rate Decrease
UDCCC UlRateDnAdjLevel lowered to the minimum rate in one step or two steps for 2_Rates, 3_Rates 2_Rates, 3_Rates None 3_Rates
Adjust Level
BE service set up on DCH in uplink.
Threshold of the UL intermediate rate when the 3-step D16, D32, D64,
Uplink Mid Bit Rate 16, 32, 64, 128,
UDCCC UlMidRateThd mode is used for UL rate adjustment and the manual D128, D144, D256, kbit/s D128
Threshold 144, 256, 384
mode is used for UL intermediate rate adjustment. D384
Threshold of the DL intermediate rate when the 3-step D16, D32, D64,
Downlink Mid Bit Rate 16, 32, 64, 128,
UDCCC DlMidRateThd mode is used for DL rate adjustment and the manual D128, D144, D256, kbit/s D128
Threshold 144, 256, 384
mode is used for DL intermediate rate adjustment. D384
RATE_8KBPS: not selected

RATE_8KBPS,
RATE_16KBPS,
RATE_32KBPS,
RATE_64KBPS,
RATE_128KBPS: selected
RATE_128KBPS,
8, 16, 32, 64, 128, RATE_144KBPS: not selected
RATE_144KBPS,
HSUPA traffic uplink rate adjustment set. It contains 144, 256, 384, RATE_256KBPS: selected
UEDCHRATEA HSUPA UpLink Rate RATE_256KBPS,
EdchRateAdjustSet rates for rate adjustment for uplink channel of HSUPA 608, 1280, 2048, None RATE_384KBPS: not selected
DJUSTSET Adjust Set RATE_384KBPS,
traffic and is used in the HSUPA DCCC algorithm. 2720, 5440, RATE_608KBPS: selected
RATE_608KBPS,
11480 RATE_1280KBPS: not selected
RATE_1280KBPS,
RATE_2048KBPS,
RATE_2720KBPS,
RATE_5440KBPS,
RATE_11480KBPS: not
RATE_11480KBPS
selected
Page 26
LDR Action (3) QoS Renegotiation for Uncontrollable Real-Time Services
The QoS renegotiation function for uncontrollable real-time services is set by the PS_STREAM_IU_QOS_NEG_SWITCH subparameter
of the PsSwitch parameter.
Uncontrollable real-time services refer to PS streaming services. The load can be reduced by adjusting the rates of real-time services
through QoS renegotiation.
The uncontrollable real-time service cannot perform rate down automatically like BE service due to the QoS requirement. That is, GBR is
specified in RAB assignment procedure and must be guaranteed.
In 3GPP R5, the RNC initiates the RAB renegotiation procedure through the “RAB MODIFY REQUEST” message on the Iu interface.
Upon reception of the “RAB MODIFY REQUEST” message, the Core Network (CN) sends the “RAB ASSIGNMENT REQUEST”
message to the RNC for RAB parameter reconfiguration. Based on this function, the RNC can adjust the rate of real-time services to
reduce the load of the current cell. The RNC will request a new MBR and GBR that are the lowest ones among the alternative
configurations in the “RAB ASSIGNMENT message” from the CN. However, the CN can decide how to react to the request upon
reception of the “RAB MODIFY REQUEST” message.

1. Based on the integrated priority, the algorithm sorts the RABs for real-time services in the PS domain in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities for QoS renegotiation. The number of selected RABs is specified
by the UlLdrPsRTQosRenegRabNum or DlLdrPsRTQosRenegRabNum parameter. If the RNC cannot find an appropriate
service for the QoS renegotiation, the action fails. The algorithm takes the next action.
3. The algorithm performs QoS renegotiation for the selected services. The GBR during the service setup is the minimum rate of the
service after the QoS renegotiation.
4. The RNC initiates the RAB MODIFY REQUEST message to the CN for the QoS renegotiation. Upon reception of the RAB MODIFY
REQUEST message, the CN sends the RAB ASSIGNMENT REQUEST message to the RNC for RAB parameter reconfiguration.
Page 27
LDR Action (4) Inter-RAT Handover in the CS Domain
This action can only be performed when the CS inter-RAT handover algorithm is enabled.
The size and coverage mode of a 2G cell are different from those of a 3G cell. Therefore, inter-RAT blind handover is not considered.
Inter-RAT handover in the CS domain involves the following actions.
Inter-RAT Should Be Load Handover in the CS Domain

1. Based on the integrated priority, the algorithm sorts the UEs with the "service handover" IE set to "handover to GSM should be
performed" in the CS domain in descending order.
2. The algorithm selects the UEs with the lowest integrated priorities. The number of selected UEs is specified by the
UlCSInterRatShouldBeHOUeNum or DlCSInterRatShouldBeHOUeNum parameter.
3. For the selected UEs, the LDR module sends the load handover command to the inter-RAT handover module, requesting the inter-
RAT handover module to hand over the UEs to the 2G system.
4. The handover module decides to trigger the inter-RAT handover, depending on the capability of the UE to support the compressed
mode.
5. If a UE that meets the handover criteria is not found, the algorithm takes the next action.
Inter-RAT Should Not Be Load Handover in the CS Domain
The algorithm for this action is the same as that for the action "Inter-RAT Should Be Load Handover in the CS Domain".
The difference is that this action only involves CS users with the "service handover" IE set to "handover to GSM should not be
performed".
The number of selected UEs is specified by the UlCSInterRatShouldNotHOUeNum or DlCSInterRatShouldNotHOUeNum

parameter.
Page 28
LDR Action (5) Inter-RAT Handover in the PS Domain
This action can only be performed when the PS inter-RAT handover algorithm is enabled.
Inter-RAT handover in the PS domain involves the following actions.
Inter-RAT Should Be Load Handover in the PS Domain
The algorithm for this action is the same as that for the action "Inter-RAT Should Be Load Handover in the CS Domain". The difference is
that this action involves only PS users with the "service handover" IE set to "handover to GSM should be performed".
The number of controlled UEs is determined by the UlPSInterRatShouldBeHOUeNum or DlPSInterRatShouldBeHOUeNum
parameter.
Inter-RAT Should Not Be Load Handover in the PS Domain
The algorithm for this action is the same as that for the action "Inter-RAT Should Not Be Load Handover in the CS Domain". The
difference is that this action involves only PS users with the "service handover" IE set to "handover to GSM should not be performed".
The number of controlled UEs is specified by the UlPSInterRatShouldNotHOUeNum or DlPSInterRatShouldNotHOUeNum
parameter.
HSPA services can be selected only when HsdpaCMPermissionInd is set to TRUE and HsupaCMPermissionInd is not set to
Limited.
Page 29
Range Range Value
PS_STREAM_IU_QOS_NEG_SWITCH: When the switch is

on, the Iu QoS Negotiation function is applied to the PS
UCORRMALGO PS_STREAM_IU_QOS_NEG_
PS_STREAM_IU_QOS_NEG_SWITCH STREAM service if Alternative RAB Parameter Values IE is OFF,ON OFF,ON None OFF
SWITCH SWITCH
present in the RANAP RAB ASSIGNMENT REQUEST or
RELOCATION REQUEST message.
Number of RABs selected in a UL/DL LDR BE traffic rate

DL LDR-BE rate reduction reduction. In the actual system, this parameter can be set on 1~10 1~10 None 1
UCELLLDR DlLdrBERateReductionRabNum
RAB number the basis of the actual circumstances. If the high-rate
subscribers occupy a high proportion, set the parameter to a
comparatively low value. If the high-rate subscribers occupy a
low proportion, set the parameter to a comparatively high
UL LDR-BE rate reduction value. Because the basic congestion control algorithm is
1~10 1~10 None 1
UCELLLDR UlLdrBERateReductionRabNum designed to slowly decrease cell load, you need to set this
RAB number
parameter to a comparatively low value.
Number of RABs selected in a UL/DL LDR uncontrolled real-

UL LDR un-ctrl RT Qos
UCELLLDR UlLdrPsRTQosRenegRabNum time traffic QoS renegotiation. The target subscribers of this 1~10 1~10 None 1
re-nego RAB num
parameter are the PS domain real-time subscribers. The
setting of this parameter is analogous to the setting of BE
service rate reduction subscriber number. Because the
DL LDR un-ctrl RT Qos number of subscribers performing QoS renegotiation may be
UCELLLDR DlLdrPsRTQosRenegRabNum 1~10 1~10 None 1
re-nego RAB num smaller than the value of this parameter.
UL CS should be HO Number of users selected in a UL/DL LDR CS domain inter-

UCELLLDR UlCSInterRatShouldBeHOUeNum 1~10 1~10 None 3
user number RAT SHOULDBE load handover. The target subscribers of
this parameter are the CS domain subscribers. Because the
CS domain subscribers are session subscribers in general
DL CS should be HO and they have little impact on load, you can set this
UCELLLDR DlCSInterRatShouldBeHOUeNum 1~10 1~10 None 3
user number parameter to a comparatively high value.
UL CS should not be HO Number of users selected in a UL/DL LDR CS domain inter-

UCELLLDR UlCSInterRatShouldNotHOUeNum RAT SHOULDNOTBE load handover. The target subscribers 1~10 1~10 None 3
user number
of this parameter are the CS domain subscribers. Because
the CS domain subscribers are session subscribers in general
UCELLLDR DlCSInterRatShouldNotHOUeNum
DL CS should not be HO and they have little impact on load, you can set this 1~10
Page 30
1~10 None 3
user number parameter to a comparatively high value.

Range Range Value
UL PS should be HO user Number of users selected in a UL/DL LDR PS domain inter-

UCELLLDR UlPSInterRatShouldBeHOUeNum RAT SHOULDBE load handover. The target subscribers of 1~10 1~10 None 1
number
this parameter are the PS domain subscribers. In the actual
system, this parameter can be set on the basis of the actual
circumstances. If the high-rate subscribers occupy a high
proportion, set the parameter to a comparatively low value. If
DL PS should be HO user the high-rate subscribers occupy a low proportion, set the
UCELLLDR DlPSInterRatShouldBeHOUeNum 1~10 1~10 None 1
number parameter to a comparatively high value.
Number of users selected in a UL/DL LDR PS domain inter-

UL PS should not be HO RAT SHOULDNOTBE load handover. The target subscribers
UCELLLDR UlPSInterRatShouldNotHOUeNum 1~10 1~10 None 1
user number of this parameter are the PS domain subscribers. In the actual
system, this parameter can be set on the basis of the actual
circumstances. If the high-rate subscribers occupy a high
DL PS should not be HO proportion, set the parameter to a comparatively low value. If
UCELLLDR DlPSInterRatShouldNotHOUeNum the high-rate subscribers occupy a low proportion, set the 1~10 1~10 None 1
user number
parameter to a comparatively high value.
Whether the compressed mode (CM) can coexist with the

HSDPA service. If this parameter is set to TRUE: 1. the RNC
can enable the CM for HSDPA services. 2. The HSDPA
CM Permission Ind on services can be enabled when the CM is enabled. If this FALSE(Forbidden),
UCMCF HsdpaCMPermissionInd FALSE, TRUE None TRUE
HSDPA parameter is set to FALSE: 1. the CM for HSDPA services TRUE(Permit)
can be enabled only after the H2D (HS-DSCH to DCH)
channel switch. 2. The HSDPA services cannot be enabled
when the CM is enabled.
Whether the compressed mode (CM) can coexist with the

HSUPA service. If this parameter is set to Permit: 1. the RNC
can enable the CM for HSUPA services. 2. The HSUPA
services can be enabled when the CM is enabled.
-If this parameter is set to Limited: 1. the CM for HSUPA Limited, Permit,
CM Permission Ind on services can be enabled only after the E2D (E-DCH to DCH) BasedOnUECap Limited, Permit, BasedOnU
UCMCF HsupaCMPermissionInd None
HSUPA channel switch. 2. The HSUPA services cannot be enabled (Based On UE BasedOnUECap ECap
when the CM is enabled. Capability)
-If this parameter is set to BasedonUECap, the RNC
determines whether CM can be enabled for HSUPA services
and whether HSUPA services can be enabled when the CM is
enabled by considering the UE capability.
Page 31
LDR Action (6) AMR Rate Reduction
This action can only be performed when the following conditions are met:
 The subparameter CS_AMRC_SWITCH of the parameter CsSwitch is set to 1.
 The parameter AMRQosPerform is set to YES.
 For uplink, the parameter UlQosAmrAdjSwitch / UlQosWAmrAdjSwitch is set to YES.
 For downlink, the parameter DlQosAmrAdjSwitch / DlQosWAmrAdjSwitch is set to YES.
AMR Rate Reduction in the Downlink

In the downlink, the LDR algorithm operates as follows:
1. Based on the integrated priority, the algorithm sorts the AMR RABs in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities and with the rates higher than the GBR for AMR services (conversational). The
number of selected RABs is specified by the DlLdrAMRRateReductionRabNum parameter. If the RNC cannot find an appropriate RAB for the
AMR rate reduction, the action fails. The algorithm takes the next action.
3. The RNC sends the rate control request message through the Iu interface to the CN to adjust the AMR rate to the GBR.
AMR Rate Reduction in the Uplink

In the uplink, the LDR algorithm operates as follows:
1. Based on the integrated priority, the algorithm sorts the AMR RABs in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities and with the rates higher than the GBR for AMR services (conversational).
The number of selected RABs is determined by the UlLdrAMRRateReductionRabNum parameter. If the RNC cannot find an appropriate RAB
for the AMR rate reduction, the action fails. The algorithm takes the next action.
3. The RNC sends the TFC CONTROL command to the UE to adjust the AMR rate to the GBR
Page 32

Range Range Value
When the switch is on and the AMRC license is activated,

UCORRMALGOSWITCH CS_AMRC_SWITCH CS_AMRC_SWITCH OFF,ON OFF,ON None OFF
the AMR control function is enabled for AMR services.
When the parameter is set to YES, the QoS control

algorithm is used for AMR services. When the parameter
is set to NO, the QoS control algorithm is not used for
UQOSACT AMRQosPerform QoS Switch for AMR Traffic NO, YES NO, YES None NO
AMR services.The QoS actions of AMR services include
rate downsizing, inter frequency handover and inter rat
handover.
Rate adjustment switch of link stability control for UL QoS

Adjustment Switch for AMR
UQOSACT UlQosAmrAdjSwitch of AMR services. When the parameter is set to YES, UL NO, YES NO, YES None NO
Uplink Rate
rate of AMR services can be adjusted.
Rate adjustment switch of link stability control for DL QoS

Switch of AMR Downlink Rate
UQOSACT DlQosAmrAdjSwitch of AMR services. When the parameter is set to YES, DL NO, YES NO, YES None NO
Adjustment
rate of AMR services can be adjusted.
Rate adjustment switch of link stability control for UL QoS

Switch of WAMR Uplink Rate
UQOSACT UlQosWAmrAdjSwitch of WAMR services. When the parameter is set to YES, UL NO, YES NO, YES None NO
Adjustment
rate of WAMR services can be adjusted.
Rate adjustment switch of link stability control for DL QoS

Switch for WAMR Downlink
UQOSACT DlQosWAmrAdjSwitch of WAMR services. When the parameter is set to YES, DL NO, YES NO, YES None NO
Rate
rate of WAMR services can be adjusted.
The mechanism of the LDR is that an action is performed

DlLdrAMRRateReduction DL LDR-AMR rate reduction in each [LDR period] and some services are selected
UCELLLDR based on the action rules to perform this action. This 1~10 1~10 None 1
RabNum RAB number
parameter defines the maximum number of RABs selected
in executing UL/DL LDR-AMR voice service rate reduction.
If the parameter value is too high, the LDR action may
fluctuate greatly and over control may occur (the state of
UlLdrAMRRateReduction UL LDR-AMR rate reduction basic congestion turns into another extreme--under load).
UCELLLDR If the parameter value is too low, the LDR action has a 1~10 1~10 None 1
RabNum RAB number
slow response and the effect is not apparent, affecting the
LDR performance.
Page 33
LDR Action (7) Code Reshuffling
To optimize the code usage efficiency, the "left most" principle is adopted in initial code allocation procedure, that is, the code with
minimum SF is reserved to ensure that the codes are available for use continuously. However, the code tree may not obey the "left most"
principle during actual use. Code reshuffling can be used to make the code tree obey "left most"’ principle.
When the cell is in the basic congestion state caused by code resource, code reshuffling can be performed to reserve sufficient code
resources for subsequent services. Code subtree adjustment refers to the switching of users from one code subtree to another.
It is used for decreasing the code fragments to release smaller codes first.
The algorithm operates as follows:

1. Initializes SF_Cur to CellLdrSfResThd.
2. Traverses all the subtrees with this SF_Cur at the root node except the subtrees occupied by common channels and HSDPA
channels, and takes the subtrees in which the number of users is not larger than the value of MaxUserNumCodeAdj as candidate
subtrees for code reshuffling.
− If such candidate subtrees are available, the algorithm goes to step 3.
− If no such candidate subtree is available, subtree selection fails. This procedure ends.
3. Selects a subtree from the candidate subtrees according to the setting of LdrCodePriUseInd.
− If this parameter is set to TRUE, the algorithm selects the subtree with the largest code number from the candidates.
− If this parameter is set to FALSE, the algorithm selects the subtree with the smallest number of users from the candidates. if multiple
subtrees have the same number of users, the algorithm selects the subtree with the largest code number.
4. Treats each user in the subtree as a new user and allocates code resources to each user.
5. Initiates the reconfiguration procedure for each user in the subtree and reconfigures the channelization codes of the users to the
newly allocated code resources.
The reconfiguration procedure on the UU interface is initiated through the PHYSICAL CHANNEL RECONFIGURATION message and
that on the Iub interface through the RL RECONFIGURATION message.
Page 34
LDR Action (7) Code Reshuffling
The following figure shows an example of code reshuffling.
In this example, CellLdrSfResThd is set to SF8, and MaxUserNumCodeAdj is set to 1.
Page 35
SF4(SF4), SF8(SF8),
This parameter specifies the Cell SF reserved threshold
SF16(SF16),
used for judging whether the code load reshuffling (LDR) is
Cell LDR SF reserved SF32(SF32), SF4, SF8, SF16, SF32,
UCELLLDR CellLdrSfResThd allowed. The code load reshuffling could be triggered only None SF8
threshold SF64(SF64), SF64, SF128, SF256
when the minimum available SF of a cell is higher than this
SF128(SF128),
threshold.
SF256(SF256)
This parameter specifies the number of users selected in

code reshuffling. Code reshuffling can be triggered only
when the number of users on a code is no greater than the
Max user number of
UCELLLDR MaxUserNumCodeAdj threshold. Code reshuffling has a big impact on the QoS. In 1~3 1~3 None 1
code adjust
addition, the reshuffled subscribers occupy two code
resources during code reshuffling. Thus, the parameter
should be set to a comparatively low value.
FALSE means not considering the code priority during the

code reshuffling. TRUE means considering the code priority
LDR code priority during the code reshuffling. If the parameter is TRUE, the FALSE(FALSE),
UCELLLDR LdrCodePriUseInd FALSE, TRUE None FALSE
indicator codes with high priority are reserved during the code TRUE(TRUE)
reshuffling. It is good for the code resource dynamic sharing,
which is a function used for the HSDPA service.
Page 36
OLC (Overload Control) – STEP(1)
After the UE access is allowed, the power consumed by a single link is adjusted by the single link power control function.
The power varies with factors such as the mobility of the UE and the changes in the environment. In some situations, the total power load
of the cell can be higher than the target load. To ensure the system stability, Overload Control (OLC) must be performed
OLC algorithm consists of three steps as following,
Overload Triggering
OLC Procedure
OLC Actions
STEP (1) Overload Triggering

Only the power resource and Iub bandwidth may result in the overload
congestion state. Hard resources such as the ENU and credit resources do
not cause overload congestion.
OLC can be enabled through the UL_UU_OLC and DL_UU_OLC subparameters

of the NBMLdcAlgoSwitch parameter.
If the UL/DL load of the cell is higher than or equal to the UlOlcTrigThd or
DlOlcTrigThd for a hysteresis time, the cell is in the overload state, and the
related overload handling action is taken.
If the current UL/DL load of the cell is lower than the UlOlcRelThd or
DlOlcRelThd for a hysteresis time, the overload state of the cell is released and
the related overload handling is stopped.
For the downlink, the hysteresis time is specified by the parameter

DlLdTrnsHysTime; for the uplink, the hysteresis time is 600ms.
 The uplink load of an HSUPA cell is calculated based on the uncontrollable load
of the cell. The downlink load of an HSDPA cell is calculated based on the load of
Page 37
non-HSPA power and GBP in the cell.
STEP (2) Overload Prcedure Turn on the OLC algorithm switch
When the cell is overloaded, the RNC takes one of the following actions in
each period specified by the OlcPeriodTimerLen parameter until the Mark "current OLC state = uncongested"
congestion is relieved:
1. Performing TF Control of BE Services Start sending the OLC congestion indication
2. Switching BE Services to Common Channels
3. Adjusting the Maximum FACH TX Power Mark "current action = first OLC action"
4. Releasing Some RABs Send congestion
relief indication to
Clear the "selected" mark of all UE OLC actions MAC (downlink
As shown in the figure, the OLC procedure is as follows: Congestion congestion)
1. The OLC takes the first action to perform TF control. state indication
Wait for congestion state indication
 If the TF control succeeds, the OLC checks whether the
system is overloaded. If yes, the OLC performs TF control
again. Yes
Current OLC state = congested?
No
 If the TF control fails, go to 2.
Performing Yes
2. The OLC takes the second action to switch BE services to Successful?
TF control
common channels. No
Keep the
 If the switching succeeds, the OLC checks whether the system action Switching BE
Successful?
Yes Wait
services to CCH Mark for the
is overloaded. If yes, the OLC switches BE services to sequence
No
unchanged "current expiration
common channels again. action = of the
and take
 If the switching fails, go to 3. the current
Adjusting Max
Successful?
Yes
success- OLC
FACH TX power
action first No ful action" period
timer
3. The OLC takes the third action to adjust the maximum Yes
Releasing
FACH transmit power. some RABs
Successful?
 If the adjustment succeeds, the OLC checks whether the No
system is overloaded. If yes, the OLC adjusts the power again.
 If the adjustment fails, the OLC takes the fourth action to No OLC action is taken
or all actions fail
release some RABs. Mark "current action = first OLC action"
Page 38
STEP (3) Overload Actions
OLC Action (1) Performing TF Control of BE Services
For the TF control in the downlink, the OLC algorithm operates as follows:
1. Based on the integrated priority, the algorithm sorts the RABs in descending order.
2. The algorithm selects the following RABs:

− DCH RABs with the rates higher than DlDcccRateThd.
− RABs with the lowest integrated priorities.
The number of RABs selected is smaller than or equal to DlOlcFTFRstrctRabNum. If the RNC cannot find an appropriate service for the TF
control, the OLC takes the next action.
3. RNC sends the TF control indication message to the MAC. Each MAC of the selected RABs will receive one TF control indication
message and will restrict the transport format combination (TFC) selection of the BE services to reduce the data rate step by step.
The MAC restricts the TFC selection according to the following formula:
TFmax(N+1) = TFmax(N) x Ratelimitcoeff
Here:
− TFmax(0) is the maximum TB number of the BE service before the service is selected for TF control.
− TFmax(N+1) is the maximum TB number during the period from (T0 + RateRstrctTimerLen x N) to (T0 + RateRstrctTimerLen x (N + 1)),
where T0 is the time when the MAC receives the TF control indication message.
− Ratelimitcoeff is specified by the RateRstrctCoef parameter.
4. If the number of times that TF control is performed exceeds DlOlcFTFRstrctTimes, the action fails. The OLC takes the next action.
Page 39
5. If the congestion is relieved, the RNC sends the congestion relief indication to the MAC. At the same time, the rate recovery timer
(RateRecoverTimerLen) is started. When this timer expires, the MAC increases the data rate step by step.
MAC recovers the TFC selection by calculating the maximum TB number according to the following formula:
TFmax(N+1) = TFmax(N) x RateRecoverCoeff
Here:
− TFmax(0) is the maximum TB number of the BE service before congestion relief indication is received.
− TFmax(N+1) is the maximum TB number during the period from (T1 + RateRecoverTimerLen x N) to (T1 + (RateRecoverTimerLen x (N +
1)), where T1 is the time when the MAC receives the congestion relief indication message.
− RateRecoverCoeff is specified by the RecoverCoef parameter.
In this example, the MAC performs TF control of a downlink 384 kbit/s
service, and RateRstrctCoef is set to 0.68.
Before point A, the cell is not in OLC state. The downlink data
transfer rate is 384 kbit/s, the corresponding TF is 12 x 336, and TFS
is {12 x 336, 8 x 336, 4 x 336, 2 x 336, 1 x 336, 0 x 336}.
At point A, the cell enters OLC state. The RNC selects this RAB for
fast TF restriction. MAC restricts the TFC selection during the period
between point A and point B by calculating the maximum TB number
as follows:
TFmax(1) = TFmax(0) x Ratelimitcoeff = 12 x 0.68 = 8.16
Compare 8.16 with the TFS. Then, the maximum TB number is 8.
The time between point A and point B is specified by the
RateRstrctTimerLen parameter.
At point B, the MAC performs further TFC restriction by calculating

maximum TB number as follows:
TFmax(2) = TFmax(1) x Ratelimitcoeff = 8 x 0.68 = 5.44
Compare 5.44 with the TFS. Then, the maximum TB number is 4.
At point C and point D, similar process is followed. Page 40
For the TF control in the uplink, the OLC algorithm operates as follows:
For a UE with the DCH service, the RNC sends a TRANSPORT FORMAT COMBINATION CONTROL message to the UE to restrict the
TFC of the UE, according to the 3GPP TS25.331.
1. Based on the integrated priority, the algorithm sorts the DCH RABs in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities and with the rates higher than UlDcccRateThd. The number of
selected RABs is specified by the UlOlcFTFRstrctRabNum parameter.
3. The RNC sends the TRANSPORT FORMAT COMBINATION CONTROL message to the UE that accesses the specified service.
This message contains the following IEs:
Transport Format Combination Set (TFCS) Identity: defines the available TFC that the UE can select, that is, the restricted TFC
sub-set. It is always the two TFCs corresponding to the lowest data rate.
TFC Control Duration: defines the period in multiples of 10 ms frames for which the restricted TFC sub-set is to be applied. It is set to a
random value from the range of 10 ms to 5120 ms, so as to avoid data rate upsizing at the same time.
After the TFC control duration is due, the UE can apply any TFC of TFCS before the TF control.
4. Each time, the RNC selects a certain number of RABs, which is specified by UlOlcFTFRstrctRabNum, for TF control. The UE of
each selected RAB will receive the TRANSPORT FORMAT COMBINATION CONTROL message. The number of times that TF
control is performed is specified by UlOlcFTFRstrctTimes.
If the RNC cannot find an appropriate service, the OLC performs the next action. Page 41
OLC Action (2) Switching BE Services to Common Channels
Whether the selected UEs can be switched to common channels depends on the setting of DRA_PS_BE_STATE_TRANS_SWITCH,
DRA_HSDPA_STATE_TRANS_SWITCH, or DRA_HSUPA_STATE_TRANS_SWITCH in the parameter DraSwitch.
For the switching of uplink BE services to common channels, if the control RTWP anti-interference function switch
(NBMCacAlgoSwitch: RTWP_RESIST_DISTURB) is turned on, the RNC checks whether the uplink equivalent user load proportion
of the cell is lower than 40%. If it is lower than 40%, the RNC does not perform this operation.
For switching BE services to common channels, the OLC algorithm operates as follows:
1. Based on the integrated priority, the algorithm sorts all the UEs in the PS domain in descending order.
2. The algorithm selects the UEs with the lowest integrated priorities. The number of selected UEs is specified by
TransCchUserNum. If the selection fails, the OLC takes the next action.
This function is disabled when the TransCchUserNum parameter is set to 0.
3. The OLC switches the selected UEs to common channels.

If both the selected UEs and the current cell support enhanced CELL_FACH, these UEs can be switched to the enhanced CELL_FACH
state.
Page 42
OLC Action (3) Adjusting the Maximum FACH TX Power
The procedure for adjusting the maximum FACH transmit power is as follows:
1. Set the maximum FACH transmit power to the target maximum transmit power. The target maximum transmit power is calculated
according to the following formula:
Ptarget = Pmax – Delta
− Ptarget is the target maximum transmit power.

− Pmax is the maximum FACH transmit power (MaxFachPower).
− Delta is the FACH power reduction step (FACHPwrReduceValue).
2. If the congestion is relieved after the power adjustment, the system starts the FACH power recovery timer, which is set to 5s.
When the timer expires, the maximum FACH transmit power is increased to the original maximum FACH transmit power if the system
is always in the normal state before the timer expires.
The preceding power adjustment is applicable to only the FACH carrying common services rather than MBMS services.
Page 43
OLC Action (4) Releasing Some RABs
OLC Algorithm for the Release of Some RABs in the Uplink

If the Control RTWP Anti-interference algorithm switch (NBMCacAlgoSwitch: RTWP_RESIST_DISTURB) is enabled, the RNC checks whether
the uplink equivalent user load proportion of the cell is lower than 40% before performing this operation. If it is lower than 40%, the RNC does not
perform this operation.
For the release of some RABs in the uplink, the OLC algorithm operates as follows:
1. Based on the integrated priority, the algorithm sorts all RABs including HSUPA and DCH services in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities. If the integrated priorities of some RABs are identical, it selects
the RAB with a higher rate in the uplink. The number of selected RABs is specified by UlOlcTraffRelRabNum.
3. The selected RABs are released directly.
OLC Algorithm for the Release of Some RABs in the Downlink

For the release of some RABs in the downlink, the OLC algorithm operates as follows:
If the SeqOfUserRel parameter is set to USER_REL, then:
1. Based on the integrated priority, the algorithm sorts all non-MBMS RABs in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities. If the integrated priorities of some RABs are identical, it selects the RAB with
a higher rate in the downlink. The number of selected RABs is specified by DlOlcTraffRelRabNum.
3. The selected RABs are directly released.
4. If all non-MBMS RABs are released but congestion persists in the downlink, MBMS RABs are selected.
If the SeqOfUserRel parameter is set to MBMS_REL, then:

1. Based on the ARP, the algorithm sorts all MBMS RABs in descending order.
2. The algorithm selects the RABs with the lowest integrated priorities. The number of selected RABs is specified by MbmsOlcRelNum.
3. The selected RABs are directly released.
4. If all MBMS RABs are released but congestion persists in the downlink, non-MBMS RABs are selected.
This function is disabled when the UlOlcTraffRelRabNum, DlOlcTraffRelRabNum, and MbmsOlcRelNum parameters are set to 0.
The higher the value of UlOlcTraffRelRabNum or DlOlcTraffRelRabNum, the more the cell load decreases, which will affect the users
experience negatively. Page 44
OLC (Overload Control) Parameters

Range Range Value
UL_UU_OLC: UL UU overload congestion control algorithm. When the cell

UL UU overload congestion
is overloaded in UL, this algorithm reduces the cell load in UL by quick TF
control
restriction or UE release.
UCELLALGOSWITCH UL_UU_OLC OFF,ON OFF,ON None OFF
DL_UU_OLC DL UU overload congestion
DL_UU_OLC: DL UU overload congestion control algorithm. When the cell
control
is overloaded in DL, this algorithm reduces the cell load in DL by quick TF
restriction or UE release.
If the ratio of UL /DL load of the cell to the uplink capacity is not lower than
UCELLLDM UlOlcTrigThd UL OLC trigger threshold this threshold, the UL overload and congestion control function of the cell is 0~100 0~1 % 95
triggered. The value of the OLC release threshold should not be much
lower than or close to the OLC trigger threshold, or the system state may
UCELLLDM DlOlcTrigThd DL OLC trigger threshold have a ping-pong effect. The recommended difference between the OLC 0~100 0~1 % 95
release threshold and the OLC trigger threshold is higher than 10%.
If the ratio of UL/DL load of the cell to the uplink capacity is lower than this
UCELLLDM UlOlcRelThd UL OLC release threshold 0~100 0~1 % 85
threshold, the UL overload and congestion control function of the cell is
stopped. The value of the OLC release threshold should not be much lower
than or close to the OLC trigger threshold, or the system state may have a
ping-pong effect. The recommended difference between the OLC release
UCELLLDM DlOlcRelThd DL OLC release threshold 0~100 0~1 % 85
threshold and the OLC trigger threshold is higher than 10%.
Identifying the period of the OLC execution. When overload occurs,

execution of OLC can dynamically reduce the cell load. When setting the
parameter, consider the hysteresis for which the load monitoring responds
to the load change. For example, when the layer 3 filter coefficient is 6, the
ULDCPERIOD OlcPeriodTimerLen OLC period timer length 100~86400000 100~86400000 ms 3000
hysteresis for which the load measurement responds to the step-function
signals is about 2.8s, namely that the system can trace the load control
effect about 3 s later after each load control. In this case, the OLC period
timer length cannot be smaller than 3s.
D8, D16, D32,
8, 16, 32, 64,
Downlink Bit Rate Threshold D64, D128,
UDCCC DlDcccRateThd 128, 144, 256, kbit/s D64
for DCCC For a BE service that has a low maximum rate, the DCCC algorithm is not D144, D256,
384
obviously effective yet it increases algorithm processing. Thus, the traffic- D384
based DCCC algorithm is applied to BE services whose maximum DL/UL D8, D16, D32,
8, 16, 32, 64,
Uplink Bit Rate Threshold for rate is greater than the threshold. D64, D128,
UDCCC UlDcccRateThd 128, 144, 256, kbit/s D64
DCCC D144, D256,
384
D384
Page 45

Range Range Value
DL fast TF restriction refers to a situation where, when the cell is overloaded and congested,
the downlink TF can be adjusted to restrict the number of blocks transported in each TTI at
the MAC layer and the rate of user data, thus reducing the cell downlink load. This parameter
DL TF rate restrict defines the downlink data rate restrict coefficient in fast TF restrict The smaller this parameter
UCELLOLC RateRstrctCoef 1~99 0.01~0.99 % 68
coefficient is, the larger the TF restrict effect. The lower the parameter is, the more severe the rate is
restricted. An excessive low parameter value, however, may affect the BE transmission delay.
A high parameter value means loose restriction, which may be ineffective in alleviating the
overload.
DL TF rate recover
UCELLOLC RecoverCoef the MAC layer and the rate of user data, thus reducing the cell downlink load. This parameter 100~200 1~2 % 130
coefficient
defines the downlink OLC fast TF rate recovery coefficient. The greater this parameter is, the
larger the TF restrict effect.
defines the time length of the downlink OLC fast TF restriction. RateRstrctTimerLen and
DL TF rate restrict RateRecoverTimerLen are effective only to the downlink. The uplink fast TF restriction is
UCELLOLC RateRstrctTimerLen 1~65535 1~65535 ms 3000
timer length performed by the UE. For the uplink fast TF restriction, the RNC only delivers a new TFCS
and randomly selects a comparatively bigger time length in the signaling value scope. The UE
automatically release the TF restriction once the time expires. The higher RateRstrctTimerLen
is, the more slowly the BE service rate decreases. The lower RateRstrctTimerLen is, the
harder it is to receive the overload release instruction.
defines the downlink data rate recover timer length in fast TF restriction. RateRstrctTimerLen
and RateRecoverTimerLen are effective only to the downlink. The uplink fast TF restriction is
DL TF rate recover performed by the UE. For the uplink fast TF restriction, the RNC only delivers a new TFCS
UCELLOLC RateRecoverTimerLen 1~65535 1~65535 ms 5000
timer length and randomly selects a comparatively bigger time length in the signaling value scope. The UE
automatically release the TF restriction once the time expires. The higher
RateRecoverTimerLen is, the more slowly the BE service rate recovers, while the lower
probability that the overload is triggered again in a short period. The lower
RateRecoverTimerLen is, the more quickly the BE service rate is recovered, but more
overloads occur.
Page 46
Range Range Value
DL/UL fast TF restriction refers to a situation where, when the cell is overloaded and
congested, the DL/UL TF can be adjusted to restrict the number of blocks transported
DL OLC fast TF restrict in each TTI at the MAC layer and the rate of user data, thus reducing the cell DL/UL
UCELLOLC DlOlcFTFRstrctTimes load.The mechanism of the OLC is that an action is performed in each [OLC period] 0~100 0~100 None 3
times
and some services are selected based on the action rules to perform this action. This
parameter defines the maximum number of downlink OLC fast TF restriction performed
in entering/exiting the OLC status.
After the overload is triggered, the RNC immediately executes OLC by first executing
fast TF restriction. The internal counter is incremented by 1 with each execution. If the
number of overloads does not exceed the OLC action threshold, the system lowers the
BE service rate by lowering TF to relieve the overload. If the number of overloads
UL OLC fast TF restrict exceeds the OLC action threshold, the previous operation has no obvious effect on
UCELLOLC UlOlcFTFRstrctTimes 0~100 0~100 None 3
times alleviating the overload and the system has to release users to solve the overload
problem. The lower the parameters are, the more likely the users are released,
resulting in negative effect on the system performance. If the parameters are
excessively high, the overload status is released slowly.
DL/UL fast TF restriction refers to a situation where, when the cell is overloaded and
DL OLC fast TF restrict congested, the DL/UL TF can be adjusted to restrict the number of blocks transported
UCELLOLC DlOlcFTFRstrctRabNum 1~10 1~10 None 3
RAB number in each TTI at the MAC layer and the rate of user data, thus reducing the cell downlink
load.The mechanism of the OLC is that an action is performed in each [OLC period]
and some services are selected based on the action rules to perform this action. This
UL OLC fast TF restrict parameter defines the maximum number of RABs selected in executing downlink OLC
UCELLOLC UlOlcFTFRstrctRabNum fast restriction.Selection of RABs of the OLC is based on the service priorities and ARP 1~10 1~10 None 3
RAB number
values and bearing priority indication.
DRA_PS_BE_STATE_TRANS_SWITCH: When the switch is on, UE RRC status

transition (CELL_FACH/CELL_PCH/URA_PCH) is allowed at the RNC.
DRA_HSDPA_STATE_TRANS_SWITCH: When the switch is on, the status of the UE

DRA_HSDPA_STATE_
RRC that carrying HSDPA services can be changed to CELL_FACH at the RNC. If a
TRANS_SWITCH
PS BE service is carried over the HS-DSCH, the switch
UCORRMALGO Dynamic Resource PS_BE_STATE_TRANS_SWITCH should be on simultaneously. If a PS real-time
DRA_HSUPA_STATE_ OFF,ON OFF,ON None ON
SWITCH Allocation Switch service is carried over the HS-DSCH, the switch
TRANS_SWITCH
PS_NON_BE_STATE_TRANS_SWITCH should be on simultaneously.
DRA_PS_BE_STATE_
DRA_HSUPA_STATE_TRANS_SWITCH: When the switch is on, the status of the UE
TRANS_SWITCH
RRC that carrying HSUPA services can be changed to CELL_FACH at the RNC. If a
PS BE service is carried over the E-DCH, the switch PS_BE_STATE_TRANS_SWITCH
should be on simultaneously. If a PS real-time service is carried over the E-DCH, the
switch PS_NON_BE_STATE_TRANS_SWITCH should be on simultaneously.
Page 47
Range Range Value
Transfer Common Channel User number

When the system is overloaded and congested, users on the DCH can be
reconfigured to the CCH in order to reduce the cell load and recover the
system.
Transfer Common Channel
UCELLOLC TransCchUserNum 0~10 0~10 None 1
User number
The mechanism of the OLC is that an action is performed in each [OLC
period] and some services are selected based on the action rules to
perform this action. This parameter defines the maximum number of users
selected in executing reconfiguration to the CCH.
UCELLOLC FACHPwrReduceValue Fach power reduce value This parameter defines the reduce value in reducing FACH power Action. 0~30 0~3 0.1dB 0
DL OLC traff release RAB User release is an extreme method in reducing the cell load and
UCELLOLC DlOlcTraffRelRabNum
number recovering the system when the cell is overloaded and congested.The 0~10 0~10 None 0
mechanism of the OLC is that an action is performed in each [OLC period]
and some services are selected based on the action rules to perform this
action. This parameter defines the maximum number of RABs released in
executing DL/UL OLC service release.
UL OLC traff release RAB For the users of a single service, the releasing of RABs means the
UCELLOLC UlOlcTraffRelRabNum
number complete releasing of the users. The releasing of RABs causes call drops, 0~10 0~10 None 0
This parameter indicates whether the MBMS service is released first or MBMS_REL, MBMS_REL,
UCELLOLC SeqOfUserRel Sequence of user release None MBMS_REL
user first when the overload occurs. USER_REL USER_REL
MBMS service release is an extreme method in reducing the cell load and
recovering the system when the cell is overloaded and congested.The
MBMS services number mechanism of the OLC is that an action is performed in each [OLC period]
UCELLOLC MbmsOlcRelNum 0~8 0~8 None 1
released and some services are selected based on the action rules to perform this
action. This parameter defines the maximum number of MBMS services
released in executing downlink OLC service release.
Page 48
Table of Contents

Page 49
Radio Bearer Specifications
Radio Bearer (RB) of a Radio Access Bearer (RAB) carries the control plane signaling or user plane data of L3 between
the UE and the UTRAN.
RB is mapped onto logical channels through the radio link control (RLC) sublayer of L2. After RLC processing, RLC protocol data units
(PDUs) are sent to the medium access control (MAC) sublayer through logical channels.
Based on the traffic class and quality of service (QoS) required by the upper layer, the MAC selects appropriate transport channels and
maps the logical channels onto the transport channels.
The physical layer (L1) maps the transport channels onto one or more physical channels. After data processing, such as spreading and
scrambling, the signaling or user data is sent through radio waves.
Radio Bearer (RB) Specifications

Based on the type of transmitted data, the radio bearers are
classified into two types:
 Signaling Radio Bearer (SRB)
 Traffic Radio Bearer (TRB)
Page 50
Page 50
SRB Specifications
Different SRB specifications, that is, 3.4/6.8/13.6/27.2 kbit/s RRC Connection
and RAB Assignment is supported to meet the QoS requirements for Maximum Bit Rate
SRB Index Direction
different services. (kbit/s)
1 3.4 DL/UL
The RNC provides different sets of configuration parameters (including SRB 2 13.6 DL/UL
maximum bit rate) for different typical services. 3 27.2 DL/UL
4 6.8 DL/UL
TRB Specifications
RNC searches for the RAB indexes of TRBs according to the attributes such as the Core Network (CN) domain, Taffic Class(TC), Source
Statistics Descriptor,(SSD) and Maximum Bit Rate (MBR). Then, the RNC configures the associated parameters for each service based on RAB
indexes.
RNC automatically checks the identity of the CN domain that provides the services. The “RAB ASSIGNMENT REQUEST” message received from
the CN contains information of traffic class, source statistics descriptor, and MBR.
Traffic Class (TC)

There are four traffic classes: conversational, streaming, interactive, and background.
The main difference between these classes is the degree of delay sensitivity, which is described as follows:
 Conversational class is the most sensitive to delay. It is used to carry real-time traffic. The real-time traffic requires shortest delay and strict time
sequence between data streams. Therefore, this traffic class has the highest QoS requirement.
 Streaming class is used to carry unidirectional data streams. It does not have a high requirement for delay, but the time sequence must be kept
within a data stream and the end-to-end delay jitter of data streams must be controlled.
 Interactive class is used to carry traditional Internet services, such as Web browsing and database query. Its round trip time (RTT) is a key
parameter, and data packets need to be transmitted transparently at low bit error rates.
 Background class is used to receive or transmit data in background mode. Such services include email, SMS, and FTP. This class does not have
a high requirement for delay, but it requires data packets to be transmitted transparently at low bit error rates.
Page 51
TRB specifications supported by Huawei RNC are listed below.
RAB Index CN Domain Traffic Class Source Statistics Descriptor MBR (kbit/s) Direction RAB Index CN Domain Traffic Class Source Statistics Descriptor MBR (kbit/s) Direction
0 CS Conversational Speech 12.2 DL/UL 64 PS Interactive Unknown 35200 DL
1 CS Conversational Speech 23.85 DL/UL 65 PS Interactive Unknown 42100 DL
2 CS Conversational Unknown 28.8 DL/UL 66 PS Interactive Unknown 11480 UL
3 CS Conversational Unknown 32 DL/UL 67 PS Interactive Unknown 512 DL
6 CS Streaming Unknown 57.6 DL/UL 70 PS Background Unknown 0 DL/UL
11 PS Conversational Unknown 8 DL/UL 71 PS Background Unknown 8 DL/UL
16 PS Conversational Unknown 38.8 DL/UL 75 PS Background Unknown 128 DL/UL
21 PS Streaming Unknown 8 DL/UL 79 PS Background Unknown 608 UL
22 PS Streaming Unknown 16 DL/UL 80 PS Background Unknown 768 DL
27 PS Streaming Unknown 256 DL/UL 85 PS Background Unknown 2048 DL/UL
40 PS Interactive Unknown 0 DL/UL 87 PS Background Unknown 3600 DL
41 PS Interactive Unknown 8 DL/UL 88 PS Background Unknown 5440 UL
49 PS Interactive Unknown 608 UL 96 PS Background Unknown 11480 UL
50 PS Interactive Unknown 768 DL 97 PS Background Unknown 512 DL
52 PS Interactive Unknown 1280 UL 99 PS Background Unknown 16000 DL
53 PS Interactive Unknown 1536 DL 100 PS Interactive Unknown 48900 DL
55 PS Interactive Unknown 2048 DL/UL 102 PS Interactive Unknown 56000 DL
107 PS Background Unknown 69900 DL
60
61
PS
PS
Interactive
Interactive
Unknown
Unknown
10100
13900
DL
DL 108 PS Interactive Unknown Page 52
77000 DL
111 PS Background Unknown 84200 DL
Transport Channel Selection
Principles of Transport Channel Selection for Signaling and Traffic
Transport Channel Selection for Signaling
During RRC connection setup, the SRB is carried on the CCH, DCH, HS- Signaling/Service
DSCH, or E-DCH, which is described as follows: Domain Transport Channel
Type
– Signaling DCH/CCH/HS-DSCH/E-DCH
After receiving the RRC CONNECTION REQUEST message from UE, RNC selects
Speech DCH/HS-DSCH/E-DCH
the channel type for the RRC connection establishment according to the establishment
cause. For different establishment causes, the selected channel types can be set CS Videophone (VP)
DCH
through the MML command SET URRCESTCAUSE. Streaming
Speech
 If the selected channel type is FACH, the SRB is carried on the CCH in both the DCH/HS-DSCH/E-DCH
Streaming
uplink and the downlink. PS
Interactive
If the selected channel type is DCH, then DCH/CCH/HS-DSCH/E-DCH
Background
− In the downlink, if SrbChlTypeRrcEffectFlag is set to TRUE and SrbChlType is
set to HSDPA or HSPA, the SRB is carried on the HS-DSCH; otherwise, the SRB is MBMS FACH/DCH/HS-DSCH
carried on the DCH.
− In the uplink, if SrbChlTypeRrcEffectFlag is set to TRUE and SrbChlType is
set to HSUPA or HSPA, the SRB is carried on the E-DCH; otherwise, the SRB is
carried on the DCH.
During TRB connection setup, the SRB is carried on the CCH, DCH, HS-DSCH, or E-DCH, which is described as follows:
 If the selected channel type is FACH, the SRB is carried on the CCH in both the uplink and the downlink.
 If the selected channel type is DCH/E-DCH/HS-DSCH, then
− In the downlink, if the channels selected by all the TRBs are HS-DSCH and SrbChlType is set to HSDPA or HSPA, the SRB is carried on the
HS-DSCH; otherwise, the SRB is carried on the DCH.
− In the uplink, if the channel types selected by all the TRBs are E-DCH and SrbChlType is set to HSUPA or HSPA, the SRB is carried on the E-DCH;
otherwise, the SRB is carried on the DCH.
If both the UE and the cell support HSPA but the SRB fails to be set up on the HSPA channel (for example, an admission failure occurs), the SRB is set
up on the DCH. At the same time, the retry timer is started for periodic retries to set up the SRB on the HSPA channel. The length of the retry timer is
specified by ChannelRetryTimerLen. Page 53
Transport Channel Selection for Traffic

The VP service and streaming service in the CS domain are carried on the DCHs.
CS speech service (CS voice service) can be carried on the DCH, HS-DSCH, or E-DCH.
 In the downlink, if CSVoiceChlType is set to HSPA, the CS speech service is carried on the HS-DSCH; otherwise, the CS speech service
is carried on the DCH.
In the uplink, if CSVoiceChlType is set to HSPA, the CS speech service is carried on the E-DCH; otherwise, the CS speech service is
carried on the DCH.
PS speech service (VoIP service) can be carried on the DCH, HS-DSCH, or E-DCH.
 In the downlink, if VoipChlType is set to HSDPA or HSPA, the VoIP service is carried on the HS-DSCH; otherwise, the VoIP service is
carried on the DCH.
 In the uplink, if VoipChlType is set to HSPA, the VoIP service is carried on the E-DCH; otherwise, the VoIP service is carried on the DCH.
The streaming service in the PS domain can be carried on the DCH, HS-DSCH, or E-DCH.
 If the maximum DL service rate is higher than or equal to DlStrThsOnHsdpa, the cell supports HSDPA, and MapSwitch:
MAP_PS_STREAM_ON_HSDPA_SWITCH is set to ON, the service is carried on the HS-DSCH.
Otherwise, the service is carried on the DCH.
If the maximum UL service rate is higher than or equal to UlStrThsOnHsupa, the cell supports HSUPA, and
MapSwitch: MAP_PS_STREAM_ON_HSUPA_SWITCH is set to ON, the service is carried on the E-DCH.
Otherwise, the service is carried on the DCH.
Page 54
PS BE services (interactive service and background

service) can be carried on the CCH, DCH, HS-DSCH, or Example PS BE services mapping (UL)
E-DCH.
Low-rate BE services carry a small amount of data.

Therefore, such PS services can be carried on the CCH to 8/8 kbps
fully utilize radio resources.
 If the maximum DL service rate is lower than

DlBeTraffDecThs, the maximum UL service rate is lower
than UlBeTraffDecThs, and the RRC connection is set up on
608/64kbps
the CCH, the service is carried on the CCH. Otherwise, the
following situations occur:
If the maximum DL service rate is higher than or

equal to DlBeTraffThsOnHsdpa, the service is
carried on the HS-DSCH; otherwise, the service is 1280kbps
carried on the DCH.
If the maximum UL service rate is higher than or

equal to UlBeTraffThsOnHsupa, the service is
carried on the E-DCH; otherwise, the service is carried
on the DCH.
If the maximum UL service rate is higher

than or equal to BeHsupa2msTtiRateThs,
the service is carried on the E-DCH (2ms TTI)
otherwise, the service is carried on the E-
DCH (10ms TTI If the maximum UL service
rate is higher than or equal to
UlBeTraffThsOnHsupa.
Page 55
Transport Channel Selection for PS BE services
For the first PS BE service, if UCORRMALGOSWITCH: RESERVED_SWITCH_0_BIT5 is set to ON and the E-DCH is selected for the
uplink, the RNC will further determines the coverage of the UE. If the UE has weak coverage, the DCH is selected to reduce the call drop
rate.
The RNC determines the signal quality of the UE based on the PCPICH Ec/No value indicated in the RRC CONNECTION REQUEST
message during the RRC setup.
 If the Ec/No value is within the valid range, that is, the timer EcN0EffectTime has not expired, and the Ec/No value is below the
threshold which can be specified by the parameter ReservedU32Para0, the signal quality is considered poor and the UE has weak
coverage. In this case, the RNC selects the DCH for the BE service. If the Ec/No value is above the threshold, the RNC selects the E-DCH
for the BE service.
 If the Ec/No value exceeds the valid range, that is, the timer EcN0EffectTime has expired, the RNC will directly set up BE services on
the E-DCH without considering the Ec/No value.
The IMS signaling can be carried on the DCH, HS-DSCH, or E-DCH.

 In the downlink, if ImsChlType is set to HSDPA or HSPA, the IMS signaling is carried on the HS-DSCH. Otherwise, the IMS signaling is
carried on the DCH.
In the uplink, if ImsChlType is set to HSPA, the IMS signaling is carried on the E-DCH. Otherwise, the IMS signaling is carried on the
DCH.
If both the cell and the UE support HSPA but a service fails to be set up on the HSPA channel (for example, an admission failure occurs),
the service is carried on the DCH. At the same time, the retry timer is started for periodic retries to access the HSPA channel. The timer
length is specified by ChannelRetryTimerLen. If an HSUPA service is set up on the DCH because of weak coverage, it will not start
periodic retries but directly switch to the E-DCH when the uplink coverage improves and traffic volume increases.
Page 56
Transport Channel Selection Parameters
Baseline
MO Parameter ID Parameter Name Parameter Description GUI Value Range Actual Value Range Unit
Value
ORIGCONVCALLEST, ORIGCONVCALLEST,
ORIGSTREAMCALLEST, ORIGSTREAMCALLEST,
ORIGINTERCALLEST, ORIGINTERCALLEST,
ORIGBKGCALLEST, ORIGBKGCALLEST,
ORIGSUBSTRAFFCALLEST, ORIGSUBSTRAFFCALLEST,
TERMCONVCALLEST, TERMCONVCALLEST,
TERMSTREAMCALLEST, TERMSTREAMCALLEST,
TERMINTERCALLEST, TERMINTERCALLEST,
Cause of RRC Cause of RRC connection establishment, that is, TERMBKGCALLEST, TERMBKGCALLEST,
EMERGCALLEST, EMERGCALLEST,
URRCESTCAUSE RrcCause connection the value of the establishment cause IE in the INTERRATCELLRESELEST, INTERRATCELLRESELEST,
None None
establishment RRC CONNECTION REQUEST message. INTERRATCELLCHGORDEREST, INTERRATCELLCHGORDEREST,
REGISTEST, DETACHEST, REGISTEST, DETACHEST,
ORIGHIGHPRIORSIGEST, ORIGHIGHPRIORSIGEST,
ORIGLOWPRIORSIGEST, ORIGLOWPRIORSIGEST,
CALLREEST, CALLREEST,
TERMHIGHPRIORSIGEST, TERMHIGHPRIORSIGEST,
TERMLOWPRIORSIGEST, TERMLOWPRIORSIGEST,
TERMCAUSEUNKNOWN, TERMCAUSEUNKNOWN,
MBMSCALLEST, DEFAULTEST MBMSCALLEST, DEFAULTEST
Effective Flag of Whether the configured type of channel that

UFRCCHLTYPEPARA SrbChlTypeRrcEffectFlag Signaling RB Channel preferably carries the signaling RB is effective in TRUE, FALSE TRUE, FALSE None FALSE
Type the case of RRC connection establishment
Type of channel that preferably carries the

signaling RB.
- DCH: Both uplink and downlink are preferably
carried on DCH. DCH(UL_DCH,DL_DCH),
Type of Channel
- HSDPA: Uplink is preferably carried on DCH, HSDPA(UL_DCH,DL_HSDSCH),
UFRCCHLTYPEPARA SrbChlType Preferably Carrying DCH, HSDPA, HSUPA, HSPA None DCH
and downlink is preferably carried on HS-DSCH. HSUPA(UL_EDCH,DL_DCH),
Signaling RB
- HSUPA: Uplink is preferably carried on E-DCH, HSPA(UL_EDCH,DL_HSDSCH)
and downlink is preferably carried on DCH.
- HSPA: Uplink is preferably carried on E-DCH,
and downlink is preferably carried on HS-DSCH.
This parameter specifies the value of the

channel retry timer. The timer will start when
Channel Retry Timer traffic is set up or reconfigured and some higher
UCOIFTIMER ChannelRetryTimerLen 0~180 0~180 s 5
Length technique is not configured by some reason
except for the capability of UE or cell. Channel
retry will be performed after this timer expires. Page 57
Actual Value Baseline
MO Parameter ID Parameter Name Parameter Description GUI Value Range Unit
Range Value
Channel type of CS voice services. DCH indicates that both uplink and
downlink are preferably carried on DCH. HSPA indicates that uplink is
DCH(UL_DCH,DL_DCH)
CS voice channel preferably carried on E-DCH, and downlink is preferably carried on HS-
UFRCCHLTYPEPARA CSVoiceChlType , HSPA(UL_EDCH,DL_ DCH, HSPA None DCH
type DSCH. In particular, both uplink and downlink will be carried on DCH if
HSDSCH)
either uplink cannot be carried on E-DCH or downlink cannot be carried
on HS-DSCH.
Type of channel that preferably carries VoIP services.

DCH(UL_DCH,DL_DCH)
- DCH: Both uplink and downlink are preferably carried on the DCH.
, HSDPA(UL_DCH,DL_
- HSDPA: Uplink is preferably carried on the DCH, and downlink is DCH, HSDPA,
UFRCCHLTYPEPARA VoipChlType VOIP channel type HSDSCH), None DCH
preferably carried on the HS-DSCH. HSPA
HSPA(UL_EDCH,DL_HS
- HSPA: Uplink is preferably carried on the E-DCH, and downlink is
DSCH)
preferably carried on the HS-DSCH.
Channel type of IMS signaling.

DCH(UL_DCH,DL_DCH)
- DCH: Both uplink and downlink are preferably carried on the DCH.
, HSDPA(UL_DCH,DL_
- HSDPA: Uplink is preferably carried on the DCH, and downlink is DCH, HSDPA,
UFRCCHLTYPEPARA ImsChlType IMS channel type HSDSCH), None DCH
preferably carried on the HS-DSCH. HSPA
HSPA(UL_EDCH,DL_
- HSPA: Uplink is preferably carried on the E-DCH, and downlink is
HSDSCH)
preferably carried on the HS-DSCH.
Rate threshold for decision to use HS-DSCH to carry DL PS streaming

D8, D16, D32, D64, 8, 16, 32, 64,
DL streaming traffic services. When the maximum DL service rate is higher than or equal to
UFRCCHLTYPEPARA DlStrThsOnHsdpa D128, D144, D256, 128, 144, 256, kbit/s D64
threshold on HSDPA this threshold, the service will be carried on HS-DSCH. Otherwise, it will
D384 384
be carried on DCH.
Rate threshold for decision to use E-DCH to carry PS streaming services. D8, D16, D32, D64, 8, 16, 32, 64,
UL streaming traffic
UFRCCHLTYPEPARA UlStrThsOnHsupa When the maximum UL rate is higher than or equal to this threshold, the D128, D144, D256, 128, 144, 256, kbit/s D256
threshold on HSUPA
service will be carried on E-DCH. Otherwise, it will be carried on DCH. D384 384
MAP_PS_STREAM_ON_HSDPA_SWITCH: When the switch is on, a PS

streaming service is mapped on the HS-DSCH if the DL maximum rate of
MAP_PS_STREAM_ON
the service is greater than or equal to the HSDPA threshold for streaming
_HSDPA_SWITCH,
Service Mapping services.
UCORRMALGOSWITCH OFF,ON OFF,ON None OFF
Strategy Switch MAP_PS_STREAM_ON_HSUPA_SWITCH: When the switch is on, a PS
MAP_PS_STREAM_ON
streaming service is mapped on the E-DCH if the UL maximum rate of the
_HSUPA_SWITCH,
service is greater than or equal to the HSUPA threshold for streaming Page 58
services.

Range Value
Rate threshold for the decision to set up a PS domain
UL BE traffic DCH background/interactive service on the DCH in the uplink/downlink. Only
UCELLFRC UlBeTraffDecThs
decision threshold when the UL/DL service rate is higher than or equal to the
uplink/downlink threshold, the service will be set up on DCH. Otherwise,
D8, D16 8, 16 kbit/s D8
the service is set up on CCH. According to the specifications, only the
background/interactive services whose rate is lower than 16 kbit/s can be
DL BE traffic DCH
UCELLFRC DlBeTraffDecThs carried on the CCH. Therefore, this parameter can only be set to 16 kbit/s
decision threshold
or 8 kbit/s.
D8, D16, D32, D64,
D128, D144, D256, 8, 16, 32, 64, 128,
Rate threshold for decision to use HS-DSCH to carry DL PS D384, D512, D768, 144, 256, 384, 512,
UFRCCHLTYP DL BE traffic threshold on background/interactive services. When the maximum DL service rate is D1024, D1536, 768, 1024, 1536,
DlBeTraffThsOnHsdpa kbit/s D64
EPARA HSDPA higher than or equal to this threshold, the service will be carried on HS- D1800, D2048, 1800, 2048, 3600,
DSCH. Otherwise, it will be carried on DCH. D3600, D7200, 7200, 8640, 10100,
D8640, D10100, 13900
D13900
This parameter specifies the rate threshold for decision to use E-DCH to D8, D16, D32, D64, 8, 16, 32, 64, 128,
UFRCCHLTYP UL BE traffic threshold on carry UL PS domain background/interactive services. When the D128, D144, D256, 144, 256, 384, 608,
UlBeTraffThsOnHsupa kbit/s D608
EPARA HSUPA maximum UL service rate is higher than or equal to this threshold, the D384, D608, D1280, 1280, 2048, 2720,
service will be carried on E-DCH. Otherwise, it will be carried on DCH. D2048, D2720, D5440 5440
This parameter specifies the rate threshold of 2ms TTI on the E-DCH for
D8, D16, D32, D64,
HSUPA BE service. When the cell capability and UE capability allows 8, 16, 32, 64, 128,
Rate threshold of BE on D128, D144, D256,
UFRC BeHsupa2msTtiRateThs 2ms TTI, 2ms TTI on the E-DCH is used if the PS BE service carried on 144, 256, 384, 608, kbit/s None
2ms TTI of HSUPA D384, D608, D1280,
HSUPA with uplink rate isn't less than this threshold. Otherwise, 10ms 1280, 2048, 2720
D2048, D2720
TTI is used.
RESERVED_SWITCH_0_BIT5: When the switch is set to ON, the RNC

establishes the BE service of a UE on R99 channels instead of on
UCORRMALG RESERVED_SWITCH_0 RESERVED_SWITCH_0_
HSUPA channels if the uplink coverage of the UE is limited. RNC OFF,ON None OFF
OSWITCH _BIT5 BIT5
determines whether the uplink coverage of a UE is limited on the basis of
the Ec/N0 reported by the UE during RRC connection establishment.
Page 59

Range Range Value
CORRM algorithm reserved U32 para 0. The para of 32 bits is reserved for
further change request use.
E2DRachRprtEcN0Thd (Ec/N0 Threshold for E-DCH-to-DCH Channel Switch):

UCORRMALG CORRM Algorithm
ReservedU32Para0 0~4294967295 0~4294967295 None 0
OSWITCH Reserved U32 Para 0 Ec/No decision threshold for BE service DCH establishment in the uplink in
case of weak coverage. If the Ec/No of the target cell reported by a UE that
meets the HSUPA bearer conditions of BE services is smaller than or equal to
the threshold, uplink BE services are carried on the DCH rather than the
E-DCH.
Time duration when the reported Ec/No is valid. The reported Ec/No is valid for
UFRC EcN0EffectTime Ec/N0 effective time the period (starting from the time when the Ec/No report is received) specified 0~65535 0~65535 ms 5000
by this parameter.
Threshold for determining the signal quality in a cell. If the reported Ec/No
UCELLFRC EcN0Ths Ec/N0 threshold exceeds the value of this parameter, you can infer that the signal quality in the 0~49 -24.5~0 0.5dB 41 (-4dB)
cell is good and a high code rate can be set for initial access.
Page 60
Channel Switching(UE State Transition)
The Channel Switching or UE state transition optimizes the use of physical radio resources by issuing reconfiguration of
an existing connection. For a particular connection, PS RAB reconfiguration is triggered by changes in the amount of data to be transmitted,
available radio resources, coverage, and mobility.
When triggered on an active UE with a large amount of data to transmit on the PS RAB, the Channel Switching (UE state transition) feature aims to
dynamically allocate the channel with highest available bit rate providing optimal common resource consumption.
A passive UE with a small amount of, or no data, to transmit is switched to the state with the least resource consumption. This increase the radio
resouces utilization efficiency and shorten the call setup time (RAB setup) when the UEs resume their activities.
The figure shows the transitions between the

UTRAN RRC mode and the GSM connected mode
for CS services and transitions between the
UTRAN RRC mode and the GSM/GPRS packet
transfer mode for PS services.
This figure also shows the state transitions

between the idle mode and the UTRAN RRC mode
and the state transitions in UTRAN RRC mode.
The principles of the UE state transition

function are as follows:
If the activity of the UE decreases, the state of the

UE transits from CELL_DCH to CELL_FACH or
from CELL_FACH to CELL_PCH/URA_PCH.
If the activity of the UE increases, the state of the

UE transits from CELL_PCH/URA_PCH to
CELL_FACH or from CELL_FACH to CELL_DCH.
Page 61
UE State Transition (Channel Switching)
After the RRC connection is set up, the RNC observes PS users activity and uses the UE state transition function to change the UE state.
UE state transistion consists of two parts: evaluation and execution. The evaluation part is responsible for monitoring the resource
needs of a UE through measurement reports received from the UE, the Node B, and the RNC. A request to change the connection state is
made to the execution part when such a need is detected.
 UE state transition is evaluated and triggered by user activity (traffic volume), user inactivity (timer) or mobility.
Traffic Volume Report from UE, Event 4a and Event 4b

Event 4a: Traffic volume is above a threshold
(High user activity).
Event 4b: Traffic volume is below a threshold during a

configurable time (Low user activity).
Cell re-selection
If the number of cell reselections exceeds configurable
threshold during a configurable period (timer), the UE is
considered to be in the state of frequent cell reselection.
Paging / Data
There is data to be transferred.
Overload
When a cell is in the overload state, RNC may transit the  If a UE has a CS service, the UE stays in CELL_DCH during the call
UE from CELL_DCH (DCH/HS-DSCH/E-DCH) to and is not allowed for D2F.
CELL_FACH (FACH) to relieve the load. If a UE has a streaming service, the UE is not allowed for F2P.
Page 62
State transition Channel switch Description
D2F CELL_DCH (DCH) to CELL_FACH
D2F E2F CELL_DCH (E-DCH) to CELL_FACH
H2F CELL_DCH (H-DSCH) to CELL_FACH
F2D CELL_FACH to CELL_DCH (DCH)
F2D F2H CELL_FACH to CELL_DCH (H-DSCH)
F2E CELL_FACH to CELL_DCH (E-DCH)
F2P - CELL_FACH to CELL_PCH
P2U - CELL_PCH to URA_PCH
P2F - CELL_PCH to CELL_FACH
U2F - URA_PCH to CELL_FACH
H2D CELL_DCH (H-DSCH) to CELL_DCH (DCH)
E2D CELL_DCH (E-DCH) to CELL_DCH (DCH)
D2D
D2H CELL_DCH (DCH) to CELL_DCH (H-DSCH)
D2E CELL_DCH (DCH) to CELL_DCH (E-DCH)
CPC2F CELL_DCH (CPC) to CELL_FACH
CPC2EFACH CELL_DCH (CPC) to Enhanced CELL_FACH
EFACH2CPC Enhanced CELL_FACH to CELL_DCH (CPC)
CPC and EFACH
EFACH2D Enhanced CELL_FACH to CELL_DCH (DCH)
D2EFACH CELL_DCH (DCH) to Enhanced CELL_FACH
F2CPC CELL_FACH to CELL_DCH (CPC)
Always online - RRC Connected mode to Idle mode
Fast dormancy - New state transition mechanism is designed for fast dormancy.
For best effort (BE) services that are not allowed for state transition,
Low Activity - RNC can instruct the BE service to enter low activity state which is
configured with lower rate. Page 63
UE State Transition Switches
SET UCORRMALGOSWITCH (DraSwitch) defines the dynamic resource allocation global parameters (RNC-level) are as follows:
 DRA_PS_BE_STATE_TRANS_SWITCH enables RRC state transitions for bearers carrying Best Effort services.
 DRA_PS_NON_BE_STATE_TRANS_SWITCH enables RRC state transitions for bearers carrying services (PS real-time service) other than Best
Effort.
 DRA_HSDPA_STATE_TRANS_SWITCH enables RRC state transition from HS-DSCH to FACH.
Only valid if either DRA_PS_BE_STATE_TRANS_SWITCH or DRA_PS_NON_BE_STATE_TRANS_SWITCH (or both) are enabled, according to the
type of service being carried on HS-DSCH, that is, Best Effort or not Best Effort, respectively.
 DRA_HSUPA_STATE_TRANS_SWITCH enables RRC state transition from E-DCH to FACH.

Only valid if either DRA_PS_BE_STATE_TRANS_SWITCH or DRA_PS_NON_BE_STATE_TRANS_SWITCH (or both) are enabled, according to the
type of service being carried on E-DCH, that is, Best Effort or not Best Effort, respectively.
In the case that the previous switches are set to 1:
 If DRA_DCCC_SWITCH is set to 1 and the dynamic channel adjustment strategy (DcccStg) is set to RATE_UP_AND_DOWN_ON_DCH, the UE state
transition function starts to work when the UL/DL rate is below or equal to the UL/DL rate threshold for DCCC (UlDcccRateThd/DlDcccRateThd).
 If DRA_DCCC_SWITCH is set to 1 and DcccStg is set to RATE_UP_ONLY_ON_DCH, the UE state transition function always works.
 If DRA_DCCC_SWITCH is set to 0, the UE state transition function always works.
For the PS inactive service, if the timer parameters for online detection, which can be set through the SET UPSINACTTIMER command, are set to be a
small value, the RRC connection may be released before UE state transition.
SET UUESTATETRANS defines the UE RRC state transition parameters (RNC-level)

SET UUESTATETRANSTIMER defines the UE RRC state transition timers (RNC-level)
Page 64
D2F State Transition
When receiving a report on event 4b, the RNC starts the timer for transition D2F and the counters for UL and DL event 4b reports.
If both the number of UL event 4b reports and the number of DL event 4b reports are above or equal to the D2F transition threshold before the timer
expires, the D2F is performed when the timer expires.
The D2F transition threshold is the rounded-down value obtained according to the following formula:
D2F transition threshold = [D2F transition time/ (Time to trigger + Pending time after trigger) x State transition traffic redundancy coefficient]
Where:
 The D2F transition time, time to trigger, and pending time after trigger are specified by the parameters listed in table below.
 The state transition traffic redundancy coefficient is used to prevent false detection that a UE is in a low-activity state,
due to the loss of measurement reports. This coefficient is set to 80%.
Event 4b
Event 4b Event 4b
Service Type D2F Transition Timer Pending Time After
Reporting threshold Time to Trigger
Trigger
BE service on
BeD2FStateTransTimer D2F2PTvmThd D2FTvmTimeToTrig D2FTvmPTAT
DCH
BE service on
BeH2FStateTransTimer BeH2FTvmThd BeH2FTvmTimeToTrig BeH2FTvmPTAT
HS-DSCH
PS real-time
RtDH2FStateTransTimer RtDH2FTvmThd RtDH2FTvmTimeToTrig RtDH2FTvmPTAT
service
BE service on E2FThrouTimeToTrig E2FThrouPTAT x

BeE2FStateTransTimer E2FThrouThd
E-DCH x E2FThrouMeasPeriod E2FThrouMeasPeriod
Page 65
D2F State Transition
The parameters of event 4b are set to their respective parameter values for the UL and the DL. The parameters include the time to trigger,
pending time after trigger, and threshold. For example, for BE services if the UL channel is the DCH and the DL channel is the HS-DSCH, the
parameter configurations for BE service on the DCH as listed in the previous table are used for the UL, and the parameter configurations for
BE service on the HS-DSCH as listed in the previous table are used for the DL.
State transition from E-DCH to FACH is similar to state transition from DCH to FACH. The difference is that the event 4b measurement for
state transition from E-DCH to FACH is based on the throughput. The throughput measurement period is specified by
E2FThrouMeasPeriod. The threshold of event 4b is also based on the throughput. The threshold is specified by E2FThrouThd.
Low Activity State

For BE services, if DRA_PS_BE_STATE_TRANS_SWITCH is set to 0 or the UE has other services that do not allow UE state transition,
UE state transition cannot be performed. In this case, if both the number of UL event 4b reports and the number of DL event 4b reports are
above or equal to the D2F transition threshold before the transition timer expires, the rate can be adjusted to the low activity rate.
For single service:
•For BE service on the DCH (both uplink and downlink), both the uplink rate and downlink rate can be adjusted to the low-activity rate threshold
(LittleRateThd).
•For BE service on the DCH (uplink) and HS-DSCH (downlink), only the uplink rate can be adjusted to LittleRateThd.
•For BE service on the E-DCH (uplink) and DCH (downlink), only the downlink rate can be adjusted to LittleRateThd.
Note that if LittleRateThd is above or equal to UlDcccRateThd or DlDcccRateThd, the rate cannot be adjusted to LittleRateThd.
If LittleRateThd is excessively high, excessive resources are occupied even if the UE does not have data to transmit for a long time, leading to network congestion.
If LittleRateThd is excessively low, the throughput will be lower than the expected value when UE starts to transmit data again, though more radio resources are saved.
For service combination:

If the service combination has a CS service, the UE stays in CELL_DCH state and is not allowed for D2F. If the PS service meets the conditions
for D2F, the RNC decrease data rate of the PS service to the low activity rate.
If the service combination has only PS services and the state transition switches are turned off, the RNC decreases the data Page 66
rate of PS services to the low activity rate.
F2D State Transition
The F2D state transition happens when the UTRAN receives a 4a report that triggered by the UL or DL traffic volume. RNC determines the transport
channel (DCH, HS-DSCH, E-DCH) and the target Target rates for UE state transitions F2D.
When the cell is in the overload congestion state, the state transition F2D is prohibited.
Service Type Threshold of Event 4a Target Rate
●For the uplink, the target rate equals the smaller one between the
MBR and the UL rate threshold for DCCC.
BE service on the DCH BeF2DTvmThd
For the downlink, the target rate equals the smaller one between
the MBR and the DL rate threshold for DCCC.
●For the uplink:
- The target rate equals the smaller one between the MBR and the
UL rate threshold for DCCC for transition from FACH to DCH.
- The target rate equals the minimum rate in the HSUPA UL rate
BE service on the HS-DSCH BeF2HTvmThd
adjustment set (EdchRateAdjustSet) for the transition from
FACH to E-DCH.
For the downlink, the target rate equals the MBR.

Real-time service in the PS
RtF2DHTvmThd MBR
domain
●For the uplink:
- The target rate equals the minimum rate in the
EdchRateAdjustSet when DRA_HSUPA_DCCC_SWITCH is
set to ON.
- The target rate equals the MBR when
DRA_HSUPA_DCCC_SWITCH is set to OFF.
BE service on the E-DCH BeF2ETvmThd
For the downlink:
- The target rate equals the smaller one between the MBR and the
DL rate threshold for DCCC when the DL channel is DCH.
- The target rate equals the MBR when the DL channel is HS-
DSCH. Page 67
F2P State Transition
UE state transition F2P is applicable to BE services in the PS domain.

UE state transition F2P is similar to UE state transition D2F.
When receiving a report that event 4b occurs and the traffic volume is zero, the RNC starts the timer for transition F2P and the counters for
UL and DL event 4b reports. If both the number of UL event 4B reports and the number of DL event 4B reports are above the F2P transition
threshold before the timer expires, the F2P happens when the timer expires.
The F2P transition threshold is the rounded-down value obtained according to the following formula:
F2P transition threshold = [F2P transition time/(Time to trigger + Pending time after trigger) x State transition traffic redundancy coefficient]
Event 4b
Event 4b Event 4b
Service Type F2P Transition Timer Pending Time After
Reporting threshold Time to Trigger
Trigger
PS BE service BeF2PStateTransTimer D2F2PTvmThd F2PTvmTimeToTrig F2PTvmPTAT
Page 68
P2U State Transition
UE state transition P2U is applicable to BE services in the PS domain.
Before the state transition, the state of the UE is CELL_PCH. During the cell reselection, the UE sends the CELL UPDATE messages.
The RNC starts a timer (CellReSelectTimer) and counts the number of CELL UPDATE messages with the cause value of cell reselection.
When the timer expires, the number of CELL UPDATE messages may be above the threshold (CellReSelectCounter).
In this case, the RNC initiates a state transition when the UE sends the CELL UPDATE message again.
The state transition P2U involves the transient state CELL_FACH, in which some necessary signaling interaction is performed.
P2F and U2F State Transition
P2F and U2F are performed when UTRAN pages the UE or when the UE needs to exchange information with the UTRAN.
There are no configuration parameters.
Page 69
D2D State Transition
For HSDPA and HSUPA, there are two more sub-state called CELL_DCH (with HS-DSCH) and CELL_DCH (with E-DCH).
Principle of state transitions of D2D is shown in the following figure.
Type Trigger (D2D)
(1) During RAB setup, RAB should be carried over E-DCH but resources are insufficient.
(2) During RAB setup, RAB should be carried over E-DCH but the coverage is
insufficient.
(3) After HSUPA RAB setup, if the radio coverage is poor during HSUPA data
transmission,
E2D E-DCH can switch to DCH based on the algorithm of coverage-based BE service
fallback from E-DCH to DCH.
Coverage-based BE service fallback from E-DCH to DCH. (4) E-DCH serving cell change (change of best cell)
During data transmission of an HSUPA BE service, if the user is located in poor radio − New best cell does not support E-DCH
− New best cell does not have enough resources for E-DCH
coverage, the BE service can fall back from E-DCH to DCH. In this way, the uplink
coverage performance of the user is improved. The algorithm is controlled by
RESERVED_SWITCH_0_BIT7. (1) In any of the scenarios above (E2D), it is possible to transit DCH to E-DCH.
− Periodic attempt to reconfigure to E-DCH (timer)
− Traffic volume (event 4A)
During data transmission of an HSUPA BE service, if the transmit power of the UE is
D2E
higher than the threshold of event 6A and the throughput is lower than the threshold of (2) If a UE reports event 6B2 and event 4A, and the UE services can be carried on the
event 4B, the service falls back from E-DCH to DCH. The threshold of event 6A is E-DCH, then the RNC switches the UE from the DCH to the E-DCH during the rate
calculated through ReservedU8Para0 and the threshold of event 4B is calculated increase process
through E2DThrou4BThd.
(1) RAB should be carried over HS-DSCH but resources are insufficient.
If an HSUPA BE service of a UE falls back from E-DCH to DCH, all uplink services of
the UE also fall back. A fallback from E-DCH to DCH occurs only in the current cell. H2D (2) HS-DSCH serving cell change (change of best cell)
− New best cell does not support HS-DSCH
− New best cell does not have enough resources for HS-DSCH
After a service falls back to the DCH, it does not periodically attempt to switch to the
E-DCH. Actually, the service directly switches from DCH to E-DCH when traffic (1) In any of the scenarios above (H2D), it is possible to transit DCH to HS-DSCH later.
volume increases and the uplink coverage is recovered. The UE reports event 4A to
indicate the traffic volume increases and event 6B to indicate the uplink coverage is D2H − Periodic attempt to reconfigure to HS-DSCH (timer)
− Traffic volume (event 4A)
recovered. The 6B threshold is calculated through ReservedU8Para1.
If the algorithm of coverage-based BE service fallback from E-DCH to DCH is enabled,
the algorithm of coverage-based dynamic TTI adjustment of BE services is
automatically disabled. Page 70
Channel Switching Parameters
Range Value
DRA_PS_BE_STATE_TRANS_SWITCH: When the switch is on, UE RRC

status transition (CELL_FACH/CELL_PCH/URA_PCH) is allowed at the
RNC.
DRA_PS_NON_BE_STATE_TRANS_SWITCH: When the switch is on, the

DRA_PS_BE_STATE
DRA_PS_BE_STATE_TRANS_status of the UE RRC that carrying real-time services can be changed to
_TRANS_SWITCH
SWITCH CELL_FACH at the RNC.
DRA_HSDPA_STATE
DRA_HSDPA_STATE_TRANS DRA_HSDPA_STATE_TRANS_SWITCH: When the switch is on, the status
_TRANS_SWITCH
_SWITCH of the UE RRC that carrying HSDPA services can be changed to
UCORRMALG
CELL_FACH at the RNC. If a PS BE service is carried over the HS-DSCH, OFF,ON OFF,ON None OFF
OSWITCH DRA_HSUPA_STATE
DRA_HSUPA_STATE_TRANS the switch PS_BE_STATE_TRANS_SWITCH should be on simultaneously. If
_TRANS_SWITCH
_SWITCH a PS real-time service is carried over the HS-DSCH, the switch
DRA_PS_NON_BE_
DRA_PS_NON_BE_STATE_
STATE_
TRANS_SWITCH: DRA_HSUPA_STATE_TRANS_SWITCH: When the switch is on, the status
TRANS_SWITCH:
of the UE RRC that carrying HSUPA services can be changed to
CELL_FACH at the RNC. If a PS BE service is carried over the E-DCH, the
switch PS_BE_STATE_TRANS_SWITCH should be on simultaneously. If a
PS real-time service is carried over the E-DCH, the switch
Strategy for adjustment of the rate of the PS BE service when the UE is in

the CELL_DCH state. RATE_UP_AND_DOWN_ON_DCH indicates that the RATE_UP_AND_D RATE_UP_AND_D
rate over the DCH can be raised or lowered. RATE_UP_ONLY indicates that OWN_ON_DCH, OWN_ON_DCH, RATE_UP_
UDCCC DcccStg DCCC Strategy the rate over the DCH can only be raised. When UE is in low activity state, it None AND_DOWN_
is no need to degrade the traffic rate to the threshold for DCCC before state RATE_UP_ONLY_ RATE_UP_ONLY_ ON_DCH
transition to CELL_FACH. That means UE can transit to CELL_FACH state ON_DCH ON_DCH
at any rate.
For a BE service that has a low maximum rate, the DCCC algorithm is not
D8, D16, D32, D64,
Uplink Bit Rate Threshold for obviously effective yet it increases algorithm processing. Thus, the traffic- 8, 16, 32, 64, 128,
UDCCC UlDcccRateThd D128, D144, D256, kbit/s D64
DCCC based DCCC algorithm is applied to BE services whose maximum UL rate is 144, 256, 384
D384
greater than the threshold.
For a BE service that has a low maximum rate, the DCCC algorithm is not
D8, D16, D32, D64,
Downlink Bit Rate Threshold for obviously effective yet it increases algorithm processing. Thus, the traffic- 8, 16, 32, 64, 128,
UDCCC DlDcccRateThd D128, D144, D256, kbit/s D64
DCCC based DCCC algorithm is applied to BE services whose maximum DL rate is 144, 256, 384
D384
greater than the threshold.
Page 71
Range Range Value
When the BE service rate decreases to the DCCC threshold rate, the UE,
however, cannot perform state transition to FACH state because, for example,
the state transition switch is OFF or there are CS services. In this case, when
D0, D8, D16,
traffic remains low for quite a long period, the service rate decreases to this rate 0, 8, 16, 32, 64,
Low Activity Bit Rate D32, D64, D128,
UDCCC LittleRateThd and D2F state transition is not performed. The timer parameters and traffic 128, 144, 256, kbit/s D64
Threshold D144, D256,
volume measurement parameters for the function of low activity rate adjustment 384
D384
are the same as those in the D2F state transition process. The relative
parameters can be set in command SET UUESTATETRANS and SET
UUESTATETRANSTIMER.
RESERVED_SWITCH_0_BIT7: When the switch is set to ON, BE services that

UCORRMALGOSWITCH RESERVED_SWITCH RESERVED_SWITCH are already established on E-DCHs can be re-established on DCHs due to OFF,ON OFF,ON None OFF
_0_BIT7 _0_BIT7 insufficient coverage.
ReservedU8Para0(Event 6A Relative Threshold for DCH-to-E-DCH Channel

CORRM Algorithm Switch):Event 6A relative threshold for triggering BE service DCH-to-E-DCH
UCORRMALGOSWITCH ReservedU8Para0 0~255 0~255 None 0
Reserved U8 Para 0 channel switch in the uplink based on coverage. This parameter is used to
calculate the event 6A absolute threshold. If the the UE transmit power for uplink
Iis greater than event 6A absolute threshold , the UE reports an event 6A.
ReservedU8Para1(Event 6B Relative Threshold for DCH-to-E-DCH Channel
Switch):Event 6B relative threshold for triggering BE service DCH-to-E-DCH
CORRM Algorithm
UCORRMALGOSWITCH ReservedU8Para1 channel switch in the uplink based on coverage. This parameter is used to 0~255 0~255 None 255
Reserved U8 Para 1
calculate the event 6B absolute threshold. If the UE transmit power for uplink is
lless than the event 6B absolute threshold , the UE reports an event 6B.
Event 4B Threshold for

Throughput event 4B threshold for BE service E-DCH-to-DCH channel switch in
UDCCC E2DThrou4BThd E-DCH-to-DCH 0~1148 0~11480 10kbit/s 30
the uplink based on coverage
Channel Switch
Threshold for triggering event 6A1 for BE services. Event 6A1 is reported
through a measurement report, and is triggered when the transmit power of a UE
is greater than the absolute threshold for triggering event 6A1. The value of this
UEDCHTTIRECFG BeThd6A1 6A1 threshold for BE 0~82 0~82 dB 3
parameter is a relative threshold, which is calculated as follows: Absolute
threshold for triggering event 6A1 = Maximum uplink transmit power - Value of
this parameter.
Threshold for triggering event 6B1 for BE services. Event 6B1 is reported
through a measurement report, and is triggered when the transmit power of a UE
is smaller than the absolute threshold for triggering event 6B1. The value of this
UEDCHTTIRECFG BeThd6B1 6B1 threshold for BE 0~82 0~82 dB 3
parameter is a relative threshold, which is calculated as follows: Absolute
threshold for triggering event 6B1 = Maximum uplink transmit power - Value of
this parameter.
Page 72
Baseline
Value
If the times the UE in the CELL_PCH state performs cell

reselection is greater than or equal to the threshold, it is
Cell Reselection regarded that the cell reselection is frequent. When the timer
UUESTATETRANS CellReSelectCounter 1~65535 1~65535 None 9
Counter Threshold expires, the target state is set to URA_PCH. In the next cell
update procedure, the UE is informed of state transition to
URA_PCH in the CELL_UPDATE_CONFIRM message.
This parameter is used to reduce incorrect UE activity

decisions caused by loss of traffic volume reports. Threshold
= maximum times of receiving traffic volume reports x traffic
State Trans Traff volume redundancy coefficient for state transition x 100%.
UUESTATETRANS StateTransTraffRedundCoef 0~100 0~1, step: 0.01 None 80
Redund Coef
Within the period specified by the state transition timer, it is
regarded that the UE is of low activity if the received traffic
volume event 4B reports is beyond the threshold.
D8, D16, D32, D64, D128,

This parameter is used to check whether the UE in the low 8, 16, 32, 64, 128, 256,
D256, D512, D1024, D2k,
activity state. If the UE is on DCH channel, the low activity 512, 1024, 2k, 3k, 4k,
D3k, D4k, D6k, D8k, D12k,
BE D2F/R or F/R2P 4B counter increases by 1 every time traffic volume event 4B is 6k, 8k, 12k, 16k, 24k,
UUESTATETRANS D2F2PTvmThd D16k, D24k, D32k, D48k, byte D64
Threshold reported. If the UE is on FACH channel, the low activity 32k, 48k, 64k, 96k,
D64k, D96k, D128k,
counter increases by 1 if the traffic volume is 0 in the traffic 128k, 192k, 256k,
D192k, D256k, D384k,
volume event 4B report. 384k, 512k, 768k
D512k, D768k
When the traffic volume is below the 4B threshold and

D0, D10, D20, D40, D60, 0, 10, 20, 40, 60, 80,
remains so for the period specified by this parameter, the
D80, D100, D120, D160, 100, 120, 160, 200,
UUESTATETRANS D2FTvmTimeToTrig BE D2F/R 4B Time event 4B is reported. This parameter can prevent ms D5000
D200, D240, D320, D640, 240, 320, 640, 1280,
unnecessary reports to be triggered, caused by traffic
D1280, D2560, D5000 2560, 5000
volume variation.
Pending time after the traffic volume event 4B for state
D250, D500, D1000,
BE D2F/R 4B Pending transition from DCH to FACH is reported. This parameter is 250, 500, 1000, 2000,
UUESTATETRANS D2FTvmPTAT D2000, D4000, D8000, ms D1000
Time used to prevent too many traffic volume events 4B being 4000, 8000, 16000
D16000
reported.
D0, D10, D20, D40, D60, 0, 10, 20, 40, 60, 80,
D80, D100, D120, D160, 100, 120, 160, 200,
UUESTATETRANS F2PTvmTimeToTrig BE F2P 4B Time event 4B is reported. This parameter can prevent ms D5000
D200, D240, D320, D640, 240, 320, 640, 1280,
D1280, D2560, D5000 2560, 5000
volume variation.
D250, D500, D1000,
BE F2P 4B Pending transition from FACH to PCH is reported. This parameter is 250, 500, 1000, 2000,
UUESTATETRANS F2PTvmPTAT D2000, D4000, D8000, ms D16000
Time used to prevent too many traffic volume events 4B being 4000, 8000, 16000
D16000
reported. Page 73
Baseline
Value
D16, D32, D64, D128, D256,
16, 32, 64, 128, 256, 512,
D512, D1024, D2k, D3k, D4k,
This parameter specifies the threshold of the traffic 1024, 2k, 3k, 4k, 6k, 8k,
D6k, D8k, D12k, D16k, D24k,
UUESTATETRANS BeF2DTvmThd BE F/R2D 4A Threshold volume of 4A event for triggering the transition of BE 12k, 16k, 24k, 32k, 48k, byte D1024
D32k, D48k, D64k, D96k,
services from FACH to DCH. 64k, 96k, 128k, 192k,
D128k, D192k, D256k,
256k, 384k, 512k, 768k
D384k, D512k, D768k
This parameter specifies the occurrence time of 4A
D0, D10, D20, D40, D60,
event for triggering the transition of BE services from 0, 10, 20, 40, 60, 80, 100,
D80, D100, D120, D160,
UUESTATETRANS BeF2DTvmTimeToTrig BE F/R2D 4A Time FACH to DCH. This parameter can prevent 120, 160, 200, 240, 320, ms D0
D200, D240, D320, D640,
unnecessary reports to be triggered, caused by traffic 640, 1280, 2560, 5000
D1280, D2560, D5000
volume variation.
D8, D16, D32, D64, D128,
8, 16, 32, 64, 128, 256,
This parameter is used to check whether the UE in the D256, D512, D1024, D2k,
512, 1024, 2k, 3k, 4k, 6k,
BE HS-DSCH2F 4B low activity state. If the UE is on HS-DSCH channel, D3k, D4k, D6k, D8k, D12k,
UUESTATETRANS BeH2FTvmThd 8k, 12k, 16k, 24k, 32k, byte D64
Threshold the low activity counter increases by 1 every time traffic D16k, D24k, D32k, D48k,
48k, 64k, 96k, 128k, 192k,
volume event 4B is reported. D64k, D96k, D128k, D192k,
256k, 384k, 512k, 768k
D256k, D384k, D512k, D768k

D0, D10, D20, D40, D60,
remains so for the period specified by this parameter, 0, 10, 20, 40, 60, 80, 100,
BE HS-DSCH2F 4B D80, D100, D120, D160,
UUESTATETRANS BeH2FTvmTimeToTrig the event 4B is reported. This parameter can prevent 120, 160, 200, 240, 320, ms D5000
Time D200, D240, D320, D640,
D1280, D2560, D5000
volume variation.

BE HS-DSCH2F 4B transition from HSPA to FACH is reported. This D250, D500, D1000, D2000, 250, 500, 1000, 2000,
UUESTATETRANS BeH2FTvmPTAT ms D1000
Pending Time parameter is used to prevent too many traffic volume D4000, D8000, D16000 4000, 8000, 16000
events 4B being reported.
D16, D32, D64, D128, D256,
16, 32, 64, 128, 256, 512,
D512, D1024, D2k, D3k, D4k,
This parameter specifies the threshold of the traffic 1024, 2k, 3k, 4k, 6k, 8k,
BE F2HS-DSCH 4A D6k, D8k, D12k, D16k, D24k,
UUESTATETRANS BeF2HTvmThd volume of 4A event for triggering the transition of BE 12k, 16k, 24k, 32k, 48k, byte D1024
Threshold D32k, D48k, D64k, D96k,
services from FACH to HS-DSCH. 64k, 96k, 128k, 192k,
D128k, D192k, D256k,
256k, 384k, 512k, 768k
D384k, D512k, D768k
This parameter specifies the occurrence time of 4A

D0, D10, D20, D40, D60,
event for triggering the transition of BE services from 0, 10, 20, 40, 60, 80, 100,
BE F2HS-DSCH 4A D80, D100, D120, D160,
UUESTATETRANS BeF2HTvmTimeToTrig FACH to HS-DSCH. This parameter can prevent 120, 160, 200, 240, 320, ms D0
Time D200, D240, D320, D640,
D1280, D2560, D5000
volume variation.
Page 74
Baseline
Value
Period of E-DCH throughput ratio measurement. This

parameter specifies the period of E-DCH throughput ratio
UUESTATETRANS E2FThrouMeasPeriod E-DCH Throu Meas Period measurement. The throughput ratio over the E-DCH is 1~10000 10~100000 10ms 30
periodically measured to implement state transition from E-
DCH to FACH and DCCC algorithm on EDCH.
This parameter is used to check whether the UE in the low

activity state. If the UE is on the E-DCH channel, the low
UUESTATETRANS E2FThrouThd E-DCH2F 4B Threshold 0~384 0~384 kbit/s 8
activity counter increases by 1 every time throughput ratio
event 4B is reported.
Number of periods before the throughput ratio event 4B for

state transition from E-DCH to FACH is triggered. When
the throughput ratio is below the lower threshold and
E-DCH2F 4B Period
UUESTATETRANS E2FThrouTimeToTrig event 4B is reported. This parameter can prevent 0~1023 0~1023 None 2
Amount
volume variation. Time to trigger the throughput ratio event
4B = value of this parameter x measurement period x 10
ms.
Number of pending periods after the throughput event 4B
for state transition from E-DCH to FACH is triggered. This
E-DCH2F 4B Pending parameter is used to prevent too many throughput events
UUESTATETRANS E2FThrouPTAT 0~1023 0~1023 None 4
Period Amount 4B being reported. Pending time after the throughput
event 4B is triggered = value of this parameter x
measurement period x 10 ms.
D16, D32, D64, D128,
D256, D512, D1024, 16, 32, 64, 128, 256,
D2k, D3k, D4k, D6k, 512, 1024, 2k, 3k, 4k,
This parameter specifies the threshold of the traffic volume
D8k, D12k, D16k, D24k, 6k, 8k, 12k, 16k, 24k,
UUESTATETRANS BeF2ETvmThd BE F2E-DCH 4A Threshold of 4A event for triggering the transition of BE services from byte D1024
D32k, D48k, D64k, 32k, 48k, 64k, 96k,
FACH to E-DCH.
D96k, D128k, D192k, 128k, 192k, 256k,
D256k, D384k, D512k, 384k, 512k, 768k
D768k
D0, D10, D20, D40, D60,
This parameter specifies the occurrence time of 4A event 0, 10, 20, 40, 60, 80,
D80, D100, D120, D160,
for triggering the transition of BE services from FACH to E- 100, 120, 160, 200,
UUESTATETRANS BeF2ETvmTimeToTrig BE F2E-DCH 4A Time D200, D240, D320, ms D0
DCH. This parameter can prevent unnecessary reports to 240, 320, 640, 1280,
D640, D1280, D2560,
be triggered, caused by traffic volume variation. 2560, 5000
D5000
Page 75
Baseline
Value
D8, D16, D32, D64, D128,

Event 4A threshold for HSDPA traffic D256, D512, D1024, D2k, 8, 16, 32, 64, 128, 256,
measurement in the optimized H2F state D3k, D4k, D6k, D8k, 512, 1024, 2k, 3k, 4k, 6k,
BE HSDPA Traffic Volume 4A
UUESTATETRANS HSDPA4ATvmThd transition algorithm. H2F state transition is a D12k, D16k, D24k, D32k, 8k, 12k, 16k, 24k, 32k, byte D1024
Threshold
state transition of HSPA UEs from CELL_DCH D48k, D64k, D96k, 48k, 64k, 96k, 128k, 192k,
to CELL_FACH. D128k, D192k, D256k, 256k, 384k, 512k
D384k, D512k
Event 4A threshold for HSUPA throughput

measurement in the optimized H2F state
BE HSUPA Throu 4A
UUESTATETRANS HSUPA4AThrouThd transition algorithm. H2F state transition is a 0~384 0~384 kbit/s 64
Threshold
state transition of HSPA UEs from CELL_DCH
to CELL_FACH.
Length of the cell reselection frequency timer.

This parameter is used together with
UUESTATETRANSTIMER CellReSelectTimer Cell Reselection Timer CellReSelectCounter to detect the frequency of 1~65535 1~65535 s 180
cell reselection of the UE in the CELL_PCH
state.
Timer for state transition from DCH to FACH of

BE DCH to FACH Transition BE services, used to check whether the UE in
UUESTATETRANSTIMER BeD2FStateTransTimer 1~65535 1~65535 s 5
Timer the CELL_DCH state with BE services is in the
stable low activity state.
Timer for state transition from FACH or E_FACH

BE FACH or E_FACH to PCH to PCH of BE services, used to check whether
UUESTATETRANSTIMER BeF2PStateTransTimer 1~65535 1~65535 s 65535
Transition Timer the UE in the CELL_FACH state is in the stable
low activity state.
Timer for state transition from HS-DSCH to

BE HS-DSCH to FACH FACH of BE services, used to check whether the
UUESTATETRANSTIMER BeH2FStateTransTimer 1~65535 1~65535 s 5
Transition Timer UE in the CELL_DCH(with HS-DSCH) state with
BE services is in the stable low activity state.
Timer for state transition from E-DCH to FACH

BE E-DCH to FACH State of BE services, used to check whether the UE in
UUESTATETRANSTIMER BeE2FStateTransTimer 1~65535 1~65535 s 5
Transition Timer the CELL_DCH(with E-DCH) state with BE
services is in the stable low activity state.
Page 76
Always Online
If there is no data transmission for the PS service of a UE, the connection for the service is released, but the CN reserves the Packet Data Protocol
(PDP) context for this PS service of the UE.
When the RNC detects that there is no data transmission for a PS service, the RNC sends a request to the CN for release of the service.
The CN initiates the release procedure, requesting the RNC to release the corresponding radio resources. The CN, however, reserves the PDP
context for this PS service of the UE. When the UE re-initiates the service of this PDP context, it does not have to apply for the PDP context again.
Thus, the UE in the PS domain is always online.
For each PS RAB, two PDCP timers, namely T1 and T2, are available.
T1 and T2 are RNC oriented. They can be set to different values for different services by the following parameters:
Conversational service T1: PsInactTmrForCon

Conversational service T2: ProtectTmrForCon
Streaming service T1: PsInactTmrForStr
Streaming service T2: ProtectTmrForStr
Interactive service T1: PsInactTmrForInt
Interactive service T2: ProtectTmrForInt
Background service T1: PsInactTmrForBac
Background service T2: ProtectTmrForBac
IMS signaling T1: PSInactTmrForImsSig
IMS signaling T2: ProtectTmrForImsSig
Page 77
Start
The PDCP entity is set up

The PDCP release procedure is as follows:
Step (1) When the PDCP entity of a service is set up, timer T1 is started.
T1 is started
Step (2) If the PDCP entity still detects no UL or DL data packet when T1
expires, the PDCP entity sends the request to the RRC layer for the service
release and starts T2.
Is there data transmitted in Yes
the uplink or downlink?
Step (3) If the CN does not initiate the service release and the PDCP entity
No still detects no UL or DL data packet when T2 expires, the PDCP entity sends
No
the request again to the RRC layer for the service release.
Does T1 expire?
If there is only one RAB for the UE in the PS domain, the RNC sends an
Yes IU RELEASE REQUEST message to the CN.
The PDCP entity sends the request
to the RRC layer for release of the If there is more than one RAB for the UE in the PS domain, the RNC
service, and T2 is started
sends an RAB RELEASE REQUEST message to the CN.
Yes
The two messages indicate the release causes in the "User Inactivity" IE.
Does the PDCP entity
initiate the release
procedure?
Step (4) When the CN receives the message, it initiates the release
No
procedure. The CN reserves the PDP context for the service and the UE
Yes
does not have to apply for the PDP context again when it re-initiates the
Is there data transmitted in
the uplink or downlink? service.
No
The service release request sent from the PDCP entity may not be processed
No
Does T2 expire? in time. The reason is that the associated processor in the RNC is busy.
Yes End
Page 78

Range Range Value
When detecting that the Ps' Conversational User had no data to transfer for
UPSINACTTIMER PsInactTmrForCon Conversational service T1 a long time which longer than this timer, the PDCP layer would request the 0~14400 0~14400 s 20
RRC layer to release this Radio Access Bear.
After sending release request to RRC layer, if the Conversational Radio

UPSINACTTIMER ProtectTmrForCon Conversational service T2 Access Bear haven't been released until the Conversational service T2 0~60 0~60 s 20
timeout, the PDCP layer will resend release request indication.
When detecting that the Ps' Streaming User had no data to transfer for a
UPSINACTTIMER PsInactTmrForStr Streaming service T1 long time which longer than this timer, the PDCP layer would request the 0~14400 0~14400 s 20
After sending release request to RRC layer, if the Streaming Radio Access
UPSINACTTIMER ProtectTmrForStr Streaming service T2 Bear haven't been released until the Streaming service T2 timeout, the 0~60 0~60 s 20
PDCP layer will resend release request indication.
When detecting that the Ps' Interactive User had no data to transfer for a
UPSINACTTIMER PsInactTmrForInt Interactive service T1 long time which longer than this timer, the PDCP layer would request the 0~14400 0~14400 s 20
After sending release request to RRC layer, if the Interactive Radio Access
UPSINACTTIMER ProtectTmrForInt Interactive service T2 Bear haven't been released until the Interactive service T2 timeout, the 0~60 0~60 s 20
PDCP layer will resend release request indication.
When detecting that the Ps' Background User had no data to transfer for a
UPSINACTTIMER PsInactTmrForBac Background service T1 long time which longer than this timer, the PDCP layer would request the 0~14400 0~14400 s 20
After sending release request to RRC layer, if the Background Radio

UPSINACTTIMER ProtectTmrForBac Background service T2 Access Bear haven't been released until the Background service T2 0~60 0~60 s 20
timeout, the PDCP layer will resend release request indication.
After sending release request to RRC layer, if the IMS Radio Access Bear
UPSINACTTIMER ProtectTmrForImsSig IMS signal T2 for signal haven't been released until the IMS service T2 timeout, the PDCP 0~60 0~60 s 20
layer will resend release request indication.
When detecting that the Ps' IMS signal had no data to transfer for a long
UPSINACTTIMER PSInactTmrForImsSig IMS signal T1 time which longer than this timer, the PDCP layer would request the RRC 0~64800 0~64800 s 20
layer to release this Radio Access Bear.
Page 79
Fast Dormancy (FD)
Smartphone Fast Dormancy

 Fast dormancy function has been introduced to some smartphones, such as iPhone OS3.0, Nokia N97 etc. to allow UEs releasing Iu signaling
connection and change to Idle mode ASAP to save power.
 When UE transfer from DCH to IDLE, 35 signaling message is interacted between UE and the network.
 Fast Dormancy function is implemented differntly in each terminal type and the OS. (iPhone 6 sec , Balck Berry 6sec, HTC Android 8 sec ,
HTC Windows no fast dormancy etc. )
Normal RRC Release Smartphone Fast Dormancy
No data transmission No data transmission
…
20 seconds (inactivity timer) Around 6 seconds (depends on UE types)
…
DCH IDLE DCH IDLE
When the application is in use, sometimes the user is doing nothing at all.
(For example, waiting for message coming when using MSN).
 During the time described as above, UE will send “Heart Beat” message to
inform the AP server that the application is still on.
Heart Beat Increases Signaling and Makes Small Data Packet
Smart phone Mass Small Packets, Always ON and Many More Applications
cause the “Signaling Storm”. Page 80
Fast Dormancy (FD)
The descriptions about the RNC fast dormancy, enhanced fast dormancy, and standard fast dormancy are as follows:
 Fast dormancy, after receiving a “Signaling Connection Release Indication (SCRI)” message from a UE, the RNC releases the UE RRC connection
regardless of whether the message contains the cause value "UE Requested PS Data session end”.
 Enhanced fast dormancy, UEs send SCRI messages that do not contain the cause value "UE Requested PS Data session end" to the RNC. After
receiving such an SCRI message, the RNC moves the UE to the CELL_PCH state.
 Standard fast dormancy is applicable to fast dormancy UEs that comply with 3GPP Release 8 (R8 for short). With standard fast dormancy, UEs send
SCRI messages that contain the cause value "UE Requested PS Data session end" to the RNC. After receiving such an SCRI message, the RNC
moves the UE to the CELL_PCH state. Standard fast dormancy is valid only when the RNC has enabled the timer T323 in system information block type
1 (SIB1).
Fast Dormancy
UEs complying with 3GPP Release 8 support the fast dormancy function, as defined in 3GPP TS 25.331 CR3483.
When a UE processing only PS services finishes PS data transmission, it sends the RNC an SCRI message that may contain the cause value "UE
Requested PS Data session end." Upon receiving the message, the RNC releases the UE’s RRC connection and enables the UE to quickly enter power-
saving mode.The following figure shows the procedure of fast dormancy.
RNC sends the timer T323 in SIB1 or UTRAN Mobility Information to the UE to control fast dormancy.
 If the value of T323 is INVALID, the timer is not sent to the UE. Accordingly, the UE does not send the
SCRI message containing the cause value "UE Requested PS Data session end.“
 If the value of T323 is 0, the UE can recognize the timer. In such a case, when to send the SCRI
messages is determined by the UE behavior.
 If the value of T323 is a value other than 0, the UE sends the release request and initiates timer
T323. The UE sends the release request again after T323 expires.
 If the UE cannot recognize the timer T323, when to send the SCRI messages is not determined by the
timer. The SCRI messages will not contain the cause value "UE Requested PS Data session end."
Page 81
Fast Dormancy (FD)
Enhanced and Standard Fast Dormancy
As the number of intelligent UEs in mobile networks continues increasing, these networks face signaling storms produced by
these UEs. Enhanced fast dormancy is applicable to networks serving a large number of intelligent UEs already supporting fast dormancy.
Enhanced fast dormancy can reduce the signaling processing efforts of the RNC and prevent the signaling processing unit in the RNC from overflowing.
To save power, when there is no PS data to transmit, an intelligent UE sends the RNC an SCRI message to request the RRC connection release or to
indicate to the UTRAN that one of its signaling connections has been released. In addition, the UE periodically sends heartbeat messages to the core
network, for example, to check for new mail on the mail server.
If fast dormancy (rather than enhanced fast dormancy) is applied, the RNC releases the RRC connection as indicated in the 3GPP specifications.
In such a case, the heartbeat messages lead to the procedures for RRC connection setups, authentication, encryption, and RAB setups.
This greatly increases the signaling processing workload of the RNC and may cause the signaling processing unit in the RNC to overflow.
Enhanced fast dormancy and standard fast dormancy are introduced to address this problem.
The following figure shows the difference between fast dormancy, enhanced fast dormancy, and standard fast dormancy
With the application of enhanced fast dormancy

or standard fast dormancy, the RNC changes
the UE state to CELL_PCH instead of idle mode
upon receiving an SCRI message from the UE.
In such a case, when the UE periodically sends

heartbeat messages, the signaling procedure
between the UE and RNC is simplified because the
RRC connection is maintained.
As the number of signaling exchange messages

decreases, significant CPU resources can be
saved for the RNC, and the UE consumes as little
battery power as a UE in idle mode.
Page 82
Fast Dormancy (FD)
Identifying Fast Dormancy UEs
Two types of fast dormancy mechanisms are implemented in UEs. UEs that support fast dormancy are referred to as fast dormancy
UEs in this document.
1) Standard fast dormancy (R8 fast dormancy):

This mechanism is standardized by 3GPP Release 8. As specified, the UE sends an SCRI message to the network with the cause value "Signaling
Connection Release Indication Cause" set to "UE Requested PS Data session end."RNC can identify this type of UE by the SCRI message with the
cause value "UE Requested PS Data session end.“
2) Pre-R8 fast dormancy:
This mechanism, which is not standardized, is used for the early proprietary implementations of fast dormancy prior to Release 8. RNC can identify
this type of UE by its IMEI and its release version. Upon receiving a Security Mode Command Complete message from the UE, the RNC identifies
enhanced fast dormancy UEs as follows:
 If the FD_TAC_MATCH_SWITCH of the parameter PROCESSSWITCH is turned on, the following UEs are identified as fast dormancy UEs:
− UE (Pre-R8 fast dormancy) that complies with 3GPP Release 5 or later releases and whose IMEI is on the white list. The IMEI range for the white list is specified by
the binary group (TAC, FastDormancy = ON) in the RNC MML command ADD UIMEITAC. If the FastDormancy switch is on, the RNC considers the UEs
specified by the IMEI (TAC) capable of enhanced fast dormancy.
− UE (R8 fast dormancy) from which the SCRI message whose cause value is "UE Requested PS Data session end" is sent.
 If the FD_TAC_MATCH_SWITCH of the parameter PROCESSSWITCH is turned off, the following UEs are identified as fast dormancy UEs.
− UEs complying with 3GPP Release 5 or later releases, excluding those on the black list. The IMEI range for the black list is specified by the binary group (TAC,
FastDormancy = OFF) in the RNC MML command ADD UIMEITAC. If the FastDormancy switch is OFF, the RNC considers the UEs specified by the IMEI (TAC)
incapable of enhanced fast dormancy.
RNC can obtain the IMEI of the UE from the UE's response to the "IDENTITY REQUEST" message. The IMEI range is specified by the parameter TAC
in the RNC MML command ADD UIMEITAC.
The IMEI (consisting of 14 decimal digits and one check digit) contains the origin, model, and serial number of the UE. The structure of the IMEI is
specified in 3GPP TS 23.003.
TAC: Type approval code, 6 bit, determined by the type approval center.
FAC: Final assembly code, 2 bit, It is determined by the manufacturer.
SNR: Serial number, 6 bits, It is issued by the manufacturer of the MS. SP: 1 bit , Not used.
(SPARE)
Check the IMEI in your MS : *#06#
Page 83
Fast Dormancy (FD)
State Transitions for Fast Dormancy UEs
If enhanced fast dormancy is disabled, the UE’s RRC connection is released after the RNC receives an SCRI message.
The enhanced fast dormancy switches are specified by the RNC-level subparameter FAST_DORMANCY_SWITCH of the parameter
PROCESSSWITCH and cell-level subparameter FAST_DORMANCY_ADCTRL of the parameter NBMCacAlgoSwitch.
If enhanced fast dormancy is enabled, the state transitions are as follows:

 If the UE is identified as R8 UE, the UE can report an SCRI message with the R8 extended cause value "UE Requested PS Data session end."
After receiving the SCRI message, the RNC moves the UE to CELL_FACH or CELL_PCH.
 If the UE in not an R8 UE and the UE is identified as fast dormancy UE, the RNC moves the UE to CELL_FACH or CELL_PCH after the RNC receives
an SCRI message or the inactivity timer for enhanced fast dormancy times out,
The following figure shows the state transitions for a fast dormancy UE.
Type Description
D2P indicates a transition from CELL_DCH to CELL_PCH. Similarly, F2P is from
CELL_FACH to CELL_PCH, and D2F is from CELL_DCH to CELL_FACH.
RNC starts the PS inactivity timer after the fast dormancy UE changes state.
-PsInactTmrForFstDrmDch specifies the PS inactivity timer for CELL_DCH.

-PsInactTmrForFstDrmFach specifies the PS inactivity timer for CELL_FACH.
-PsInactTmrForPreFstDrm specifies the PS inactivity timer for CELL_PCH.
D2P
If the UE is a fast dormancy UE and is not processing CS services, when the PS inactivity
D2F timer for fast dormancy expires or the RNC receives the SCRI message or 4B event from
the UE, the UE changes states as follows:
F2P
If the UE is in the CELL_DCH state,
-If the switch RSVDBIT1_BIT29 of RsvdPara1 in the SET URRCTRLSWITCH command is
turned off, then the RNC will trigger the D2F state transition, where the UE transitions from CELL_DCH
to CELL_FACH.
− If the switch RSVDBIT1_BIT29 of RsvdPara1 in the SET URRCTRLSWITCH command is turned
on, then the RNC will trigger the D2P state transition, where the UE transitions from CELL_DCH to
CELL_PCH.
If the UE is in the CELL_FACH state,

− The RNC will trigger F2P state transition, where the UE transitions from CELL_FACH to CELL_PCH.
Page 84
Fast Dormancy (FD)
Type Description
D2I indicates a transition from CELL_DCH to IDLE.
When the FACH is congested, D2F will not be triggered for inactive CELL_DCH users. Instead, the RNC triggers a D2I transition in order to release the UE
connection.
D2I
The default D2I function switch is turned off by default. Run the following command to switch it on:
SET UCORRMALGOSWITCH: ReservedSwitch0=RESERVED_SWITCH_0_BIT16-1;
P2I indicates a transition from CELL_PCH to IDLE.

P2I If the UE is in the CELL_PCH state, when the PS inactivity timer for fast dormancy (specified by PsInactTmrForPreFstDrm) expires, the RNC releases the
RRC connection and then the UE enters idle mode.
P2D indicates a transition from CELL_PCH to CELL_DCH.

-If the switch RSVDBIT1_BIT20 of RsvdPara1 in the SET URRCTRLSWITCH command is turned on:
When the UE tries to originate or terminate a CS service and the FACH is congested, the RNC instructs the UE to perform P2D state transition instead of P2F
and then F2D.
SET URRCTRLSWITCH: RsvdPara1=RSVDBIT1_BIT20-0;
P2D
Note: For the RSVDBIT1_BIT20 switch, 0 indicates on and 1 indicates off.
-If the switch FD_P2D_SWITCH of PROCESSSWITCH in the ADD UIMEITAC command is turned on:
If the TAC of the UE is the same as that configured on the RNC and if the UE is in the CELL_PCH and the UE is going to transmit uplink data or respond to the
paging message, the RNC instructs the UE to perform a P2D transition directly.
P2F indicates a transition from CELL_PCH to CELL_FACH.

P2F
When the UE transmits heartbeat messages or traffic data, the RNC instructs the UE to enter the CELL_FACH state.
F2D indicates a transition from CELL_FACH to CELL_DCH.

When the fast dormancy UE is in the CELL_FACH state, the UE undergoes the following transitions:
-If the UE reports event 4A to the RNC, the RNC instructs the UE to enter the CELL_DCH state (F2D transition).
-If the UE reports event 4B to the RNC, the RNC instructs the UE to enter the CELL_PCH state (F2P transition).
F2D
When the fast dormancy UE periodically sends small heartbeat messages, the normal state transition mechanism changes the UE state from CELL_FACH to
CELL_DCH. To prevent state transitions from being triggered by a small number of heartbeat messages, a new threshold for Event 4A is introduced:
FastDormancyF2DHTvmThd.
To prevent too many UEs from remaining in the CELL_FACH state, fast dormancy ensures that the state transition switch does not affect the 4A and 4B
measurement or state transition from CELL_FACH to another state for fast dormancy UEs.
Page 85
Fast Dormancy Parameters
Range Value
If the UE checks that no PS/CS data is transmitted, the UE will send the
SIGNALLING CONNECTION RELEASE INDICATION message to the RNC,
informing the RNC to terminate the PS data session.
D0, D5, D10, D20, 0, 5, 10, 20, 30,
UCONNMODETIMER T323 Timer 323 D30, D60, D90, 60, 90, 120, s INVALID
Meanwhile, the UE starts T323. The UE shall be inhibited from sending the
D120, INVALID INVALID
SIGNALLING CONNECTION RELEASE INDICATION message with the IE
"Signalling Connection Release Indication Cause" set to "UE Requested PS
Data session end" whilst timer T323 is running.
Whether to enable or disable state transition of users in the CELL-DCH state,

who are enabled with fast dormancy, to ease FACH congestion in a cell. If this
FAST_DORMANCY_ FAST_DORMANCY_
UCELLALGOSWITCH switch is turned off in a cell, state transition of such users is disabled. Note that OFF, ON ON, OFF None OFF
ADCTRL ADCTRL
when this switch is turned off in multiple cells under an RNC, signaling storm
may occur. As a result, the CPU usage of the RNC, NodeB, and SGSN
increases greatly, leading to service setup failure.(Parameter withdrawal)
The function of the TAC.If Fast_Dormancy is selected, the TAC specifies the
Fast_Dormancy, Fast_Dormancy,
The function of the UEs that are enabled with the Fast Dormancy feature; if
UIMEITAC TAC_FUNC HSDPA_RB_ HSDPA_RB_ None None
TAC HSDPA_RB_Setup_Cmp_CRC is selected, the TAC specifies the UEs that
Setup_Cmp_CRC Setup_Cmp_CRC
occasionally fail during the HSDPA RAB setup process.
Type Allocation Code (TAC) is part of the International Mobile Station

UIMEITAC TAC TAC 0~99999999 0~99999999 None None
Equipment Identity (IMEI), it is used to identify the type of the UE.
1~64 length of
UIMEITAC Description Description Description UE type. None None None
byte
Fast Dormancy
UIMEITAC FastDormancy The switch of Fast Dormancy function for one type UE. OFF, ON ON, OFF None None
Switch
FD_P2D_SWITCH (TAC-Level P2D Control Switch)

When the switch is turned on, state transition from CELL_PCH/URA_PCH to
CELL_DCH of a UE is triggered if the following conditions are met:
1. The TAC of the UE is the same as that configured on the RNC.
2. The UE is in the CELL_PCH or URA_PCH state, and the RNC receives a
UIMEITAC FD_P2D_SWITCH FD_P2D_SWITCH CELL UPDATE message containing a cause value of "uplink data OFF, ON ON, OFF None None
transmission" or "PS paging response."
When the switch is turned off, state transition from CELL_PCH/URA_PCH to

CELL_DCH is not triggered by the CELL UPDATE message containing "uplink
data transmission" or "PS paging response."
Page 86
Fast Dormancy Parameters

Range Value
FAST_DORMANCY_SWITCH.
FAST_DORMANCY_ FAST_DORMANCY_ When it is checked, RNC will apply FAST DORMANCY function.
SWITCH SWITCH FD_TAC_MATCH_SWITCH.
URRCTRLSWITCH When it is checked, RNC will apply FAST DORMANCY function for UE OFF,ON OFF,ON None OFF
FD_TAC_MATCH_ FD_TAC_MATCH_ whose TAC is configured in database. When it is not checked, RNC
SWITCH SWITCH will apply FAST DORMANCY function for UE whose version is R5 or
later.
When the switch is turned off, the RNC performs the CELL_DCH-to-
CELL_FACH (D2F for short) procedure on UEs that support fast
dormancy.
URRCTRLSWITCH RSVDBIT1_BIT29 RSVDBIT1_BIT29 OFF, ON ON, OFF None OFF
When the switch is turned on, the RNC performs the CELL_DCH-to-
CELL_PCH (D2P for short) procedure on UEs that support fast
dormancy.
When this switch is turned on, the RNC performs a state transition
from CELL_PCH or URA_PCH to CELL_FACH (P2F) upon receiving a
CELL UPDATE message with the cause value "uplink data
transmission" or "paging response".
When this switch is turned off, if receiving a CELL UPDATE message
with the cause value "uplink data transmission" or "paging response,"
the RNC performs a state transition from CELL_PCH or URA_PCH to
URRCTRLSWITCH RSVDBIT1_BIT20 RSVDBIT1_BIT20 OFF,ON OFF,ON None OFF
CELL_DCH (P2D) in either of the following scenarios:
(1)Congestion occurs on the FACH.

(2)Congestion occurs on the DCCH, and the value of the
Establishment cause information element (IE) in the message is
Originating Conversational Call, Terminating Conversational Call, or
Emergency Call.
D16, D32, D64,

16, 32, 64, 128,
D128, D256, D512,
256, 512, 1024,
D1024, D2k, D3k,
2k, 3k, 4k, 6k,
D4k, D6k, D8k,
Fast Dormancy User This parameter specifies the threshold of the traffic volume of 4A event 8k, 12k, 16k,
D12k, D16k, D24k,
UUESTATETRANS FastDormancyF2DHTvmThd FACH/E_FACH2DCH/ for triggering the transition of fast dormancy user from FACH/E_FACH 24k, 32k, 48k, byte D3k
D32k, D48k, D64k,
HSPA 4A Threshold to DCH/HSPA. 64k, 96k, 128k,
D96k, D128k,
192k, 256k,
D192k, D256k,
384k, 512k,
D384k, D512k,
768k
D768k
Page 87
Table of Contents

Page 88
Handover Overview
The purpose of handover control is to manage the mobility aspect of a Radio Resource Control (RRC) connection.
This means keeping track of the user equipment (UE) as it moves around in the network, and ensuring that its connections are
uninterrupted and meet the negotiated Quality ofService (QoS) requirements.
Besides supporting the mobility of the UE, handovers play a key role in maintaining high capacity in the network.
In order to keep the power of its signal constant, the UE must raise its
transmission power as it moves further away from the WCDMA BTS.
To minimise transmission powers, and consequently interference,

the UE should at all times be connected to the strongest cell.
Handover can be classified into the following types:
Intra-UMTS handover
Intra-frequency handover
Inter-frequency handover
Inter-RAT handover
UMTS to GSM handover
GSM to UMTS handover
Page 89
Neighbour cell definitions
Neighbour Cells
When the UE is in connected mode, the RNC follows it on cell level. Once it knows in which cell the UE is located, the RNC checks
information about all the neighbouring cells and transmits the data back to the UE. RNC updates continuously the neighbour cell lists in
order to reflect the changing neighbourhood of a moving mobile station in connected mode.
The neighbouring cells are defined on a cell-by-cell basis, that is, each cell can have its own set of neighbouring cells.
The UE monitors three separate cell categories:
1) Active Set cells: Radio links are established between active set cells and the UE. All cells in the active set send user information.
The cells in the active set are participating in soft handover and they are included in the intra-frequency cell list of the UE.
2) Monitored Set cells: Cells included in the intra-frequency, interfrequency and inter-system cell lists of the UE and monitored
according to these lists. The intra-frequency cells in the monitored set are not participating in soft handovers.
3) Detected Set cells: The cells in the detected set have been detected by the UE outside the intra-frequency cell list of the UE.
Neighbor Cells Configuration
UMTS
Maximum of 31 intra-frequency neighboring cells can be configured for a cell.
Maximum of 64 inter-frequency neighboring cells can be configured for a cell.
(maximum of 32 neighboring cells can have the same downlink frequency and uplink frequency).
Maximum of 32 GSM neighboring cells can be configured for a UMTS cell.
GSM
Maximum 32 UMTS neighbor cells can be configured for a GSM cell.
Page 90
Intra-Frequency Handover
Intra-frequency handover is of the following two types:
Intra-frequency soft handover: means that multiple radio links (RL) are connected to the UE at the same time.
Intra-frequency hard handover: means that only one radio link is connected to the UE at the same time.
Soft Handover
Item Hard Handover
(Softer Handover)
Number of radio links (RL) in the Active Set Several (3 by default) One
Interruption during handover No Yes

Intra-frequency
Frequencies in the cells before and after
Intra-frequency Inter-frequency or
handover Inter-RAT
Maximum-ratio Combining (MRC) or Selective

Handover gain Combining which can reduce the impact of fading None
and transmission power of the UE.
Disadvantage Occupies more resources. Call drop ratio is relatively higher
Intra-Frequency Soft Handover

Intra-frequency soft handover is more commonly used than intra-frequency hard handover.
The types of intra-frequency soft handover are as follows:

Intra-NodeB soft handover (also known as softer handover)
Intra-RNC inter-NodeB soft handover
Inter-RNC soft handover
Intra-frequency soft handover is characterized by the function that the UE can be connected to multiple Universal Terrestrial Radio
Access Network (UTRAN) access points at the same time. Addition and/or release of radio links are controlled by the ACTIVE SET
UPDATE procedure.
Page 91
Handover Overview
Soft handover (SHO) is an important feature of WCDMA which avoids excessive interference from neighboring cells.
During the soft handover , the power control command (UL/DL) will be decided based on the measurement of all cells in Active Set,
therefore the power level of the radio links are optimum.
Page 92
Item Softer Handover Soft Handover
When the UE is in the overlapped coverage area
When the UE is in the overlapped coverage area of two
of multiple neighboring cells of a NodeB with
neighboring cells of different NodeBs
combined RLs. (Radio Link Set : RLS)
Scenario
When the UE communicates with multiple cells by When the UE communicates with different cells by
setting up multiple channels over the Uu interface. setting up multiple channels over the Uu interface
Uplink signal Using maximum-ratio combination (MRC) Using selection combination (SC)
Downlink signal Using maximum-ratio combination (MRC) Using maximum-ratio combination (MRC)
Resource use Occupying less Iub bandwidth Occupying more Iub bandwidth
The HoSwitch: HO_INTRA_FREQ_SOFT_HO_SWITCH parameter is

used to determine whether to enable both soft handover and softer
handover. By default, this switch is set to ON, indicating that both soft
handover and softer handover are enabled. After the RNC receives the
event 1A, 1B, 1C, or 1D report, it initiates the corresponding soft handover
procedure for the UE.
The DivCtrlField parameter indicates whether Maximum-Ratio

Combination (MRC) is enabled in the uplink during softer handover.
When the NodeB decides not to perform Maximum-Ratio Combination
(softer combination), the RNC performs selective combination.
NodeB requires that all cells in softer handover for a UE must be

configured in one uplink resource group.
For 3900 series, NodeB requires that not only all cells in softer handover
for a UE are configured in one uplink resource group but also in one
downlink resource group.
Page 93
Intra-Frequency Hard Handover
Intra-frequency hard handover refers to a handover where all the old radio links are released before the new radio links are established. Compared
with soft handover, intra-frequency hard handover uses fewer resources.
The scenarios of intra-frequency hard handover are as follows:

No Iur interface is present between RNCs. In this scenario, intra-frequency hard handover instead of soft handover can be performed between two
RNCs.
The Iur interface is congested between RNCs. In this scenario, also intra-frequency hard handover instead of soft handover can be performed
between two RNCs.
The intra-frequency soft handover fails and intra-frequency hard handover is allowed.
When intra-frequency soft handover fails because of a congestion problem of the target cell, the RNC tries an intra-frequency hard handover with a
lower service bit rate.
The HO_INTRA_FREQ_HARD_HO_SWITCH parameter is used to determine whether to enable intra-frequency hard handover. By default, this
switch is set to ON.
Inter RNC Handover

When the target cell under the target RNC fulfils the criteria for intra-frequency soft handover, intra- or inter-frequency hard handover, the conditions
that the handover over Iur is triggered are as follows:
Whether the Iur interface is available depends on the setting of the following two parameters according to handover types:
 Whether intra- or inter- frequency hard handover is allowed over the Iur interface depends on the setting of the HHOTRIG parameter.
 Whether inter RNC soft Handover is allowed depends on the setting of the SHOTRIG parameter.
The SHOTRIG parameter consists of three sub-switches:
 CS_SHO_SWTICH: If this switch is checked, soft handover for CS service over the Iur interface is allowed.
 HSPA_SHO_SWTICH: If this switch is checked, soft handover for HSPA service over the Iur interface is allowed.
 NON_HSPA_SHO_SWTICH: If this switch is checked, soft handover for non-HSPA PS service over the Iur interface is allowed.
If the RRC connection has been set up but the Radio Bearers (RBs) have not, whether a cross-Iur soft handover can be executed is determined by
HoSwitch: HO_MC_SIGNAL_IUR_INTRA_SWITCH parameter. Only if the switch is set to ON, can the cross-Iur soft handover be executed.
Page 94
Intra-Frequency Handover procedure
The intra-frequency handover procedure is divided into three phases:
Handover measurement, Handover decision, and Handover execution.
After the UE transits to the CELL_DCH state in connected mode during a call, the RNC sends a MEASUREMENT CONTROL message
to instruct the UE to take measurements and report the measurement event results.
The MEASUREMENT CONTROL message carries the following information:

Event trigger threshold
Hysteresis value
Event trigger delay time
Neighboring cell list
Page 95
Intra-Frequency Handover measurement
In the measurement phase, the UE takes measurements according to the MEASUREMENT CONTROL message
received from the RNC. When the event triggering conditions are met, the UE sends measurement reports to the RNC according to the
rules defined in the MEASUREMENT CONTROL message.
Intra-Frequency Handover Measurement Quantities

Intra-frequency handover uses Ec/No or RSCP of the CPICH as the measurement value. Intra-frequency handover measurement events
can be configured through the parameter IntraFreqMeasQuantity. UE performs layer 3 filtering on measurement values before it
decides measurement events and sends measurement reports. The measurement model, as shown below, is defined in 3GPP25.302.
A: measurement value of the physical layer

B: measurement value obtained after layer-1 filtering.
C: measurement value obtained after layer 3 filtering. This value is controlled by
the higher layer. Filtering coefficient C is applicable to event reports and periodic
reports.
C': another measurement value. C' and C are measured in the same way.
D: measurement report information (message) of Uu or Iub transmission.
Measurement time of layer 1 is 200 ms. The vendor can decide the filtering
method.
The calculation is based on the following formula:
Fn: new measurement result after filtering
Fn＝(1-α)Fn-1+αMn Fn-1: the last measurement result after filtering

Mn: the latest measurement result from the physical layer of the UE
α＝0.5k/2, k is the parameter in filter coefficient (FilterCoef )in the measurement control message.
If k is set to 0 , layer 3 filtering is unavailable.
Page 96
 Accuracy of measurement results on the UE and handover delay should be considered in setting the value of k.
 Usually, at a low speed, k should be larger so that more inaccurate measurement results caused by quick fading can be filtered; at
a high speed, k should be smaller to reduce handover delay and to improve handover success rate.
Decrease effect of noise

Filter response
0
Filter response
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
t
-2
-4
Delay to fast changes

Value, dB
Measured
-6 Filtered, k = 3
Filtered, k = 5 Filter response
-8
0
9
11
13
15
17
19
21
23
25
27
29
31
33
t
-10
-5
-12
Measurements -10
Value, dB
Measured
Filtered, k = 3
Filtered, k = 5
-15
-20
-25
Measurements
Page 97
Intra-Frequency Handover Measurement Events
In intra-frequency handover, the UE reports measurement results to the RNC through event reporting.
When the measurement reporting criteria are met, the UE reports the results to the RNC as an event.
Event report
 The measurement report is sent when the reporting criteria are met.
Periodic report (event to periodic report)
 If RNC does not respond after several events are reported, the UE
switches to periodic report.The interval between report and total
report times are controlled through parameters.
Event Description
A primary CPICH enters the reporting range. This indicates that the quality of a cell is close to the quality
1A of the best cell in the active set. A relatively high combined gain can be achieved when the cell is added to
the active set.
A primary CPICH leaves the reporting range. This indicates that a cell has a lower quality than the best
1B
cell in the active set. The cell has to be deleted from the active set.
A non-active primary CPICH becomes better than an active primary CPICH. This indicates that the quality
1C of a cell is better than the quality of the worst cell in the active set. The RNC replaces a cell in the active
set with a cell in the monitored set.
1D The best cell changes.

The measurement results of the cells in the active set are lower than the absolute threshold.
1F
(Event 1F is used to trigger emergency blind handover )
Page 98
Triggering of Event 1A
Event 1A is triggered under the following condition:
MNew is the measurement value of the cell in the reporting range.

CIONew is equal to the sum of CIO and CIOOffset, which adjusts the cell boundary in the handover algorithms. This parameter is determined by
network planning according to actual environment configuration. To facilitate handover in neighboring cell configuration, the parameter is set as a positive
value; otherwise, the parameter is set as a negative value.
W represents weighted factor, which is determined by the parameter Weight. The total quality of the best cell and the active set is specified by W.
Mi is the measurement value of a cell in the active set.
NA is the number of cells not forbidden to affect the reporting range in the active set. The parameter CellsForbidden1A indicates whether adding the
cell to the active set affects the relative threshold of event 1A.
MBest is the measurement value of the best cell in the active set.
R1a is the reporting range or the relative threshold of soft handover. The threshold parameters of the CS non-VP service, VP service, and PS service are
as follows:
− IntraRelThdFor1ACSVP
− IntraRelThdFor1ACSNVP
− IntraRelThdFor1APS
 For the PS and CS combined services, the threshold for CS services is used.
 For the single signaling connection of the UE, the threshold for CS services is used.
 H1a represents HystFor1A the hysteresis value of event 1A
Page 99
If the signal quality of a cell that is in the monitor set is higher than Th1A for a period of time specified by TrigTime1A (that is, Time to trigger in Figure
below), the UE reports event 1A.
Th1A = (CPICH Ec/No of the best cell in the active set) - (reporting range for event 1A)
 If Weight > 0, then Th1A = (general signal quality of all the cells in the active set) - (reporting range for event 1A).
 Reporting range for event 1A is equal to the value of IntraRelThdFor1ACSVP, IntraRelThdFor1ACSNVP, or IntraRelThdFor1APS.
The report mode of 1A is event trigger report.
Generally the event 1A is reported only once. However, to avoid

measurement report loss, the event 1A reporting can be turned to
periodical reporting which is controlled by the parameters,
ReportIntervalfor1A ,PeriodMRReportNumfor1A
Page 100
Triggering of Event 1B
Event 1B is triggered under the following condition:
MOld is the measurement value of the cell that becomes worse.

CIOOld is equal to the sum of CIO and CIOOffset, which is the offset between the cell in the reporting range and the best cell in the active set.
W represents weighted factor, used to weight the quality of the active set. The total quality of the best cell and the active set is specified by the
parameter Weight.
Mi is the measurement value of a cell in the active set.
NB is the number of cells not forbidden to affect the reporting range in the active set. The parameter CellsForbidden1B indicates whether adding the
cell to the active set affects the relative threshold of event 1B.
R1b is the reporting range or the relative threshold of soft handover. The threshold parameters of the CS non-VP service, VP service, and PS services
are as follows:
− IntraRelThdFor1BCSVP
− IntraRelThdFor1BCSNVP
− IntraRelThdFor1BPS
For the PS and CS combined services, the threshold for CS services is used.
If the UE currently has only signaling connections, the threshold for CS services is used.
H1b is the hysteresis value of event 1B, which is determined by the parameter Hystfor1B.
Configuration rule and restriction

The value of IntraRelThdFor1BCSNVP has to be larger than that of IntraRelThdFor1ACSNVP.
The value of IntraRelThdFor1BCSVP has to be larger than that of IntraRelThdFor1ACSVP.
The value of IntraRelThdFor1BPS has to be larger than that of IntraRelThdFor1APS.
Page 101
Th1B = (CPICH Ec/No of the best cell in the active set) - (reporting range for event 1B)
 Reporting range for event 1B is equal to the value of IntraRelThdFor1BCSVP, IntraRelThdFor1BCSNVP, or IntraRelThdFor1BPS.
 If Weight > 0, then Th1B = (general signal quality of all the cells in the active set) - (reporting range for event 1B).
If the signal quality of a cell in the active set is lower than Th1B for a period of time specified by TrigTime1B (Time to trigger in the figure), the UE
reports event 1B.
Page 102
Triggering of Event 1C
 MNew is the measurement value of the cell in the reporting range.

CIONew is the cell individual offset value of the cell in the reporting range. It is equal to the sum of CIO and CIOOffset, which is the offset between the
cell in the reporting range and the best cell in the active set.
MInAS is the measurement value of the worst cell in the active set.
H1c is the hysteresis value of event 1C, which is determined by the parameter Hystfor1C.
Th1C = (CPICH Ec/No of the worst cell in the active set) + (hysteresis/2)
Where,
 Hysteresis is equal to the value of Hystfor1C.
If the signal quality of a cell not in the active set is higher than Th1C
for a period of time specified by TrigTime1C (Time to trigger in
the figure), the UE reports event 1C, as shown in the figure.
The UE reports event 1C for qualified cells after the number of cells
in the active set reaches the maximum value. The maximum
number of cells in the active set can be set by the
MaxCellInActiveSet parameter.
The report mode of 1C is event trigger report.
Generally the event 1C is reported only once. However, to avoid

measurement report loss, the event 1C reporting can be turned to
periodical reporting which is controlled by the parameters,
ReportIntervalfor1C ,PeriodMRReportNumfor1C
Page 103
Triggering of Event 1D
 MNotBest is the measurement value of a cell that is not the best cell.
CIONotBest is equal to the sum of CIO and CIOOffset, which is the offset between the cell in the reporting range and the best cell in the active set.
CIOBest is the cell individual offset value of the best cell. This parameter is not used for event 1D.
H1d is the hysteresis value of event 1D, which is determined by the parameter Hystfor1D.
If the signal quality of a cell not in the active set is higher than
Th1D for a period of time specified by TrigTime1D (Time to
trigger in the figure), the UE reports event 1D.
Page 104
Neighboring cell combination
After the active set is updated, the RNC updates the neighboring cell list by using the neighboring cell combination algorithm
according to the status of the active set. This list includes the new intra-frequency, inter-frequency, and inter-RAT neighboring cells.
The combination methods of intra-frequency handover, inter-frequency handover, and inter-RAT handover are the same.
Neighboring cell combination switch:HO_MC_NCELL_COMBINE_SWITCH is the neighboring cell combination switch.

If the switch is set to ON, measurement objects are chosen from the neighboring cells of all the cells in the active set.
If the switch is set to OFF, measurement objects are chosen from the neighboring cells of the best cell.
Neighboring cell combination procedure:

After obtaining the intra-frequency neighboring cells of each cell in the active set, the RNC calculates the combined neighboring cell set of the intra-
frequency cells, which is referred as Sall, This method can also be used to generate the Sall of inter-frequency or inter-RAT cells.
The intra-frequency, inter-frequency, and inter-RAT neighboring cells of each cell

in the current active set are obtained.
RNC sort the cells in the active set in descending order of CPICH Ec/No
according to the latest measurement report (event 1A, 1B, 1C, or 1D) from the UE.
The best cell is based on event 1D, whereas other cells are based on the latest
measurement report.
The cells in the active set are added to Sall.
The neighboring cells of the best cell in the active set are added to Sall.
NprioFlag (the flag of the priority) and Nprio (the priority of the neighboring cell),
which are set for each neighboring cell, are used to change the order of adding the
neighboring cells to Sall.
The neighboring cells of other cells in the active set are added to Sall in
descending order by CPICH Ec/No values of these cells in the active set. The
neighboring cells of the same cell in the active set are added according to Nprio Sall
and the number of repeated neighboring cell is recorded.
If there are more than 32 intra-frequency neighboring cells in Sall, delete the
repeated neighboring cells whose number in Sall is less. The top 32 neighboring
cells are grouped into the final Sall. The limit of 32 neighboring cells is the same
for the inter-frequency neighboring cells. Page 105
Intra-Frequency Handover Decision and Execution
The intra-frequency handover decision and execution procedure depends on the different measurement events that the RNC receives.
When receiving an event 1A, 1C, or 1D report, the RNC adds a target cell to the active set only when the CPICH Ec/No of the target cell is
higher than the absolute threshold SHOQualmin.
Event Decision and Execution
When receiving an event 1A report, the RNC decides whether to add a cell.
For event 1A, the UE can report more than one cell in the event list in one measurement report. These cells are in the list of the
MEASUREMENT CONTROL message, and they are sorted in descending order of measurement quantity.
1A
For the cells in the list, the RNC adds the radio link to the active set only if the number of cells in the active set does not reach the
maximum value. This operation is not required if the number of cells in the active set reaches a specified value.
When receiving an event 1B report, the RNC decides whether to delete a cell.
1B For event 1B, if there is more than one radio link in the active set, the RNC decides whether to delete a radio link. This operation is
not required if there is only one radio link in the active set.
When receiving an event 1C report, the RNC decides whether to change the worst cell.
For event 1C, the UE reports a list that contains good cells and the cells to be replaced, and sequences the cells in descending order
1C
by measurement quantity. After receiving the list from the UE, the RNC replaces the bad cells in the active set with the good cells in
the list.
An event 1D report includes information about only one cell. This cell can be listed in an active set or a monitored set. RNC learns
that the quality of this cell is better than that of the serving cell and takes one of the following actions:
- If the reported cell is in the active set, the RNC decides whether to change the best cell or reconfigure measurement control.
- If the reported cell is in the monitored set, then:
- If the number of cells in the active set has not reached the maximum value, the RNC adds the cell to the active set.
1D
- If the number of cells in the active set has reached the maximum value, the RNC replaces the worst cell in the active set with the
reported cell.
-The best cell is changed to the reported cell.
RNC determines whether the intra-frequency hard handover scenarios are applicable. If any scenario is applicable, the RNC performs
an intra-frequency hard handover.
Page 106
Signaling Procedures for Intra-Frequency Handover
Intra-NodeB Intra-Frequency Soft Handover Signaling Procedure (Softer Handover)
The connections involved in the intra-NodeB intra-frequency softer

handover change are as follows:
Before the softer handover, only cell 1 is connected to the UE.

During the softer handover, both cell 1 and cell 2 are connected to
the UE.
Before the softer handover, only cell 2 is connected to the UE.

Cell 1 is removed from the active set.
Page 107
Intra-RNC Inter-NodeB Intra-Frequency Soft Handover Signaling Procedure
Before the soft handover, only NodeB 1 is connected to the UE.

During the soft handover, both NodeBs are connected to the UE.
After the soft handover, only NodeB 2 is connected to the UE.

The active set of NodeB 1 is removed.
Page 108
Inter-RNC Intra-Frequency Soft Handover Signaling Procedure
Before the soft handover, the UE is connected to NodeB 1 and

NodeB 2.
After the SRNC makes a soft handover decision, it sets up a

connection between NodeB 3 under another RNC and the UE, and
releases the connection between NodeB 1 and the UE.
Page 109
Intra-RNC Inter-NodeB Intra-Frequency Hard Handover Signaling Procedure
Page 110
Inter-RNC Intra-Frequency Hard Handover Signaling Procedure
Page 111
Intra-Frequency Handover Parameters

Range Value
HO_INTRA_FREQ_HARD_HO_SWITCH: When the switch is

on, the RNC is allowed to initiate the intra-frequency hard
handover.
HO_INTRA_FREQ_SOFT_HO_SWITCH: When the switch is

on, the cells on the RNC can active the soft handover. When
HO_INTRA_FREQ_HARD_HO_ HO_INTRA_FREQ_HARD_HO_ the RNC receives reports on the events 1A, 1B, 1C, or 1D,
SWITCH SWITCH associated addition, removal, and replacement of handover cell
of the soft handover are initiated.
HO_INTRA_FREQ_SOFT_HO_ HO_INTRA_FREQ_SOFT_HO_
SWITCH SWITCH HO_MC_NCELL_COMBINE_SWITCH: When the switch is on,
the neighboring cell combined algorithm is used during the
UCORRMALGO
HO_MC_NCELL_COMBINE_ HO_MC_NCELL_COMBINE_ delivery of the objects to be measured. When the switch is not OFF,ON OFF,ON None ON
SWITCH
SWITCH SWITCH on, the optimal cell algorithm is used.
HO_MC_SIGNAL_IUR_INTRA_ HO_MC_SIGNAL_IUR_INTRA_ HO_MC_SIGNAL_IUR_INTRA_SWITCH: When the switch is

SWITCH SWITCH on, intra-frequency handover is allowed over the Iur interface if
the UE has only signaling.
HO_MC_SIGNAL_SWITCH: HO_MC_SIGNAL_SWITCH:
HO_MC_SIGNAL_SWITCH: When the switch is on, quality
measurement on the active set is delivered after signaling setup
but before service setup. If the UE is at the cell verge or
receives weak signals after accessing the network, the RNC
can trigger inter-frequency or inter-RAT handover when the UE
sets up the RRC.
There are two combination methods for uplink combination of

soft handover: one is maximum ratio combination at the NodeB
Rake receiver, which gives the highest combination gain; the
other is selective combination at the RNC, which gives a
MAY(NodeB chooses
relatively smaller combination gain.
whether combine at
NodeB),
Softer Handover Combination The default value of the indication switch is MAY, which means MAY, MUST,
UHOCOMM DivCtrlField MUST(Must combine None MAY
Indication Switch the NodeB decides whether to implement maximum ratio MUST_NOT
at NodeB),
combination according to its own physical conditions; when
MUST_NOT(Must
MUST is selected, the NodeB is forced to carry out maximum
combine at RNC)
ratio combination which is usually used in tests; when
MUST_NOT is selected, the NodeB is forbidden to carry out
maximum ratio combination.
Page 112
Baseline
Value
Indicating whether to trigger soft handover cross the Iur interface

between the RNC and the neighboring RNC:
1) CS_SHO_SWITCH. Indicating whether to trigger soft CS_SHO_SWITCH, CS_SHO_SWITCH,
handover for CS cross the Iur interface. HSPA_SHO_SWITCH, HSPA_SHO_SWITCH,
UNRNC SHOTRIG SHO cross IUR trigger None None
2) HSPA_SHO_SWITCH. Indicating whether to trigger soft NON_HSPA_SHO_ NON_HSPA_SHO_
handover for HSPA cross the Iur interface. SWITCH SWITCH
3) NON_HSPA_SHO_SWITCH. Indicating whether to trigger soft
handover for PS(R99) cross the Iur interface.
Indicating whether to trigger hard handover cross the Iur

UNRNC HHOTRIG HHO cross IUR trigger OFF, ON OFF, ON None None
interface between the RNC and the neighboring RNC.
Quantity of the triggered measurements for intra-frequency
handovers. This parameter specifies the reference measurement
quantity for the UE to report event 1x.
For events 1A, 1B, 1C, and 1D which adopt relative thresholds,
the measurement quantity does not have great impact on the
threshold values.
For event 1F which adopts absolute threshold, the measurement
Intra-frequency quantity should be set based on the reality of the cell. CPICH_EC/NO, CPICH_EC/NO,
UINTRAFREQHO IntraFreqMeasQuantity None CPICH_EC/NO
Measurement Quantity CPICH_RSCP CPICH_RSCP
If the cell is located in the center of the coverage area, the main
factor that affects signal quality is intra-frequency interference. In
this case, you are advised to set this parameter to
CPICH_EC/NO. If the cell is located at the border of the
coverage area, the main factor that affects signal quality is the
distance between the cell and the base station. In this case, you
are advised to set this parameter to CPICH_RSCP.
According to 3GPP TS 25.331, the filtered measurement value

is calculated using the following formula, where k represents this
parameter:
Fn = (1 - a) * Fn - 1 + a * Mn D0, D1, D2, D3, D4, D5,

Intra-frequency L3 Filter 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,
UINTRAFREQHO FilterCoef a = (1/2)(k/2) D6, D7, D8, D9, D11, None D3
Coefficient 11, 13, 15, 17, 19
Here, D13, D15, D17, D19
Fn: filtered measurement result
Fn - 1: previously filtered measurement result
Mn: latest received measurement result from physical layer
When a is set to 1, L3 filtering is not applied.
Page 113
Range Value
Maximum number of reporting event 1A after the reporting mode is

Event 1A to Periodical D1, D2, D4, D8, D16, 1, 2, 4, 8, 16, 32,
UINTRAFREQHO PeriodMRReportNumfor1A changed to periodical reporting. When the number of reporting event 1A None D16
Report Number D32, D64, INFINITY 64, INFINITY
exceeds this parameter, the periodical reporting is disabled.
Interval at which event 1A is reported after the reporting mode is

changed to periodical reporting. Generally, event 1A is reported only
once. To avoid the loss of measurement reports, the UE sets the mode NON_PERIODIC_
NON_PERIODIC_
of reporting event 1A to periodical if the cell reporting event 1A is not REPORT, 250,
Event 1A to Periodical REPORT, D250, D500,
UINTRAFREQHO ReportIntervalfor1A added to the active set in a specified period of time. Event 1A is 500, 1000, 2000, ms D4000
Report Period D1000, D2000, D4000,
reported for "PeriodMRReportNumfor1A" times at the interval specified 4000, 8000,
D8000, D16000
by this parameter. If the mode of reporting event 1A is set to 16000
"NON_PERIODIC_REPORT", the UE cannot change the event 1A
reporting mode to periodical.
Maximum number of reporting event 1C after the reporting mode is

Event 1C to Periodical D1, D2, D4, D8, D16, 1, 2, 4, 8, 16, 32,
UINTRAFREQHO PeriodMRReportNumfor1C changed to periodical reporting. When the number of reporting event 1C None D16
Report Number D32, D64, INFINITY 64, INFINITY
exceeds this parameter, the periodical reporting is disabled.
Interval at which event 1A is reported after the reporting mode is

changed to periodical reporting. Generally, event 1C is reported only
once. To avoid the loss of measurement reports, the UE set the mode of NON_PERIODIC_
NON_PERIODIC_
reporting event 1C to periodical if the cell reporting event 1C is not REPORT, 250,
Event 1C to Periodical REPORT, D250, D500,
UINTRAFREQHO ReportIntervalfor1C added to the active set in a specified period of time. Event 1C is 500, 1000, 2000, ms D4000
Report Period D1000, D2000, D4000,
reported for "PeriodMRReportNumfor1C" times at the interval specified 4000, 8000,
D8000, D16000
by this parameter.If the mode of reporting event 1C is set to 16000
"NON_PERIODIC_REPORT", the UE cannot change the event 1C
reporting mode to periodical.
VP Service Event 1A
UINTRAFREQHO IntraRelThdFor1ACSVP 0~29 0~14.5 0.5dB 6
Relative Threshold Relative threshold for event 1A decision when VP service is performed.
If this parameter is set to a greater value, the probability of triggering
event 1A increases. If this parameter is set to a smaller value, the
probability of triggering event 1A reduces.
CS Non-VP Service If this parameter is set to a greater value, the probability of adding a cell
UINTRAFREQHO IntraRelThdFor1ACSNVP 0~29 0~14.5 0.5dB 6
Event 1A Relative THD to the active set increases. In this case, more UEs are in soft handover
status; however, more forward resources are occupied.If this parameter
is set to a smaller value, the probability of adding a cell to the active set
reduces. Under this situation, the communication quality cannot be
PS Service Event 1A guaranteed, and smooth handover may be affected.
UINTRAFREQHO IntraRelThdFor1APS 0~29 0~14.5 0.5dB 6
Relative Threshold
Page 114

Range Range Value
Relative threshold for event 1B decision when VP service is

VP Service Event 1B Relative
UINTRAFREQHO IntraRelThdFor1BCSVP performed. If this parameter is set to a smaller value, the 0~29 0~14.5 0.5dB 12
Threshold
probability of triggering event 1B increases. If this parameter is set
to a greater value, the probability of triggering event 1B reduces.
CS Non-VP Service Event 1B If this parameter is set to a greater value, the probability of adding
UINTRAFREQHO IntraRelThdFor1BCSNVP 0~29 0~14.5 0.5dB 12
Relative Threshold a cell to the active set increases. In this case, more UEs are in soft
handover status; however, more forward resources are occupied. If
this parameter is set to a smaller value, the probability of adding a
cell to the active set reduces. Under this situation, the
PS Service Event 1B Relative
UINTRAFREQHO IntraRelThdFor1BPS communication quality cannot be guaranteed, and smooth 0~29 0~14.5 0.5dB 12
Threshold
handover may be affected.
Ec/No absolute threshold for event 1F in the SHO algorithm. This

parameter must be set to the value that guarantees the quality of
basic services. In addition, the value of this parameter affects event
1F triggering.
UINTRAFREQHO IntraAblThdFor1FEcNo Event 1F Absolute Ec/No Threshold -24~0 -24~0 dB -24
Event 1F refers to the event reported when an urgent blind
handover is triggered. If event 1F is reported in a cell belonging to
the active set, the signal quality of the active set is poor. Under this
situation, blind handover is triggered to prevent call drops.
RSCP absolute threshold for event 1F in the SHO algorithm. This

parameter must be set to the value that guarantees the quality of
basic services. In addition, the value of this parameter affects event
1F triggering.
UINTRAFREQHO IntraAblThdFor1FRSCP Event 1F Absolute RSCP Threshold -115~-25 -115~-25 dBm -115
Event 1F refers to the event reported when an urgent blind
handover is triggered. If event 1F is reported in a cell belonging to
the active set, the signal quality of the active set is poor. Under this
situation, blind handover is triggered to prevent call drops.
This parameter specifies the hysteretic value for event 1A. The
value of this parameter is associated with the slow fading. If this
parameter is set to a greater value, the number of ping-pong
handovers and the number of incorrect handovers reduce;
however, the handover may not be triggered in time. To ensure
UINTRAFREQHO HystFor1A 1A Hysteresis that the handover is triggered in time, the hysteresis for event 1A 0~15 0~7.5 0.5dB 0
should be a little smaller than the hysteresises used for events 1B,
1F, 1C, and 1D. If the difference between the hysteresises is too
great, the ratio of soft handovers (SHOs) may be affected. In
addition, filter coefficient and interval time must be considered in
setting this parameter.
Page 115
Range Value
This parameter specifies the hysteretic value for event 1B. The value of
UINTRAFREQHO HystFor1B 1B Hysteresis this parameter is associated with the slow fading, and it can be used to 0~15 0~7.5 0.5dB 0
reduce ping-pong handovers and incorrect handovers.
This parameter specifies the hysteretic value for event 1C. The value of
UINTRAFREQHO HystFor1C 1C Hysteresis this parameter is associated with the slow fading, and it can be used to 0~15 0~7.5 0.5dB 8
This parameter specifies the hysteretic value for event 1D. The value of
UINTRAFREQHO HystFor1D 1D Hysteresis this parameter is associated with the slow fading, and it can be used to 0~15 0~7.5 0.5dB 8
This parameter specifies the hysteretic value for event 1F. The value of
UINTRAFREQHO HystFor1F 1F Hysteresis this parameter is associated with the slow fading, and it can be used to 0~15 0~7.5 0.5dB 8
Used for calculating the relative threshold of the soft handover based on
the measurement report of each cell in the active set. If this parameter is
UINTRAFREQHO Weight Weighted factor set to a greater value, the relative threshold of the soft handover is greater 0~20 0~2, step:0.1 None 0
with the other conditions unchanged. When this parameter is set to 0, the
general quality of the active set is the quality of the best cell.
This parameter specifies the interval time between detection of event 1A
D0, D10, D20, D40, 0, 10, 20, 40, 60,
and sending of the measurement report. The value of this parameter is
D60, D80, D100, D120, 80, 100, 120, 160,
associated with slow fading. If this parameter is set to a greater value, the
UINTRAFREQHO TrigTime1A Event 1A Triggering Delay D160, D200, D240, 200, 240, 320, ms D320
probability of incorrect decision becomes low; however, the handover
D320, D640, D1280, 640, 1280, 2560,
algorithm becomes slow in responding to signal change.
D2560, D5000 5000
This parameter specifies the interval time between detection of event 1B
D0, D10, D20, D40, 0, 10, 20, 40, 60,
D60, D80, D100, D120, 80, 100, 120, 160,
UINTRAFREQHO TrigTime1B Event 1B Triggering Delay D160, D200, D240, 200, 240, 320, ms D640
D320, D640, D1280, 640, 1280, 2560,
D2560, D5000 5000
This parameter specifies the interval time between detection of event 1C
D0, D10, D20, D40, 0, 10, 20, 40, 60,
D60, D80, D100, D120, 80, 100, 120, 160,
UINTRAFREQHO TrigTime1C Event 1C Triggering Delay D160, D200, D240, 200, 240, 320, ms D640
D320, D640, D1280, 640, 1280, 2560,
D2560, D5000 5000
This parameter specifies the interval time between detection of event 1D
D0, D10, D20, D40, 0, 10, 20, 40, 60,
D60, D80, D100, D120, 80, 100, 120, 160,
UINTRAFREQHO TrigTime1D Event 1D Triggering Delay D160, D200, D240, 200, 240, 320, ms D640
D320, D640, D1280, 640, 1280, 2560,
D2560, D5000 5000 Page 116
Range Value
0, 10, 20, 40,
This parameter specifies the interval time between detection of event 1F and D0, D10, D20, D40,
60, 80, 100,
sending of the measurement report. The value of this parameter is associated D60, D80, D100, D120,
Event 1F Triggering 120, 160, 200,
UINTRAFREQHO TrigTime1F with slow fading. If this parameter is set to a greater value, the probability of D160, D200, D240, ms D640
Delay 240, 320, 640,
incorrect decision becomes low; however, the handover algorithm becomes D320, D640, D1280,
1280, 2560,
slow in responding to signal change. D2560, D5000
5000
If events 1A, 1C, and 1D are reported to the RNC, the target cell can be added
to the active set only when CPICH Ec/No is greater than this parameter.
This parameter is used to set a baseline for the quality of the radio links to be
added. The current events of soft handovers are defined on the basis of
relative thresholds and have no requirement for the absolute quality. If the
signal quality of added cells is too poor, the general quality of the active set
cannot be improved. In this case, however, more downlink resources are
Min Quality THD for occupied and more TPC errors are caused, and the soft handover
UINTRAFREQHO SHOQualmin -24~0 -24~0 dB -24
SHO performance is worsened. Therefore, this parameter is required.
This parameter should be adjusted, based on the planned Ec/No that the cell
soft handover area is expected to reach. If this parameter is set to a greater
value, the probability for adding a neighboring cell to the active set decreases.
In this case, the service quality of the cell to be added is guaranteed. If this
parameter is set to a smaller value, the probability for adding a neighboring
cell to the active set increases. In this case, however, the service quality of the
cell to be added is not guaranteed.
Maximum number of cells in an active set. This parameter is used to achieve

the balance between the signal quality in the SHO area and the system
performance. If more cells are in the active set, a user can obtain a greater
macro diversity gain; the intra-frequency interference is smaller. In this case,
however, the user occupies resources of several cells. This increases the
Max Number of Cell in
UINTRAFREQHO MaxCellInActiveSet amount of data to be processed and lowers the system performance. To 1~6 1~6 None 3
Active Set
reduce intra-frequency interference, it is recommended that the number of
cells in the active sets should be increased in the hot-spot and micro cells,
where the signals are complicated. For the cell that provides high speed
services, the decrease in cells in the active set can reduce the consumption of
system resources.
This parameter works with the offset of neighboring cell-oriented Cell

Cell Oriented Cell
UCELL CIO Individual Offset (CIO). It is used for intra- or inter-frequency handover -20~20 -10~10 0.5dB 0
Individual Offset
decisions. For details, refer to 3GPP TS 25.331.
Page 117

Range Value
Neighboring cell oriented CIO.
This parameter specifies the neighboring cell oriented Cell Individual
Neighboring Cell Oriented Offset (CIO). The sum of the value of this parameter, the value of the cell
UINTRAFREQNCELL CIOOffset -20~20 -10~10 0.5dB 0
CIO oriented CIO, and the actual measurement quantity is used for the event
evaluation of the UE or used for the handover decision procedure on the
RNC side.
Determines whether event 1A threshold is affected when the cell is added

to the active set. If this parameter is set to "NOT_AFFECT", the relative
threshold of event 1A is not affected when the cell is added to the active
set. That is, the cell signals are not involved when the UE evaluates AFFECT, AFFECT,
UINTRAFREQNCELL CellsForbidden1A Affect 1A Threshold Flag None AFFECT
whether event 1A should be triggered. If this parameter is set to NOT_AFFECT NOT_AFFECT
"AFFECT", the relative threshold of event 1A is affected when the cell is
added to the active set. That is, the cell signals are involved when the UE
evaluates whether event 1A should be triggered.
Flag of whether adding a cell into the active set will affect the relative
threshold of the event 1B.
This parameter is a flag of whether adding the cell into the active set will
affect the relative threshold of the event 1B NOT_AFFECT: Adding the
AFFECT, AFFECT,
UINTRAFREQNCELL CellsForbidden1B Affect 1B Threshold Flag cell into the active set does not affect the relative threshold of the event None AFFECT
NOT_AFFECT NOT_AFFECT
1B. The cell signal will affect the UE evaluate whether event 1B should
occur. AFFECT: Adding the cell into the active set affects the relative
threshold of the event 1B. The cell signal will not affect the UE evaluate
whether event 1B should occur.
Priority flag of neighboring cells

The value TRUE indicates that the neighboring cell priority is valid, and
Neighboring Cell Priority
UINTRAFREQNCELL NPrioFlag the value FALSE indicates that the neighboring cell priority is invalid. In FALSE, TRUE FALSE, TRUE None FALSE
Flag
the algorithm of neighboring cell combination, the cell with an invalid
priority is the last one to be considered as the measurement object.
Priority of neighboring cells. The priority assigned to a neighboring cell is

valid only when this parameter is set to "TRUE". The neighboring cell with
Priority of Neighboring a lower priority is more possibly delivered as the measurement object. For
UINTRAFREQNCELL NPrio 0~30 0~30 None None
Cell example, the neighboring cell with priority 1 is more possible to be
selected as the measurement object than the neighboring cell with priority
2.
Page 118
Inter-Frequency and Inter-RAT Handover
Based on the handover triggering causes, the inter-frequency and Inter-RAT handover falls into five categories, as described below.
Inter-RAT handover refers to the handover performed between 3G network and 2G network.
This section mainly describes the 3G-to 2G coverage handover and QoS handover. Except the cause of the triggering is different, the handover
procedure is the same for the coverage handover and QoS handover.
Type Description
Coverage handover involves the following features:
Coverage -WRFD-020302 Inter Frequency Hard Handover Based on Coverage

handover -WRFD-020303 Inter-RAT Handover Based on Coverage
If a moving UE leaves the coverage of the current frequency, the RNC needs to trigger the coverage-based inter-frequency or inter-
RAT handover to avoid call drops.
QoS handover involves the following features:
-WRFD-020304 Inter Frequency Hard Handover Based on DL QoS

QoS handover -WRFD-020309 Inter-RAT Handover Based on DL QoS
If the link quality becomes worse, the Link Stability Control Algorithm may trigger the QoS-based inter-frequency or inter-RAT
handover to avoid call drops.
When the LDR function detects the basic congestion, the RNC chooses some UEs and performs the inter-frequency or inter-RAT
LDR handover
handover according to user priorities and service priorities to balance the load between inter-frequency or inter-RAT cells.
Service handover involves the feature WRFD-020305 Inter-RAT Handover Based on Service
Service handover
Based on layered services, the traffic of different classes is handed over to different systems. For example, when an Adaptive Multi
Rate (AMR) speech service is requested, this service can be handed over to the 2G network.
HCS handover involves the feature WRFD-021200 HCS.
HCS handover Inter-frequency or inter-RAT handover can be triggered by the UE speed estimation algorithm of the HCS. To reduce frequent
handovers, the UE at a higher speed is handed over to a cell under a larger coverage, whereas the UE at a lower speed is handed
over to a cell under a smaller coverage.
Page 119
Inter-frequency and Inter-RAT Handover Switches
Some switches are important for inter-frequency handover because they decide whether the handover can be performed successfully.
These switches are the parameter values of handover algorithm switches in the command SET UCORRMALGOSWITCH, as described below.
HOSWITCH > HO_INTER_FREQ_HARD_HO_SWITCH:

The switch decides whether the RNC allows inter-frequency handover.
HOSWITCH > HO_INTER_RAT_PS_OUT_SWITCH:
The switch decides whether the RNC allows inter-RAT handover of the PS domain from the UTRAN.
HOSWITCH > HO_INTER_RAT_CS_OUT_SWITCH:
The switch decides whether the RNC allows inter-RAT handover of the CS domain from the UTRAN.
HHOTRIG: The switch decides whether intra- or inter- frequency hard handover is allowed over the Iur interface.
HOSWITCH > HO_MC_SIGNAL_SWITCH:

The switch decides when the RNC performs the active set signal quality measurement before the RB setup. If the UE is at the cell verge or the signal is
weak signals after accessing the network, the RNC can trigger inter-frequency or inter-RAT handover when the UE sets up the RRC.
 If the switch is set to ON, the RNC initiates the active set quality measurement after the RRC connection setup is completed (before the RB setup).
 If the switch is set to OFF, the RNC initiates the active set quality measurement after the RB setup is completed.
 The switch is set to OFF by default.
HOSWITCH > HO_MC_MEAS_BEYOND_UE_CAP_SWITCH:

The switch decides whether the neighboring cell will be sent in the inter-frequency measurement control message when the frequency of the
neighboring cell is not included in the measurement capability of the UE. The reported measurement capability of the UE is not the same as the actual
measurement capability of the UE. Measurement capability at some frequencies may not be reported due to the limitation of the version of UE protocol.
 If the switch is set to ON, the RNC sends the inter-frequency measurement control message with the neighboring cell, whose frequency is not
included in the measurement capability of the UE.
If the switch is set to OFF, the RNC sends the inter-frequency measurement control message without the neighboring cell, whose frequency is not
included in the measurement capability of the UE.
The switch is set to OFF by default.
Page 120
Coverage or QoS Handover procedure
Coverage or QoS Handover Procedure
If a moving UE leaves the coverage area of the current frequency or the link quality degrades, the RNC triggers the coverage-based or QoS-based
inter-frequency or inter-RAT handover to avoid call drops. The QoS handover procedure is the same as the coverage handover procedure.
In the triggering phase

-If the CPICH of the current cell becomes worse, the UE reports the event
2D. Then the RNC initiates the coverage handover.
-If the link quality of the current service deteriorates, the Link Stability Control
Algorithm makes a QoS handover decision. Then RNC initiate the QoS
handover.
In the measurement phase

RNC sends an inter-frequency measurement control message to the UE,
requesting the NodeB and UE to start the compressed mode. RNC also
requests the UE to perform the inter-frequency or inter-RAT handover
measurement.
-If the CPICH of the current cell becomes better, the UE reports the event 2F.
Then the RNC stops the compressed mode and the coverage handover.
In this phase, the method of either periodical measurement report or
event-triggered measurement report can be used.
In the decision phase.

After the UE reports event 2B or 3A,RNC performs the handover. Or the UE
periodically generates measurement reports, and the RNC makes a decision
after evaluation.
In the execution phase.

RNC executes the handover procedure.
Page 121
Coverage or QoS Handover measurement
Coverage or QoS Handover Measurement
In the measurement phase of inter-frequency or inter-RAT handover, the UE takes measurement according to the MEASUREMENT CONTROL
message received from the RNC. When the measurement report conditions are met, the UE sends measurement reports to the RNC according to the
rules defined in the MEASUREMENT CONTROL message.
Coverage or QoS Handover Measurement Report Modes

The coverage-based and QoS-based handover use event-triggered or periodical measurement report mode.
InterFreqReportMode: The measurement report mode of inter-frequency handover.

InterRatReportMode: The measurement report mode of inter-RAT handover.
PrdReportInterval: The periodical measurement report interval of inter-frequency handover.
InterRATPeriodReportInterval: The periodical measurement report interval of inter-RAT handover.
The advantage of periodical measurement report is that if the handover fails, the RNC reattempts the handover to the same cell after receiving the
periodical measurement report from the UE. This increases the probability of the success of inter-frequency handover.
Based on the measurement control message received from the RNC, the UE periodically reports the measurement quality of the target cell. Then,
based on the measurement report, the RNC makes the handover decision and performs handover.
Coverage or QoS Handover Measurement Quantity

Measurement quantities vary according to the type of inter-frequency or inter-RAT handover.
The used frequency belongs to a UMTS cell and the used frequency measurement quantities are set as follows:
In inter-frequency or inter-RAT coverage or QoS handover, event 2B/2D/2F or periodical measurement takes both CPICH Ec/No and RSCP as
measurement quantities. In coverage-based inter-RAT handover, the system delivers both CPICH Ec/N0 and CPICH RSCP for 2D/2F measurement.
In inter-RAT coverage handover, the event 3A measurement quantity is set through the parameter MeasQuantityOf3A.
 In inter-RAT QoS handover, the event 3A measurement quantity is set through the parameter UsedFreqMeasQuantityForQos3A.
The unused frequencies belong to a GSM cell, and the unused frequency measurement quantity is referred to as to GSM RSSI.
Page 122
Coverage or QoS Handover measurement events
Coverage or QoS Handover Measurement Events
When the measurement thresholds are reached, the UE reports the events to the RNC to trigger related handover procedures.
Event Description
2D The estimated quality of the currently used frequency is below a certain threshold.
2F The estimated quality of the currently used frequency is above a certain threshold.
The estimated quality of the currently used frequency is below a certain threshold and the
2B
estimated quality of a non-used frequency is above a certain threshold.
The estimated quality of the currently used UTRAN frequency is below a certain threshold
3A
and the estimated quality of the other system is above a certain threshold.
Triggering of Event 2D
After the conditions of event 2D are fulfilled and maintained until the TimeToTrig2D is reached, the UE sends the event 2D measurement report
message.Event 2D is triggered on the basis of the following formula:
QUsed is the measured quality of the used frequency.
QUsed ≤ TUsed2d - H2d/2
TUsed2d is the absolute quality threshold of the cell that uses the current frequency.
Based on the service type and measurement quantity, TUsed2d threshold can be configured through one of the following parameters:
Inter-frequencyparameters Inter-RAT parameter
− InterFreqCSThd2DEcN0 − InterRATCSThd2DEcN0
− InterFreqR99PsThd2DEcN0 − InterRATR99PsThd2DEcN0
− InterFreqHThd2DEcN0 − InterRATHThd2DEcN0
− InterFreqCSThd2DRSCP − InterRATCSThd2DRSCP
− InterFreqR99PsThd2DRSCP − InterRATR99PsThd2DRSCP
− InterFreqHThd2DRSCP − InterRATHThd2DRSCP
− For the PS and CS combined services, the threshold is set to the higher one of CS or PS services.
− If the UE has only signaling connections currently, the thresholds for CS services are used.
− H2d is the event 2D hysteresis value set through the parameter HystFor2D.
Page 123
Triggering of Event 2F
After the conditions of event 2F are fulfilled and maintained until the parameter TimeToTrig2F is reached, the UE reports the event 2F measurement
report message.Event 2F is triggered on the basis of the following formula:
QUsed ≥ TUsed2f + H2f/2 QUsed is the measured quality of the used frequency.
TUsed2f is the absolute quality threshold of the cell that uses the current frequency.
Based on the service type and measurement quantity, TUsed2f threshold can be configured through one of the following parameters:
Inter-frequency parameters
− InterFreqCSThd2FEcN0
− InterFreqCSThd2FRSCP
− InterFreqR99PsThd2FEcN0
− InterFreqR99PsThd2FRSCP
− InterFreqHThd2FEcN0
− InterFreqHThd2FRSCP
Inter-RAT parameters
− InterRATCSThd2FEcN0
− InterRATR99PsThd2FEcN0
− InterRATHThd2FEcN0
− InterRATCSThd2FRSCP
− InterRATR99PsThd2FRSCP
− InterRATHThd2FRSCP
− For the PS and CS combined services, the threshold is set to the higher one of CS or PS services.
− If the UE has only signaling connections currently, the thresholds for CS services are used.
− H2f is the event 2F hysteresis value set through the parameter HystFor2F.
Page 124
After the conditions of event 2B are fulfilled and maintained until the parameter TimeToTrig2B is reached, the UE reports the
event 2B measurement report message.Event 2B is triggered on the basis of the following formula:
QNoused ≥ TNoused2b + H2b/2  QNoused is the measured quality of the cell that uses the other frequencies.
 QUsed is the measured quality of the used frequency.
QUsed ≤ TUsed2b - H2b/2
 H2b is the event 2B hysteresis value set through the parameter HystFor2B.
 TNoused2b is the absolute quality threshold of the cell that uses the other frequencies.
Based on the service type and measurement quantity, TNoused2b threshold can be configured through one of the following parameters:
− TargetFreqCsThdEcN0
− TargetFreqCsThdRscp
− TargetFreqR99PsThdEcN0
− TargetFreqR99PsThdRscp
− TargetFreqHThdEcN0
− TargetFreqHThdRscp
TUsed2b is the absolute quality threshold of the cell that uses the current frequency. TUsed2b is set in the following way:
Based on the service type and measurement quantity, this threshold can be configured through one of the following parameters:
If event 2D with the CPICH RSCP value is received by the RNC:

− TUsed2b of event 2B with the CPICH RSCP value can be:
UsedFreqCSThdRSCP
UsedFreqR99PsThdRSCP
UsedFreqHThdRSCP
− TUsed2b of event 2B with the CPICH Ec/No value is configured as the maximum value 0 dB.
If event 2D with the CPICH Ec/No value is received by the RNC:

− TUsed2b of event 2B with the CPICH Ec/No value can be:
UsedFreqCSThdEcN0
UsedFreqR99PsThdEcN0
UsedFreqHThdEcN0
− TUsed2b of event 2B with the CPICH RSCP value is configured as the maximum value –25 dBm.
For the PS and CS combined services, the threshold is set to the higher one of CS or PS services.
If the UE has only signaling connections currently, the thresholds for CS services are used.
Page 125
When the conditions for event 3A are met and maintained in time-to-trigger specified by TrigTime3A the UE sends the
measurement report of event 3A.Event 3A is triggered on the basis of the following formula:
QUsed is the measurement value of the cell at the currently used frequency.
 TUsed is the absolute quality threshold of the cell that uses the current
frequency.
QUsed ≤ TUsed - H3a/2 and MOtherRAT + CIOOtherRAT ≥ TOtherRAT + H3a/2 MOtherRAT is the measurement value of the cell (in another RAT) in the
reporting range.
 CIOOtherRAT is the cell individual offset value of the cell (in another RAT) in
the reporting range that is equal to the sum of CIO and CIOOffset.
TOtherRAT is the absolute inter-RAT handover threshold.
Based on the service type and measurement quantity in the coverage-based handover, TUsed can be configured through the following parameters
− UsedFreqCSThdEcN0
− UsedFreqCSThdRSCP
− UsedFreqHThdEcN0
− UsedFreqHThdRSCP
− UsedFreqR99PsThdEcN0
− UsedFreqR99PsThdRSCP
In the Coverage-based handover, based on the measurement quantity (CPICH Ec/No or RSCP), TUsed is configured as follows:
− If the measurement quantity is CPICH Ec/No:

If 2D is triggered by RSCP, TUsed is configured as the maximum value 0 dB.
If 2D is triggered by Ec/No, TUsed is configured as the Ec/No threshold specified by the previous parameters.
− If the measurement quantity is CPICH RSCP:

If 2D is triggered by RSCP, TUsed is configured as the RSCP threshold specified by the previous parameters..
If 2D is triggered by Ec/No, TUsed is configured as the maximum value -25 dBm.
Page 126
In the uplink QoS-based handover, based on the measurement quantity (CPICH Ec/No or RSCP), TUsed is configured as the maximum value
according to 3GPP specifications, as described below:
− If the measurement quantity is CPICH Ec/No, TUsed is configured as the maximum value 0 dB.
− If the measurement quantity is CPICH RSCP, TUsed is configured as the maximum value –25 dBm.
In the downlink QoS-based handover:

− If the measurement quantity is CPICH Ec/No, TUsed is configured as the maximum value 0 dB.
− If the measurement quantity is CPICH RSCP, based on the service type, TUsed can be configured as one of the following sums:
− UsedFreqCSThdRSCP and DlRscpQosHyst

− UsedFreqR99PsThdRSCP and DlRscpQosHyst
− UsedFreqHThdRSCP and DlRscpQosHyst
Based on different service types (CS, PS domain R99 service, or PS domain HSPA service), TOtherRAT threshold can be configured through the
following parameters:
− TargetRatCsThd
− TargetRatR99PsThd
− TargetRatHThd
H3a is 3A hysteresis, the hysteresis value of event 3A.

For the PS and CS combined services, the threshold for CS service is used.
Page 127
Coverage or QoS Handover Decision and Execution
BSIC Verification Requirements for 2G Cells
During inter-RAT measurement, it is recommended that the UE report the 2G cell to the RNC after the Base Transceiver Station Identity
Code (BSIC) of the cell is verified. This greatly enhances the reliability of handover.
The parameter BSICVerify is the control switch for the BSIC verification.
Inter-Frequency Coverage or QoS Handover Decision and Execution

The coverage-based and QoS-based inter-frequency handovers are categorized into two types according to the following two measurement report
modes: periodical measurement report mode and event-triggered measurement report mode. Each mode corresponds to a different decision and
execution procedure.HOCovPrio specifies the neighboring cell priority for coverage based inter-frequency handover.
Inter-Frequency Handover in Periodical Measurement Report Mode

After receiving the periodical measurement report of the inter-frequency cell, the RNC starts the following decision procedures:
1. Decide whether both the CPICH Ec/No value and CPICH RSCP value of the pilot signal of the target cell meet the requirement of inter-
frequency handover.The evaluation formula is listed below:
Mother_Freq is the CPICH Ec/No or CPICH RSCP measurement value of the target cell reported by the
UE. Both of the two measurement values of the inter-frequency cell must satisfy the formula.
Mother_Freq + CIOother_Freq ≥ Tother_Freq + H/2 CIOother_Freq is the cell individual offset value of the target cell. It is equal to the sum of CIO and
CIOOffset.
 Tother_Freq is the decision threshold of inter-frequency hard handover.
Based on the service type and measurement quantity, Tother_Freq threshold can be configured through one of the following parameters:
TargetFreqCsThdEcN0
TargetFreqCsThdRscp
TargetFreqR99PsThdEcN0
TargetFreqR99PsThdRscp
TargetFreqHThdEcN0
TargetFreqHThdRscp
− H is the inter-frequency hard handover hysteresis value set through the parameter HystForPrdInterFreq.
Page 128
2. Start the hard handover time-to-trigger timer, which is configured through the parameter TimeToTrigForPrdInterFreq.
3. If Mother_Freq + CIOother_Freq < Tother_Freq - H/2, stop the timer.
4. Select the cells in sequence, that is, from high quality cells to low quality ones, to initiate inter-frequency handover in the cells where
the hard handover time-to-trigger timer expires.
Each cell in the measurement report shall be evaluated as mentioned previously. When the hard handover time-to-trigger timers of more
than one cell expire at the same time, the latest measurement report is used for selecting the best inter-frequency neighboring cell for
handover. For example, the cell with the highest CPICH RSCP in the latest measurement report is selected, as shown in figure below.
Inter-Frequency Handover in Event-Triggered Measurement

Report Mode
After receiving the event 2B measurement reports of CPICH RSCP

and CPICH Ec/No of the inter-frequency cell, the RNC starts the
following procedure:
1. Add all the pilot cells that trigger event 2B to a cell set and
arrange the cells according to the measurement quality of
CPICH_Ec/No in descending order.
2. Select the cells in turn from the cell set to perform inter-
frequency handover.
Page 129
3G-to-2G Coverage and QoS Handover Decision and Execution

The coverage-based and QoS-based 3G-to-2G handover is categorized into two types according to the following two measurement report modes:
periodical measurement report mode and event-triggered measurement report mode. Each mode corresponds to a different decision and execution
procedure.
3G-to-2G Coverage and QoS Handover in Periodical Report Mode

After receiving the periodical measurement report of GSM cells, the RNC performs the following decision and execution procedures:
1. Decide whether the quality of 2G cells meets the conditions of inter-RAT handover.The evaluation formula is listed below:
− Mother_RAT is the measurement result of inter-RAT handover received by the RNC.
Mother_RAT + CIOother_RAT ≥ Tother_RAT + H/2 − CIOother_RAT is the cell individual offset value of the target cell. It is equal to the sum of CIO and
CIOOffset.
− Tother_RAT is the decision threshold of inter-RAT hard handover.
Based on the service type and measurement quantity, Tother_RAT threshold can be configured through the following parameters:
TargetRatCsThd
TargetRatR99PsThd
TargetRatHThd
− H is the inter-RAT handover hysteresis value set through HystforInterRAT.
− For the PS and CS combined services, one or more handover thresholds for CS services are used.
2. Start the evaluation of the cells that meet the quality requirement and start the time-to-trigger timer. If the measurement report meet the following
formula and time-to-trigger timer does not expire, stop the time-to-trigger timer.
Mother_RAT + CIOother_RAT < Tother_RAT - H/2
The length of the time-to-trigger timer is configured through the parameter TimeToTrigForVerify (with BSIC acknowledged) or the parameter
TimeToTrigForNonVerify (with BSIC unacknowledged).
3. Select the cells in sequence, that is, from high quality cells to low quality ones, to initiate 3G-to-2G handover in the cells where the handover time-
to-trigger timer expires.
Page 130
3G-to-2G Coverage and QoS Handover Decision and Execution
3G-to-2G Coverage and QoS Handover in Event Report Mode

After receiving the event 3A measurement report of 2G cells, the RNC performs the following decision and execution procedures:
1. Put all the 2G cells that trigger event 3A into a cell set and arrange the cells according to the measurement quality in descending order.
2. Select the cells in sequence from the cell set to perform inter-RAT handover.
Rules for 3G-to-2G Coverage or QoS Handover

The rules for enabling the 3G-to-2G handover are based on the Service Cell Service Capability (Required by 2G)
Handover Indicator and the capability requirement. The rules vary according to UE Capability
Capability
the types of inter-RAT handover. EDGE GPRS GSM
EDGE Allowed Allowed Allowed
Capability Requirement GPRS Not allowed Allowed Allowed
Before deciding the 3G-to-2G handover, the RNC considers 2G cell capability, EDGE
service capability and UE capability. GSM Not allowed Not allowed Allowed
Not supported by 2G Not allowed Not allowed Not allowed
2G cell capability: 2G cell capability is configured through the parameter EDGE Not allowed Allowed Allowed
RatCellType. This parameter indicates whether the cell supports the GSM,
GPRS Not allowed Allowed Allowed
GPRS, or EDGE. GPRS
GSM Not allowed Not allowed Allowed
Service required capability: The Required 2G Capability (Req2GCap) Not supported by 2G Not allowed Not allowed Not allowed
specifies the capability of 2G cells required by inter-RAT handover. This
EDGE Not allowed Not allowed Allowed
indicates whether the service is supported by the GSM, GPRS, or EDGE.
GPRS Not allowed Not allowed Allowed
GSM
UE capability: Upon the reception of the UE capability information message, GSM Not allowed Not allowed Allowed
the RNC decides whether to start the inter-RAT measurement. The information Not supported by 2G Not allowed Not allowed Not allowed
indicates whether the UE supports the GSM, GPRS, or EDGE.
The tables describe the impacts of different types of capability on handover

decision. If the capability of all 2G neighboring cells does not meet the
requirement, the inter-RAT measurement will not be triggered.
Page 131
Service Handover Indicator
The RNC initiates the coverage- or QoS-based UMTS-to-GSM handover only when Service Handover Indicator is as follows:
 HO_TO_GSM_SHOULD_BE_PERFORM
 HO_TO_GSM_SHOULD_NOT_BE_PERFORM
The IE Service Handover Indicator indicates the CN policy for the service handover to the 2G network. This IE is indicated in the Radio Access Bearer
(RAB) assignment signaling assigned by the CN, or provided by the RNC side.
The algorithm switch HoSwitch: HO_INTER_RAT_RNC_SERVICE_HO_SWITCH decides whether the service attribute of inter-RAT handover is
based on the RNC or the CN.
 If the switch is set to ON, the service attribute of inter-RAT handover is based on the parameter configured on the RNC side.
 If the switch is set to OFF, the service attribute of inter-RAT handover is first based on the CN when the indicator is contained in the RAB assignment
signaling assigned by the CN. If the CN does not allocate a service indicator, the service attribute of inter-RAT handover is based on the RNC side.
Through the SHIND parameter, the service handover indicators are set as follows:
 HO_TO_GSM_SHOULD_BE_PERFORM: means that the handover to the 2G network is performed when 2G signals are available.
 HO_TO_GSM_SHOULD_NOT_BE_PERFORM: means that the handover to the 2G network is performed when 3G signals are weak but 2G signals
are strong.
 HO_TO_GSM_SHALL_NOT_BE_PERFORM: means that the handover to the 2G network is not performed even when 3G signals are weak but 2G
signals are strong.
By default, the RNC does as follows:

For a UE with a single signaling RAB, the RNC supports the handover to the GSM. But it is not recommended.
For the UE accessing combined services (with CS services), the RNC sets the service handover indicator of the UE to that of the CS service, because
the CS service has the highest QoS priority.
For the UE accessing combined services (with only PS services), the RNC sets the service handover indicator of the UE to that of the PS service with
the highest QoS priority
If the service handover indicators are not configured by the CN, each indictor can be set to the service parameter index of a service on the RNC.
Each service parameter index is the index of one typical service RAB, which involves a set of service type, source description, CN domain ID, and
maximum rate (bit/s).
Page 132
3G-to-2G NACC
The Network Assisted Cell Change (NACC) function can efficiently reduce the delay of UMTS-to-GSM handover.
Some services have requirements for the delay. If the handover takes too long, TCP may start slowly or data transmission of the service
stream may be interrupted due to the overflow of the UE buffer. The introduction of NACC enables the system information exchange
between different BSSs, or between BSS and RAN. The inter-system delay, especially inter-system delay in PS domains, can be reduced.
With NACC, the RNC sends the UE a cell change order, which contains the GSM EDGE Radio Access Network (GERAN) system
information, when the 3G-to-2G handover in the PS domain is triggered.
To enable the NACC function, do as follows:

Run the SET UCORRMALGOSWITCH command to set HoSwitch: HO_INTER_RAT_PS_3G2G_CELLCHG_NACC_SWITCH to ON.
Run the ADD UEXT2GCELL / MOD UEXT2GCELL command to set SuppRIMFlag to TRUE.
3G-to-2G PS Handover
PS handover is similar to the inter-RAT handover in the CS domain.
If the HoSwitch: HO_INTER_RAT_PS_3G2G_RELOCATION_SWITCH is turned on, the PS handover from the 3G network to the 2G
network is performed in the relocation process. When the switch is not on, the PS handover from the 3G network to the 2G network is
performed in the cell change order process.
To enable the PS HO function, do as follows:

Run the SET UCORRMALGOSWITCH command to set HoSwitch: HO_INTER_RAT_PS_3G2G_RELOCATION_SWITCH parameter to ON.
Run the ADD UEXT2GCELL / MOD UEXT2GCELL command to set SuppPSHOFlag to TRUE.
2G-to-3G Handover
The 2G-to-3G handover is initiated by the 2G network, where the dual-mode (GSM and WCDMA) mobile terminals are required. Both the
GSM MSC and the GSM BSS must support the GSM-to-UMTS handover.
Page 133
Interoperability between Inter-Freq & Inter-RAT HO
During the coverage-based and QoS-based 3G to 2G handover, the measurements on both inter-frequency and inter-RAT neighboring
cells can be made, which enables the cells to provide continuous coverage and high quality.
The preconditions for the measurements are as follows:

Both inter-frequency and inter-RAT neighboring cells are available.
InterFreqRATSwitch is set to SIMINTERFREQRAT.
If InterFreqRATSwitch is set to:
Inter-frequency measurement (INTERFREQ), the RNC allows the UE to perform only this type of measurement.
Inter-RAT measurement (INTERRAT), the RNC allows the UE to perform only this type of measurement.
Concurrent inter-frequency and inter-RAT measurement (SIMINTERFREQRAT), the RNC allows the UE to perform both
types of measurement at the same time.
During the concurrent inter-frequency and inter-RAT measurement, the values of the parameter CoexistMeasThdChoice for event 2D
are chosen as follows:
When the value COEXIST_MEAS_THD_CHOICE_INTERFREQ is chosen, the inter-frequency measurement threshold for
event 2D is used.
When the value COEXIST_MEAS_THD_CHOICE_INTERRAT is chosen, the inter-RAT measurement threshold for event 2D is
used.
Page 134
Coverage or QoS Handover Parameters
Range Range Value
HO_INTER_FREQ_HARD_HO_SWITCH: When the switch is on,
the RNC is allowed to initiate inter-frequency measure control or the
load-based inter-frequency hard handover upon the handover
HO_INTER_FREQ_HARD_HO HO_INTER_FREQ_HARD_HO_
decision on inter-frequency load.
_SWITCH, SWITCH,
HO_INTER_RAT_CS_OUT_SWITCH: When the switch is on, the
HO_INTER_RAT_CS_OUT_ HO_INTER_RAT_CS_OUT_
UCORRMALGOSWITCH RNC is allowed to initiate inter-frequency measure control and the OFF,ON OFF,ON None ON
SWITCH, SWITCH,
CS inter-RAT hard handover from the 3G network to the 2G network.
HO_INTER_RAT_PS_OUT_ HO_INTER_RAT_PS_OUT_
HO_INTER_RAT_PS_OUT_SWITCH: When the switch is on, the
SWITCH SWITCH
RNC is allowed to initiate inter-frequency measure control and the
PS inter-RAT hard handover from the 3G network to the 2G network.
HO_INTER_RAT_PS_3G2G_CELLCHG_NACC_SWITCH: When
the switch is on, the NACC function is supported during the PS inter-
RAT handover from the 3G network to the 2G network in the cell
change order process. When the switch is not on, the NACC function
is not supported. When PS_3G2G_RELOCATION_SWITCH is ON,
HO_INTER_RAT_PS_3G2G_CEL this switch is useless. When the NACC function is supported, the UE
HO_INTER_RAT_PS_3G2G_
LCHG_NACC_ skips the reading procedure as the SI/PSI of the target cell is
CELLCHG_NACC_SWITCH,
SWITCH, provided after the UE accesses the 2G cell. Thus, the delay of inter-
cell handover is reduced.
HO_INTER_RAT_PS_3G2G_
HO_INTER_RAT_PS_3G2G_REL
UCORRMALGOSWITCH RELOCATION_SWITCH, OFF,ON OFF,ON None OFF
OCATION_SWITCH, HO_INTER_RAT_PS_3G2G_RELOCATION_SWITCH: When the
switch is on, the PS inter-RAT handover from the 3G network to the
2G network is performed in the relocation process. When the switch
HO_MC_MEAS_BEYOND_UE
HO_MC_MEAS_BEYOND_UE_C is not on, the PS inter-RAT handover from the 3G network to the 2G
_CAP_SWITCH
AP_SWITCH network is performed in the cell change order process.
HO_MC_MEAS_BEYOND_UE_CAP_SWITCH: When the switch is

on, the neighboring cell whose frequency band is beyond the UE's
capabilities can also be delivered in the inter-frequency
measurement list.
Priority for neighboring cell supporting coverage-based inter-

frequency handover.
The value "0" indicates that the coverage-based inter-freqency
Priority of Coverage-Based
UINTERFREQNCELL HOCovPrio handover is not supported. The value "1" indicates that the cell is 0~3 0~3 None 2
Inter-Frequency Handover
assigned with the highest priority for the handover. The value "3"
indicates that the cell is assigned with the lowest priority for the
handover.
Page 135
Baseline
Value
If this parameter is set to PERIODICAL_REPORTING, measurement
reports are periodically reported. If this parameter is set to
EVENT_TRIGGER, measurement reports are reported by triggering the
event.
Event-triggered report mode
In this mode, event 2B is used to decide whether to trigger inter-
frequency handover. This prevents the ping-pong handover (The quality
of the currently used frequency is lower than the absolute threshold
"used frequency quality threshold", and the quality of the unused
frequency is higher than another absolute threshold "target frequency
trigger threshold"). Event 2B cannot change from event-triggered mode
to periodical mode. When event-triggered measurement report mode is
selected, Ec/No and RSCP are both used as the measurement quantity
for inter-frequency measurement.The advantage of event-triggered
PERIODICAL_
report mode is that the signaling transmission and processing load are PERIODICAL_
REPORTING
saved. The disadvantage of event-triggered report mode is that the REPORTING,
UINTERFREQ IInter-frequency Measure (Periodical reporting), PERIODICAL_
InterFreqReportMode event is reported only once and cannot be changed to periodical mode. None
HOCOV Report Mode REPORTING
Periodical report mode EVENT_
EVENT_TRIGGER
In this mode, event 2D/2F is used to start and stop the compressed TRIGGER
(Event trigger)
mode, and to periodically report the inter-frequency cell measurement
result in compressed mode. When the cell quality reported by the UE is
higher than the absolute threshold plus hysteresis, the triggering delay
timer is started. If the conditions are always met before the timer
expires, the inter-frequency handover is started after the timer expires.
If the handover fails, the handover decision is performed, according to
the periodical inter-frequency measurement report.
The advantage of the periodical measurement report mode is that it can
repeatedly perform direct retry on the same cell when the handover
fails, and that the following algorithms can be flexibly developed. The
disadvantage of the periodical measurement report mode is that it
requires large amount of signaling and increases the load on the air
interface and for signaling processing.
This parameter specifies the Layer 3 filter coefficient for the inter-
frequency measurement This parameter has the same physical
D0, D1, D2, D3, D4,
significance and measurement model as the layer 3 filter coefficient for 0, 1, 2, 3, 4, 5, 6, 7, 8,
UINTERFREQ IInter-frequency Measure D5, D6, D7, D8, D9,
InterFreqFilterCoef the intra-frequency measurement. 9, 11, 13, 15, 17, None D3
HOCOV Filter Coeff D11, D13, D15, D17,
The difference is that the report period of the inter-frequency 19
D19
measurement is 480 ms while the report period of the intra-frequency
measurement is 200 ms.
Page 136
Range Value
Interval between periodic reporting for the inter-frequency NON_PERIODIC_REPORT(Non
handover. periodical reporting), D250(D250),
In periodic reporting mode, the inter-frequency handover D500(D500), D1000(D1000), NON_PERIODIC_
attempts is reported at the preset interval. It is not D2000(D2000), D3000(D3000), REPORT, 250,
Inter-frequency recommended that this parameter be set to D4000(D4000), D6000(D6000), 500, 1000, 2000,
Measure Periodical "NON_PERIODIC_REPORT" since the UE behavior may be D8000(D8000), D12000(D12000), 3000, 4000, 6000,
UINTERFREQHOCOV PrdReportInterval ms D500
Measurement Report unknown. This parameter has impact on the Uu signaling flow. D16000(D16000), 8000, 12000,
Period If the interval is too short and the frequency is too high, the D20000(D20000), 16000, 20000,
RNC may have high load when processing signaling. If the D24000(D24000), 24000, 28000,
interval is too long, the network cannot detect the signal D28000(D28000), 32000, 64000
changes in time. This may delay the inter-frequency D32000(D32000),
handover, thus causing call drops. D64000(D64000)
Hysteresis for triggering event 2B.

The value of this parameter is associated with the slow fading.
If this parameter is set to a greater value, the ability of
0.5dB/
UINTERFREQHOCOV HystFor2B 2B Hysteresis resisting signal fluctuation improves and the number of ping- 0~29 0~14.5 4
0.5dBm
pong handovers decreases; however, the handover algorithm
becomes slow in responding to the signal change and thus
event 2B may not be triggered in time.
Hysteresis for triggering event 2D.
If this parameter is set to a greater value, the probability of
0.5dB/
UINTERFREQHOCOV HystFor2D 2D Hysteresis ping-pong reporting of event 2D and event 2F reduces and 0~29 0~14.5 4
0.5dBm
the number of incorrect decisions on event 2D decreases;
however, event 2D may not be triggered in time.
Hysteresis for triggering event 2F.

If this parameter is set to a greater value, the probability of
0.5dB/
UINTERFREQHOCOV HystFor2F 2F Hysteresis ping-pong reporting of event 2D and event 2F reduces and 0~29 0~14.5 4
0.5dBm
the number of incorrect decisions on event 2F decrease;
however, event 2F may not be triggered in time.
Hysteresis in the inter-frequency hard handover triggered by

the periodic measurement report.
This parameter is used to estimate the inter-frequency
handover on the RNC side. If this parameter is set to a greater 0.5dB/
UINTERFREQHOCOV HystForPrdInterFreq HHO Hysteresis 0~29 0~14.5 0
value, the ability of resisting signal fluctuation improves and 0.5dBm
the number of ping-pong handovers decreases; however, the
handover algorithm becomes slow in responding to the signal
change and thus the handover may not be triggered in time. Page 137
Range Range Value
The parameter WeightForUsedFreq is the frequency weighting factor
used to calculate the quality of the current frequency.
If this parameter is set to a greater value, the higher quality of the active
Weight for Used set is obtained. If this parameter is set to 0, the general quality of the
UINTERFREQHOCOV WeightForUsedFreq 0~20 0~2, step:0.1 None 0
frequency active set is considered the quality of the best cell in this set. This
parameter is used for event-triggered reporting of inter-frequency
handovers for events 2D, 2F, 2B and 2C, but not used for periodical
reporting of inter-frequency handovers.
UINTERFREQHOCOV TimeToTrig2B Event 2B Trigger Delay Interval time between the detection of event 2B, 2D or 2F and sending D0, D10, D20,
of measurement report. This parameter correlates with slow fading. If D40, D60, 0, 10, 20, 40, D0
this parameter is set to a greater value, the probability of incorrect D80, D100, 60, 80, 100,
UINTERFREQHOCOV TimeToTrig2D Event 2D Trigger Delay decision becomes low; however, the handover algorithm becomes slow D120, D160, 120, 160, 200,
ms
in responding to signal change. The emulation results show that setting D200, D240, 240, 320, 640, D320
this interval can effectively reduce the average number of handovers D320, D640, 1280, 2560,
UINTERFREQHOCOV TimeToTrig2F Event 2F Trigger Delay and the number of incorrect handovers, preventing unnecessary D1280, 5000
handovers. D2560, D5000 D1280
Interval between reception of periodical reports and triggering of the
inter-frequency handover.
Only the inter-frequency cell in which the signal quality is above a
HHO Period Trigger certain threshold in all periodic reports during a time equal to this
UINTERFREQHOCOV TimeToTrigForPrdInterFreq 0~64000 0~64000 ms 0
Delay parameter can be selected as the target cell for the inter-frequency
handover. If this parameter is set to a greater value, the probability of
incorrect decision becomes low; however, the handover algorithm
becomes slow in responding to signal change.
Inter-freq CS Measure Start Ec/No threshold of triggering the inter-frequency measurement for CS
UINTERFREQHOCOV InterFreqCSThd2DEcN0 ,PS non-HSPA or HSPA services. -24~0 -24~0 dB -14
Ec/No THD
The threshold of triggering the inter-frequency measurement (namely,
Inter-freq R99 PS Measure threshold of enabling the compression mode) is a key parameter in the
UINTERFREQHOCOV InterFreqR99PsThd2DEcN0
Start Ec/No THD inter-frequency handover policy. The setting of this parameter affects -24~0 -24~0 dB -14
the proportion of the UEs in compression mode in a cell and the
Inter-freq H Measure Start
UINTERFREQHOCOV InterFreqHThd2DEcN0 success rate of the hard handover. To set this parameter, consider the
Ec/No THD -24~0 -24~0 dB -14
following factors: moving speed of the UE and cell radius.
Inter-freq CS Measure Stop
UINTERFREQHOCOV InterFreqCSThd2FEcN0 Ec/No threshold of stopping the inter-frequency measurement for CS -24~0 -24~0 dB -12
Ec/No THD
,PS non-HSPA or HSPA services.
The threshold of stopping the inter-frequency measurement (namely,
Inter-freq R99 PS Measure threshold of disabling the compression mode) is a key parameter in the
UINTERFREQHOCOV InterFreqR99PsThd2FEcN0 -24~0 -24~0 dB -12
Stop Ec/No THD inter-frequency handover policy. The setting of this parameter affects
the proportion of the UEs in compression mode in a cell and the
Inter-freq H Measure Stop success rate of the hard handover. To set this parameter, consider the
UINTERFREQHOCOV InterFreqHThd2FEcN0 -24~0 -24~0 dB -12
Ec/No THD following factors: moving speed of the UE and cell radius.
Page 138
Range Range Value
Inter-freq CS Measure Start

RSCP threshold of triggering the inter-frequency
UINTERFREQHOCOV InterFreqCSThd2DRSCP -115~-25 -115~-25 dBm -95
RSCP THD measurement for CS ,PS non-HSPA or HSPA services.
The threshold of triggering the inter-frequency measurement
(namely, threshold of enabling the compression mode) is a
Inter-freq R99 PS Measure Start
UINTERFREQHOCOV InterFreqR99PsThd2DRSCP key parameter in the inter-frequency handover policy. The -115~-25 -115~-25 dBm -95
RSCP THD
setting of this parameter affects the proportion of the UEs in
compression mode in a cell and the success rate of the hard
Inter-freq H Measure Start RSCP handover. To set this parameter, consider the following
UINTERFREQHOCOV InterFreqHThd2DRSCP factors: moving speed of the UE and cell radius. -115~-25 -115~-25 dBm -95
THD
RSCP threshold of stopping the inter-frequency measurement

Inter-freq CS Measure Stop
UINTERFREQHOCOV InterFreqCSThd2FRSCP -115~-25 -115~-25 dBm -92
RSCP THD for CS ,PS non-HSPA or HSPA services.
The threshold of stopping the inter-frequency measurement
(namely, threshold of disabling the compression mode) is a
Inter-freq R99 PS Measure Stop
UINTERFREQHOCOV InterFreqR99PsThd2FRSCP key parameter in the inter-frequency handover policy. The -115~-25 -115~-25 dBm -92
RSCP THD
setting of this parameter affects the proportion of the UEs in
compression mode in a cell and the success rate of the hard
Inter-freq H Measure Stop RSCP handover. To set this parameter, consider the following
UINTERFREQHOCOV InterFreqHThd2FRSCP -115~-25 -115~-25 dBm -92
THD factors: moving speed of the UE and cell radius.
Inter-freq CS Target Frequency Threshold of the target frequency for triggering inter-frequency
UINTERFREQHOCOV TargetFreqCsThdEcN0 -24~0 -24~0 dB -12
Trigger Ec/No THD measurement based on measurement quantity of Ec/No for
CS,PS R99 or HSPA services. If the inter-frequency reporting
mode is set to "EVENT_TRIGGER", this parameter is used to
Inter-freq R99 PS Target
UINTERFREQHOCOV TargetFreqR99PsThdEcN0 set the criterion for triggering event 2B. That is, event 2B can -24~0 -24~0 dB -12
Frequency Trigger Ec/No THD
be triggered when the quality of the target frequency is higher
than this threshold. If the inter-frequency reporting mode is set
Inter-freq HSPA Target to "PERIODIC_TRIGGER", this parameter is used for
UINTERFREQHOCOV TargetFreqHThdEcN0 -24~0 -24~0 dB -12
Frequency Trigger Ec/No THD handover decision.
Inter-freq CS Target Freqency Threshold of the target frequency for triggering inter-frequency
UINTERFREQHOCOV TargetFreqCsThdRscp -115~-25 -115~-25 dBm -92
Trigger RSCP THD measurement based on measurement quantity of RSCP for
CS,PS R99 or HSPA services.If the inter-frequency reporting
mode is set to "EVENT_TRIGGER", this parameter is used to
Inter-freq R99 PS Target
UINTERFREQHOCOV TargetFreqR99PsThdRscp set the criterion for triggering event 2B. That is, event 2B can -115~-25 -115~-25 dBm -92
Frequency Trigger RSCP THD
be triggered when the quality of the target frequency is higher
than this threshold. If the inter-frequency reporting mode is set
Inter-freq HSPA Target Freqency to "PERIODIC_TRIGGER", this parameter is used for
UINTERFREQHOCOV TargetFreqHThdRscp handover decision. -115~-25 -115~-25 dBm -92
Trigger RSCP THD
Page 139
Range Range Value
Inter-freq CS UsedThreshold of used frequency quality for triggering inter-frequency

measurement based on measurement quantity of Ec/No for CS ,PS non- -24~0 -24~0 dB -12
UINTERFREQHOCOV UsedFreqCSThdEcN0 frequency trigger Ec/No
THD HSPA or HSPA services. If the value of "Inter-frequency Measurement
Report Mode" is set to EVENT_TRIGGER, this parameter is used to set the
measurement control of event 2B. One of the necessary conditions for -24~0 -24~0 dB -12
Inter-freq R99 PS Used triggering event 2B can be met only when the quality of the uesd frequency is
UINTERFREQHOCOV UsedFreqR99PsThdEcN0 frequency trigger Ec/No smaller than this threshold. In addition, event 2B is triggered only when both
THD the necessary conditions are met.
After handover, even if the inter-frequency measurement is triggered again, it
is very difficult to hand over the UE again to the cell of currently used -24~0 -24~0 dB -12
Inter-freq H Used frequency
UINTERFREQHOCOV UsedFreqHThdEcN0 frequency. That is, this parameter is usually set smaller than the start
trigger Ec/No THD
threshold for event 2F or equal to the threshold of event 2D.
Threshold of used frequency quality for triggering inter-frequency

Inter-freq CS Used
UINTERFREQHOCOV UsedFreqCSThdRSCP measurement based on measurement quantity of RSCP for CS ,PS non-
frequency trigger RSCP -115~-25 -115~-25 dBm -92
THD HSPA or HSPA services.If the value of "Inter-frequency Measurement Report
Mode" is set to EVENT_TRIGGER, this parameter is used to set the
Inter-freq R99 PS Used measurement control of event 2B. One of the necessary conditions for -115~-25 -115~-25 dBm -92
UINTERFREQHOCOV UsedFreqR99PsThdRSCP frequency trigger RSCP triggering event 2B can be met only when the quality of the used frequency is
THD smaller than this threshold. In addition, event 2B is triggered only when both
the necessary conditions are met.
After handover, even if the inter-frequency measurement is triggered again, it
Inter-freq H Used frequency is very difficult to hand over the UE again to the cell of currently used -115~-25 -115~-25 dBm -92
UINTERFREQHOCOV UsedFreqHThdRSCP frequency. That is, this parameter is usually set smaller than the start
trigger RSCP THD
threshold for event 2F or equal to the threshold of event 2D.
Minimum threshold for EcN0 of the target measurement or handover

frequency. This parameter is used for the periodic measurement reports
involved in the three processes: coverage-based inter-frequency handovers,
Inter-freq Handover Min measurement-based periodic DRD, and inter-frequency load handovers
UINTERFREQHOCOV HHOEcNomin -24~0 -24~0 dB -16
Access Ec/No THD based on measurement. In a periodic measurement report, the quality
threshold for the target frequency for the preceding processes is the value of
this parameter or the value of "TargetFreqThdEcN0" specific to one of the
preceding processes However, only the larger value is used.
Minimum threshold for RSCP of the target measurement or handover

frequency. This parameter is used for the periodic measurement reports
involved in the three processes: coverage-based inter-frequency handovers,
Inter-freq Handover Min measurement-based periodic DRD, inter-frequency load handovers based on
UINTERFREQHOCOV HHORSCPmin -115~-25 -115~-25 dBm -115
Access RSCP THD measurement. In a periodic measurement report, the quality threshold for the
target frequency for the preceding three processes is the value of this
parameter or the value of "TargetFreqThdRscp" specific to one process of
the preceding processes. However, only the larger value is used
Page 140
Range Range
Length of the timer for the inter-frequency measurement.
If the inter-frequency handover is not performed before this timer expires, the inter-
Inter-freq Measure frequency measurement is stopped and the compression mode is disabled (if enabled
UINTERFREQHOCOV InterFreqMeasTime 0~512 0~512 s 60
Timer Length before). The value 0 indicates that this timer is not to be started.This parameter is used
to prevent the long duration of the inter-frequency measurement state (compression
mode) due to unavailability of a target cell that meets the handover criteria.
Inter-RAT measurement reporting mode.
RNC provides two inter-RAT measurement reporting modes, event-triggered reporting
and periodical reporting.
Event-triggered reporting
To avoid the ping-pong effect before and after the inter-RAT handover, use event 3A
(quality of the currently used frequency is lower than the absolute threshold and the
signal level of the GSM cell is higher than another absolute threshold) as the triggering
PERIODICAL_
event that determines the inter-RAT handover.
REPORTING
The advantage of event-triggered reporting is that the signaling transmission and PERIODICAL_
(Periodical
processing load are saved. The drawback of event-triggered reporting is that the event REPORTING,
reporting), PERIODICAL_
UINTERRATHOCOV InterRatReportMode Inter-RAT Report Mode is reported only once and cannot be changed to periodical reporting. None
REPORTING
Periodical reporting EVENT_
EVENT_
When the quality of the GSM cell reported by the UE meets the criteria for inter-RAT TRIGGER
TRIGGER
handover, the delay trigger timer is started. If the quality of the GSM cell always meets
(Event trigger)
the criteria for inter-RAT handover before timeout, the inter-RAT handover is triggered
after the delay trigger timer expires.
The advantage of periodical reporting is that it can be used for repeated handover re-
attempts on the same cell when the handover fails, and that subsequent algorithms can
be flexibly developed. The drawback of periodical reporting is that it requires large
amount of signaling and increases the load on the air interface and for signaling
processing.
L3 filtering coefficient for event 2D or event 2F measurement
If this parameter is set to a greater value, the effect of smoothing signals and the ability
D0, D1, D2, D3,
of anti-fast-fading improve. However, the ability of tracing the signal change becomes
D4, D5, D6, D7, 0, 1, 2, 3, 4, 5,
low and call drops are likely to occur because handover is not performed in time. If this
UINTERRATHOCOV FilterCoefOf2D2F 2D/2F Filter Coefficient D8, D9, D11, 6, 7, 8, 9, 11, None D3
parameter is set to a smaller value, the number of unnecessary handovers increases.
D13, D15, D17, 13, 15, 17, 19
The parameter has the same physical meaning and measurement model as the inter-
D19
frequency L3 filter coefficient. The reporting period is 480 ms. For specific analysis, see
the intra-frequency L3 filter coefficient and inter-frequency L3 filter coefficient.
Measurement quantity used in coverage-based inter-RAT measurement in event 3A-
triggered reporting mode.
When CPICH_Ec/No is selected, it indicates that the Ec/No measurement quantity is
CPICH_
used for event 3A measurement. The physical unit is dB. CPICH_EC/NO,
EC/NO,
UINTERRATHOCOV MeasQuantityOf3A 3A Measure Quantity When CPICH_RSCP is selected, it indicates that the RSCP measurement quantity is CPICH_RSCP, None AUTO
CPICH_RSCP,
used for event 3A measurement. The physical unit is dBm. AUTO
AUTO
When AUTO is selected, it indicates that the Ec/No measurement quantity is used for
event 3A measurement if the RNC receives Ec/No 2D firstly. If the RNC receives the
RSCP 2D firstly, the RSCP measurement quantity is used for event 3A measurement. Page 141
Range Value
This parameter specifies the Layer 3 filter coefficient for the inter-RAT
measurement.
If this parameter is set to a greater value, the effect of smoothing signals
and the ability of anti-fast-fading improve. However, the ability of tracing the
signal change becomes low and call drops are likely to occur because 0, 1, 2, 3, 4, 5,
UINTERRATHO Inter-RAT Filter D0, D1, D2, D3, D4, D5, D6, D7, D8,
InterRATFilterCoef handover is not performed in time. If this parameter is set to a smaller 6, 7, 8, 9, 11, None D3
COV Coefficient D9, D11, D13, D15, D17, D19
value, the number of unnecessary handovers increases. 13, 15, 17, 19
The parameter has the same physical meaning and measurement model as
the inter-frequency L3 filter coefficient. The reporting period is 480 ms. For
specific analysis, see the intra-frequency L3 filter coefficient and inter-
frequency L3 filter coefficient.
The parameter WeightForUsedFreq is the frequency weighting factor used

to calculate the quality of the current frequency.
UINTERRATHO Weight for Used This parameter is used for event 3A evaluation. For detailed information of
WeightForUsedFreq 0~20 0~2, step:0.1 None 0
COV Frequency this parameter, refer to 3GPP TS 25.133.
To set this parameter, see the method for setting the intra-frequency
handover weighting factor "Weight".
Interval that the UE reports inter-RAT measurement results to the RNC. NON_PERIODIC_REPORT(Non NON_PERIODI
This parameter specifies the interval that the UE sends inter-RAT periodical reporting), D250(D250), C_REPORT,
measurement results to the RNC in periodical reporting mode. It is not D500(D500), D1000(D1000), 250, 500, 1000,
recommended that this parameter is set to NON_PERIODIC_REPORT D2000(D2000), D3000(D3000), 2000, 3000,
UINTERRATHO InterRATPeriodRepor Inter-RAT Period
since the UE behavior may be unknown.The GSM RSSI measurement D4000(D4000), D6000(D6000), 4000, 6000, ms D1000
COV tInterval Reporting Interval
period is 480 ms. Therefore, the inter-RAT periodical reporting interval D8000(D8000), D12000(D12000), 8000, 12000,
should be longer than 480 ms. If the periodical reporting interval is D16000(D16000), D20000(D20000), 16000, 20000,
excessively high, the handover decision time will be long, and handovers D24000(D24000), D28000(D28000), 24000, 28000,
will be slow. D32000(D32000), D64000(D64000) 32000, 64000
UINTERRATHO Hysteresis for event 2D. 0.5dB/

Hystfor2D 2D Hysteresis 0~29 0~14.5 4
COV This parameter is used to avoid the ping-pong reporting of event 2D (the 0.5dBm
estimated quality of the currently used frequency is below a certain
threshold). The value of this parameter is associated with slow fading. If this
parameter is set to a greater value, the probability of ping-pong reporting or
UINTERRATHO wrong decision is lower, but the event may not be triggered in time. If this 0.5dB/
Hystfor2F 2F Hysteresis 0~29 0~14.5 4
COV parameter is set to a smaller value, ping-pong reporting of event 2D is likely 0.5dBm
to occur.
The setting of this parameter should consider the radio conditions (slow
fading), actual handover distance, and moving speed of the UE. The value
UINTERRATHO of this parameter ranges from 2 dB to 5 dB. In addition, filter coefficient and
Hystfor3A 3A Hysteresis 0~15 0~7.5 0.5dB 4
COV triggering delay must be considered in setting this parameter.
Page 142
Range Value
Hysteresis for inter-RAT periodical reporting.
This parameter is used to avoid incorrect decisions caused by unexpected
Inter-RAT
UINTERRATHOCOV HystforInterRAT jitters of signals during inter-RAT handover decisions. HystforInterRAT and the 0~15 0~7.5 0.5dB 0
Hysteresis
inter-RAT handover decision threshold determine whether to trigger inter-RAT
handovers.
Interval time between detection of event 2D and sending of the measurement
D0, D10, D20, D40, D60, 0, 10, 20, 40, 60,
report.
D80, D100, D120, D160, 80, 100, 120, 160,
2D Event Trigger The value of this parameter is associated with slow fading. If this parameter is
UINTERRATHOCOV TrigTime2D D200, D240, D320, 200, 240, 320, ms D320
Delay Time set to a greater value, the probability of incorrect decision becomes low;
D640, D1280, D2560, 640, 1280, 2560,
however, the handover algorithm becomes slow in responding to signal
D5000 5000
change.
Interval time between detection of event 2F and sending of the measurement
D0, D10, D20, D40, D60, 0, 10, 20, 40, 60,
report.
D80, D100, D120, D160, 80, 100, 120, 160,
2F Event Trigger The value of this parameter is associated with slow fading. If this parameter is
UINTERRATHOCOV TrigTime2F D200, D240, D320, 200, 240, 320, ms D1280
Delay Time set to a greater value, the probability of incorrect decision becomes low;
D640, D1280, D2560, 640, 1280, 2560,
however, the handover algorithm becomes slow in responding to signal
D5000 5000
change.
Interval time between detection of event 3A and sending of the measurement
report.
D0, D10, D20, D40, D60, 0, 10, 20, 40, 60,
The value of this parameter is associated with the slow fading. If this parameter
D80, D100, D120, D160, 80, 100, 120, 160,
3A Event Trigger is set to a greater value, the probability of incorrect handover decision
UINTERRATHOCOV TrigTime3A D200, D240, D320, 200, 240, 320, ms D0
Delay Time becomes low; however, the handover algorithm becomes slow in responding to
D640, D1280, D2560, 640, 1280, 2560,
signal change. If this parameter is set to a smaller value, the handover
D5000 5000
algorithm becomes fast in responding to signal change; however, the
probability of incorrect decision becomes high.
Time delay for triggering handovers to GSM cells with non-verified BSIC.
During the period of time specified by this parameter, if the signal quality in a
Time to Trigger neighboring GSM cell fulfills inter-RAT handover criteria and the neighboring
UINTERRATHOCOV TimeToTrigForNonVerify Handover to Non- GSM cell is not verified, an inter-RAT handover is triggered. When this 0~64000, 65535 0~64000, 65535 ms 0
Verified GSM Cell parameter value is 65535, the RNC does not perform inter-RAT handovers to
non-verified GSM cells. If this parameter is set to a larger value, the average
number of handovers decreases, but call drops may occur.
Time delay for triggering handovers to GSM cells with verified BSIC.
Time to Trigger
During the period of time specified by this parameter, if the signal quality in a
UINTERRATHOCOV TimeToTrigForVerify Handover to 0~64000 0~64000 ms 0
neighboring GSM cell fulfills inter-RAT handover criteria and the neighboring
Verified GSM Cell
GSM cell is verified, an inter-RAT handover is triggered
Switch for verifying the Base Station Identity Code (BSIC).

This parameter is used to control cells where inter-RAT measurement reports
REQUIRED(Verify
are triggered. When the parameter is set to "REQUIRED", the measurement
mode), REQUIRED,
UINTERRATHOCOV BSICVerify BSIC Verify Switch reporting is triggered after the BSIC of the measured cell is decoded correctly. None REQUIRED
NOT_REQUIRE(Non- NOT_REQUIRE
When the parameter is set to "NOT_REQUIRE", the measurement reporting is
verify mode)
triggered regardless of whether the BSIC of the measured cell is decoded
correctly.
Page 143
Range Range Value
Inter-RAT CS Measure Start

UINTERRATHOCOV InterRATCSThd2DEcN0 -24~0 -24~0 dB -14
Ec/No THD Threshold of triggering inter-RAT measurement for CS ,PS non-HSPA
or HSPA services. when measurement quantity is Ec/No. The
threshold of triggering the inter-RAT measurement (the threshold of
Inter-RAT R99 PS Measure Start enabling the compressed mode) is a key parameter in the inter-RAT
UINTERRATHOCOV InterRATR99PsThd2DEcN0 -24~0 -24~0 dB -15
Ec/No THD handover policy. The setting of this parameter affects the proportion of
the UEs in compressed mode in a cell and the success rate of the
hard handover. The setting of this parameter should consider two
Inter-RAT HSPA Measure Start factors, moving speed of the UE and cell radius.
UINTERRATHOCOV InterRATHThd2DEcN0 -24~0 -24~0 dB -15
Ec/No THD
Inter-RAT CS Measure Stop

Threshold of stopping inter-RAT measurement for CS ,PS non-HSPA
UINTERRATHOCOV InterRATCSThd2FEcN0 -24~0 -24~0 dB -12
Ec/No THD or HSPA services. when measurement quantity is Ec/No.
The threshold of stopping the inter-RAT measurement (the threshold
of disabling the compressed mode) is a key parameter in the inter-
Inter-RAT R99 PS Measure Stop
UINTERRATHOCOV InterRATR99PsThd2FEcN0 RAT handover policy. The setting of this parameter affects the -24~0 -24~0 dB -13
Ec/No THD
proportion of the UEs in compressed mode in a cell and the success
rate of the hard handover. The setting of this parameter should
Inter-RAT HSPA Measure Stop consider two factors, moving speed of the UE and cell radius.
UINTERRATHOCOV InterRATHThd2FEcN0 -24~0 -24~0 dB -13
Ec/No THD
Threshold of triggering inter-RAT measurement for CS ,PS non-HSPA

Inter-RAT CS Measure Start
UINTERRATHOCOV InterRATCSThd2DRSCP or HSPA services. when measurement quantity is RSCP. When RSCP -115~-25 -115~-25 dBm -100
RSCP THD
is used as the measurement quantity for CS ,PS non-HSPA or HSPA
services,the UE reports event 2D when the measured RSCP value is
lower than this threshold. Then, the RNC sends the signaling to
Inter-RAT R99 PS Measure Start
UINTERRATHOCOV InterRATR99PsThd2DRSCP enable the compressed mode and start the inter-RAT measurement. -115~-25 -115~-25 dBm -110
RSCP THD
To enable the compressed mode earlier, increase the threshold of
triggering event 2D; otherwise, decrease the threshold of triggering
event 2D. To prevent the frequent enabling and disabling of the
Inter-RAT HSPA Measure Start compressed mode, increase the difference between the thresholds of
UINTERRATHOCOV InterRATHThd2DRSCP -115~-25 -115~-25 dBm -110
RSCP THD triggering event 2D and event 2F.
Inter-RAT CS Measure Stop

UINTERRATHOCOV InterRATCSThd2FRSCP Threshold of stopping inter-RAT measurement for CS ,PS non-HSPA -115~-25 -115~-25 dBm -97
RSCP THD
or HSPA services. when measurement quantity is RSCP. When RSCP
Inter-RAT R99 PS Measure Stop is used as the measurement quantity for CS ,PS non-HSPA or HSPA
UINTERRATHOCOV InterRATR99PsThd2FRSCP -115~-25 -115~-25 dBm -107
RSCP THD services, the UE reports event 2F when the measured RSCP value is
larger than this threshold. Then, the RNC sends the signaling to
Inter-RAT HSPA Measure Stop disable the compressed mode and stop the inter-RAT measurement.
UINTERRATHOCOV InterRATHThd2FRSCP -115~-25 -115~-25
RSCP THD PagedBm
144 -107
GUI Value Baseline
MO Parameter ID Parameter Name Parameter Description Actual Value Range Unit
Range Value
Quality requirement for the cell of another RAT during inter-RAT handover lower than -110, -
for CS ,PS non-HSPA or HSPA services. 110~-48(Actual value
Inter-RAT CS Handover If the value of "Inter-RAT Reporting Mode" is set to "EVENT_TRIGGER",
UINTERRATHOCOV TargetRatCsThd 0~63 meets the condition: dBm 16
Decision THD this parameter is used to set the criterion for triggering event 3A. That is, Actual Value = GUI
event 3A is triggered only when the quality of the target frequency is higher Value - 111)
than this threshold and the measurement quality of the current CS services
lower than -110, -
is lower than the quality of the currently used frequency. If the value of
110~-48(Actual value
Inter-RAT R99 PS Handover "Inter-RAT Reporting Mode" is set to "PERIODICAL_REPORTING", this
UINTERRATHOCOV TargetRatR99PsThd 0~63 meets the condition: dBm 16
Decision THD parameter is used to evaluate inter-RAT coverage handovers at the RNC.
Actual Value = GUI
Note that 0 means a value smaller than -110 dBm.
Value - 111)
If the value of "Inter-RAT Reporting Mode" is set to
"PERIODICAL_REPORTING", this parameter is used for the assessment of lower than -110, -
110~-48(Actual value
Inter-RAT HSPA Handover inter-RAT coverage handover. That is, Tother_RAT in the following
UINTERRATHOCOV TargetRatHThd formulas. This parameter is the absolute threshold of the cell of another 0~63 meets the condition: dBm 16
Decision THD
RAT (Received Signal Strength Indicator: RSSI) for the inter-RAT handover Actual Value = GUI
decision. Value - 111)
Inter-RAT CS Used
UINTERRATHOCOV UsedFreqCsThdEcN0 Frequency Trigger Ec/No -24~0 -24~0 dB -12
Ec/No quality threshold for used frequency of CS ,PS non-HSPA or HSPA
THD
services.
For CS ,PS non-HSPA or HSPA services. , if Ec/No is used as the
Inter-RAT R99 PS Used
measurement quantity for inter-RAT measurement and the value of "Inter-
UINTERRATHOCOV UsedFreqR99PsThdEcN0 Frequency Trigger Ec/No -24~0 -24~0 dB -13
RAT Reporting Mode" is set to "EVENT_TRIGGER", this parameter is used
THD
to set the measurement control of event 3A. That is, event 3A is triggered
only when the quality of the currently used frequency is lower than this
Inter-RAT HSPA Used
threshold and the TargetRatCsThd criterion is fulfilled.
UINTERRATHOCOV UsedFreqHThdEcN0 Frequency Trigger Ec/No -24~0 -24~0 dB -13
THD
Inter-RAT CS Used
UINTERRATHOCOV UsedFreqCsThdRscp Frequency Trigger RSCP RSCP quality threshold for used frequency of CS ,PS non-HSPA or HSPA -115~-25 -115~-25 dBm -97
THD services.
For CS ,PS non-HSPA or HSPA services. , if RSCP is used as the
Inter-RAT R99 PS Used
measurement quantity for inter-RAT measurement and the value of "Inter-
UINTERRATHOCOV UsedFreqR99PsThdRscp Frequency Trigger RSCP -115~-25 -115~-25 dBm -107
RAT Reporting Mode" is set to "EVENT_TRIGGER", this parameter is used
THD
to set the measurement control of event 3A. That is, event 3A is triggered
Inter-RAT HSPA Used only when the quality of the currently used frequency is lower than this
UINTERRATHOCOV UsedFreqHThdRscp Frequency Trigger RSCP threshold and the TargetRatCsThd criterion is fulfilled. -115~-25 -115~-25 dBm -107
THD
Length of the timer for inter-RAT measurement.

After inter-RAT measurement starts, if no inter-RAT handover is performed
Inter-RAT Measure Timer
UINTERRATHOCOV InterRATMeasTime when this timer expires, the inter-RAT measurement is stopped. In addition, 0~512 0~512 s 60
Length
the compressed mode (if activated) should be deactivated. The value 0
indicates that the inter-RAT measurement timer is not started. Page 145
Cell-level switch for controlling inter-frequency and inter-RAT

handover.
When inter-frequency and inter-RAT cells coexist, the
parameter should be configured according to actual handover
INTERFREQ
policies.
(inter-frequency handover only),
INTERFREQ,
When INTERFREQ is selected, only inter-frequency
Inter-Frequency and INTERRAT
neighboring cells are measured and inter-frequency handover SIMINTERFREQ
UCELLHOCOMM InterFreqRATSwitch Inter-RAT Coexist (inter-RAT handover only), INTERRAT, None
is performed. RAT
Switch
When INTERRAT is selected, only GSM neighboring cells are
SIMINTERFREQRAT( SIMINTERFREQRAT
measured and inter-RAT handover is performed.
inter-frequency and inter-RAT
handover)
When SIMINTERFREQRAT is selected, both inter-frequency
and inter-RAT cells are measured, and the handover is
performed according to the type of the cell that first meets the
handover decision criteria.
This parameter specifies the event 2D/2F measurement

threshold of which measurement type (inter-frequency or
inter-RAT measurements) will be used, when inter-frequency
and inter-RAT measurements coexist.
When COEXIST_MEAS_THD_CHOICE_INTERFREQ is
selected, event 2D/2F measurement thresholds oriented
towards inter-frequency configuration are selected.
COEXIST_MEAS_THD_CHOIC
When COEXIST_MEAS_THD_CHOICE_INTERRAT is
E_INTERFREQ
selected, event 2D/2F measurement thresholds oriented
(Choosing the inter-frequency COEXIST_MEAS_THD_
towards inter-RAT configuration are selected.
Inter-Freq and Inter- oriented parameters), CHOICE_INTERFREQ, COEXIST_MEAS
UCELLHOCOMM CoexistMeasThdChoice RAT Coexist Measure None _THD_CHOICE_
The factors such as the event 2D/2F measurement thresholds
Threshold Choice COEXIST_MEAS_THD_CHOIC COEXIST_MEAS_THD_ INTERFREQ
for inter-frequency measurement and inter-RAT
E_INTERRAT CHOICE_INTERRAT
measurement, inter-frequency and inter-RAT handover
(\Choosing the inter-RAT
decision thresholds, and current handover policy should be
oriented parameters)
considered during setting. For example, if the event 2D
threshold for inter-RAT measurement is higher than that for
inter-frequency measurement, and inter-frequency
neighboring cells are preferred when inter-RAT and inter-
frequency neighboring cells coexist, then
COEXIST_MEAS_THD_CHOICE_INTERFREQ should be
selected.
Page 146
Load Handover
Load handover is used to balance the load among inter-frequency or inter-RAT cells. Load handover falls into the following categories:
 Inter-frequency LDR handover

 Inter-RAT LDR handover
 Inter-RAT service handover
Inter-Frequency LDR Handover

Inter-Frequency LDR Handover Procedure
The inter-frequency LDR handover (WRFD-020103 Inter Frequency Load Balance) suits co-sited cells covering the same area.
In the triggering phase

The Load Reshuffling (LDR) module detects that the current cell is in basic congestion and then initiates an inter-frequency handover.
In the decision phase

For Inter-frequency LDR blind handover, the RNC decides to trigger an inter-frequency blind handover if the corresponding conditions are
met. After the inter-frequency handover is triggered, the RNC chooses a decision algorithm according to whether the conditions of direct
blind handover are met. For inter-frequency LDR measure-based handover, the RNC requests the UE to perform the inter-frequency
measurement. Based on the measurement results, the RNC chooses a target cell to perform inter-frequency hard handover.
In the execution phase

The RNC performs the handover according to the decision result.
Inter-Frequency LDR Handover Measurement

The Inter-frequency LDR handover uses the periodical report mode.
In Inter-frequency LDR blind handover, CPICH RSCP of the used frequency is measured.
In inter-frequency LDR measure-based handover, both CPICH_RSCP and CPICH_Ec/No of the target frequency are measured.
Page 147
Inter-frequency LDR handover
Inter-Frequency LDR Handover Decision and Execution
The LDR algorithm may trigger an inter-frequency handover. The following describes the procedure for handover decision and execution.
The inter-frequency LDR handover can be performed based on blind handover or measurement that can be decided by the parameter
InterFreqLDHOMethodSelection.
Inter-Frequency LDR Blind Handover Decision

1. The LDR algorithm learns that a cell is in basic congestion and provides target cells and the UE with low priority for handover.
2. The RNC determines to trigger an inter-frequency blind handover.
If the UE is not in soft handover state, the RNC directly performs Inter-frequency LDR blind handover.
If the UE is in soft handover state, the RNC operates based on the following conditions:
If the HoSwitch: HO_ALGO_LDR_ALLOW_SHO_SWITCH is set to ON,
RNC determines whether the cell that triggers LDR is the best cell.
− If this cell is the best cell, the RNC initiates an intra-frequency measurement for Inter-frequency LDR blind handover. The intra-frequency
measurement is used to estimate quality of the inter-frequency cell of the same coverage.
− If this cell is not the best cell, the RNC does not initiate a Inter-frequency LDR blind handover.
If the HoSwitch: HO_ALGO_LDR_ALLOW_SHO_SWITCH is set to OFF, the RNC does not initiate a Inter-frequency LDR blind
handover.
Inter-Frequency LDR Blind Handover Execution

The inter-frequency cells with the same coverage area have the same CPICH RSCP values. By measuring the CPICH RSCP of the cell,
the quality of the cells with the same coverage area can be determined, which increases the probability of successful blind handover.
1. RNC initializes the timer of intra-frequency measurement for blind handover.,the timer is specified by internal algorithm and need not be configured.
2. RNC initiates a periodical intra-frequency measurement.
The measurement report mode is set to periodical report.
− The reporting period is BlindHOIntrafreqMRInterval.
− The number of measurement reports is BlindHOIntrafreqMRAmount.
− The intra-frequency handover measurement quantity is CPICH RSCP.
− The list of measured cells contains only the cells that trigger LDR. Page 148
Inter-frequency LDR handover
3. After receiving from the UE the intra-frequency measurement reports for conditional blind handover, the RNC checks whether the following
condition is met:
CPICH RSCP of the cell in the measurement report >= BlindHOQualityCondition
− If the condition is met, the RNC increments the counter of the number of intra-frequency measurement reports for blind handover by 1.
− If the condition is not met, the RNC does not perform a blind handover to the cell that triggers LDR and stops intra-frequency measurement for blind
handover.
4. When the counter reaches the value of BlindHOIntrafreqMRAmount, the RNC initiates a blind handover to the target cell.
If the counter does not reach this value, the RNC waits for the next intra-frequency measurement report from the UE.
5. If the timer of intra-frequency measurement for blind handover expires, the RNC does not perform a blind handover to the target cell and stops
intra-frequency handover measurement for blind handover.
If the inter-frequency handover based on coverage or QoS is triggered, the RNC stops the intra-frequency measurement for conditional blind handover.
Inter-Frequency LDR Measurement-based Handover Decision and Execution

1. The LDR algorithm learns that a cell is in basic congestion and provides target cells and the UE with low priority for handover.
2. The RNC selects the target cell based on the measurement results. The report period is specified by the parameter PrdReportInterval.
The target cell must meet the following conditions:

− The CPICH RSCP value of the target cell is larger than TargetFreqThdRscp.
− The CPICH Ec/No value of the target cell is larger than TargetFreqThdEcN0
− The target cell is not in the basic congestion state.
The RNC performs an inter-frequency hard handover to the target cell directly.
Page 149
Inter-RAT LDR handover
When the load of the 3G network is heavy and all the RABs of the UE are supported by the 2G network,
the Inter-RAT (3G-to-2G) LDR handover (WRFD-020306 Inter-RAT Handover Based on Load) is triggered.
Inter-RAT LDR handover procedure

In the triggering phase
When the load of the 3G cell that the UE accesses is higher than the related
threshold, the Load Reshuffling (LDR) algorithm makes a handover decision.
In the measurement phase

RNC enables the compressed mode and starts the inter-RAT handover
measurement.
 In the decision phase

After the UE reports event 3C, the RNC makes a handover decision.
In the execution phase

RNC initiates a handover procedure.
Page 150
Inter-RAT LDR handover
Inter-RAT LDR Handover Measurement
For Inter-RAT LDR handover, the measurement report is triggered only by event.
In inter-RAT LDR handover, event 3C takes only GSM RSSI as measurement quantity.
When the estimated quality of other system is higher than a certain threshold, the UE reports event 3C to the RNC.
When the conditions for event 3C are met and the delay requirement specified by the TrigTime3C parameter can be satisfied, the UE
sends the measurement report of event 3C.
Event 3C is triggered on the basis of the following formula:
−MOtherRAT is the measurement value of the cell (in another RAT) in the reporting range.
MOtherRAT + CIOOtherRAT ≥ TOtherRAT + H3c/2 − CIOOtherRAT is the cell individual offset value of the cell (in another RAT) in the reporting range,
which is equal to the sum of CIO and CIOOffset.
− TOtherRAT is the absolute inter-RAT handover threshold.
Based on different service types (CS, PS domain R99 service, or PS domain HSPA service), TOtherRAT threshold can be configured through the
following parameters:
− InterRATNCovHOCSThd
− InterRATNCovHOPSThd
H3c is HYSTFOR3C, the hysteresis value of event 3C.
For the PS and CS combined services, the threshold (s) for CS services is (are) used.
Inter-RAT LDR Handover decision and Execution

After receiving the event 3C measurement report of 2G cells, the RNC performs the following handover decision and execution procedure:
1. Put all the 2G cells that trigger event 3C into a cell set and arrange the cells according to the measurement quality in descending order.
2. Select the cells in sequence from the cell set.
To avoid the impact of the UE (in long-term measurement of compressed mode) on the radio network, the parameter
InterRATHOAttempts is set to restrict the maximum attempts of the 3G-to-2G LDR-based or service-based handover.
The parameter indicates the handover attempts made to the same cell or different cells. If the number of attempts exceeds the parameter
value, RNC does not initiate the handover.
Page 151
Inter-RAT LDR handover Parameters
Range Value
HO_ALGO_LDR_ALLOW_SHO_SWITCH: When the switch is on,

UCORRMALGOSWITCH HO_ALGO_LDR_ALLOW_ HO_ALGO_LDR_ALLOW OFF,ON OFF,ON None ON
the LDR inter-frequency handover is allowed during soft handover.
SHO_SWITCH _SHO_SWITCH
Intra-Frequency
This parameter specifies the interval between intra-frequency
UINTRAFREQHO BlindHOIntrafreqMRInterval Measurement Report D250, D500 250, 500 ms D250
measurement reports for Load Reshuffling (LDR) blind handover.
Interval of Blind HO
Used in the algorithm of the load reshuffling (LDR) intra-frequency
blind handover. This parameter specifies the number of the
required measurement reports that fulfill the handover criterion
Intra-Frequency before the blind handover decision is triggered. The RNC initiates
UINTRAFREQHO BlindHOIntrafreqMRAmount Masurement Report blind handover only when the UE continuously reports adequate D1, D2, D4, D8 1, 2, 4, 8 None D2
Amount of Blind Handover intra-frequency measurement reports that fulfill the quality criterion
for blind handover. If the UE reports an unqualified measurement
report, the measurement process is stopped. In this case, the RNC
does not initiate blind handover.
Measurement signal level threshold for blind handovers involved in
inter-frequency load balancing. When the measurement signal
level of a cell is higher than this threshold, the cell can be selected
as a target cell for a blind handover. Otherwise, the cell cannot be
Blind Handover Quality selected as the target cell.
UINTERFREQNCELL BlindHOQualityCondition -115~-25 -115~-25 dBm -92
Condition If this parameter is not set to -115, a conditional blind handover
can be triggered in an inter-frequency neighboring cell with the
same coverage. If this parameter is set to -115, a direct blind
handover can be triggered in an inter-frequency neighboring cell
with the larger coverage.
D250, D500,
250, 500, 1000,
The interval between two reports is the configured value. D1000, D2000,
2000, 3000, 4000,
This parameter has impact on the Uu signaling flow. If the interval D3000, D4000,
6000, 8000,
Inter-freq Measure is too short and the frequency is too high, the RNC may have D6000, D8000,
UMCLDR PrdReportInterval 12000, 16000, ms D3000
Periodical Report Period burden in processing signaling. If the interval is too long, the D12000, D16000,
20000, 24000,
network cannot detect the signal change in time, which may delay D20000, D24000,
28000, 32000,
the inter-frequency handover. D28000, D32000,
64000
D64000
Page 152
Inter-RAT LDR handover Parameters
Range Range Value
Estimate the signal quality of the periodic reports. The inter-frequency

Inter-freq Target handover is triggered only when the signal quality of the target cell is higher
UMCLDR TargetFreqThdRscp Frequency Trigger RSCP than this parameter. Note: The threshold can be reached only when RSCP -115~-25 -115~-25 dBm -92
THD and EcNo of the target cell are above the RSCP and EcNo that are set in the
command.
Estimate the signal quality of the periodic reports. The inter-frequency
Inter-freq Target handover is triggered only when the signal quality of the target cell is higher
UMCLDR TargetFreqThdEcN0 Frequency Trigger Ec/No than this parameter. Note: The threshold can be reached only when RSCP -24~0 -24~0 dB -12
THD and EcNo of the target cell are above the RSCP and EcNo that are set in the
command.
The value of this parameter is associated with the slow fading. If this
parameter is set to a greater value, the ability of resisting signal fluctuation
improves and the number of ping-pong handovers decreases; however, the
UINTERRATHONCOV Hystfor3C 3C Hysteresis 0~15 0~7.5 0.5dB 0
handover algorithm becomes slow in responding to the signal change. If this
parameter is set to a too large value, the cell of another RAT where the UE
needs to be handed over to must be of good quality.
Interval time between detection of event 3C and sending of the measurement D0, D10, D20,
report. D40, D60,
0, 10, 20, 40, 60,
The value of this parameter is associated with the slow fading. If this D80, D100,
80, 100, 120, 160,
parameter is set to a greater value, the probability of incorrect handover D120, D160,
UINTERRATHONCOV TrigTime3C Event 3C Trigger Delay 200, 240, 320, ms D640
decision becomes low; however, the handover algorithm becomes slow in D200, D240,
640, 1280, 2560,
responding to signal change. If this parameter is set to a smaller value, the D320, D640,
5000
handover algorithm becomes fast in responding to signal change; however, D1280, D2560,
the probability of incorrect decision becomes high. D5000
Quality requirement for the cell of another RAT during inter-RAT handover for
lower than -110,
CS domain services.
Inter-RAT CS Handover -110~-48
UINTERRATHONCOV InterRATNCovHOCSThd This parameter is used for measurement control on event 3C. When the target 0~63 dBm 21
Decision THD (Actual Value =
frequency quality is higher than this threshold, event 3C is triggered. The value
GUI Value - 111)
0 means a value smaller than -110 dBm.
Quality requirement for the cell of another RAT during inter-RAT handover for
lower than -110,
CS domain services.
Inter-RAT PS Handover -110~-48
UINTERRATHONCOV InterRATNCovHOPSThd This parameter is used for measurement control on event 3C. When the target 0~63 dBm 21
Decision THD (Actual Value =
frequency quality is higher than this threshold, event 3C is triggered. The value
GUI Value - 111)
0 means a value smaller than -110 dBm.
Maximum number of inter-RAT handover attempts after the non-coverage-

Inter-RAT Handover Max
UINTERRATHONCOV InterRATHOAttempts based handover failures. For each handover attempt, a new target cell where 1~16 1~16 None 16
Attempt Times
the UE has not been handed over to and fulfills the criterion can be selected. Page 153
Inter-RAT Service Handover
Inter-RAT service handover supports 3G-to-2G handover based on service attributes.
When 3G and 2G networks coexist, this feature enables the 3G traffic to be directed to the 2G network.
This feature can balance the load between the two systems by transferring some kind of appropriate services to GSM/GPRS.
Switches for Inter-RAT Service Handover

To perform the service-based 3G-to-2G handover, the RNC must turn on the related switches for services in the CS and PS domains.
When a single CS service is initially set up by the UE, the RNC allows the 3G-to-2G service-based handover if CSServiceHOSwitch is
set to ON.
 When a single PS service is initially set up by the UE, the RNC allows the service-based 3G-to-2G handover if PSServiceHOSwitch is
set to ON.
Inter-RAT Service Handover Procedure

For the combined services, no service-based handover is triggered.
 Except for the triggering phase, the procedure of Inter-RAT service

Handover is the same as that of the Inter-RAT LDR handover.
 When a service is established, the RNC requests the handover to the

GSM based on the service type and service handover indicator assigned by
the CN.
Page 154
Inter-RAT Service Handover
Rules for Enabling Inter-RAT LDR or Service Handover
RNC initiates the 3G-to-2G load handover only when Service Handover Indicator (SHIND) is as follows:
 HO_TO_GSM_SHOULD_BE_PERFORM
 HO_TO_GSM_SHOULD_NOT_BE_PERFORM
RNC initiates the service-based 3G-to-2G handover only when the Service Handover Indicator is
HO_TO_GSM_SHOULD_BE_PERFORM.
Before deciding the 3G-to-2G handover, the RNC considers 2G cell capability (RatCellType), service required capability (Req2GCap)
and UE capability.
Example of rules for indicator of 3G-to-2G

handover based on load and service
Page 155
Inter-RAT Service Handover Parameters
Whether the cell allows inter-RAT handover for CS or

PS services.
Inter-RAT CS When the switch is set to ON, the inter-RAT handover
UCELLHOCOMM CSServiceHOSwitch OFF, ON OFF, ON None OFF
Handover Switch for CS or PS services is enabled. When the switch is
set to OFF, the inter-RAT handover for CS or PS
services is disabled.
Based on the Service Handover Indicator of a service
and the related parameter configurations on the
network side, related measurements and inter-RAT
Inter-RAT PS handover are triggered immediately once a service is
UCELLHOCOMM PSServiceHOSwitch set up. This switch is set to ON only when service OFF, ON OFF, ON None OFF
Handover Switch
handover is required. Generally, the switch is set to
OFF.
Service handover attribute. If the

HO_INTER_RAT_RNC_SERVICE_HO_SWITCH is set
to ON, the service handover setting of this parameter is
applied. Otherwise, the service handover setting HO_TO_GSM_SHOULD_ HO_TO_GSM_SHOULD_
assigned by the CN is applied. BE_PERFORM, BE_PERFORM,
- HO_TO_GSM_SHOULD_BE_PERFORM: Handover
HO_TO_GSM_
Service Handover to the 2G network is performed so long as 2G signals HO_TO_GSM_SHOULD_ HO_TO_GSM_SHOULD_
UTYPRABBASIC SHInd None SHOULD_NOT_
Indicator are available. NOT_BE_PERFORM, NOT_BE_PERFORM,
BE_PERFORM
- HO_TO_GSM_SHOULD_NOT_BE_PERFORM:
Handover to the 2G network is performed when 3G HO_TO_GSM_SHALL_NOT_ HO_TO_GSM_SHALL_NOT
signals are weak but 2G signals are strong. BE_PERFORM _BE_PERFORM
- HO_TO_GSM_SHALL_NOT_BE_PERFORM:
Handover to the 2G network is not performed even if
3G signals are weak but 2G signals are strong.
Minimum capability of the 2G cell required for inter-

RAT handover from 3G network to 2G network. If the
2G Cell Capability
2G cell is not capable, the handover will not be
UTYPRABBASIC Req2GCap Required for Inter-RAT GSM, GPRS, EDGE GSM, GPRS, EDGE None None
performed. An EDGE cell is more capable of
Handover
supporting PS services than a GPRS cell, whereas a
GPRS cell more capable than a GSM cell.
Identifying the type of the inter-RAT cell

NO_CAPABILITY indicates that the capacity of the NO_CAPABILITY(Cell
GSM cell remains unknown. It is not recommended to capability unknown), NO_CAPABILITY, GSM,
UEXT2GCELL RatCellType Inter-RAT cell type None None
set the cell type to NO_CAPABILITY. If the cell type is GSM(GSM), GPRS(GPRS), GPRS, EDGE
NO_CAPABILITY, the inter-RAT handover cannot be EDGE(EDGE)
triggered. Page 156
Signaling Procedures for Inter-Frequency Handover
Signaling procedure for inter-frequency handover between NodeBs within one RNC
Page 157
Signaling Procedures for Inter-Frequency Handover
Signaling procedure for inter-frequency handover between NodeBs between RNCs
Page 158
Signaling Procedures for Inter-RAT Handover
3G-to-2G Handover in CS Domain 3G-to-2G Handover in PS Domain
Page 159
Signaling Procedures for Inter-RAT Handover
2G-to-3G Handover in CS Domain 2G-to-3G Handover in PS Domain
Page 160
Table of Contents

Page 161
Bearer Mapping
During the service setup, the RNC selects appropriate channels based on the UE capability, cell capability, and service
parameters to optimize the use of cell resources and ensure the QoS.
Can Be Carried on
CN Domain Service Type Optional Feature?
HS-DSCH? Huawei RAN supports the setting
- Signaling (SRB) Yes Yes of the types of RABs carried on
the HS-DSCH according to
Voice Yes Yes
CS Videophone No No
service requirements.
Streaming No No
Conversational Yes Yes
Streaming Yes Yes
Interactive Yes No
PS
Background Yes No
IMS signaling Yes Yes
MBMS PTP Yes Yes
To support SRB over HSDPA, SrbChlType should be set to HSDPA or HSPA.

If SrbChlType is set to HSDPA or HSPA, SrbChlTypeRrcEffectFlag will affect when the downlink signaling is carried by HSDPA.
- If SrbChlTypeRrcEffectFlag is set to TRUE, after “RRC Connection Setup” the downlink signaling will be carried by HS-DSCH.
- If SrbChlTypeRrcEffectFlag is set to FALSE, after “RB Setup” the downlink signaling will be carried by HS-DSCH.
During service setup procedure DL PS interactive and background services (BE service) can be mapped onto the CCH, DCH, HS-DSCH based
on maximum service rate (MBR).
Low-rate PS services have relatively small amount of data. Therefore, such PS services can be carried on the CCH to save radio resources.
If the maximum DL service rate is lower than DlBeTraffDecThs, the maximum UL service rate is lower than UlBeTraffDecThs, and the RRC
connection is set up on the CCH, then the service is carried on the CCH.
Otherwise, the following situations occur:
If the maximum DL service rate is higher than or equal to DlBeTraffThsOnHsdpa, the service is carried on the HS-DSCH. Otherwise,
the service is carried on the DCH. Page 162
HSDPA Code Resource Management
Code resource management allocates code resources to the HS-SCCH and HS-PDSCH.
The NodeB supports HS-DSCH transmissions to multiple users in parallel in a TTI. If more than one HS-PDSCH code can be allocated by
the NodeB, then code multiplexing can be used to allocate the codes to multiple users so as to improve resource usage and system
throughput.
HS-SCCH Code Resource Management

Each HS-SCCH uses an SF128 code. The number of HS-SCCHs determines the maximum number of HSDPA users that can be
scheduled simultaneously in a TTI. Generally, the number of HS-SCCHs depends on the traffic characteristics of the cell. The default
number is 4, which is specified by the parameter HsScchCodeNum on the RNC side.
If the default setting is used, the HS-PDSCH can use only 14 SF16 codes. To enable the HS-PDSCH to use 15 SF16 codes, you are
advised to configure 2 HS-SCCHs.
HS-PDSCH Code Resource Management

The transport channel HS-DSCH is mapped on one or several High-Speed Physical Downlink Shared Channels (HS-PDSCHs) which are
simultaneously received by the UE. As indicated in 3GPP specifications, there are up to 15 HS-PDSCHs per cell with the spreading factor
fixed to 16.
The number of the HS-PDSCHs per NodeB is configurable and dependent on the license. The NodeB can dynamically allocate license
codes to the HS-PDSCHs between cells based on the actual requirements. The number of available HS-PDSCH codes for a cell is the
number of license codes allocated by the NodeB or the number of idle SF16 codes in the cell, whichever is smaller. The DPCH and the
HS-PDSCH coexist in a cell. Therefore, sharing the cell code resources between them to improve the resource usage is of critical
importance in HSDPA code resource management.
Huawei supports both RNC-level and NodeB-level code resource management. RNC-controlled static or dynamic code allocation is
enabled through the parameter AllocCodeMode. NodeB-controlled dynamic code allocation is enabled through the parameter
DynCodeSw. The dynamic code allocation controlled by the NodeB is more flexible than that controlled by the RNC. It shortens the
response time and saves the Iub signaling used for code reallocation.
Page 163
Huawei support three code resource management methods:
1) RNC-Controlled Static Code Allocation

− The number of HS-PDSCH codes is specified by the parameter HsPdschCodeNum.
2) RNC-controlled Dynamic Code Allocation

− The minimum number of HS-PDSCH codes is specified by the parameter HsPdschMinCodeNum.
− The maximum number of HS-PDSCH codes is specified by the parameter HsPdschMaxCodeNum.
Adding an HS-PDSCH Code

After a DCH RL is released or reconfigured (for example, because the spreading
factor becomes larger), the RNC adds an HS-PDSCH code if the following
conditions are met:
The code adjacent to the allocated HS-PDSCH codes is idle.
After the code is added, the minimum spreading factor of the remaining codes is
smaller than or equal to the value of CellLdrSfResThd.
The parameter CellLdrSfResThd set on the RNC side is used to reserve codes
for new users, to avoid congestion due to code insufficiency, and to avoid
unnecessary reshuffling of the code tree.
Page 164
Releasing an HS-PDSCH Codes
If idle DPCH codes are insufficient when a DCH RL is set up, added, or
reconfigured (for example, because the spreading factor becomes smaller), the
RNC preempts HS-PDSCH codes in the shared codes for the DPCH.
In addition, if the minimum spreading factor of idle DPCH codes is greater than
the value of CellLdrSfResThd, the RNC can also reallocate some
HS-PDSCH codes to the DPCH. The reallocated code number must be the
smallest one of the available shared codes.
.
3) NodeB-Controlled Dynamic Code Allocation ("Dynamic Code Allocation Based on NodeB" is an optional feature)
Generally, the NodeB can use the HS-PDSCH codes only allocated by the RNC. The NodeB-controlled dynamic code allocation, however,
allows the NodeB to temporarily allocate idle codes to the HS-PDSCH.
Every TTI, the NodeB detects the SF16 codes that are not allocated to the HS-PDSCH. If such an SF16 code or any of its subcodes is
allocated by the RNC to the DCH or a common channel, this SF16 code is regarded as occupied. Otherwise, it is regarded as unoccupied.
Therefore, the available HS-PDSCH codes include the codes reserved by the RNC and the idle codes adjacent to the allocated HS-PDSCH
codes. Every time the RNC allocates or release HS-PDSCH codes, it notifies the NodeB through Iub signaling and the NodeB performs the
corresponding processes.
Huawei recommends the following code allocation modes, where the first mode is preferred:
 Configure the RNC to use static code allocation and the NodeB to use dynamic code allocation.
 If the NodeB does not support dynamic code allocation, configure the RNC to use dynamic code allocation.
 If not all the NodeBs controlled by an RNC support dynamic code allocation, the RNC-controlled dynamic code allocation is
recommended. In this case, the NodeB-controlled dynamic code allocation can also be enabled for those supporting NodeBs.
Page 165
Dynamic Code Tree Reshuffling
Regardless of whether dynamic code allocation is controlled by the RNC or the NodeB, the number of continuous
codes available for the HS-PDSCH shall be maximized. The dynamic code tree reshuffling function can achieve this goal by reallocating
DPCH codes.
Dynamic code tree reshuffling takes effect only when the following conditions are met:
 The cell is not in the basic congestion state which is triggered by code resource.
 The switch parameter CodeAdjForHsdpaSwitch is set to ON.
When dynamic code tree reshuffling takes effect, the RNC reshuffles the codes used by the DPCH to provide more continuous SF16 codes
for HSDPA through this function. This function is described as follows:
Every time the codes used by the DPCH are changed, the RNC will choose an SF16 subtree which are not used by HS-PDSCH from right
to left. The selected subtree must meet the following conditions:
 The selected subtree belongs to the code trees that can be shared between HS-PDSCH and DPCH.
 The number of users on the selected subtree is smaller than or equal to the threshold specified by the parameter
CodeAdjForHsdpaUserNumThd.
The parameter CodeAdjForHsdpaUserNumThd limits the number of users that can be reshuffled each time, to prevent too many users
from being reshuffled in a short time and therefore to avoid affecting user experience.When the above conditions are met, the RNC will
select this subtree for reshuffling and relocate the users to the positions where the codes are idle.
Page 166
Code Resource Management Parameters
Range Range Value
If Manual is chosen, parameter " Code Number for HS-PDSCH "
determines HS-PDSCH code number to be allocated. If Automatic is
Manual, Manual,
UCELLHSDPA AllocCodeMode Allocate Code Mode chosen, allocate HS-PDSCH code number between configured " Code None Automatic
Automatic Automatic
Max Number for HS-PDSCH " and " Code Min Number for HS-PDSCH
".
Code Number for The parameter specifies the number of HS-DPSCH codes. This
UCELLHSDPA HsPdschCodeNum 1~15 1~15 None 5
HS-PDSCH parameter is valid only when "Allocate Code Mode" is set to "Manual".
The parameter determines the maximum number of HS-PDSCH codes

(SF=16). This parameter is valid only when "Allocate Code Mode" is
Code Max Number for set to "Automatic". The number of codes used by the HS-PDSCH is
UCELLHSDPA HsPdschMaxCodeNum 1~15 1~15 None 5
HS-PDSCH dynamically set between "Code Min Number for HS-PDSCH" and
"Code Max Number for HS-PDSCH", based on whether the code tree
is idle or busy.
The parameter specifies the minimum number of the HS-PDSCH

codes (SF=16). This parameter is valid only when "Allocate Code
Code Min Number for Mode" is set to Automatic. The number of codes used by the HS-
UCELLHSDPA HsPdschMinCodeNum 1~15 1~15 None 1
HS-PDSCH PDSCH is dynamically set between "Code Min Number for HS-
PDSCH" and "Code Max Number for HS-PDSCH", based on the idle
state of the code tree.
This parameter decides the maximum number of subscribers that the
UCELLHSDPA HsScchCodeNum Code Number for HS-SCCH 1~15 1~15 None 4
NodeB can schedule in a TTI period.
This parameter specifies code reshuffling switch for HDSPA. If the

switch is set as ON, codes occupied by the R99 service can be
Code Adjust Switch for adjusted toward codes with small numbers to release the sharing OFF(OFF),
UCELLHSDPA CodeAdjForHsdpaSwitch OFF, ON None ON
HSDPA codes adjacent to HSDPA code. When " Allocate Code Mode " is set ON(ON)
to Automatic or the NodeB automatic code algorithm is enabled, the
released codes can be used by HSDPA .
H-based code tree reshuffle user number threshold. When the switch
"Code Adjust Switch for HSDPA"is enabled, if the number of users on
User Number for Code the tree to be reshuffled is no greater than this parameter, the
UCELLHSDPA CodeAdjForHsdpaUserNumThd 1~16 1~16 None 3
Adjust for HSDPA reshuffle is allowed. Otherwise, the reshuffle is given up. This
parameter limits the number of users involved in one reshuffle so that
reshuffle on lots of users at a time is avoided.
Punish Timer Length for This parameter is used for setting the punish timer length for code
UCELLHSDPA HCodeAdjPunshTimerLength 0~300 0~300 s 5
Code Adjust for HSDPA adjust for hsdpa.
OPEN(open), OPEN~0,
MACHSPARA DYNCODESW Dynamic Code Switch Indicates the Dynamic Code Switch.
CLOSE(close) CLOSE~1 Page
None 167
OPEN(open)
HSDPA Power Resource Management
Power resource management determines the transmit power of the HS-PDSCH, HS-SCCH, and HS-DPCCH.
Generally, an HSDPA cell has the same coverage as the corresponding R99 cell. To improve the resource usage in this case, the downlink
power resources of HSDPA can be dynamically allocated as follows:
The downlink power resources are first reserved for common physical channels and allocated to the DPCH. The remaining power
resources are available for HSPA, including HSUPA and HSDPA.
The HSPA power resources are first allocated to the HSUPA downlink control channels, including the E-AGCH, E-RGCH, and E-HICH. The
remaining power resources are available for HSDPA.
The HSDPA power resources are first allocated to the downlink control channel HS-SCCH. The remaining power resources are allocated to
the traffic channel HS-PDSCH.
Every TTI, the NodeB detects the power usage of R99 channels to
determine the power available for HSPA.
To reserve the power for R99 power control itself, the power margin
PwrMgn needs to be set on the NodeB side.
In addition, the power allocated to HSPA must not exceed the maximum
permissible power HspaPower, which can be set on the RNC side.
Page 168
HSDPA Power Resource Management
Power Control on HS-SCCH
The power control method for the HS-SCCH can be fixed (fixed transmit power control) or CQI-based (dynamic transmit power control).
The method is specified by the parameter HSSCCHPWRCMINDCH.
Fixed transmit power control: The power of the HS-SCCH is determined by the offset relative to the P-CPICH power. The offset is
specified by the parameter SCCHPWR. The transmit power of the HS-SCCH is fixed without consideration of the channel quality but with
consideration of the receive quality of UEs at the cell edge.
Dynamic transmit power control based on CQI: The NodeB dynamically adjusts the transmit power of the HS-SCCH based on the CQI
feedback from UE.
Page 169
Power Resource Management Parameters

Range Range
MACHSPARA PWRMGN Power Margin Indicates the Power Margin Ratio. 0~100 0~100 % 5
HS-SCCH Power Control Indicates the HS-SCCH Power Control Method in CQI(CQI), CQI,
MACHSPARA HSSCCHPWRCMINDCH None CQI(CQI)
Method in CELL DCH state CELL DCH state. FIXED(FIXED) FIXED
Indicates the power of the HS-SCCH is an offset in -10~10,

MACHSPARA SCCHPWR HS-SCCH Power 0~80 0.25dB 28
dB to the transmit power of the PCPICH. step:0.25
MACHSPARA MXPWRPHUSR Max Power Per Hs-user Indicates the Max Power Per Hs-user. 1~100 1~100 % 100
This parameter specifies the offset between the

total HSPA power and the maximum transmission
The Offset of HSPA Total power of a cell. The total HSPA power is the
UCELLHSDPA HspaPower -500~0 -50~0 0.1dB 0
Power maximum value of HSPA dynamical power can be
adjusted. For details about this parameter, refer to
3GPP TS 25.308.
Page 170
MAC-hs Scheduling
With the limited Uu resources for HSDPA in a cell, the user expects to maximize the service rate while the
telecom operator expects to maximize the system capacity. MAC-hs scheduling is used to coordinate the Uu resources, user experience,
and system capacity. It is implemented at the NodeB MAC-hs.
The scheduling algorithm consists of two steps.
At first, the algorithm determines which initial transmission queues or retransmission processes can be put into the candidate set for
scheduling.
Then, the algorithm calculates their priorities based on factors such as the CQI, user fairness, and differentiated services. If the algorithm is
weighted more towards the channel quality of the UE, the HSDPA cell can have a higher capacity but user fairness and differentiated
services may be affected. If the algorithm is weighted more towards user fairness and differentiated services, the system capacity may be
affected.
Huawei provides four scheduling algorithms: maximum C/I (MAXCI), round-robin (RR), proportional fair (PF), and Enhanced
Proportional Fair (EPF). The EPF algorithm is optional.
Page 171
MAC-hs Scheduling
Determining the Candidate Set
The candidate for scheduling contains new data packets (initial transmission queues) or data packets to be retransmitted (retransmission
processes), with the following exceptions:
 If the UE starts the compressed mode, its data cannot be put into the candidate set during the GAP.
 If the UE category requires the UE to wait for several TTIs before it can be scheduled again, its data cannot be put into the candidate set in this period.
The UE of category 1 or 2 needs to wait for 3 TTIs, and the UE of category 3, 4, and 11 must wait for 2 TTIs.
 If the number of retransmissions of a data packet reaches or exceeds the maximum number, the data of this UE cannot be put into the candidate set.
The data should be discarded.
Huawei supports that the maximum number of retransmissions is set on a service basis:
− MaxNonConverHarqRt: the maximum number of non-conversational service retransmissions in the CELL_DCH state
− MaxEfachHarqRt: the UE in the enhanced CELL_FACH state does not report ACK, NACK, or CQI in the uplink. The HARQ processes of the UE
use the blind retransmission mechanism.
Factors which considered in the four scheduling algorithms
Other user data can be put into the candidate set.
Factor MAXCI RR PF EPF

Calculating Priorities Service type No No No Yes
Four algorithms are available for calculating the priorities of data
Initial transmission or
packets in the candidate set. The scheduling policies vary according to Yes Yes Yes Yes
retransmission
the algorithms for calculating the priorities of data packets.
Maximum power No No No Yes
The algorithm to be used is specified by the parameter SM on the Waiting time No Yes No Yes
NodeB LMT. CQI Yes No Yes Yes
Actual throughput No No Yes Yes
SPI No No No Yes
GPR No No No Yes
HBR No No No Yes
Page 172
MAC-hs Scheduling
MAXCI Algorithm
The retransmission processes unconditionally have higher priorities than the initial transmission queues. The retransmission processes are sorted in
first-in first-out (FIFO) mode. The initial transmission queues are sorted in the CQI order. A higher CQI means a higher data priority.
The MAXCI algorithm aims to maximize the system capacity but cannot ensure user fairness and differentiated services.
The UE estimates the CQI based on the assumption that the transmit power of the HS-PDSCH on the network side is as follows:
RR Algorithm
FIFO mode. The initial transmission queues are sorted in the order of the waiting time in the MAC-hs queue. A longer waiting time means a higher data
priority. The RR algorithm aims to ensure user fairness but cannot provide differentiated services. Not considering the CQI reported by the UE leads to
lower system capacity.
PF Algorithm
FIFO mode. The initial transmission queues are sorted in the order of R/r. Here, R represents the throughput corresponding to the CQI reported by the
UE, and r represents the throughput achieved by the UE. A greater R/r value means a higher data priority.
The PF algorithm aims to make a trade off between system capacity and user fairness. It provides the user with an average throughput that is
proportional to the actual channel quality. The system capacity provided by PF is between the system capacity provided by RR and that provided by
MAXCI.
Page 173
MAC-hs Scheduling
EPF Algorithm
The EPF algorithm (WRFD-01061103 Scheduling based on EPF and GBR) can meet the requirements of telecom operators related to user
fairness and differentiated services and also provide a high system capacity.
By calculating the priority of each queue, the scheduling algorithm achieves the following:
 When the system resources are sufficient to meet the basic QoS requirements of all users, the transmission delay of delay-sensitive data is within the
permissible range and the transmission rate of throughput-sensitive data is not lower than the GBR. High-priority users can obtain more resources for
higher QoS.
 When the system resources are insufficient to meet the basic QoS requirements of all users, delay-sensitive data has higher priorities than throughput-
sensitive data. High-priority users can obtain more resources to ensure the basic QoS.
Queue priorities are determined on the basis of service types. The EPF algorithm distinguishes between delay-sensitive data and throughput-sensitive
data based on the QoS requirements. The following factors are considered: the waiting time, CQI reported by the UE, throughput achieved by the UE,
guaranteed bit rate (GBR), scheduling priority indicator (SPI) weight, happy bit rate (HBR), and power consumed in the queue for a certain period.
User fairness is implemented in EPF as follows:

 EFP algorithm guarantees the user fairness in the same way as that PF algorithm. HBR and Resource Limit is used in EPF to limit the use of single
users and improve fairness.
 HBR is used to determine the throughput expected by the user based on a study on user experience.
When the rate for a user reaches the HBR, the scheduling probability for the user is decreased. The HBR is specified by the parameter HappyBR on
the RNC side.
Resource Limit is used to prevent the users in areas with poor coverage from consuming too many cell resources so that there is no decrease in
system capacity.
− When the resource limitation switch (RscLmSw) is on, the algorithm allocates the lowest priority to a queue whose power consumption exceeds the
threshold. The ratio of the maximum available power of a queue to the total power of the cell is specified by the NodeB command SET RSCLMTPARA.
Differentiated service is implemented in EPF as follows:

Differentiated services are provided based on SPI and SPI weights.
− SPI is a parameter specified on the basis of service types and users priorities.
− SPIweight can be specified according to the SPI to provide differentiated services.
The SPI weight affects the calculation of queue priorities. It is used to quantify the differentiated services. If resource is insufficient, the proportion of SPI
weights determines the approximate proportion of rates among users. For example, for three throughput-sensitive service users with the same channel
quality, the same GBR and the proportion of SPI weights is 100:50:30, the proportion of actual rates is close to 100:50:30. Page 174
MAX EFACH HARQ Indicates the MAX HARQ Retransmission

MACHSPARA MAXEFACHHARQRT 0~10 0~10 None 2
Retransmission Times Times of E_FACH user.
MAX HARQ
Indicates the MAX HARQ Retransmission
Retransmission Times of
MACHSPARA MAXNONCONVERHARQRT Times of Non-Conversational service in CELL 0~10 0~10 None 4
Non-Conversational
DCH state.
service in CELL DCH state
OPEN(open),
MACHSPARA RSCLMSW Resource Limiting Switch Indicates the Resource Limiting Switch. OPEN, CLOSE None OPEN(open)
CLOSE(close)
EPF(Enhanced PF),
EPF(Enhanced
MACHSPARA SM Scheduling Method Indicates the HSDPA Scheduling Method. PF(PF), RR(Round Robin), EPF, PF, RR, MAXCI None
PF)
MAXCI(Max C/I)
OPEN(open),
MACHSPARA RSCLMSW Resource Limiting Switch Indicates the Resource Limiting Switch. OPEN, CLOSE None OPEN(open)
CLOSE(close)
-3.0dB, -2.5dB, -2.0dB, -

3.0dB, -2.5dB, -2.0dB, -
1.5dB, -1.0dB, -0.5dB,
1.5dB, -1.0dB, -0.5dB,
This parameter named Measure Power Offset 0.0dB, 0.5dB, 1.0dB,
0.0dB, 0.5dB, 1.0dB, 1.5dB,
Constant is used to compute measurement 1.5dB, 2.0dB, 2.5dB,
2.0dB, 2.5dB, 3.0dB, 3.5dB,
power offset. Measurement power offset is 3.0dB, 3.5dB, 4.0dB,
4.0dB, 4.5dB, 5.0dB, 5.5dB,
used by UE to obtain total received HS-PDSCH 4.5dB, 5.0dB, 5.5dB,
6.0dB, 6.5dB, 7.0dB, 7.5dB,
power. The calculation for Measure Power 6.0dB, 6.5dB, 7.0dB,
HS-PDSCH MPO 8.0dB, 8.5dB, 9.0dB, 9.5dB,
UCELLHSDPA HsPdschMPOConstEnum Offset is as shown below: 7.5dB, 8.0dB, 8.5dB, dB 2.5dB
Constant 10.0dB, 10.5dB, 11.0dB,
Measure Power Offset = Max(-6, 9.0dB, 9.5dB, 10.0dB,
11.5dB, 12.0dB, 12.5dB,
Min(13,CellMaxPower - PcpichPower - 10.5dB, 11.0dB, 11.5dB,
13.0dB, 13.5dB, 14.0dB,
Measure Power OffsetConstant)). For details of 12.0dB, 12.5dB, 13.0dB,
14.5dB, 15.0dB, 15.5dB,
the IE "Measure Power Offset", refer to 3GPP 13.5dB, 14.0dB, 14.5dB,
16.0dB, 16.5dB, 17.0dB,
TS 25.214. 15.0dB, 15.5dB, 16.0dB,
17.5dB, 18.0dB, 18.5dB,
16.5dB, 17.0dB, 17.5dB,
19.0dB
18.0dB, 18.5dB, 19.0dB
Page 175

Range Range Value
Defines the happy bit rate of the best effort (BE) service with different user
priorities(user priorities can be set by parameter UserPriority). This Happy
bit rate is sent to NodeB by RNC through the Iub interface. When the
NodeB resource is limited and the HS-DSCH bit rate of the user exceeds
UOPERUSERHAPPYBR HappyBR Happy bit rate the Happy bit rate, the HS-DSCH scheduling priority will be decreased. 0~27900 0~27900 kbit/s 0
When this parameter is set to zero, it indicates that NodeB will not adjust
the HS-DSCH scheduling priority.If the value of the parameter HappyBR in
command ADD UOPERUSERHAPPYBR is larger than 5000, it will be set to
the minimum of the HappyBR value in SET UUSERHAPPYBR and 5000.
Scheduling priority of interactive and background services. Value 11

Scheduling Priority
UOPERSPIWEIGHT SPI indicates the highest priority, while value 2 indicates the lowest priority. 0~15 0~15 None None
Indicator
Values 0, 1, 12, 13, 14, and 15 are reserved for the other services.
Specifies the weight for service scheduling priority. This weight is used in
two algorithms. In scheduling algorithm, it is used to adjust the handling
priority for different services. In Iub congestion algorithm, it is used to
UOPERSPIWEIGHT SpiWeight SPI Weight 1~100 1~100 % 100
allocate bandwidth for different services. If the weight is higher, it is more
possible to increase the handling priority of the user or get more Iub
bandwidth, respectively.
Page 176
HSDPA Mobility Management
HSDPA Intra-Frequency Handover
Handling of Event 1A
After receiving an event 1A report, the RNC proceeds as follows:
 If the number of cells in the active set does not reach the maximum value, the RNC adds the cell to the active set.
 If the number of cells in the active set reaches the maximum value, the RNC does not add the radio link to the active set.
Handling of Event 1B
After receiving an event 1B report, the RNC determines whether to delete a cell.
If the cell to be deleted is not an HSDPA serving cell, the cell is directly removed.
If the cell to be deleted is an HSDPA serving cell, then:
− If the new best cell supports HSDPA, the new best cell is reconfigured to be an HSDPA serving cell. If the reconfiguration fails, the
service is reconfigured onto DPCH.
− If the new best cell does not support HSDPA, the service is reconfigured onto DPCH to ensure the continuity of the service.
Handling of Event 1C
After receiving an event 1C report, the RNC decides whether to change the worst cell.
If the cell to be replaced is not an HSDPA serving cell, the cell is directly removed.
If the cell to be replaced is an HSDPA serving cell, then:
− If the best cell supports HSDPA, the best cell is reconfigured to be an HSDPA serving cell. If the reconfiguration fails, the service is
reconfigured onto DPCH.
− If the best cell does not support HSDPA, the service is reconfigured onto DPCH to ensure the continuity of the service.
Handling of Event 1D
After receiving an event 1D report, the RNC proceeds as follows:
 If the downlink service is carried on the HSDPA, then:
− If the new best cell in the active set supports HSDPA and the HSPA hysteresis timer (HspaTimerLen) expires, the new best cell is
reconfigured to be an HSDPA serving cell.
− If the new best cell in the active set does not support HSDPA, the downlink service is directed to the DCH through the reconfiguration.
Page 177
Figure below shows an example of how to handle event 1D in this situation.
Assume that the UE moves from HSDPA cell 1 to HSDPA cell 2, that the two cells are intra-frequency neighboring cells, and that all the cells
in the active set support HSDPA. RNC updates the HSDPA serving cell according to the reported event and keeps the HSDPA serving cell
consistent with the best cell.
If the downlink service is carried on the DCH, then:

To avoid frequent handovers at the boundary between an HSDPA cell and an R99 cell, a protection timer is used. After an intra-frequency
handover, the timer starts. After this timer expires, the RNC reconfigures the service of the UE onto the HS-PDSCH of the HSDPA cell if
either of the following two conditions is met:
The target cell supports HSDPA.
The target cell does not support HSDPA but has a DRD neighboring cell.
The timer length is specified by the parameter ChannelRetryHoTimerLen.
In the execution procedure mentioned above, CMP_UU_SERV_CELL_CHG_WITH_ASU_SWITCH of the CmpSwitch parameter is used to
determine whether the update of the active set and the change of the serving cell are synchronized. This switch is applicable to only R6
UEs.
If the switch is ON, the UE supports the synchronization of the update of the active set and the change of the serving cell.
If the switch is OFF, the UE reconfigures the change of the serving cell by allocating physical channels after updating the active set.
Page 178
HSDPA Inter-Frequency Handover
Scenario1 of HSDPA Inter-frequency Handover
When the UE moves from cell 1 (an HSDPA cell) to cell 2

(an R99 cell), the inter-frequency handover conditions are met
and measurement control is triggered.
The RNC makes a handover decision for HSDPA services
according to the neighboring cell measurement report, and
reconfigures the service onto the DPCH of cell 2.
When the UE moves from cell 2 (an R99 cell) to cell 1 (an HSDPA
cell), the inter-frequency handover conditions are met and
measurement control is triggered.
The RNC hands over the UE to the DPCH first, and then to the
HS-PDSCH if the service can be carried on the HS-PDSCH.
Page 179
HSDPA Inter-Frequency Handover
When the UE moves from cell 1 (an HSDPA cell) to cell 2 (an
HSDPA cell) and the inter-frequency handover conditions are met,
the UE is handed over from cell 1 to cell 2.
The target cell becomes the new serving HS-PDSCH cell.
Page 180
HSDPA Mobility Parameters

Range Range Value
HSPA serving cell change is usually triggered by event 1D. To

avoid frequent serving cell change which may affect the system
performance, a protection timer TimerHSPA is needed. This
timer shall be started upon HSPA serving cell changed, i.e. the
serving HSDSCH cell changed or the EDCH serving cell
HSPA Hysteresis Timer
UHOCOMM HspaTimerLen changed. And no more 1D-triggered serving cell change shall 0~1024 0~1024 s 0
Length
be performed until the timer expires. However, this timer shall
not restrict HSPA serving cell change to another cell in the
active set if current serving cell is to be removed from the active
set. The value 0 means this timer shall not be started and
serving cell change be initiated immediately.
This parameter specifies the value of the channel retry

handover timer.
When handover is performed and some higher HSPA or HSPA
plus technique is supported, UTRAN will trigger the
reconfiguration for the higher techniques.
Channel Retry Ho Timer
UCOIFTIMER ChannelRetryHoTimerLen Pingpang will happen when the reconfiguration is triggered 0~999 0~999 s 2
Length
immediately when handover succeeds, because handover
procedure is frequently.
In order to avoid the pingpang, this timer will start after
handover procedure is performed, and the reconfiguration will
not be triggered until the timer expires.
CMP_UU_SERV_CELL_CHG_WITH_ASU_SWITCH: When the

switch is on, the active set update is in the same procedure as
CMP_UU_SERV_CELL_ CMP_UU_SERV_CELL_ the change of the serving cell. When the switch is not on, the
UCORRMALGOSWITCH OFF,ON OFF,ON None OFF
CHG_WITH_ASU_SWITCH CHG_WITH_ASU_SWITCH serving cell is changed after the UE updates the active set and
delivers reconfiguration of physical channels. This switch is
applicable only to R6 or aboveUEs.
Page 181
Table of Contents

Page 182
Bearer Mapping
The transport channel carrying HSUPA services is the E-DCH. It can carry services of multiple types and service combinations.
During the service setup, the RNC selects appropriate channels based on the UE capability, cell capability, and service parameters to
optimize the use of cell resources and ensure the QoS. Huawei RAN supports the setting of the types of RABs carried on the E-DCH
according to service requirements.
To support SRB over HSUPA, SrbChlType should be set to
Can Be Carried HSUPA or HSPA.
CN
Service Type on Optional Feature?
Domain
E-DCH? If SrbChlType is set to HSUPA or HSPA,
Yes SrbChlTypeRrcEffectFlag will affect when the uplink signaling will
– Signaling (SRB) Yes
Feature name: SRB over HSUPA be carried by HSUPA.
Yes
If SrbChlTypeRrcEffectFlag is set to TRUE, after “RRC
Voice Yes Feature name: CS Voice over
HSPA/HSPA+ Connection Setup” the uplink signaling will be carried by E-
Videophone No No DCH.
CS If SrbChlTypeRrcEffectFlag is set to FALSE, after “RB
Streaming No No
Yes Setup” the uplink signaling will be carried by E-DCH
Conversational Yes Feature name: VoIP over
HSPA/HSPA+ PS interactive and background services (i.e. BE service) can be
Yes mapped onto the CCH, DCH, HS-DSCH, or E-DCH
Streaming Yes Feature name: Streaming Traffic Low-rate PS services have relatively small amount of data.
Class on HSUPA Therefore, such PS services can be carried on the CCH to save
Interactive Yes No radio resources
PS
Background Yes No If the maximum DL service rate is lower than DlBeTraffDecThs,
Yes
the maximum UL service rate is lower than UlBeTraffDecThs,
IMS signaling Yes Feature name: IMS Signaling over
HSPA and the RRC connection is set up on the CCH, then the service is
carried on the CCH.
Otherwise, the following situations occur:

If the maximum UL service rate is higher than or equal to
UlBeTraffThsOnHsupa, the service is carried on the E-
DCH. Otherwise, the service is carried on the DCH.
Page 183
HSUPA Dynamic CE Management
CE resources are hard resources used for channel modulation and demodulation. Generally, if the service rate is higher, more CE
resources are required. To ensure the QoS, CE resources need to be allocated on the basis of the maximum bit rate of users. In this case,
if fixed CE resources are allocated to users, the resources are wasted when the rate of the service source is low.
The rate of the services that are carried on the HSUPA channel and configured in scheduling mode can be controlled by the scheduling
module. Thus, dynamic CE management is used. That is, CE resources to be allocated to users are adjusted dynamically according to the
conditions of CE resources and the change of user requirements to improve the utilization of CE resources.
Without dynamic CE
 Fixed CE resource allocation based on Maximum Bit Rate
management
 Low CE Utilization Effieciency
With dynamic CE  Dynamic CE resources allocation based on the condition of CE

management resoruces and the change of users requirements.
 High CE Utilization Effieciency
The rules for dynamic CE resource allocation are as follows:

 When the rate of the service source decreases, the redundant CE resources are called back.
 When there is a need to increase the service rate, CE resources are reserved.
 When there are insufficient available CE resources, CE resources are allocated to users in the serving RLS preferentially because the QoS of users
depends on the resource allocation of the serving cell.
 When the available CE resources are insufficient to meet the requirements of all the users in the serving RLS, user priorities need to be considered to
provide differentiated services.
In addition, the dynamic CE management module needs to process messages from external functional modules, such as the resource allocation
request during the establishment of a new connection and the channel reassignment request. In such a case, the QoS requirement of users and user
priorities must be considered. Dynamic CE management has no function switch parameter. Page 184
HSUPA Dynamic CE Management
HSUPA Dynamic CE Procedures
(1) Calling back the CE resoruces of the serving RLS.

If the NodeB detects that the allocated CE resources are not fully used in the recent
time (100 ms for 10 ms TTI users and 20 ms for 2 ms TTI users), it calls back the
unused CE resources. For users whose CE resources are called back, the algorithm
notifies the MAC-e scheduler of the SGmax.
(2) Allocating CE resources for the new RL

After a new RL is admitted, the new RL requests CE resources based on the CE init.
If CE resources are insufficient, CE preemption is triggered.
(3) Processing CE resources among serving RLS for fairness

The algorithm performs a fairness judgment every 160 ms and takes the GBR and
SPI weight (FACTOR) into comprehensive consideration. The algorithm selects the
user of the serving RLS with the largest value of priority and calculates the number
of required CEs.
(4) Allocating CE resources for the AG or RG UP user of the serving RLS.

If the AG or RG UP user of the serving RLS requests CE resources, the CE
scheduler allocates the required CE resources. The GBR and SPI weight (FACTOR)
are considered in the allocation. When allocating CE resources to the user of the
serving RLS, the CE scheduler can preempt the CE resources of the user of the
non-serving RLS.
(5) Allocating CE resources to the RG UP user of the non-serving RLS

If the RG UP user of the non-serving RLS requests CE resources, the CE scheduler
allocates the required CE resources considering the GBR and SPI weight
(FACTOR).
Page 185
HSUPA Mobility Management
HSUPA Intra-Frequency Handover
Handling of Event 1A
After receiving the measurement report, the RNC proceeds as follows:
If the target cell supports HSUPA and the uplink service is carried on the E-DCH, then:
 If the current number of cells in the E-DCH active set is smaller than the value of MaxEdchCellInActiveSet, the target cell is added to
both the DCH and E-DCH active sets.
 Otherwise, the target cell is added to only the DCH active set.
After deciding that a cell can be added to the E-DCH active set,
If the admission in the downlink fails, the cell is added to neither the E-DCH active set nor the DCH active set. It waits for the next event
1A report for retry.
 Otherwise, if the admission in the downlink succeeds, the RNC perform the HSUPA admission in the uplink.
− If HSUPA admission in the uplink succeeds, the cell is added to the E-DCH active set and the DCH active set.
− If HSUPA admission in the uplink fails, the cell is added only to the DCH active set. If the DCH admission in the uplink still fails, the cell is added to
neither the E-DCH active set nor the DCH active set. It waits for the next event 1A report for retry.
Handling of Event 1B
If the number of radio links in the DCH active set is larger than one, then:
If the cell to be removed is not an HSUPA serving cell, the cell is directly removed.
If the cell to be removed is an HSUPA serving cell, then:
− If the new best cell supports HSUPA, the new best cell is reconfigured to be an E-DCH serving cell.
− If the new best cell does not support HSUPA, the uplink service is redirected to the DCH through the RB reconfiguration.
If the current service is CS Voice over HSPA and the cell to be removed is an E-DCH serving cell, then:
− If the new best cell supports both HSDPA and HSUPA, the cell is reconfigured to be an HS-DSCH serving cell or an E-DCH serving cell.
− If the new best cell supports HSDPA but not HSUPA, the current service is changed from CS Voice over HSPA to CS Voice over DCH. If there
remain other HSDPA services, the serving cell should be updated and the new best cell should be the HS-PDSCH serving cell.
− If the new best cell supports neither HSUPA nor HSDPA, the current service is reconfigured to be CS Voice over DCH.
Page 186
Handling of Event 1C
When event 1C is triggered, the UE reports the event-triggered list that contains good cells and the cells to be replaced, and sequences
the cells from the highest to the lowest quality according to measurement quantity.
After receiving the measurement report, the RNC proceeds as follows:

If the new cell supports HSUPA, then:
− If the current number of cells in the E-DCH active set is smaller than the value of MaxEdchCellInActiveSet, the new cell is added to
the E-DCH active set.
− If the current number of cells in the E-DCH active set is equal to the value of MaxEdchCellInActiveSet and the cell to be replaced is
also included in the E-DCH active set, the new cell joins the E-DCH active set through replacement.
− If the current number of cells in the E-DCH active set is equal to the value of MaxEdchCellInActiveSet and the cell to be replaced is
not included in the E-DCH active set, the new cell is added only to the DCH active set.
If the current service is CS Voice over HSPA and the cell to be removed is an E-DCH serving cell, then:
− If the new best cell supports both HSDPA and HSUPA, the cell is reconfigured to be an HS-DSCH serving cell or an E-DCH serving cell.
− If the new best cell supports HSDPA but not HSUPA, the current service is changed from CS Voice over HSPA to CS Voice over DCH
through the reconfiguration. If there remain other HSDPA services, the serving cell should be updated and the new best cell should be the
HS-PDSCH serving cell.
− If the new best cell supports neither HSUPA nor HSDPA, the current service is reconfigured to be CS Voice over DCH. If the cell to be
removed is an E-DCH serving cell, the HS-DSCH serving cell must be removed at the same time. Therefore, a new E-DCH serving cell or
HS-DSCH serving cell must be determined. The method is the same as that of removing the serving cell, as described in the previous
section of Handling of Event 1B.
If the new cell does not support HSUPA, the cell is added to only the DCH active set.
Page 187
Handling of Event 1D
If the criteria for intra-frequency hard handover are fulfilled, the RNC performs intra-frequency hard handover after receiving the event 1D
report. If the criteria for intra-frequency hard handover are not fulfilled, the RNC performs intra-frequency soft handover based on the
measurement report.
If the uplink service is carried on the E-DCH, then:

 If the new best cell in the active set supports HSUPA and the HSPA hysteresis timer (HspaTimerLen) expires, the E-DCH serving cell
becomes the best cell. The HSPA hysteresis timer is restarted after the cell change.
If the new best cell in the active set does not support HSUPA,
− The uplink service is directed to the DCH through the reconfiguration. If the reconfiguration fails, the service is still carried on the E-DCH.
The UE is connected to the new best cell only on the DCH.
If the uplink service is carried on the DCH, then:

The uplink service is still carried on the DCH. After handover, the RNC may retry to configure the service on E-DCH.
If the current service is CS Voice over HSPA, then:

 If the new best cell supports both HSUPA and HSDPA, the cell is reconfigured to be an HSPA serving cell.
 Otherwise, the service is changed from CS Voice over HSPA to CS Voice over DCH through the reconfiguration.
Page 188
HSUPA Intra-Frequency Handover
Scenario1 of HSUPA Intra-frequency Handover

Cell 2 is the intra-frequency adjacent cell of Cell 1.
With the movement of UE, the signal of Cell 2 becomes

better and 1a or 1c event is triggered. Cell2 is added to the
active set. Through the RB reconfiguration, the service is
reconfigured from E-DCH to DCH.
Scenario2 of HSUPA Intra-frequency Handover
UE moves from Cell 1 to Cell 2. Cell 2 is the intra-frequency

adjacent cell of Cell 1.
With the movement of UE, when the signal of Cell 2 becomes

better, 1d event is reported. RNC reconfigures the service from
E-DCH to DCH through the RB.
Page 189
HSUPA Inter-Frequency Handover
Scenario1 of HSUPA Inter-frequency Handover
Assume that the UE moves from HSUPA cell 1 to HSUPA cell 2
and that the two cells are inter-frequency neighboring cells.
During this procedure, event 2B is triggered (Suppose event-
triggered measurement report is used). The HSUPA admission
request is made for the target cell by the RNC.
If the admission request is accepted, the handover is
performed through the physical channel reconfiguration
message.
If the admission request is rejected, the RB
reconfiguration is implemented, and the services on the E-
DCH are reconfigured for the DCH.

Assume that the UE moves from HSUPA cell 1 to non-HSUPA cell
2 and that the two cells are inter-frequency neighboring cells

triggered measurement report is used).
The DCH admission request is made for the target cell. After the
request is accepted, the RB reconfiguration is implemented, and
the services on the E-DCH are reconfigured for the DCH.
Page 190
HSUPA Inter-Frequency Handover

Assume that the UE moves from non-HSUPA cell 1 to HSUPA
cell 2 and that the two cells are inter-frequency neighboring
cells.

triggered measurement report is used). The inter-frequency
handover on DCH is performed, and the HSUPA admission
request is made for the target cell by the RNC.
After the request is accepted, the RB reconfiguration is

implemented, and the services on the DCH are reconfigured for
the E-DCH.
Page 191
HSUPA Mobility Parameters
This parameter determines the maximum number of

links in the EDCH active set. When the RNC acts as
the SRNC, the number of links in the EDCH active set
for all the UEs under the RNC cannot exceed the
parameter value. I
f the parameter value is too large, a lot of resources on

Max Number of Cell in EDCH
UHOCOMM MaxEdchCellInActiveSet the RAN side will be occupied as the same data is 1~4 1~4 None 3
Active Set
transferred over multiple EDCH links in macro diversity,
thus affecting the system performance. I
f this parameter value is too small, insufficient

combination gain can be achieved in macro diversity by
the EDCH, which causes excessive retransmissions
and affects the UE speed.
Page 192

UMTS/HSPA RRM Algorithms and Parameters Technical Training Part-II

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

UMTS/HSPA RRM Algorithms and Parameters Technical Training Part-II

Uploaded by

Copyright:

Available Formats

UMTS/HSPA RRM Algorithms

1 Overview of Radio Resource Management (RRM)

2 Idle Mode Procedures and Parameters

3 Random Access & Paging Procedures and Parameters

4 Power Control Algorithm and Parameters

5 Admission Control Algorithm and Parameters

6 Load Control Algorithm and Parameters

7 Channel Switching Algorithm and Parameters

8 Handover Algorithm and Parameters

9 HSDPA RRM Algorithm and Parameters

10 HSUPA RRM Algorithm and Parameters

LDR algorithm consists of three steps as following,

Basic Congestion Triggering

Assume that the parameters are set as follows:

Then, the basic congestion is relieved in the following sequence:

STEP (1) Basic Congestion Triggering

If the parameter NBMUlCacAlgoSelSwitch /

For an R99 cell:

For an HSPA cell:

In the downlink, when the cell does not support DC-HSDPA,

In the downlink, when the cell support DC-HSDPA,

For a path, the following congestion-related parameters are applicable:

Congestion detection parameters: FWDCONGBW , BWDCONGBW

Congestion resolving parameters:FWDCONGCLRBW , BWDCONGCLRBW

Then, the mechanism of congestion detection for the port is as follows:

NodeB Credit Resource

Type A: Basic congestion at local cell level

Type B: Basic congestion at local cell group level (if any)

Load Control Algorithm Switch LDC Algorithm Switch

Type A LC_CREDIT_LDR_SWITCH CELL_CREDIT_LDR

Type B LCG_CREDIT_LDR_SWITCH LCG_CREDIT_LDR

Type C NODEB_CREDIT_LDR_SWITCH NODEB_CREDIT_LDR

GUI Value Actual Value

NODEB_CREDIT_LDR_SWITCH: NodeB credit congestion control algorithm.

LCG_CREDIT_LDR_SWITCH: Local cell group credit congestion control

LC_CREDIT_LDR_SWITCH: Local cell credit congestion control algorithm.

UL_UU_LDR and DL_UU_LDR: UL /DL UU load reshuffling algorithm. When

GUI Value Actual Value Baseline

STEP (2) LDR Procedure

The RNC periodically takes actions if the basic congestion is detected.

Inter-frequency load handover

Iu QoS renegotiation

Mark "current LDR state = uncongested"

Start sending the LDM congestion indication LDR Procedure

No LDR action is taken Mark Page 13

GUI Value Actual Value Baseline

UCELLLDR DlLdrFirstAction DL LDR first action CodeAdj

UCELLLDR DlLdrSecondAction DL LDR second action InterFreqLDHO

UCELLLDR UlLdrSeventhAction UL LDR seventh action NoAct

GUI Value Actual Value Baseline

HO_ALGO_LDR_ALLOW_SH When the switch is on, the LDR inter-frequency

This switch is valid only when the inter-frequency

This parameter specifies load handover

Whether to perform blind handover.

The inter-frequency neighboring cell could be

Code resource usage difference threshold.

The UE can be selected to process load

The UE can be selected to process load

The algorithms with the above values represent

Specify the flags of the cells that the DRD measurement or

D250, D500, D1000,

RSCP Threshold for the target cell. This parameter is used