Customer UMTS RF TAN U03.03 Power Control

Lucent Technologies Proprietary
This document contains proprietary information of

Lucent Technologies and is not to be disclosed or used
except in accordance with applicable agreements.
Use pursuant to Company instructions.
Page 1 of 1

Subject: UMTS RF Translation Application Notes:
Power Control U03.03
Version 8.01 Date: June 2006

Abstract
This document provides with the description and implementation of power control
algorithms for the UMTS Network Release U03.03 and covers the relevant translation
parameters and their recommended values.

Power Control is in charge of the regulation of the transmitted power level of the Node B
and the User Equipment. It ensures that the quality of service required for every radio link
is achieved irrespectively of the radio channel conditions and that the interferences are
reduced.

The new feature introduced in this document mainly consists in 384k uplink bearers. The
Translation Application Notes specifically dedicated to HSDPA will contain information
related to the Power Control for that feature.

Subject:
UMTS RF Translation Application
Notes Power Control U03.03
Date:
June 2006

Lucent Technologies - Proprietary

Page 2 of 2

Version History
Version Date Changes
1.0 11.15.2003 Draft version U01.03
Author: Sbastien Fritsch
1.1 02.01.2004 Correction of the parameter values based on the latest Release
of U01.03.07.01 and opened MRs.
Insertion of the comments provided by Stefan Brueck
2.0 02.20.2004 Correction of some Glossary terms.
Insertion of the comments provided by Hussein Fawaz
Final Release of the document.
3.0 02.25.2004 Draft version U01.04
4.0 06.15.2004 Final version U01.04:
Set default Parameters values according to latest certified load
5.0 12.06.2004 Draft Version U02.01
7.0 06.01.2005 Draft Version U03.01
7.1 06.13.2005 Insertion of comments provided by reviewers.
8.0 05.15.2006 Version U03.03
8.01 05.17.2006 Change dLReferencePower from -20dB to -10dB (voice:
UEGTFCS=2)

Subject:
Date:
June 2006


Page 3 of 3

TABLE OF CONTENTS
1. GLOSSARY OF TERMS AND ABBREVIATIONS ..........................................................................6
2. REFERENCES...................................................................................................................................... 10
3. INTRODUCTION TO POWER CONTROL.................................................................................... 11
3.1. POWER CONTROL FOR DOWNLINK DEDICATED CHANNELS ......................................................... 11
3.1.1. Downlink Open Loop Power Control ...................................................................................... 11
3.1.2. Downlink Outer Loop Power Control ..................................................................................... 11
3.1.3. Downlink Inner Loop Power Control ...................................................................................... 12
3.1.4. Power Imbalance ..................................................................................................................... 12
3.1.5. DPCCH and associated DPDCHs........................................................................................... 13
3.1.6. Deactivation of the downlink inner loop power control ......................................................... 13
3.1.7. Downlink power control in compressed mode ........................................................................ 14
3.2. POWER SETTINGS FOR DOWNLINK COMMON CHANNELS ............................................................. 15
3.3. POWER CONTROL FOR UPLINK DEDICATED CHANNELS ............................................................... 17
3.3.1. Uplink Open Loop Power Control........................................................................................... 17
3.3.2. Uplink Outer Loop Power Control .......................................................................................... 17
3.3.3. Uplink Inner Loop Power Control........................................................................................... 17
3.3.4. Impact of Compressed Mode on uplink transmit power control............................................. 19
3.4. POWER CONTROL FOR UPLINK COMMON CHANNELS................................................................... 20
3.5. POWER OVERLOAD CONTROL ....................................................................................................... 20
4. POWER CONTROL FOR DOWNLINK DEDICATED CHANNEL........................................... 21
4.1. DOWNLINK DEDICATED CHANNEL FRAME STRUCTURE............................................................... 21
4.2. INITIAL TRANSMIT POWER FOR DOWNLINK DEDICATED CHANNEL............................................. 21
4.2.1. Downlink DCH Open Loop Power Control ............................................................................ 21
4.2.1.1. Downlink DCH Open Loop Power Control at Call set-up ............................................................21
4.2.1.2. Downlink DCH Open Loop Power Control in Cell_DCH state....................................................22
4.2.1.3. Translation Settings ........................................................................................................................23
4.3. OUTER LOOP POWER CONTROL FOR DOWNLINK DEDICATED CHANNELS ................................... 23
4.3.1. Downlink DCH Outer Loop Power Control Principle............................................................ 23
4.3.2. Translation Settings ................................................................................................................. 24
4.3.3. Outer Loop Power Control and User Elements ...................................................................... 24
4.4. INNER LOOP POWER CONTROL FOR DOWNLINK DEDICATED CHANNEL ...................................... 25
4.4.1. Downlink DCH Inner Loop Power Control Principle ............................................................ 25
4.4.2. SIR Estimation and TPC Command Generation..................................................................... 26
4.4.2.1. TPC bit patterns ..............................................................................................................................26
4.4.2.2. TPC pattern during Radio Link synchronization ...........................................................................27
4.4.2.3. Downlink Inner Loop Timing.........................................................................................................28
4.4.2.4. Downlink Inner Loop Power Control Modes.................................................................................29
4.4.3. Downlink Transmit Power Limits............................................................................................ 29
4.4.3.1. Principle ..........................................................................................................................................29
4.4.3.2. Transition between xx_inactive and xx_active states ....................................................................30
4.4.3.3. Downlink Inner Loop Power Control Algorithm and Execution ..................................................31
4.4.3.4. Downlink Inner Loop Transmit Power Step Size ..........................................................................32

Subject:
Date:
June 2006


Page 4 of 4

4.4.4. Downlink Dynamic Range ....................................................................................................... 33
4.4.4.1. Downlink Transmit power limits from the NodeB........................................................................33
4.4.4.2. Downlink Transmit power limits from the RNC ...........................................................................34
4.4.4.3. Downlink Transmit Power Limits in Softer Handover..................................................................34
4.4.4.4. Tuning of Pmax and Pmin..............................................................................................................34
4.4.4.5. Translation settings.........................................................................................................................35
4.4.5. Downlink power control in compressed mode ........................................................................ 36
4.4.5.1. Impact of compressed mode on the Power Control function.........................................................36
4.4.5.2. DL power step size in compressed mode .......................................................................................37
4.4.5.3. Calculation of P
SIR
(k)......................................................................................................................37
4.4.5.4. DL transmit power range aspects ...................................................................................................38
4.4.5.5. Impact of DL compressed mode on the behavior of the UE..........................................................38
4.4.6. Power Imbalance ..................................................................................................................... 39
4.4.6.1. Introduction.....................................................................................................................................39
4.4.6.2. Power Balance Procedure at the RNC............................................................................................41
4.4.6.3. Imbalance correction at the Drift RNC ..........................................................................................42
4.4.6.4. Estimation of P
bal
............................................................................................................................43
4.4.6.5. Limitation at the NodeB.................................................................................................................44
4.4.7. Power Offset between Downlink DPCCH and its DPDCHs .................................................. 46
4.4.7.1. Principle ..........................................................................................................................................46
4.4.7.2. Impact of PO1 and PO3 on the radio link ......................................................................................47
4.4.7.3. Dynamic management of the downlink power offset PO2............................................................47
4.4.8. Deactivation of the downlink ILPC ......................................................................................... 49
5. POWER SETTINGS FOR DOWNLINK COMMON CHANNELS.............................................. 51
5.1. INTRODUCTION .............................................................................................................................. 51
5.2. POWER CONTROL PROCEDURES .................................................................................................... 51
5.3. COMMON PILOT CHANNEL AND SYNCHRONIZATION CHANNEL POWER....................................... 52
5.4. PRIMARY COMMON CONTROL PHYSICAL CHANNEL POWER........................................................ 53
5.5. SECONDARY COMMON CONTROL PHYSICAL CHANNEL POWER................................................... 53
5.6. PAGING INDICATOR AND ACQUISITION INDICATOR CHANNELS POWER....................................... 55
5.7. DOWNLINK LOAD DUE TO COMMON CHANNELS POWER SETTINGS ............................................. 56
5.8. TRANSLATION SETTINGS ............................................................................................................... 57
6. POWER CONTROL FOR UPLINK DEDICATED CHANNELS................................................. 58
6.1. UPLINK DEDICATED CHANNELS FRAME STRUCTURE ................................................................... 58
6.2. INITIAL TRANSMIT POWER FOR UPLINK DEDICATED CHANNELS ................................................. 58
6.2.1. Uplink DCH Open Loop Power Control ................................................................................. 58
6.2.2. Uplink DPCCH Power Control Preamble .............................................................................. 59
6.3. OUTER LOOP POWER CONTROL FOR UPLINK DEDICATED CHANNELS........................................... 61
6.3.1. Uplink DCH Outer Loop Power Control Principle ................................................................ 61
6.3.2. Inter-RNC soft handover.......................................................................................................... 62

Subject:
Date:
June 2006


Page 5 of 5

6.3.3. Radio Link Quality Estimation ................................................................................................ 63
6.3.3.1. BLER metric ...................................................................................................................................63
6.3.3.2. Quality Estimation metrics .............................................................................................................63
6.3.4. Frame Selector Function ......................................................................................................... 66
6.3.5. Uplink Outer Loop Power Control Algorithm......................................................................... 66
6.3.5.1. Principle ..........................................................................................................................................66
6.3.6. Outer Loop Power Control Execution..................................................................................... 70
6.4. INNER LOOP POWER CONTROL FOR UPLINK DEDICATED CHANNELS.............................................. 72
6.4.1. Uplink Inner Loop Power Control Principle........................................................................... 72
6.4.2. SIR Estimation and TPC Commands Generation ................................................................... 72
6.4.2.1. SIR Estimation................................................................................................................................72
6.4.2.2. Transmit Power Control commands...............................................................................................73
6.4.3. Uplink Inner Loop Power Control Algorithms........................................................................ 74
6.4.3.1. Overview.........................................................................................................................................74
6.4.3.2. Algorithm 1 for processing TPC commands..................................................................................75
6.4.3.3. Algorithm 2 for processing TPC commands..................................................................................76
6.4.4. Uplink Inner Loop Power Control Execution ......................................................................... 78
6.4.4.1. Principle ..........................................................................................................................................78
6.4.4.2. Transmit Power Step Size...............................................................................................................79
6.4.4.3. Transmit power limits Maximum transmit power ......................................................................79
6.4.4.4. Transmit power limits Minimum transmit power .......................................................................79
6.4.5. Uplink DPCCH/DPDCH power gains .................................................................................... 80
6.4.5.1. Effect of bit rate variability on uplink power control ....................................................................80
6.4.5.2. Gain factors controlling procedures ...............................................................................................82
6.4.5.3. Translations settings .......................................................................................................................84
6.4.6. Impact of Compressed Mode on uplink transmit power control............................................. 86
6.4.6.1. Impact at the UE.............................................................................................................................86
6.4.6.2. Impact on the uplink SIR target......................................................................................................89
7. POWER CONTROL FOR UPLINK COMMON CHANNELS..................................................... 90
8. POWER OVERLOAD CONTROL ................................................................................................... 91
8.1. UPLINK OVERLOAD CONTROL........................................................................................................ 91

Subject:
Date:
June 2006


Page 6 of 6

1. Glossary of Terms and Abbreviations

3GPP 3
rd
Generation Partnership Project
AI Acquisition Indicator
AICH Acquisition Indicator Channel
AOC Aggregate Overload Control
BCCH Broadcast Control Channel
BCH Broadcast Channel
BER Bit Error Rate
BLER Block Error Rate
CFN Connection Frame Number
CM Compressed Mode
CN Core Network
CPICH Common Pilot Channel
CRC Cyclic Redundancy Checksum
CRCI CRC Indicator
C-RNC Control-RNC
CS Circuit Switched
DCH Dedicated Channel
DL Downlink
DPCCH Dedicated Physical Control Channel
DPCH Dedicated Physical Channel
DPDCH Dedicated Physical Data Channel
D-RNC Drift-RNC
DSP Digital Signal Processor
DTX Discontinuous Transmission
FACH Forward Access Channel
FP Framing Protocol
HSDPA High Speed Downlink Packet Access
IAOC Improved Aggregate Overload Control
ID Identity
IE Information Element
ILPC Inner Loop Power Control
ITP Initial Transmit Power
LMT Local Maintenance Tool / Terminal
MAC Medium Access Control
MAOC Modified Aggregate Overload Control
MIB Master Information Block
NBAP NodeB Application Part
NDP Network Data Provision
OAM Operation And Maintenance
OLPC Outer Loop Power Control
OMC Operations and Maintenance Centre
OMC-U OMC for UMTS
P-CCPCH Primary Common Control Physical Channel
PCH Paging Channel
PCA Power Control Algorithm
PCP Power Control Preamble
P-CPICH Primary CPICH

Subject:
Date:
June 2006


Page 7 of 7

PHY Physical layer
PICH Page Indicator Channel
PS Packet Switched
P-SCH Primary SCH
QE Quality Estimation
QoS Quality of Service
QPSK Quadrature (Quaternary) Phase Shift Keying
RAB Radio Access Bearer
RACH Random Access Channel
RF Radio Frequency
RLC Radio Link Control
RLS Radio Link Set
RNC Radio Network Controller
RNSAP Radio Network Subsystem Application Part
RPL Recovery Period Length
RPP Recovery Period Power Control Mode
RRC Radio Resource Control
RSCP Received Signal Code Power
SAE System Architecture and Performance Engineering
S-CCPCH Secondary Common Control Physical Channel
S-CPICH Secondary CPICH
SER Symbol Error Rate
SIB System Information Block
SIR Signal-to-Interference Ratio
S-RNC Serving RNC
S-SCH Secondary Synchronization Channel
TFC Transport Format Combination
TFCI Transport Format Combination Indicator
TFCS Transport Format Combination Structure
TPC Transmit Power Control
UE User Equipment
UEGTFC User Equipment Group Transport Format Combination
UEGTFCS User Equipment Group Transport Format Combination Set
UL Uplink
UMTS Universal Mobile Telecommunications System
UTRAN UMTS Terrestrial Radio Access Network

Note on the UMTS ParCat Attributes:
RNC-CLI the parameter is a local RNC parameter accessible via CLI script
RW-EL read-write parameter only when the top-level object locked
RW-LO read-write parameter only when locked
RW-PL read-write parameter only when parent object locked
OMC the parameter is accessible via OMC-U
RW-LO read-write parameter only when locked
RW-PL read-write parameter only when parent object locked
Note on the Parent Objects with Multiple Instances:

Subject:
Date:
June 2006


Page 8 of 8

The parameters related to power control parameters can be defined differently either on a
radio bearer or service combination basis. The instances related to defined radio bearer
(via object DedicatedTransportChannel) and service combination (via object UEGTFCS)
are reported in Table 2 and Table 1.

Logical Object
UEGTFCS Instances
Service Combination
Uplink
Service Combination
Downlink
1 3.4kbps signalling alone 3.4kbps signalling alone
2 3.4kbps signalling + speech 12kbps 3.4kbps signalling + speech 12kbps
3 3.4kbps signalling + CS 64kbps 3.4kbps signalling + CS 64kbps
4 3.4kbps signalling + PS 64kbps 3.4kbps signalling + PS 64kbps
6 3.4kbps signalling + PS 64kbps
+ speech 12kbps
3.4kbps signalling + PS 64kbps
+ speech 12kbps
+ speech 12kbps
+ speech 12kbps
+ speech 12kbps
+ speech 12kbps
+ speech 12kbps
+ speech 12kbps
19 3.4kbps signalling + PS 64kbps 3.4kbps signalling + HSDPA
+ speech 12kbps
3.4kbps signalling + HSDPA
+ speech 12kbps
28
+ speech 12kbps
+ speech 12kbps
34
+ speech 12kbps
+ speech 12kbps
35
+ speech 12kbps
+ speech 12kbps
36
+ speech 12kbps
+ speech 12kbps

Subject:
Date:
June 2006


Page 9 of 9

Logical Object
UEGTFCS Instances
Service Combination
Uplink
Service Combination
Downlink
51
+ speech 12kbps
+ speech 12kbps
52
+ speech 12kbps
+ speech 12kbps
53
+ speech 12kbps
+ speech 12kbps
54
+ speech 12kbps
+ speech 12kbps
Table 1: UEGTFCS Instances and associated service combination.

Object number Defines
1 Standalone Signalling 3.4kbps
2 Speech 12kbps RAB sub flow #1
5 64kbps CS
6 64kbps PS
7 384kbps PS
9 128kbps PS
19 DL HSDPA UL 128k PS Int/BG
20 DL 64k UL 128k PS Int/BG
30 DL HSDPA UL 384k PS Int/BG
Table 2:DedicatedTransportChannel instances and associated radio bearer.

Subject:
Date:
June 2006


Page 10 of 10

2. References
[1] 3GPP TS25.211 Physical channels and mapping of transport channels onto physical channels
(FDD)
[2] 3GPP TS25.214 Physical Layer Procedures (FDD)
[3] 3GPP TS25.215 Physical Layer; Measurements (FDD)
[4] 3GPP TS25.101 "User Equipment (UE) Radio transmission and reception (FDD)"
[5] 3GPP TS25.104 "Base Station (BS) Radio transmission and reception (FDD)"
[6] 3GPP TS25.133 "Requirements for Support of Radio Resource Management (FDD)"
[7] 3GPP TS25.331 Radio Resource Control (RRC) protocol specification
[8] 3GPP TS25.423 UTRAN Iur interface Radio Network Subsystem Application Part (RNSAP)
signalling
[9] 3GPP TS25.427 UTRAN Iub/Iur interface user plane protocols for DCH data streams
[10] 3GPP TS25.433 UTRAN Iub interface NBAP Specification
[11] 3GPP TS25.321 Medium Access Control (MAC) protocol Specification
[12] 3GPP TS34.108 Common Test Environments for User Equipment (UE) Conformance Testing
[13] UMTS RF Translation Application Notes: Load Control Algorithms
[14] UMTS RF Translation Application Notes: Handover
[15] UMTS RF Translation Application Notes: Access Procedures
[16] UMTS RF Translation Application Notes: HSDPA

Subject:
Date:
June 2006


Page 11 of 11

3. Introduction to Power Control

Power Control is implemented in the Downlink Dedicated Channels, Downlink Common
Channels, Uplink Dedicated Channels and Uplink Common Channels.
At radio link establishment, the transmit power level is estimated without any information
from the counter part. This procedure is known as open loop power control.
Once a link is established, the power is continuously optimized using a loop including
UE, NodeB and RNC. This procedure is known as closed loop and is split into two sub-
control loops, the inner loop and the outer loop.

3.1. POWER CONTROL FOR DOWNLINK DEDICATED CHANNELS
3.1.1. Downlink Open Loop Power Control
At radio link establishment, the initial downlink transmit power of the DCH is determined
by the open loop power control, depending on whether the UE has already another
existing radio link or not.
The parameters related to the downlink open loop power control are listed in Table 3.
Top Object:
Parent Object
Parameter name Short Description Default
EcNotarget
(prev. slRTarget)
Used for the calculation of the initial transmit power of
the first radio link.
dLPowerMargin This margin is used in order to allow changes in the
linkbudget between the time the CPICH measurements
are performed at the UE and the time the first radio link
is active.
UEGTFCS:
DLDPCHInfo
newLegPowerOffset Used for the calculation of the initial transmit power of
the radio links other than the first one.
See
4.2.1.3
Table 3: Power control at radio link establishment downlink related parameters

3.1.2. Downlink Outer Loop Power Control
The downlink outer loop power control sets the target Signal to Interference Ratio that
needs to be achieved at the UE to ensure the performances, in terms of Block Error Rate,
requested for the downlink DCH. The parameter is described in Table 4 and can be set on
a per radio bearer basis.
Parent Object Parameter name Short Description Default
DedicatedTransportChannel bLERQualityTarget It provides the Downlink BLER
target
. See 4.3.2
Table 4: Downlink outer loop power control - Performance target parameter

Subject:
Date:
June 2006


Page 12 of 12

3.1.3. Downlink Inner Loop Power Control

The inner loop power control generates the transmit power control commands based on
the SIR
target
provided by the outer loop power control. This ensures a fast adaptation of the
NodeB transmit power for the considered downlink DCH.
Different downlink power control modes can be supported by the UE to reduce the rate at
which the NodeB adjust the power based on the TPC commands.
In order to restrict the downlink interference, a maximum and minimum transmit power
level with respect to the P-CPICH are defined per radio link for the downlink DCH
irrespectively of the TPC commands.
The minimum downlink power level can also be used e.g. to artificially maintain the call
at a higher-than-needed level in order to improve the performances of the system (reduce
call drop rates etc).
The parameters controlling the downlink inner loop power control are listed in Table 5.

Top Object:
Parent Object
Parameter
name
Short Description Default
DedicatedTran
sportChannel
limitedDLPower
Increase
Indicates whether DL power increase limitation is
allowed for the DPDCH where it is mapped
PowerRaiseLimit Parameter for the DL Power Increase Limitation
algorithm.
LCell
DLPowerAvWin
dowSize
Parameter for the DL Power Increase Limitation
algorithm.
tPCStepSize It defines the adjustment step specified in the
downlink power control commands.
See 4.4.3.5
dPCMode It defines the Downlink Power Control modes
supported by the UE.
See 4.4.2.5
minDLPower Minimum downlink power level for a given code
power.
UEGTFCS:
DLDPCHInfo
maxDLPower Maximum downlink power level for a given code
power.
PminCorrectionF
actor
Local correction to the value of minDLPower used
to define the DL dynamic range of dedicated radio
links on a cell basis.
LCell
PmaxCorrection
Factor
Local correction to the value of maxDLPower
used to define the DL dynamic range of dedicated
radio links on a cell basis.
See 4.4.4.5
Table 5: Downlink inner loop power control - DPC mode parameter.
3.1.4. Power Imbalance
In some conditions the error rate of the power control command detection becomes high
on the weakest link of a connection. When such an error occurs, the power control

Subject:
Date:
June 2006


Page 13 of 13

command may be interpreted as a Power_Up (wrong value) whilst on the other leg(s) it
is interpreted as a Power_Down (correct value); this results in downlink DCH power
levels diverging by twice the power adjustement step size (e.g. 2dB). This effect is known
as power imbalance. The parameters involved in the procedure to correct it are listed in
Table 6.

Top Object:
Parent Object
ImbalanceCorrection Defines if imbalance correction of downlink
transmit power should be activated or not.
maxAdjustmentStep Defines a time period, in terms of number of
slots, in which the accumulated power
adjustment shall be maximum 1dB. This value
does not include the DL inner loop PC
adjustment.
adjustmentPeriod Maximum time allowed for the application of
the balance correction which value has been
calculated at the beginning of the correction
period.
scaleAdjustmentRatio Defines the convergence rate used for the
associated Adjustment Period.
UEGTFCS:
DLDPCHInfo
dLReferencePower Reference Power level for imbalance correction
algorithm
See
4.4.6.6
Table 6: Power Imbalance
3.1.5. DPCCH and associated DPDCHs
The downlink transmit power control procedure controls simultaneously the power of a
Dedicated Physical Common Control Channel and the corresponding Dedicated Physical
Data Control Channels. The power of the DPCCH is adjusted with respect to the power of
DPDCHs using the parameters listed in Table 7.

3.1.6. Deactivation of the downlink inner loop power control
The downlink Inner Loop Power Control can be deactivated at the NodeB for a group of
radio links belonging to the same Node B communication context. This feature is not
expected to be used in the Lucent Technologies RRM strategy. However the message
could be generated by a non-Lucent Technologies Drift-RNC or could be of some use for
test purposes. This is done using the parameter innerLoopPCStatus described in Table 8.

Subject:
Date:
June 2006


Page 14 of 14

Top Object:
Parent Object
pO1 Power offset for the TFCI bits.
Values greater than zero for TFCI power offset
may improve the BLER performances.
pO2ForOneRLSet Power offset for the TPC bits when there is only
one radio link set connected to the UE.
Values greater than zero for this offset may
reduce the TPC command error detection rate.
pO2ForTwoRLSet Power offset for the TPC bits when there are
two radio links set connected to the UE.
pO2ForThreeRLSet
AndMore
Power offset for the TPC bits when there are
three radio links set connected to the UE.
UEGTFCS:
DLDPCHInfo
pO3 Power offset for the pilot bits.
Current default value for pilot power offset
improves the UE synchronization and increases
the downlink capacity gain.
See 4.4.7.4
Table 7: Downlink inner loop power control - Power offsets parameters

Top Object:
Parent Object
UEGTFCS:
DLDPCHInfo
innerLoopPCStatus Activate / deactivated the inner loop power
control feature.
See 4.4.8
Table 8: Activation / Deactivation of the Inner Loop Power Control

3.1.7. Downlink power control in compressed mode
The downlink power control procedure in compressed mode has for purpose to recover as
fast as possible a signal-to-interference ratio close to the E
c
N
otarget
after each transmission
gap. The UE is behaving on a similar way as it does in non-compressed mode, generating
the TPC commands based on a comparison of the DL E
c
N
oestimate
and the DL E
c
N
otarget.

However during the radio frames when compressed mode idle period takes place an
alternative E
c
N
o
target value is used. This is calculated using a new variable DELTA SIR,
which is calculated based on the new parameters delta SIR1, delta SIR1After, delta SIR2
and deltaSIR2After introduced in Table 9.

Subject:
Date:
June 2006


Page 15 of 15

Parent Object Parameter
name
Value
deltaSIR1 Delta SIR target value to be set in the NodeB
during the frame containing the start of the first
transmission gap in the transmission gap pattern
(without including the effect of the bit-rate
increase). This Delta will be applied to the
uplink/downlink transmission power control
procedure.
deltaSIRAfter1 Delta SIR target value to be set in the Node-B one
frame after the frame containing the start of the
first transmission gap in the transmission gap
pattern.
deltaSIR2 Delta SIR target value to be set in the Node-B
during the frame containing the start of the second
increase). When omitted, DeltaSIR2 = DeltaSIR1.
LRNC:
Transmission
GapPattern
deltaSIRAfter2 Delta SIR target value to be set in the Node-B one
second transmission gap in the transmission gap
pattern. When omitted, DeltaSIRAfter2 =
DeltaSIRAfter1
See
4.4.5.6
Table 9 Compressed Mode Parameters

3.2. POWER SETTINGS FOR DOWNLINK COMMON CHANNELS
The following common channels for physical downlink are supported in U03.03:
Common Pilot Channel with sub-channels Primary CPICH (P-CPICH)
Primary Common Control Physical Channel (P-CCPCH) that carries the BCH
Secondary Common Control Physical Channel (S-CCPCH) that carries the FACH
and PCH.
Synchronization Channel (SCH) with sub-channels Primary SCH and Secondary
SCH.
Paging Indicator Channel PICH
Acquisition Indicator Channel AICH
The power level of all the downlink common channels is controlled using the parameters
listed in Table 10.

Subject:
Date:
June 2006


Page 16 of 16

Top Object:
Parent Object
pCPICHpower Primary CPICH power, in dBm
The setting strongly depends on the cell planning.
This setting influence the power of the other DL
common channels, the DL DCH power limits values
and the IAOC scaling factors. CPICH power has also
strong impact on synchronization as well as on
soft/softer handover procedures.
pSCHpower Primary SCH power, in dB with respect to P- CPICH
power
Current setting ensures high success rate in different
steps of cell search/synchronization procedure as
well as low DL power consumption (duty cycle
10%).
sSCHpower Secondary SCH power, in dB with respect to P-
CPICH power. Current setting ensures high success
rate in different steps of cell search/synchronization
procedure as well as low DL power consumption
(duty cycle 10%).
LCell
bCHPower BCH power , in dB with respect to P-CPICH power
Current setting ensures high success rate in decoding
SIB and MIB messages as well as low DL power
consumption (duty cycle 90%).
See 5.8
secondaryCCPCH.
powerOffset1
Power offset of the Secondary CCPCH TFCI field
relative to the transmission power of the data part
(PCH or FACH). Values greater than zero for TFCI
power offset may improve the BLER performances
secondaryCCPCH.
powerOffset2
Power offset of the Secondary CCPCH pilot field
relative to the transmission power of the data part
(PCH or FACH).
Values greater than zero for pilot power offset may
help the UE to achieve synchronization and perform
a better reception of the S-CCPCH.
pCHPower PCH power, in dB with respect to P-CPICH power
LCell:
SCCPCHPch
pICHPower PICH power, in dB with respect to P-CPICH power
fACHTrafPower FACH power, in dB with respect to P-CPICH power.
fACHSigPower FACH power, in dB with respect to P-CPICH power
secondaryCCPCH.
powerOffset1
Same as for the SCCPCHPch
LCell:
SCCPCHFach
secondaryCCPCH.
powerOffset2
Same as for the SCCPCHPch
LCell:
PRACH
aICHPower Defines the output power level of the AICH physical
channel, in dB with respect to P-CPICH power.
Current setting ensure successful decoding at the UE
of AI messages along RACH Access procedures.
See 5.8
Table 10: Downlink common channels power parameters

Subject:
Date:
June 2006


Page 17 of 17

3.3. POWER CONTROL FOR UPLINK DEDICATED CHANNELS
As described for the downlink channels, an open loop and a closed loop power control are
used for the uplink dedicated channels.
3.3.1. Uplink Open Loop Power Control
During the radio link establishment the open loop control procedure sets an initial
transmitted power in the UE for the uplink dedicated channel. Additionally a power
control preamble following the transmission of uplink DPCCH can be used to make sure
that the ILPC has converged before the start of the uplink DPDCH transmission. The
related parameters are included in Table 11.
3.3.2. Uplink Outer Loop Power Control
In the NodeB(s), the uplink outer loop sets and adjusts the SIR
target
in order to achieve the
performance targeted for the uplink DCH. This target is expressed in terms of Block Error
Rate. The method used to calculate the optimal SIR target value for providing the
requested BLER is Lucent Technologies proprietary and the parameters related to this
algorithm are listed in Table 12. Except for the first parameter all the other parameters are
set on a per radio bearer basis.

Top Object:
Parent Object
dPCCHPowerOffset Used to determine initial transmit power of a new
DTCH. Value expressed in dB.
UEGTFCS:
ULDPCHInfo
pCPreamble Indicates length of power control preamble.
Expressed in number of radio frames.
No PC preamble is currently used to delay uplink
DPDCH transmission after uplink DPCCH
transmission. The reason is that some UEs crash if
the value is other than 0.
See 6.2.3
Table 11: Power control at radio link establishment uplink related parameters.
3.3.3. Uplink Inner Loop Power Control
The uplink inner loop power control adjusts the transmit power of the UE in order to
achieve the SIR
target
provided by the uplink outer loop for a considered uplink DCH.
The NodeB generates transmit power control commands that are sent to the UE.
Irrespectively of the commands, the transmit power of the UE must remain within a
power range defined by a minimum power and the maximum power. The minimum
power depends only on the UE capability while the maximum power is specified by the
RNC per radio bearer combination. There are two algorithms supported by the UE to
combine different TPC commands when in soft/softer handover into a single one. The
related parameters are listed in Table 13.

Subject:
Date:
June 2006


Page 18 of 18

Top Object:
Parent Object
qESelector It defines the metric to be selected for radio
link quality estimation.
This settings means that the selected metric
for radio link quality estimation is the
Physical Channel BER.
See
6.3.3.2.3
downSIRStep It defines the uplink SIRtarget decrement
step in dB.
Current values ensure quick OLPC reaction
in highly variable RF environment without
causing spikes in SIRtarget.
upSIRStep It defines the uplink SIR target increment
step in dB.
Current values ensure quick OLPC reaction
in highly variable RF environment without
causing spikes in SIRtarget.
qEThreshold It defines the QE threshold for allowing
decrement of SIRtarget in dB.
This value disables the check in OLPC on
the Quality Estimation, thus OLPC decision
on SIRtarget calculation are based only on
CRC check results.
sIRAllowance It defines the hysteresis factor used to
restrict SIR target update in dB.
sIRDownAdditional Extra UPLINKSIR target decrement when a
Transport block is received with CRCI =
ok. Used when High Speed SIR Target
Decrease is active.
DedicatedTransport
Channel
NbOfConsecutiveGo
odFramesThreshold
Threshold for activation of High Speed SIR
Target Decrease
See 6.3.6
initialSIRTarget Initial SIRtarget for the considered radio
bearer expressed in dB.
Setting depends on power imbalance and
DL OLPC issues. Should be further
optimized in order to minimize the
convergence time, hence the uplink power
consumption & noise rise during the
transient period.
minSIRTarget Minimum SIR target for considered radio
bearer in dB.
UEGTFCS:
ULDPCHInfo
maxSIRTarget Maximum SIR target for considered radio
bearer in dB.
See 6.3.5.2
Table 12: Uplink outer loop power control parameters.

Subject:
Date:
June 2006


Page 19 of 19

Top Object:
Parent Object
tPCStepSizeUL Adjustment step specified in the
uplink power control commands 0
means 1dB, 1 means 2dB steps.
maxULDPCHPower This is the highest permitted value
ensuring maximum uplink
coverage. It may be decreased in
case of uplink interference issues.
See 6.4.4.5
UEGTFCS:
ULDPCHInfo
PowerControlAlgorithm It specifies the algorithm to be
used by UE to interpret TPC
commands
See 6.4.3.4
Table 13: Uplink inner loop power control - TPC related parameters.

Top Object:
Parent Object
gainFactorBc It defines the DPCCH gain given for the highest
rate of the Transport Format Combination
Structure (TFCS).
Current values help to mitigate the power
imbalance effects by increasing both DPCCH and
DPDCH performances without causing high
uplink power consumption.
gainFactorBd It defines the DPDCH gain given for the highest
rate of the TFCS.
Current values help to mitigate the power
imbalance effects by increasing both DPCCH and
DPDCH performances without causing high
uplink power consumption
gainFactorType It defines the method to either directly signalling
or computing the gain factors.
UEGTFC
referenceTFC It defines the reference number in case computed
method is used.
See 6.4.5.3
Table 14: Uplink inner loop power control - DPCCH/DPDCH power gains parameters.
In order to make the quality of service provided to a DPCH independent from the instant
data rate, power gains between the uplink DPDCH and DPCCH have been introduced for
every uplink service combination. The parameters are listed in Table 14.

3.3.4. Impact of Compressed Mode on uplink transmit power control
As in the Downlink the principle of the Power Control in the uplink when in compressed
mode is not fundamentally different from the normal case. The power will be adapted
before and after the transmission gaps in order to ensure the continuity of the service and
the recovery. The SIR Target will, as for the Downlink, be adapted during the
transmission gaps using new parameters as defined in Table 15

Subject:
Date:
June 2006


Page 20 of 20

name
Value
deltaSIR1 Delta SIR target value to be set in the UE during
the frame containing the start of the first
increase). This Delta will be applied to the
uplink/downlink transmission power control
procedure.
deltaSIRAfter1 Delta SIR target value to be set in the UE one
first transmission gap in the transmission gap
pattern.
deltaSIR2 Delta SIR target value to be set in the UE during
the frame containing the start of the second
increase). When omitted, DeltaSIR2 = DeltaSIR1.
LRNC:
Transmission
GapPattern
deltaSIRAfter2 Delta SIR target value to be set in the UE one
second transmission gap in the transmission gap
pattern. When omitted, DeltaSIRAfter2 =
DeltaSIRAfter1
See
4.4.5.6

3.4. POWER CONTROL FOR UPLINK COMMON CHANNELS
The power control procedure for the uplink common channels applies only to the RACH
channel since this is the only uplink common channel supported in U03.03. All the
RACH procedures including the power control mechanism are covered in [15].

3.5. POWER OVERLOAD CONTROL
A power overload control function has been introduced in order to keep the total
downlink transmitted power level in the short term within an acceptable range. This
function is located in the NodeB and is implemented by the use of the Improved
Aggregate Overload Control procedures. These functions are controlled by NodeB
internal parameters and by the UTRAN parameter listed in Table 16.

Top Object:
Parent Object
LCell
maxTransmissionPower Maximum downlink power, in dBm allowed
in the cell.
43.0dBm
Table 16: Power overload control parameter.

Subject:
Date:
June 2006


Page 21 of 21

4. Power Control for Downlink Dedicated Channel
4.1. DOWNLINK DEDICATED CHANNEL FRAME STRUCTURE
As specified in [1] the Dedicated Physical Channel is the only type of downlink dedicated
physical channel. Within one downlink DPCH, dedicated data generated at RLC/MAC
Layer and upper layers, i.e. the Dedicated Transport Channel, is transmitted in time-
multiplex with control information generated at Physical Layer. This control information
includes known pilot bits, the Transmit Power Control commands and an optional
Transport Format Combination Indicator. Figure 1 shows the frame structure of the
downlink DPCH.

One radio frame, T
f
= 10 ms
TPC
N
TPC
bits
Slot #0 Slot #1 Slot #i Slot #14
T
slot
= 2560 chips, 10*2
k
bits (k=0..7)
Data2
N
data2
bits
DPDCH
TFCI
N
TFCI
bits
Pilot
N
pilot
bits
Data1
N
data1
bits
DPDCH DPCCH DPCCH
Figure 1: Downlink DPCH frame structure.

Each frame of length 10 ms is split into 15 slots, each of length T
slot
= 2560 chips,
corresponding to one power-control period. Therefore the maximum rate of power control
commands is equal to 1500 Hz.
The parameter k in Figure 1 determines the total number of bits per downlink DPCH slot.
More information concerning DPCH frame structure can be found in [1].
4.2. INITIAL TRANSMIT POWER FOR DOWNLINK DEDICATED CHANNEL
4.2.1. Downlink DCH Open Loop Power Control
4.2.1.1.Downlink DCH Open Loop Power Control at Call set-up
At call set-up, the RNC creates a new link for a UE without any existing radio link. The
downlink transmit power is derived using measurements reported by the UE and RNC
parameters.
In particular the RNC extracts from the RRC Connection Request message the primary
measurement reported by the UE and then calculates the initial downlink transmit power
according to (1)

Subject:
Date:
June 2006


Page 22 of 22

.
ARGIN DL_POWER_M /I CPICH_E - EcNoTarget
, P
, P
Min = power TX DL Initial
o c
Min
Max
+
Max

(1)
Where
Initial DL TX Power is expressed in dB with respect to the P-CPICH power
P
Min
and P
Max
define the minimum and maximum allowed downlink transmitted
power for the considered radio bearer configuration. They are expressed in dB
with respect to the P-CPICH power
EcNoTarget is referred to the considered radio bearer configuration and expressed
in dB.
CPICH_Ec/Io is the primary measurement reported by the UE within the RRC
Connection Request message and is expressed in dB upon proper mapping as
specified in [6]
DL_Power_Margin is a safety margin value allowing some change in the link
budget between the time the measurement is performed and the time the radio link
is actually active. It is expressed in dB

As described in [14] the UE in idle state receives the measurement quantity to be
measured via SIB11 message broadcasted on the BCCH.
In the case where there is no CPICH Ec/Io measurement available in the RRC Connect
Request or the Cell Update message, the maximum allowed transmit power level should
be set to P
Max
.

4.2.1.2.Downlink DCH Open Loop Power Control in Cell_DCH state
When the RNC adds a new radio link to existing ones (i.e. UE in Cell_DCH state moves
into soft/softer handover mode), the NodeB transmit power is calculated according to (2).
Initial DL TX Power = P
Min
+ P
new_leg_offset
(2)
Where:
P
Min
is the minimum downlink transmit power allowed at the NodeB on for
the considered UE
P
new_leg_offset
is an offset value read from the RNC MIB

The initial downlink transmitted power information is provided to the NodeB via NBAP
signalling used during e.g. Radio Link Setup or Radio Link Addition procedures. Refer to
[10] for more information.

Once the NodeB receives any of these NBAP messages, it applies the given power on
each downlink Channelization Code of the radio link when starting transmission. Upon

Subject:
Date:
June 2006


Page 23 of 23

reception of a Radio Link Addition Request message, if no initial downlink Transmission
Power is included, the NodeB shall use the power level currently used for the considered
UE.

4.2.1.3.Translation Settings
The parameters defined in chapters 4.2.1.1 and 4.2.1.2 are summarized in Table 17. The
range of these parameters is indicated as well as their default value. Since the top object
for all these parameters is UEGTFCS they can all be set with different values depending
on the service combination supported in the downlink.

Parent
Object
Parameter name Attribute Syntax
& Range
UEGTFCS
Instance
Default
Value
Access
1, 2, 3, 4, 5,
6, 7, 9, 11,
13, 14
-19.0dB
31, 32, 33,
34, 35, 50,
51,52,54
-18.0dB
EcNoTarget
(prev. slRTarget)
Real - [dB]
-24.0, -23.9...+1.5
36, 49, 53 -15.0dB
All +3dB dLPowerMargin Real - [dB]
-10, -9.9, +10 13, 14 0
1, 2, 4, 5, 6, 7
9, 11
+5dB
DLDPCHInfo
newLegPowerOffset Real - [dB]
0, 0.1, .25.5
All others +6dB
RNC-CLI
RW-EL
Table 17: Downlink open loop power control parameters

4.3. OUTER LOOP POWER CONTROL FOR DOWNLINK DEDICATED CHANNELS
4.3.1. Downlink DCH Outer Loop Power Control Principle
The downlink Outer Loop Power Control is in charge of updating the SIR
target
in order to
provide the Block Error Rate performances corresponding to the requested Quality of
Service. This control loop runs autonomously in the UE with a maximum speed of
100Hz. The principle is depicted in Figure 2.
The target is sent by the RNC to the UE via RRC signalling, using the downlink dedicated
control channel.
From this provided BLER value the UE derives the initial SIR
target
value. The instant
BLER is continuously monitored and the SIR
target
is re-adjusted in order to ensure the
requested quality is achieved in any radio channel scenario (speed, environment, etc.).
The method on how to set SIR
target
in order to provide the requested BLER is not specified
in the 3GPP standard. The algorithm is specific to the terminal manufacturers. Minimal
UE performances in given RF conditions are specified in [4].

Subject:
Date:
June 2006


Page 24 of 24

UE
Node B
Link quality estimate:
BER, BLER, SIR,
channel, speed ...
Inner
Loop
PC
SIR
target
= f(link quality)
RNC
Pmax, Pmin
= f(QoS, load)
BLER
target
,
Measurements:
BLER, SIR
Overall Tx power,.
up/down
data
NBAP
NBAP
DCCH
DCCH
DL Outer Loop PC

Figure 2: Downlink outer loop power control principle
4.3.2. Translation Settings
The parameter defining the BLER target value is named bLERQualityTarget. This
parameter is defined per radio bearer; i.e. one for downlink signalling, and one for
downlink data service. The conversion between the real values provided in the OAM
format and the typical values of the BLER is provided in (3) and (4). The basic
information is reported in Table 18.

Parent Object Parameter name Attribute
Syntax
& Range
Parent
Object
Instances
Default Value Access
1 -2 (1%)
2, 3, 4 -2.1 (0.7%)
5 -3 (0.1%)
Dedicated
TransportChannel
bLERQualityTarget Real
-6.3, -6.2 0
All others -1.3 (5%)
RNC-CLI
RW-EL
Table 18: Downlink Outer loop Power control - BLER target parameter

|
.
|
\
|
=
100
Value
Value
%
10 OAM
Log (3)
OAM
Value
%
10 100 Value = (4)
4.3.3. Outer Loop Power Control and User Elements
During optimization activities it has been observed that some UEs have Outer Loop
Power Control not properly working.

Subject:
Date:
June 2006


Page 25 of 25

This means that the SIR
target
is not updated according to the BLER measured on the
downlink DCH, i.e. the Outer Loop provides SIR
target
values to the Inner Loop based
apparently on a look-up table.

This results in following issues:
In good RF conditions (i.e. pilot Ec/Io better than 7dB) the UE requests higher
downlink DCH power than required to maintain around 0-2% BLER which is
lower than the BLER
target
of 5% defined via parameter bLERQualityTarget.
In poor RF conditions, the SIR
Target
is not increased in order to maintain the target
BLER. As a consequence, the downlink BLER shoots up even if downlink DCH
power is available.
Irrespective of the downlink BLER calculated, the UE requests higher downlink
power than required which leads to higher downlink DCH power consumption and
reduced downlink capacity.
This misbehavior is strongly impacting the current setting for Uplink/Downlink Closed
Loop power control parameters.

4.4. INNER LOOP POWER CONTROL FOR DOWNLINK DEDICATED CHANNEL
4.4.1. Downlink DCH Inner Loop Power Control Principle
The Inner Loop Power Control purpose is fast adaptation of the NodeB transmit power in
order to achieve a targeted signal to interference ratio for the considered downlink radio
channel.
Because of the speed of the control loop (up to 1500 Hz), the only elements involved in
the inner loop power control are the UE and the NodeB. Figure 3 describes the principle
of downlink inner loop power control.
UE node B
TPC = Power_up
or Power_down
transmit power
=
f(TPC,deltaP,
Pmin, Pmax,
....)
compare
with SIR
target
SIR estimate
TPC
c
data

Figure 3: Downlink inner loop power control principle

Subject:
Date:
June 2006


Page 26 of 26

The Inner Loop Power Control procedures can be separated in four basic steps:
1. The NodeB transmits user data on the downlink DCH to the UE.
2. The UE performs the SIR estimation and compares the SIR
est
with the SIR
Target
.
3. Depending on the results of the comparison, the UE uses a dedicated uplink
control channel conveying, the Transmit Power Control commands to the NodeB.
The rate of these commands is up to 1500 Hz.
According to the information included in the TPC command, the NodeB decides to either
increase or decrease its transmit power. Irrespective of the TPC commands the NodeB
transmit power must remain within power ranges defined by [P
min ;
P
max
].
4.4.2. SIR Estimation and TPC Command Generation
For each received downlink DCH radio frame slot (2560 chip) the UE performs the SIR
estimation. The measured value SIR
est
is the pilot bits of the DPCCH to noise ratio. The
measurement technique is not specified in the 3GPP standard but is UE vendor specific.
Some performance requirements are defined in [6].
Based on the SIR
est
value, the UE generates TPC commands according to the following
rule:

If SIR
est
>= SIR
target

Then
TPC command = "0" [requesting a NodeB transmit power decrease]
Else
TPC command = "1" [requesting a NodeB transmit power increase]
4.4.2.1.TPC bit patterns
Depending on the slot format the TPC command may take one of two TPC bit patterns
presented in Table 19. This bit pattern is derived from the Uplink DPCCH Slot Format
information element transported in the Radio Link Setup Request and Radio Link Setup
Reconfiguration NBAP messages.

.TPC Bit Pattern Transmitter power control
command
N
TPC
= 1
(TPC_BIT_PATTERN = 0)
N
TPC
= 2
(TPC_BIT_PATTERN = 1)

1
0
11
00
1
0
Table 19:TPC bit pattern and transmitter power control command.

The relation between the slot format value and the TPC_BIT_PATTERN values, N
TPC
,
are defined in Table 20 extracted from [1].

Subject:
Date:
June 2006


Page 27 of 27

Slot Format #i Channel Bit
Rate (kbps)
Channel Symbol
Rate (ksps)
SF N
TPC
N
TFCI
N
FBI

0 15 15 256 2 2 0
1 15 15 256 2 0 0
2 15 15 256 2 2 1
3 15 15 256 2 0 1
4 15 15 256 2 0 2
5 15 15 256 1 2 2
Table 20:Slot format value and DPCCH fields
The TPC commands are sent in TPC field of the uplink DPCCH. Only slot format number
0 is used and therefore N
TPC
equal to 2 is always used with the corresponding TPC bit
pattern.
The reason to have only one slot format is due to the limited support of closed loop
diversity (therefore no FBI bits is needed) and no support of uplink blind rate detection
(therefore TFCI bits are always needed).

4.4.2.2.TPC pattern during Radio Link synchronization
When commanded by NBAP signalling, the TPC commands are sent on a downlink radio
link from NodeBs that have not yet achieved uplink synchronization. These TPC
commands follow a pattern as specified in the downlink TPC pattern 01 count IE.

If the value included in the First Radio Link Set indicator IE indicates that the radio link
is part of the first radio link set sent to the UE the following procedure applies:
A value equal to n is obtained from the downlink TPC pattern 01 count IE;
The TPC pattern consists of n instances of "01" plus one instance of "1";
The TPC pattern continuously repeat but is forced to re-start at the beginning of
each frame where CFN mod 4 = 0.
Otherwise if the First Radio Link Set indicator IE indicates the radio link is not part of the
first radio link set then the TPC pattern consists of all "1".

The TPC pattern terminates once uplink synchronization is achieved.
If out of sync is detected at the NodeB after establishment of uplink synchronization, the
downlink transmit power that has been supplied immediately before the detection of out
of sync is held supplied if power balancing is set to Off. In the case power balancing is
activated, the power is set to the reference power as defined in the NBAP power control
message.

If synchronization is established again, transmit power control is started according to the
TPC bit pattern.

Subject:
Date:
June 2006


Page 28 of 28

From U03.03 onwards this will be a tunable parameter ranging from 0 to 255. The default
value is 30 (i.e. corresponding to a speed of ~ 25dB/s). The value 0 is a special value used
for the request of the alternance of "0" and "1" with no extra "1" at any time.

4.4.2.3.Downlink Inner Loop Timing
The standard does not specify the timing for the application on the downlink channel of
the TPC commands received from the UE. Indicative information is provided in [2]. The
decision for the timing of the update of the DPCH output power is left to each NodeB
manufacturer.

Discussion lead on Standard Committee level recommended that in the case of the
downlink the inner loop power function should update the downlink DPCH output power
in the NodeB at the start of the downlink pilot field. This would allow decoding the full
TPC command received from the UE as illustrated in Figure 4.

The SIR measurement periods illustrated here are examples; other ways of measurement
are allowed to achieve accurate SIR estimation. If there is not enough time for the
UTRAN to respond to the TPC, the action can be delayed until the next slot at the
beginning of the next pilot field.
To be noticed that this mechanism of downlink inner loop timing has only non-tunable
parameter.

Data2 Data1 Data1
T
P
C
Data1
T
P
C
PILOT PILOT
PILOT
Response
To TPC
T
P
C
DL SIR
measurement
PILOT TFCI TPC
DL-UL timing
offset (1024 chips)
Slot (2560 chips)
PILOT PILOT Data2 Data1
T
P
C
PILOT PILOT TFCI TPC
Slot (2560 chips)
Propagation delay
DL DPCCH
at UTRAN
Propagation delay
DL
DPCCH
at UE
UL DPCCH
at UTRAN
UL
DPCCH
at UE
512 chips
TF
CI
TF
CI

Figure 4: Downlink inner loop power control timing

Subject:
Date:
June 2006


Page 29 of 29

4.4.2.4.Downlink Inner Loop Power Control Modes
According to the first versions of the 3GPP standards Release 99 [2], there are two
downlink inner loop power control modes available:

single TPC the UE sends a unique TPC command in each slot and the TPC
command is transmitted in the first available TPC field in the
uplink DPCCH. The power control speed is equal to 1500 Hz
TPC triplet the UE repeats the same TPC command over 3 slots and the new
TPC command is transmitted such that there is a new command at
the beginning of the frame. The power control speed thus is
reduced to 500Hz.
This power control mode is configured in the UE via RRC signalling messages, i.e. Radio
Bearer Reconfiguration or Radio Bearer Set-up messages. More information is included
in [7].
The parameter dPCMode defines the downlink power control mode and is a UE specific
parameter controlled by the UTRAN. It is used to allow the UE to reduce the rate at
which the NodeB adjusts its power. Since only single TPC mode is supported, this
parameter has a fixed setting. More information on this parameter can be taken from
Table 21.

name
Attribute Syntax
& Range
UEGTFCS
Instance
Default
Value
Access
DLDPCHInfo DPCMode Enumerated
(sINGLETPC)
All sINGLETPC RNC-CLI
RW-EL
Table 21:Downlink power control mode parameter
4.4.3. Downlink Transmit Power Limits
4.4.3.1.Principle
Simulations have shown that in the case of non real time transfers, i.e. when the RLC
mode is set to Acknowledge Mode, a capacity gain can be achieved by limiting the
increased speed of the inner power control loop. The gain is achieved by limiting the
transmit power peaks whilst the resulting errors are recovered by RLC retransmissions.
The two following sections describe the expected behaviour from both the RNC and the
NodeB in order to support this feature.
The power increase limitation is a function that uses following set of parameters that are
defined on a per cell basis:
Power_Raise_Limit
DL_power_averaging_window_size

Subject:
Date:
June 2006


Page 30 of 30

Upon creation of a new cell, the C-RNC shall extract both the Power_Raise_Limit and
DL_power_averaging_window_size OAM parameters from the cell OAM profile and
populate the corresponding fields of the Cell Setup Request NBAP message with these
values.
No strategy involving the dynamic management of these parameters is foreseen for
U03.03. The Radio Link Reconfiguration Request message can also support this
information element but it is not supported in U03.03.

The actual IE that triggers the activation of this function is the Limited Power Increase
IE. More information can be taken from [10]. This parameter can take following values:
Limited_Power_Increase_ACTIVE [True]
Limited_Power_Increase_INACTIVE [False]

The Limited_Power_Increase_xxACTIVE state is defined in the RNC on a per call basis.
The default state value is always Limited_Power_Increase_INACTIVE
If, as a result of the active set update procedure, the S-RNC adds initial radio links to a
NodeB/D-RNC, the "Limited Power Increase" IE shall be populated according to the
procedure described in the next chapter.
4.4.3.2.Transition between xx_inactive and xx_active states
Limited Power Increase is activated when all the radio bearers mapped onto the radio link
allow the use of this feature. This means that e.g. if a call supports a voice service,
Limited Power Increase will not be allowed.
In order to manage this requirement the boolean Limited_Power_Increase_Allowed OAM
flag is defined on a per radio bearer basis. The RNC shall store the value of the parameter
for each of the radio bearers of each ongoing call.

When in Cell_DCH mode, upon creation or deletion of a radio bearer the RNC shall
check the updated list of Limited_Power_Increase_Allowed flags associated with the call:
If the call is in Limited_Power_Increase_INACTIVE state and all the
Limited_Power_Increase_Allowed flags are set to allowed the call shall switch to
Limited_Power_Increase_ACTIVE state and the RNC shall activate the Limited
Power Increase function by setting the Limited Power Increase IE to used in the
subsequent NBAP message sent to the NodeBs involved in the call.
If the call is in Limited_Power_Increase_ACTIVE state and at least one of the
Limited_Power_Increase_Allowed flags is set to not allowed the call shall switch
to Limited_Power_Increase_INACTIVE state and the RNC shall de-activate the
Limited Power Increase function by setting the Limited Power Increase IE to
not used in the subsequent NBAP message sent to the NodeBs involved in the call.
When entering the Cell_DCH mode, the RNC shall check the
Limited_Power_Increase_Allowed flag of each of the radio bearers mapped on the
DPDCH:

Subject:
Date:
June 2006


Page 31 of 31

If all the Limited_Power_Increase_Allowed flags are set to allowed the call shall
switch to Limited_Power_Increase_ACTIVE state and the RNC shall activate the
Limited Power Increase function by setting the Limited Power Increase IE to
used in the subsequent Radio Link Setup Request NBAP message sent to the
NodeBs involved in the call.
If at least one of the Limited_Power_Increase_Allowed flags is set to not allowed
the call shall be set to Limited_Power_Increase_INACTIVE state and the RNC
shall either:
not populate,
set to not used,
the Limited Power Increase IE to used in the subsequent Radio Link Setup Request
NBAP message sent to the NodeBs involved in the call.

4.4.3.3.Downlink Inner Loop Power Control Algorithm and Execution
The algorithm is located in the NodeB and is used to update the transmitter power
according to the TPC command received by the UE. For the k:th received TPC command
TPC
est
(k), the NodeB adjusts the current downlink power P(k-1) to a new power P(k)
according to (5).

P(k) = P(k-1) + P
TPC
(k) + P
bal
(k) (5)
where
P
TPC
(k) is the k:th power adjustment due to the inner loop power control
calculated in [dB] as follows:

If the value of Limited Power Increase is set to False

=
= +
=
0 ) ( TPC if
1 ) ( TPC if
) ( P
est TPC
est TPC
TPC
k
k
k
(6)
If the value of Limited Power Increase is set to True

=
+ =
< + = +
=
0 ) ( TPC if
_ _ ) ( 1 ) ( TPC if
_ _ ) ( 1 ) ( TPC if
) ( P
est TPC
TPC sum est TPC
TPC sum est TPC
TPC
k
Limit Raise Power k and k
Limit Raise Power k and k
k
(7)
where
+ =
=
1
1 _ _ _ _
TPC sum
) ( P ) (
k
Size window Averaging Power DL k i
i k
(8)

P
bal
(k) is a correction in [dB] according to the downlink power control procedure
for balancing radio link power towards a common reference power in soft/softer
handover scenarios. The estimation of this parameter is described in following
chapters.

Subject:
Date:
June 2006


Page 32 of 32

For the first (DL_Power_Averaging_Window_Size 1) adjustments after the activation
of the limited power raise method, formula (6) shall be used instead of formula (7).
The NodeB shall store the value of the DL_power_averaging_window_size and
Power_Raise_Limit parameters contained in the Cell Setup Request message received
during the cell set up procedure. These parameter values are used each time a request for
the activation of the downlink Power Increase Limitation function on one of the radio
links supported by the cell is received.
Upon reception of a Limited Power Increase IE set to Used, the NodeB shall activate
for the related radio link(s) the Limited Power Increase function as described. In case of
softer handover, all the softer legs associated with to the UE context shall be affected by
the activation.

Upon reception of a Limited Power Increase IE set to not used, the Node B shall de-
activate for the related radio link(s) the Limited Power Increase function, if it is active. In
case of softer handover, all the softer legs associated with the UE context shall be affected
by the de-activation.

In case of softer handover, the downlink requested transmit level shall be the same for all
the softer legs. Therefore, the Limited Power Increase function shall be either active or
inactive for the whole set of softer legs.
4.4.3.4.Downlink Inner Loop Transmit Power Step Size
The power step size is set by the RNC through an NBAP primitive. The value is conveyed
by the TPC DL step size information element, which can only be found in the Radio
Link Setup Request primitive. Upon reception of a TPC downlink step size information
element the NodeB shall store and apply the new step size value when the message
becomes effective.
Upon creation of a radio link, if the TPC downlink step size IE is not available, the
NodeB shall use the default value. Upon creation or reconfiguration of a radio link the
RNC shall populate the TPC Downlink Step Size IE in the associated NBAP message
with the value extracted from the OAM MIB table corresponding to the service profile
mapped onto the downlink physical channel.
The 3GPP standard states in [2] that it is mandatory for UTRAN to support TPC
commands adjustments steps
TPC
of 1dB, while support of 0.5, 1.5 and 2.0dB is optional.
All the step sizes are supported in U03.03.
The step size accuracy complies with the 3GPP requirements as described in [5]. These
specify the transmitter power control step tolerance and the transmitter aggregate power
control step range as described in Table 22 and Table 23 respectively.

Power control commands
in the downlink
Transmitter power control step tolerance
0.5 dB step size 1 dB, 1.5dB and 2dB step size
Lower Upper Lower Upper

Subject:
Date:
June 2006


Page 33 of 33

Up (TPC command "1") +0.25 dB +0.75 dB +0.5 dB +1.5 dB
Down (TPC command "0") -0.25 dB -0.75 dB -0.5 dB -1.5 dB
Table 22: NodeB transmitter power control step tolerance
Power control commands
in the downlink
Transmitter aggregated power control step range after 10
consecutive equal commands
0.5 dB step size 1 dB, 1.5dB and 2dB step size
Lower Upper Lower Upper
Up (TPC command "1") +4 dB +6 dB +8 dB +12 dB
Down (TPC command "0") -4 dB -6 dB -8 dB -12 dB
Table 23:NodeB transmitter aggregated power control step range
The parameter to which the Limited Power Increase IE is mapped is
limitedDLPowerIncrease. The values for this parameter as well as for the
powerRaiseLimit, dLPowerAvWindowSize and tPCStepSize are described in Table 24.
Additionally, information on their range and access can be found in this table.

Parent Object Parameter name Attribute Syntax
& Range
Instances Default
Value
Access
All others Y Dedicated
Transport
Channel
limitedDLPowerIncrease Boolean
2, 3, 4, 5 N
DLDPCHInfo
tPCStepSize Enumerated
{TPC0-5 = 0.5,
TPC1-0 = 1,
TPC1-5 = 1.5,
TPC2-0 = 2}
All TPC1-0
RNC CLI
RW-EL
powerRaiseLimit Integer
[0,10]
- 10
LCell
dLPowerAvWindowSize Integer
[1,60]
- 15
OMC
RO-SC,
RNC CLI
RO-SC
Table 24: Downlink power control step size parameter
4.4.4. Downlink Dynamic Range
4.4.4.1.Downlink Transmit power limits from the NodeB
In [5] the requirements for the downlink power control dynamic range are specified. This
range is defined as the difference between the maximum and the minimum transmit
output power of a downlink channelization code for a specified reference condition and
should be at least 18dB.
Depending on the Power Amplifier configuration, the maximum output power is equal to
46dBm (i.e. 40Watt) or 43dBm (i.e. 20Watt) per carrier.
Therefore, the minimum transmit power should be lower than 25dBm for a 20Watts
NodeB and 28dBm for a 40Watts NodeB.

Subject:
Date:
June 2006


Page 34 of 34

4.4.4.2.Downlink Transmit power limits from the RNC
In order to restrict the downlink interference generated on one sector by one call, the RNC
defines a maximum transmitted power level per radio link P
Max
.
Similarly a minimum transmitted power level P
Min
is defined. This can be used to
maintain the call at a level higher than the one required e.g. in a zone where coverage
variations are very sharp and the call drop rate is high. In most cases however it is
recommended to set the minimum downlink power level to the minimum code power
value than can be achieved by the NodeB hardware.
Once the NodeB receives these power limits via NBAP messages (i.e. Radio Link Setup
Request or Radio Link Addition Request) it will not transmit with a level outside of the
range defined by [P
Min
; P
Max
]. This applies to the average power of the data part only, i.e.
the DPDCH symbols. Transmitted DPDCH symbol means here a complex QPSK symbol
before spreading, which does not contain DTX. More details are provided in [10].
4.4.4.3.Downlink Transmit Power Limits in Softer Handover
In the case where two or more sectors are in softer handover, the allowed dynamic range
[P
mini
, P
maxi
] of each sector i is therefore defined independently from one softer sector to
the other.
In order to avoid downlink imbalance within a set of softer radio links, the RNC requires
that the initial downlink power level of any added radio link is the same as the one of the
already existing links. This is achieved by not populating the Initial DL Tx Power IE in
any Radio Link Addition Request NBAP message.
Upon creation of a new radio leg, the RNC shall populate the Maximum downlink power
and the Minimum downlink power IE of the Radio Link Setup Request NBAP message.
The values to be used shall be those read from the OAM MIB table corresponding to the
combination of radio bearers mapped onto the downlink DPDCH.
If no Maximum downlink power IE is included in the NBAP message, the Maximum
downlink power value stored for already existing Radio Links for this UE shall be
applied. The same procedure is applied to determine the value of the Minimum downlink
power if it is not present in the Radio Link Setup Request message.
4.4.4.4.Tuning of Pmax and Pmin
A correction factor is introduced in U03.03 to correct both Maximum DL power and
Minimum DL power. These correction factors are OAM parameters defined on a per cell
basis. They allow to fine tune P
min
and P
max
on per cell basis thereby making possible to
solve or mitigate local cell planning issues.

Upon creation of a new radio leg, the S-RNC shall populate the Maximum DL power and
the Minimum DL power IEs of the Radio Link Setup Request and Radio Link Addition
Request NBAP messages according to (9) and (10).

Minimum DL power = Pmin + PminCorrectionFactor (9)
Maximum DL power = Pmax + PmaxCorrectionFactor (10)

Subject:
Date:
June 2006


Page 35 of 35

Upon addition, deletion or reconfiguration of one or more of the downlink radio bearers
that are mapped onto the physical channel, the RNC shall populate the Maximum DL
power and the Minimum DL power IEs of the related Radio Link Reconfiguration
message.
4.4.4.5.Translation settings
The translations of P
max
and P
min
are mapped into parameters maxDLPower and
minDLPower respectively.

Subject:
Date:
June 2006


Page 36 of 36

name
Attribute Syntax
& Range
UEGTFCS Default Value Access
minDLPower Real - dB
-35, -34.9 +15
All -19dB
1, 2, 13, 50,
54
+3dB
DLDPCHInfo
maxDLPower Real - dB
-35, -34.9 +15
Others +6dB
PminCorrection
Factor
Real dB
-10.0, -9.9 +10
- 0dB LCell
PmaxCorrection
Factor
Real dB
-10.0, -9.9 +10
- 0dB
RNC-CLI
RW-EL
Table 25: Downlink Power Range Parameters.

To be noted that setting downlink DCH Power to 39dBm (i.e. 8Watt) results in having
40% of the Total Downlink Power assigned to a single user assuming that the UE requests
most of the time the maximum allowed downlink DCH power to the NodeB.
The basic settings for both transmit power limit parameters are included in Table 25.

4.4.5. Downlink power control in compressed mode
4.4.5.1.Impact of compressed mode on the Power Control function

The aim of downlink power control in downlink compressed mode is to recover as fast as
possible a signal-to-interference ratio close to the E
c
N
otarget
after each transmission gap.

In every slot during compressed mode except during downlink transmission gaps, the
Node B shall estimate the k
th
TPC command and adjust the current downlink power P(k-
1) [dB] to a new power P(k) [dB] according to (11)

P(k) = P(k-1) + P
TPC
(k) + P
SIR
(k) + P
bal
(k) (11)

where:
P
TPC
(k) is the k
th
power adjustment due to the inner loop power control
P
SIR
(k) is the k
th
power adjustment due to the downlink SIR
target
variation
P
bal
(k) [dB] is a correction according to the downlink power control procedure for
balancing radio link powers towards a common reference power.

If no uplink TPC command is received because of uplink compressed mode gaps, the
value P
TPC
(k) derived by the Node B shall be set to zero.

Subject:
Date:
June 2006


Page 37 of 37

Otherwise, P
TPC
(k) shall be calculated the same way as in normal mode but with a step
size
STEP
, as described in section 4.4.5.2, instead of
TPC
.

4.4.5.2.DL power step size in compressed mode

The power control step size
STEP
shall take the value of
RP-TPC
during RPL slots after
each transmission gap and
STEP
shall take the value of
TPC
otherwise, where:

RPL is the recovery period length and is expressed as a number of slots. RPL is
equal to the minimum value out of the transmission gap length and 7 slots. If a
transmission gap is scheduled to start before RPL slots have elapsed, then the
recovery period shall end at the start of the gap, and the value of RPL shall be
reduced accordingly.

TPC
is the power control step size in normal mode.

RP-TPC
is called the recovery power control step size [dB] and defined in (12).

.
. 2
3
Min =
TPC )
`
TPC RP

(12)

4.4.5.3.Calculation of P
SIR
(k)

The power offset is defined in (13).

( )
prev curr
P P = k P
SIR
(13)

where P
curr
and P
prev
are respectively the value of P in the current slot and the most
recently transmitted slot and is computed according to (14)

( ) coding P coding P _ 2 _ 1 sion Pn_compres , ... sion, P1_compres Max P + + = (14)

where
n is the number of different TTI lengths amongst TTIs of all TrChs of the CCTrCh
P1_coding and P2_coding are computed from the uplink parameters
DeltaSIR1, DeltaSIR2, DeltaSIRafter1, DeltaSIRafter2 as defined in (15) and
(16).
Pi_compression is defined by (17).

P1_coding = DeltaSIR1 if the start of the first transmission
gap in the transmission gap pattern is within the current frame.
(15)

Subject:
Date:
June 2006


Page 38 of 38

P1_coding = DeltaSIRafter1 if the current frame just follows a
frame containing the start of the first transmission gap in the
transmission gap pattern.
P1_coding = 0 dB in all other cases

P2_coding = DeltaSIR2 if the start of the second transmission
gap in the transmission gap pattern is within the current frame.
P2_coding = DeltaSIRafter2 if the current frame just follows a
frame containing the start of the second transmission gap in the
transmission gap pattern.
P2_coding = 0 dB in all other cases.
(16)

Pi_compression = 3 dB for downlink frames compressed by
reducing the spreading factor by 2.
Pi_compression = 10 log ( 15*Fi / ( 15*Fi -TGLi)) if there is a
transmission gap created by puncturing method within the
current TTI of length Fi frames, where TGLi is the gap length in
number of slots ( either from one gap or a sum of gaps) in the
current TTI of length Fi frames.
Pi_compression = 0 dB in all other cases.
(17)

In case several compressed mode patterns are used simultaneously, a P offset is
computed for each compressed mode pattern and the sum of all P offsets is applied to the
frame.

The P
SIR
(k) results in Node B raising its power during the compressed slots in a frame
containing idle period in order to compensate the raised DL EcNo target (SIRtarget_CM)
by UE.

4.4.5.4.DL transmit power range aspects

For all time slots except those in transmissions gaps, the average power of transmitted
DPDCH symbols over one timeslot shall not exceed Maximum_DL_Power [dB] by more
than P
curr
, nor shall it be below Minimum_DL_Power [dB].

4.4.5.5.Impact of DL compressed mode on the behavior of the UE

Subject:
Date:
June 2006


Page 39 of 39

The UE behavior is the same in compressed mode as in normal mode, i.e. TPC commands
should be generated based on the comparison of DL E
c
N
oestimate
and DL E
c
N
otarget
.
However an alternative EcNo target value is used in the radio frames where the
compressed mode idle period takes place. The SIR target value is computed every radio
frame and is defined in (18).

SIR normal et t CM et t
Delta SIR SIR + =
_ arg _ arg
(18)

Delta
SIR
is defined in (19).

( ) coding P coding P Delta
SIR
_ 2 _ 1 sion Pn_compres , ... sion, P1_compres Max + + = (19)

where

n is the number of different TTI lengths amongst TTIs of all TrChs of the
CCTrCh,
P1_coding and P2_coding are computed from the parameters DeltaSIR1,
DeltaSIR2, DeltaSIRafter1, DeltaSIRafter2 signalled by RRC signalling.

With only one gap being used in U03.03, the relevant parameters are DeltaSIR1,
DeltaSIRafter1. These parameters affect the DL/UL ILPC in the frames with
transmissions gaps or the frames with gaps in previous frame. The default values are 0 dB
but simulation/measurement results are required to assess BLER performance impact of
DeltaSIR1/after1 parameters. The corresponding translation settings are described in
Table 26.

& Range
UEGTFCS
Instance
Default
Value
Access
deltaSIR1
Integer [dB]
0, 0.01, 3
0
deltaSIRAfter1
Integer [dB]
0, 0.01, 3
0
deltaSIR2
Integer [dB]
0, 0.01, 3
0
Transmission
GapPattern
deltaSIRAfter2
Integer [dB]
0, 0.01, 3
All
0
RNC-CLI
RW

4.4.6. Power Imbalance
4.4.6.1.Introduction
In some conditions the TPC command detection error rate becomes high on the weakest
link of a connection. When such a TPC error occurs, the TPC command is interpreted as a
Power_Up (wrong value) whilst on the other leg(s) it is interpreted as a Power_Down

Subject:
Date:
June 2006


Page 40 of 40

(correct value); this results in downlink DCH power levels diverging by twice the power
adjustement step size (e.g. 2dB). The next time an error occurs, the weakest leg may
diverge by an extra 2dB or converge back if the errors cancel each other. As a result the
involved sector transmits with a higher power than needed. This leads to a random walk
effect for the transmit power of this link, and therefore there is a non-negligible chance
that their downlink transmission power reaches much higher level than needed.
This effect is known as power imbalance. An example is depicted in Figure 5. The
variations in downlink Transmitted Code Power of two legs in soft handovers is caused
by power imbalance.
Power imbalance can also happen when one of the transmit legs has a maximum transmit
power value lower than the other legs. In such cases, the power up TPC commands may
be ignored on this leg whilst the TPC down commands would result in power decrease.
Power imbalance does not deteriorate the performance of the imbalanced links; it may
improve it slightly. The main issue is that the overpowered transmission pollutes
unnecessarily the other radio links in the vicinity, thus leading to a lower radio network
capacity. In order to correct power imbalance, the transmit power balance procedure is
used.

Figure 5: Downlink Power Imbalance Effect in Soft Handover Scenario.

The NodeB shall be able to support downlink Power Control procedures initiated by the
C-RNC at any time when the NodeB communication context exists, irrespective of other
ongoing C-RNC initiated dedicated NBAP procedures towards this NodeB
communication context. The only exception occurs when the C-RNC has requested the
deletion of the last radio link via this NodeB, in which case the downlink Power Control
procedure shall no longer be initiated.

Subject:
Date:
June 2006


Page 41 of 41

4.4.6.2.Power Balance Procedure at the RNC
The transmit power balance procedure is triggered by the S-RNC which requests to all the
NodeBs involved in the active legs of a connection to activate a process that makes sure
that the downlink power level of the weakest links does not shoot up to some unwanted
level.
From a RNC perspective, the Imbalance Correction procedure for a given UE shall be
activated or deactivated at the S-RNC by setting the Imbalance Correction parameter in
the RNC MIB either to Active or Inactive.
There shall be one Imbalance Correction MIB parameter per downlink Radio Access
Bearer combination, i.e. one for Standalone Signalling, one for Voice + Signalling, etc ...
There shall be also one downlink reference Power, Adjustment Period, Adjustment Ratio
and Max Adjustment Step parameters per downlink Radio Access Bearer combination.
The activity or inactivity of Imbalance Correction does not depend on the number of
active radio legs.
4.4.6.2.1. Activation of the Imbalance Correction
The activation of the Imbalance Correction for one particular UE shall occur either:
At the call set up
Upon modification of the number of radio bearer services multiplexed onto the
physical channel.
Upon modification of the number of radio bearer services of an ongoing call where
Imbalance Correction is inactive, or at call setup, the S-RNC shall check the Imbalance
Correction MIB parameter associated with the configuration which is about to be
activated.

If the Imbalance Correction is set to Active:
The S-RNC shall trigger an Imbalance Correction activation procedure by sending
a Downlink Power Control Request message to every NodeB connected to the S-
RNC and supporting one or more radio links involved in the call. If the call
involves one or more D-RNC, one message shall also be sent to each of these D-
RNC (one per D-RNC).
The downlink reference Power, Adjustment Period, Adjustment Ratio and Max
Adjustment Step used in the related signalling message shall correspond to those
of the new setting in the RNC MIB.
The Power Adjustment Type IE shall be set to Common
Imbalance Correction shall then be considered to be active for that call.
4.4.6.2.2. De-activation of the Imbalance Correction
The de-activation of the Imbalance Correction for one particular UE only occurs upon
modification of the number of radio bearer services multiplexed onto the physical
channel.

Subject:
Date:
June 2006


Page 42 of 42

Imbalance Correction is active the S-NRC shall check the Imbalance Correction MIB
parameter associated with the configuration, which is about to be activated.

If the Imbalance Correction is set to Inactive:
The S-RNC shall trigger an Imbalance Correction de-activation procedure by
sending a downlink power control request message to every NodeB connected to
the S-RNC and supporting one or more radio links involved in the call. If the call
involves one or more D-RNC, one message shall also be sent to each of these D-
RNC (one per D-RNC).
The Power Adjustment Type IE shall be set to none.
Imbalance Correction shall then be considered to be inactive for that call.
4.4.6.2.3. Reconfiguration of the Imbalance Correction
The reconfiguration of the Imbalance Correction parameters for one particular UE only
occurs upon modification of the number of radio bearer services multiplexed onto the
physical channel.

Imbalance Correction is active, the S-NRC shall check the Imbalance Correction MIB
parameter associated with the configuration which is about to be activated.

If the Imbalance Correction is set to Active':
The S-RNC shall trigger an Imbalance Correction activation procedure by sending
a downlink power control request message to every NodeB connected to the S-
RNC and supporting one or more radio links involved in the call. If the call
involves one or more drift RNC, one message shall also be sent to each of these
drift RNCs (one per drift RNC).
The downlink reference Power, Adjustment Period, Adjustment Ratio and Max
Adjustment Step used in the related signalling message shall correspond to those
of the new setting in the RNC MIB.
The Power Adjustment Type IE shall be set to Common'
Imbalance Correction shall be considered to remain active for that call
4.4.6.3.Imbalance correction at the Drift RNC
In the case where it acts as a D-RNC, the RNC has to support Imbalance Correction
requests from the S-RNC in both individual and common mode. That is for the case
where the S-RNC is a non-Lucent Technologies product and it requires the use of BTP in
individual mode.

At the creation of a UE context and prior to the reception of any downlink power control
request RNSAP message, the D-RNC shall set its Imbalance Correction parameters to the
following default values:

Subject:
Date:
June 2006


Page 43 of 43

Power Adjustment Type = none
DL Reference Power = 0
Max Adjustment Step = 1
Adjustment Period = 1
Adjustment Ratio = 0

Upon reception of a downlink power control request RNSAP message from the S-RNC
with Power Adjustment set to any value, the D-RNC shall store the new Power Imbalance
Correction parameters values in the related UE context.

In the case where the Power Adjustment Type IE of the message is set to common or
none, the D-RNC shall forward a copy of the "DL Power Control" RNSAP message to
each NodeB involved in the call and connected to the D-RNC via a downlink power
control request NBAP message.

In the case where the D-RNC is aware that a NodeB is already in "none" mode, it shall
not forward towards that NodeB any subsequent "DL Power Control" RNSAP message
containing a Power Adjustment Type IE set to none, unless the DL inner loop PC
status IE requires a change from the current "inner loop PC status" value for the call.
At Radio Link Setup the BTP mode shall be assumed to be none and the DL inner
loop PC status mode is provided in the RL Setup Request RNSAP message.

In the case where the Power Adjustment Type IE of the message is set to individual,
the D-RNC shall also forward a copy of the "DL Power Control" RNSAP message to
each NodeB involved in the call and depending of it via a downlink power control request
NBAP message.
Note, that if there is more than one radio link on a NodeB, which is in the "individual"
list, the RNC shall send only one message to that NodeB and that message should contain
the list of all the radio links on that NodeB listed in the RNSAP message.
4.4.6.4. Estimation of P
bal

If power balancing is active on a radio link, the power balancing adjustment is
superimposed on the inner loop power control adjustment as described in (20). The
accuracy should be of 0.5 dB.

) )( 1 (
init CPICH P ref bal
P P P r P + =

(20)

Where:
The sum is performed over an adjustment period corresponding to a number of
frames equal to the value of the Adjustment Period IE,
Pref is the value of the DL Reference Power IE,

Subject:
Date:
June 2006


Page 44 of 44

P
P-CPICH
is the power used on the primary CPICH,
Pinit is the code power of the last slot of the previous adjustment period and
r is given by the scale Adjustment Ratio IE.

4.4.6.5.Limitation at the NodeB
The RNC does not wait for the radio link to establish uplink synchronization to send the
downlink power control request message to the NodeB. At call setup, if the Imbalance
Correction is applied before closed loop power control synchronization is achieved, the
odds are very high that the downlink power level will converge towards P
ref
(which is
usually a low power level) and the call setup procedure will fail.
As a workaround for this issue, the following implementation is required to the NodeB for
U03.03.
For the radio links established with a Radio Link Setup Request message
containing a First RLS indicator IE set to true, the NodeB shall delay the
application the downlink power balance corrections P
bal
until the uplink has
reached synchronization.
For the radio links established with a Radio Link Setup Request message
containing a First RLS indicator IE set to false, the downlink power balance
corrections P
bal
shall be applied immediately after reception of the downlink
power control request message, and independently of the uplink synchronization
status. The power adjustments shall be started at the first slot of a frame with CFN
modulo the value of Adjustment Period IE equal to 0, shall be repeated for every
adjustment period and restarted at the first slot of a frame with CFN equal to 0,
until a new downlink power control request message is received or the RL is
deleted.
The algorithm presented in Figure 6 is used for application of the correction at the NodeB
level.

Subject:
Date:
June 2006


Page 45 of 45

CFN = 0?
or
CFN mod Adjustment Period = 0?
begining of a new radio frame?
Period_Correction = (1-r) (Pref + PCPICH - Pinit)
Already_Corrected = 0
Pbal = sgn(Period_Correction )
x min(1/ Max Adjustment Step ,
abs(Period_Correction - Already_Corrected))
Already_Corrected = Already_Corrected + Pbal
new timeslot
yes
yes
no
no

Figure 6 Downlink Power Balance Algorithm at NodeB level

Upon reception of a downlink power control request message:
If the value of the Power Adjustment Type parameter is Common, the Power
Balancing Adjustment Type of the NodeB Communication Context shall be set to
Common. As long as the Power Balancing Adjustment Type of the NodeB
Communication Context is set to Common, the NodeB shall perform the power
adjustment for all existing and future radio links associated with the context identified
by the NodeB Communication Context ID IE and use a common downlink reference
power level.
If the value of the Power Adjustment Type IE is Individual, the Power Balancing
Adjustment Type of the NodeB Communication Context shall be set to Individual.
The Node B shall perform the power adjustment for all radio links addressed in the
message using the given downlink Reference Powers per radio link. If the Power
Balancing Adjustment Type of the NodeB Communication Context was set to
Common before this message was received, power balancing on all radio links not
addressed by the downlink power control request message shall remain to be executed
in accordance with the existing power balancing parameters which are now
considered RL individual parameters. Power balancing will not be started on future
radio links without a specific request.

Subject:
Date:
June 2006


Page 46 of 46

If the value of the Power Adjustment Type IE is None, the Power Balancing
Adjustment Type of the Node B Communication Context shall be set to None and
the Node B shall suspend on going power adjustments for all radio links for the UE
Context.

The parameters involved in the downlink power imbalance correction and described in the
previous chapters are presented in Table 27.
Parent
Object
& Range
UEGTFCS
Instance
Default
Value
Access
ImbalanceCorrection Bolean All Y
maxAdjustmentStep Integer
[1, 10] slots
All 2
adjustmentPeriod Integer
[1, 2, 256] frames
All 1
scaleAdjustmentRatio Integer
[0 100]
All 50
1 10dB
2, 13, 50, 54 -20dB
3, 4, 6, 10, 31,
47, 51
-14dB
5, 7, 33, 36,
49, 53
-8dB
14 -17dB
9, 32, 34, 35,
48, 52
-11dB
DLDPCHInfo
dLReferencePower Real [dB]
[-35, -34.9+15]
11 -10dB
RNC-CLI
RW-EL
Table 27: Imbalance Correction Parameters

4.4.7. Power Offset between Downlink DPCCH and its DPDCHs
4.4.7.1.Principle
The downlink transmit power control procedure controls simultaneously the power of a
DPCCH and its associated DPDCHs. The power of the different channels is adjusted in a
way that keeps the relative power difference between DPCCH and DPDCH constant.
Figure 7 depicts the single slot structure of a downlink DPCH including the different
power offsets. The power levels of TFCI, TPC and pilot fields of the DPCCH are defined
as offsets relative to the DPDCHs power by PO1, PO2 and PO3 [dB] respectively.

Subject:
Date:
June 2006


Page 47 of 47

One radio frame, T
f
= 10 ms
TPC
N
TPC
bits
T
slot
= 2560 chips, 10*2
k
bits (k=0..7)
Data2
N
data2
bits
DPDCH
TFCI
N
TFCI
bits
Pilot
N
pilot
bits
Data1
N
data1
bits
DPDCH DPCCH DPCCH
PO2 PO1 PO3

Figure 7: Power offsets in downlink DCH.
These values are set by the S-RNC, which reads the values from the OAM managed
tables. Upon reception of the related NBAP messages (Common Transport Channel Setup
Request / Radio Link Setup Request) the NodeB applies the requested downlink power
offset to the relevant DPCCH block.
To be noticed that PO1 and PO3 can be signaled to the NodeB only at Radio Link Setup
([10]). This means that they cannot be modified once the radio link is set up. PO2 can be
modified only through DCH Framing Protocol signalling as defined in [9]. Based on the
BLER
target
sent by the RNC for the downlink DCH, the UE is able to deduce, through the
Outer Loop procedure, a SIR
target
for the data part, but not for the pilot part. In order to
calculate the SIR
target
for the pilot, the UE needs to add up the power offset PO3 between
pilot field and data field. This is achieved via RRC signalling on the FACH as described
in details in [7].
4.4.7.2.Impact of PO1 and PO3 on the radio link
In marginal conditions, increasing the value of PO1 may help to improve the BLER
performance by reducing the probability of TFCI decoding errors. Note that only one
TFCI per radio frame has to be decoded so the probability to fail is relatively small.
Otherwise if the TFCI decoding fails the whole radio frame will be discarded.
Increasing PO3 could help the UE to achieve synchronization and consequently its
receiver to perform better at a given SIR set point, thus leading to downlink capacity gain.
4.4.7.3.Dynamic management of the downlink power offset PO2
4.4.7.3.1. Introduction
Experience with IS95 networks showed that the radio link quality is substantially
enhanced when the power offset between the data and the TPC parts of the downlink
frame increases with the number of radio link sets connected to the UE. The same
strategy has been ported to the UMTS context.

Three parameters are defined in the RNC MIB for each service combination supported:

Subject:
Date:
June 2006


Page 48 of 48

PO2_for_one_RL_set when the number of Radio Link Active Sets is equal to 1
PO2_for_two_RL_set when the number of Radio Link Active Sets is equal to 2
PO2_for_three_RL_set_and_more when the number of Radio Link Active Sets is
equal to 3 and more.

These values shall be passed to the TPU at call set up and each time a new service is
added or removed on the downlink. This is achieved through DCH FP signalling, using
the Radio Interface Parameter Update message as described in [9].
The PO2 field in the NBAP/RNSAP Radio Link Setup Request message is mandatory.
Unfortunately, the serving RNC only knows the number of RLS after reception of the
subsequent Radio Link Setup Response message. Therefore when it prepares the Radio
Link Setup Request message, the S-RNC cannot determine which one of the PO2_xxx
parameters it should use.
Therefore the S-RNC populate always the PO2 IE in any NBAP/RNSAP Radio Link
Setup Request message with the value of the PO2_for_one_RL_set corresponding to
the related service profile.
Once the transport bearer of the new Radio Link has established synchronization, the
RNC shall calculate the number of Radio Link sets and subsequently using the Radio
Interface Parameter Update DCH Framing Protocol message:
Update PO2 for the newly created RLS if only this RLS requires to be updated
Update PO2 for all the RLS involved in the call, if they all require an update
Perform no PO2 update, if none is required.
4.4.7.3.2. Algorithm to determine the value of the PO2 field
The S-RNC will determine the value of the PO2 field according to following cases.

PO2 is populated with the value PO2_for_one_RL_set when either:
The number of radio link sets decrease and is equal to 1.
PO2_for_one_RL_set has been updated at the RNC and the number of RLS is 1
PO2 is populated with the value PO2_for_two_RL_set when either:
The number of radio link sets changed and is now equal to 2.
PO2_for_two_RL_set has been updated at the RNC and the number of RLS is 2.
PO2 is populated with the value PO2_for_three_RL_set_and more when either:
The number of radio link sets increase and is equal to 3.
PO2_for_three_RL_set_and more has been updated at the RNC and the number of
radio link sets is equal to 3 or more.
Each of the parameters defining the power offsets can be defined with different values
depending on the selected downlink service combination specified through different
instances of top object UEGTFCS.

Subject:
Date:
June 2006


Page 49 of 49

& Range
UEGTFCS
Instances
Default
Value
Access
pO1 Real - [dB]
(0.0, 0.25,6.0)
All 0dB
pO2ForOneRLSet Real- [dB]
(0.0,0.25,6.0)
All 0dB
pO2ForTwoRLSet Real- [dB]
(0.0,0.25,6.0)
All 3dB
pO2ForThreeRFSetAnd
More
Real- [dB]
(0.0,0.25,6.0)
All 6dB
DLDPCHInfo
pO3 Real- [dB]
(0.0,0.25,6.0)
All 2dB
RNC-CLI
RW-EL
Table 28: Downlink DCH power offsets parameters.
During lab optimization tests as well as in the field setting of pO3 parameter equal to 2dB
has been validated for all supported service combinations. This setting helps to minimize
random walk issues as well as to solve issues related to loss of downlink synchronization
observed before the DPCH power reaches the upper limit. The default values of these
power offsets are included in Table 28.
4.4.8. Deactivation of the downlink ILPC
The downlink Inner Loop Power Control can be deactivated at the NodeB for a group of
radio links belonging to the same Node B communication context. This is achieved by the
use of the Inner Loop downlink power control IE in the NBAP / RNSAP downlink Power
Control Request message. There is no possibility of disabling the Inner Loop Power
Control on one Radio Link and keep active the other Radio Links of the same NodeB-UE
context.
This feature is not expected to be used in the Lucent Technologies RRM strategy.
However the message could be generated by a non-Lucent Technologies Drift-RNC or
could be of some use for test purposes. For these reasons, the IE needs to be supported on
the I
ub
/ I
ur
interface and DL ILPC inactivation needs to be supported at the Node B.
The S-RNC shall never populate the Inner Loop Downlink Power Control Status IE in the
downlink power control request message.
Upon reception over the I
ur
of a downlink power control request the C-RNC shall forward
it to the Node Bs involved in the call
If the S-RNC is a non-Lucent Technologies RNC, it may or may not have populated the
Inner Loop downlink Power Control Status IE. If the IE is present, it shall be replicated in
the downlink power control request message sent to the NodeB(s).
If the NodeB receives a downlink power control request Message and the Inner Loop
downlink Power Control Status IE is
Present and set to Inactive , the NodeB shall deactivate inner loop downlink
power control for all radio links for the Node B communication context.
Set to Active , the NodeB shall activate inner loop downlink power control for
all radio links for the NodeB communication context.
Note that [2] defines that the new power P(k) is to be calculated as follows:

Subject:
Date:
June 2006


Page 50 of 50

Deactivation of downlink power control for a radio link shall be accomplished by
excluding the component P
TPC
(k) from (6) and (7). All other components in this
equations remain unaffected, especially the power imbalance correction function
P
bal
(k).
The parameter used to activate / deactivate the inner loop power control is described in
Table 29.
Parent
Object
Parameter name Attribute Syntax &
Range
UEGTFCS
Instances
Default
Value
Access
DLDPCHInfo innerLoopPCStatus Boolean All Y RNC-CLI
RW-EL
Table 29: Deactivation of the Inner Loop Power Control

Subject:
Date:
June 2006


Page 51 of 51

5. Power Settings for Downlink Common Channels
5.1. INTRODUCTION
Following downlink physical common channels are supported:
Common Pilot Channel (CPICH) with sub-channels Primary CPICH (P-CPICH)
and Secondary CPICH (S-CPICH).
Primary Common Control Physical Channel (P-CCPCH): this channel is used to
carry the BCH.
Secondary Common Control Physical Channel (S-CCPCH): this channel is used
to carry the FACH and PCH.
Synchronization Channel (SCH) with two sub-channels the Primary SCH (P-SCH)
and the Secondary SCH (S-SCH).
Paging Indicator Channel PICH
Acquisition Indicator Channel AICH
To be noticed that S-CPICH is supported but not used in U03.x. Since in U03.03 only one
S-CCPCH is supported, both PCH and FACH channels are multiplexed over this channel.
More detailed information on downlink common channels can be found in [1].
5.2. POWER CONTROL PROCEDURES
The RNC controls the transmit power for all these channels. For all these channels except
the FACH, there is no control loop. Therefore the transmit power values remain constant
until the relevant RNC parameters are modified requiring a different transmit power.
For the FACH channel, an open loop function can be implemented according to the
specifications. This is not the case in U03.03. The power of the FACH channels is set to
the maximum power allowed for this channel in the cell, i.e. the FACH channel is
broadcast in the whole cell.
The P-CPICH level is defined in dBm. The other downlink common channel power levels
are defined in dB with respect to the P-CPICH level. This implies that when the P-CPICH
level is modified the power level of the other downlink common channels is also
modified accordingly.
The configuration as well as the re-configuration (in case of changes in settings) of the
transmit power of these downlink common channels is performed from the RNC by
sending a Cell Setup Request NBAP message to the NodeBs to set the power of the p-
SCH, S-SCH, S-CPICH and BCH channels. For the PICH and AICH, the level is set by
sending a Common Transport Channel Setup Request NBAP message. All these NBAP
procedures are described in detail in [10].
In the following sections a description of the downlink common channels as well as
proper recommendation on the power settings are provided.

Subject:
Date:
June 2006


Page 52 of 52

5.3. COMMON PILOT CHANNEL AND SYNCHRONIZATION CHANNEL POWER
The P-CPICH is transmitted continuously and represents the power overhead consumed
for this channel. Most descriptions give the pilot power as a ratio of the total output power
of the amplifier (percentage). However the translation parameter is specified as absolute
power in dBm to the system.
The two synchronization channels are time multiplexed with the P-CCPCH having a duty
cycle of 10% for one slot of 0.67ms. The power settings specified refer to a power on chip
level and therefore must be multiplied with the duty cycle to get the real overhead powers
for these channels. The power of the P-CCPCH, occupying the remaining 90% of each
slot, can be specified separately.

Power P
2
Power P
1
Power P
3
One frame (10 ms)
P-SCH:
S-SCH:
P-CPICH:
One slot (0.67 ms)
Figure 8: Frame and Slot Structure of CPICH, P-SCH and S-SCH.
Commonly both synchronization power settings are given relative to the pilot power
setting. The powers P
1
and P
2
are equivalent to about 5% of the total amplifier power
assuming a pilot power setting of 10%. To get the total overhead power of each channel
the power must be multiplied by the duty cycle of 10% and thus only 0.5% of the total
amplifier power is consumed for each channel (P-SCH and S-SCH). The corresponding
translation parameters are provided in Table 30.
All these parameter settings have a wide impact on the RF coverage as well as on basic
procedures such as cell search (i.e. synchronization) and soft/softer handover.
In particular the P-CPICH power should be set to a value that ensures proper RF coverage
by achieving high success rate along synchronization procedures. On the other hand, too
high power setting results in downlink capacity issues.
The P-SCH and S-SCH power settings have to ensure high success rate along the
different steps of the synchronization procedure as well as preserving system capacity
though the impact is much lower than in P-CPICH case due to the low duty cycle of both
channels. For this reasons current recommendations for these power settings are based on
results observed along tests focused on synchronization procedures in different radio
propagation environments and with different UEs.

Subject:
Date:
June 2006


Page 53 of 53

5.4. PRIMARY COMMON CONTROL PHYSICAL CHANNEL POWER
The P-CCPCH is the downlink physical channel used to carry the BCH transport channel.
The BCH is used to broadcast information to the UE such as RRC System Information
Broadcast and Master Information Block messages that includes several parameters used
by the UE along idle mode procedures (e.g. RACH access and cell reselection) as well as
connected mode procedures (e.g. soft/softer handover). More details on SIB and MIB
procedures are provided in [7]. Figure 9 shows the frame structure of the Primary
CCPCH.

Figure 9:Primary CCPCH Frame Structure.
The P-CCPCH is not transmitted during the first 256 chips of each slot. As each P-
CCPCH slot consists of 2560 chips, the P-CCPCH duty cycle is equal to 90%. The power
of the P-CCPCH is set through the parameter bCHPower defined with respect to the P-
CPICH as presented in Table 30.
Taking into account the 90% duty cycle if, e.g., P-CCPCH power is set to 30 dBm (i.e. 1
Watt or 5% of the 20W PA), the total overhead power is calculated as 90% of 1 Watt,
hence it is equal to 0.9 Watt or 4.5% of the 20W PA.
An optimal P-CCPCH power setting should ensure high success rate in decoding SIB and
MIB messages at the UE while preserving system capacity in terms of downlink power
available for Dedicated Channels.
Along evaluation of achieving high decoding success rate, the time spent to decode the
messages (that may consist of several Transport Blocks) as well as the time spent in
decoding retransmitted messages due to CRC check failure have also to be taken into
consideration. Theoretical assumptions as well as physical layer simulations have
indicated that keeping BCH BLER target equal to 10% ensures the best trade-off between
high decoding success rate (thus low decoding delay) and low power consumption.
5.5. SECONDARY COMMON CONTROL PHYSICAL CHANNEL POWER
The S-CCPCH is used to carry the FACH and PCH that are time-multiplexed. The FACH
and the PCH channels are used to carry RRC signalling messages along, e.g., RRC
Connection Establishment procedures and Paging procedures respectively. More details
on these procedures are provided in [7].

Subject:
Date:
June 2006


Page 54 of 54

In general, as depicted in Figure 10 below, the S-CCPCH can support three different
fields within its frame: the TFCI (unlikely the P-CCPCH), the data and the pilot. It is the
UTRAN that determines if a TFCI should be transmitted, hence making it mandatory for
all UE(s) to support the use of TFCI. Also the use of the pilot field is optional.

One radio frame, T
f
= 10 ms
T
slot
= 2560 chips, 20*2
k
bits (k=0..6)
Data
N
data2
bits
TFCI
N
TFCI
bits
Pilot
N
pilot
bits

Figure 10:Secondary CCPCH Frame Structure.
When the FACH is active, DTX is enabled on the inactive channel (i.e. PCH) and vice
versa. However due to the time-multiplexed structure of the FACH and PCH within S-
CCPCH, there is no power saving at all if at least one of the two channels are active
because the power on the data part within one Timeslot has to be the same.
General optimization criteria for both FACH and PCH power settings should be based on
minimizing the message error rate.

T
slot
= 2560 chips, 20*2
k
bits (k=0..6)
Data
N
data2
bits
TFCI
N
TFCI
bits
Pilot
N
pilot
bits
PO1
PO2

Figure 11: Power offsets in S-CCPCH.
Concerning FACH power settings optimization tests results have provided +4dB (i.e. 37
dBm) as the current recommended value to achieve BLER
target
of 15%.
This value ensures higher decoding success rate of RRC Connection Setup message sent
on the FACH along RRC Connection Establishment procedure, thus higher Network
Attach as well as Call Set-Up Success Rate.
This value results fairly high and in case high S-CCPCH power peaks it may occur that
IAOC get activated even in medium load conditions and high BLER observed over
periods longer than a UMTS frame due to IAOC. Separated power settings can be defined
for FACH and PCH.
The parameter k in Figure 10 and Figure 11 determines the total number of bits per
downlink S-CCPCH slot. To be noted that TFCI bits are continuously transmitted but
duty cycle is much lower than the one from the data field (i.e. FACH and PCH). The

Subject:
Date:
June 2006


Page 55 of 55

power levels of TFCI and pilot fields of the S-CCPCH can be defined as offsets relative
to the power used in the data field (PCH or FACH) by PO1 and PO2 respectively. Figure
11 depicts the single slot structure of a S-CCPCH including the different power offsets.
Increasing the value of the PO1 may help to improve the FACH/PCH BLER performance
by reducing the probability of TFCI decoding errors. Increasing PO2 may help the UE to
achieve synchronization and consequently perform a better reception of the S-CCPCH
frame.
Upon reception of the relevant NBAP messages the NodeB applies the requested
downlink power offset to the relevant S-CCPCH block as described in [10]. The
corresponding translations for power offsets PO1 and PO2 are presented in Table 30.

Figure 12:PICH Frame Structure.
5.6. PAGING INDICATOR AND ACQUISITION INDICATOR CHANNELS POWER
The PICH is a fixed rate (SF=256) physical channel used to carry the paging indicators.
Figure 12 illustrates the frame structure of the PICH. One PICH radio frame of length
10 ms consists of 300 bits (b0, b1,, b299). Of these, 288 bits (b0, b1 b287) are used
to carry paging indicators. The remaining 12 bits are not formally part of the PICH and
are not transmitted. PICH is always associated with an S-CCPCH to which a PCH
transport channel is mapped. Due to UE limitations PICH power settings have not been
investigated in details; current default value is set to 5dB (i.e. 28 dBm).
The AICH is a fixed rate (SF=256) physical channel used to carry Acquisition Indicators
that corresponds to signature s on the PRACH. The phase reference for the AICH is the
Primary CPICH. Figure 13 illustrates the structure of the AICH. The AICH consists of a
repeated sequence of 15 consecutive access slots, each of length 5120 chips. Each access
slot consists of two parts, an Acquisition-Indicator part consisting of 32 real-valued
symbols a0, , a31 and a part of duration 1024 chips with no transmission that is not
formally part of the AICH. Optimization tests focused on RACH and AICH have verified
that current default value equal to 5dB (i.e. 28 dBm) ensures high success rate of
Acquisition Indication message received at the UE.
The PICH is continuously transmitted with a duty cycle of 96% whereas AICH is
transmitted only when RACH preambles need to be acknowledged.

Subject:
Date:
June 2006


Page 56 of 56

Figure 13:AICH Frame Structure.
Parameters pICHPower and AICHPower define the power setting of PCH and AICH
respectively and are presented in Table 30.
5.7. DOWNLINK LOAD DUE TO COMMON CHANNELS POWER SETTINGS
In order to provide an indication on the power available for downlink dedicated channels,
it is important to calculate the total downlink common channels load as a percentage with
respect to the total downlink power defined by the PA size.
Taking into consideration the Common Channels power settings and duty cycles
described in previous sections, the total downlink Common Channels power (upon
conversion in Watt) is equal to 8.7 Watts. This value is valid only if it is assumed that the
FACH is always transmitted while PCH is inactive. In case of 20 Watts PA the total
downlink Common Channels Load is given by following power ratio in Watts:
Total downlink Common Channels Load= 8.7/20 [Watts] = 43.5%
This value is very high and definitely impacts the cell capacity when FACH is used.
However taking into consideration that currently FACH power is transmitted only along
call set-up or call re-establishment procedures (e.g. when RRC Connection Setup message
is sent), re-calculation of total downlink Common Channels power without taking into
consideration the S-CCPCH power results in 3.7 Watts. Therefore the downlink Common
Channels load results:
Total DL Common Channels Load= 3.7/20 [Watts] = 18.5%
This is the value that is normally assumed for Total downlink Load evaluation.

Subject:
Date:
June 2006


Page 57 of 57

5.8. TRANSLATION SETTINGS

Table 30 summarizes the different parameters involved in the power setting of the
downlink common channels.

Parent
Object
Parameter name Attribute Syntax &
Range
Default
Value
Access
pCPICHpower Real [dBm]
-10, -9.9, +50
+33.0dBm
pSCHpower Real - [dB]
-35.0, -34.9, +15.0
-3.0dB
sSCHpower Real - [dB]
-35.0, -34.9, +15.0
-5.0dB
LCell
bchPower Real - [dB]
-35.0, -34.9, +15.0
-3.0dB
OMC
RW-LO
RNC-CLI
RW-LO
pCHPower Real - [dB]
-35.0, -34.9, +15.0
+4dB
pICHPower Integer - [dB]
-10,-9.9, +5
-5.0dB
SecondaryCCPCH.
powerOffset1
Real - [dB]
0.0, 0.25, 6.0
+0.0dB
SCCPCH
Pch
SecondaryCCPCH.
powerOffset2
Real - [dB]
0.0, 0.25, 6.0
+0.0dB
fACHTrafPower Real - [dB]
-35.0, -34.9, +15.0
-5.0dB
fACHSigPower Real - [dB]
-35.0, -34.9, +15.0
+4.0dB
SecondaryCCPCH.
powerOffset1
Real - [dB]
0.0, 0.25, 6.0
+0.0dB
SCCPCH
Fach
SecondaryCCPCH.
powerOffset2
Real - [dB]
0.0, 0.25, 6.0
+0.0dB
PRACH
aICHPower Integer - [dB]
-22, -21.9, +5
-5.0dB
OMC
RW-PL
RNC-CLI
RW-PL
Table 30: Power Settings for Downlink Common Channels

Subject:
Date:
June 2006


Page 58 of 58

6. Power Control for Uplink Dedicated Channels
6.1. UPLINK DEDICATED CHANNELS FRAME STRUCTURE
There are two types of uplink Dedicated Physical Channels DPCH, the uplink Dedicated
Physical Data Channel (uplink DPDCH) and the uplink Dedicated Physical Control
Channel (uplink DPCCH), which are I/Q code multiplexed within each radio frame as
depicted in Figure 14.
The uplink DPDCH is used to carry the Dedicated transport Channel. The number of
uplink DPDCH transmitted per each radio link can vary.
The uplink DPCCH is used to carry control information that consists of known pilot bits
to support channel estimation for coherent detection, transmit power-control commands,
feedback information, and an optional transport-format combination indicator. There is
one and only one uplink DPCCH on each radio link.

Pilot
N
pilot
bits
TPC
N
TPC
bits
Data
N
data
bits
T
slot
= 2560 chips, 10 bits
1 radio frame: T
f
= 10 ms
DPDCH
DPCCH
FBI
N
FBI
bits
TFCI
N
TFCI
bits
T
slot
= 2560 chips, N
data
= 10*2
k
bits (k=0..6)
Figure 14: Frame structure for uplink DPDCH/DPCCH
The parameter k in Figure 14 determines the number of bits per uplink DPDCH slot.
More information concerning uplink DPCH are specified in [1].
6.2. INITIAL TRANSMIT POWER FOR UPLINK DEDICATED CHANNELS
At radio link establishment, an Uplink Dedicated Channel has to be set-up only at call set-
up. The uplink DCH is set-up as soon as downlink DCH synchronization has been
achieved. The DPCCH is always transmitted before the data part DPDCH.
The initial transmit power for uplink DPCCH is determined by the open loop power
control procedure.
To delay the start of uplink DPDCH transmission, a power control preamble procedure
can be used. In that case the transmission start between DPCCH and DPDCH is delayed.
6.2.1. Uplink DCH Open Loop Power Control
When establishing the first uplink DPCCH the UE sets the power level according to (21).

Subject:
Date:
June 2006


Page 59 of 59

DPCCH_Initial_power = DPCCH_Power_offset CPICH_RSCP (21)
Where
DPCCH_Power_offset is a parameter in dB.
CPICH_RSCP is measured by the UE.
The reason to use uplink open loop power control is due to the high speed of uplink Inner
Loop Power Control function. The UE calculates an initial transmit power value and once
the new radio link has been synchronized, the inner loop power control function will
make the uplink power converge.
6.2.2. Uplink DPCCH Power Control Preamble
During the call set-up phase, a Power Control Preamble can be used to make sure that the
inner loop power control has converged before the actual start of the data transmission.
This uplink DPCCH power control preamble is a period of N
pcp
uplink DPCCH radio
frames transmitted prior to the start of the uplink DPDCH.
The length N
pcp
of the power control preamble is a UE-specific parameter signaled by the
RNC and can take the values between 0 and 7 radio frames.
If the length of the power control preamble is greater than zero, the power control
procedure used during the power control preamble differs from the ordinary power
control that is used afterwards. After the first slot of the power control preamble, the
change in uplink DPCCH transmit power is initially given by formula (22).

DPCCH
=
TPC
TPC_cmd (22)

Where TPC_cmd is derived by the UE according to Algorithm1 for processing TPC
commands (i.e. every timeslot) as described in 6.4.3.2.
After the end of the uplink DPCCH power control preamble, the UE is expected to use the
inner loop power control with one of the power control algorithms described in 6.4.3.3.
Since the NodeB is not informed of the length of the Power Control Preamble period, the
NodeB is designed to cope with the following requirements:
The TFCI are filled with zeros during the PCP period as specified in [1].
No transmission is done on the uplink DPDCH during the PCP period as specified
in [2].
The DPCCH power control preamble mode in the UE is configured by the RNC through
RRC signalling as described in [7].

Subject:
Date:
June 2006


Page 60 of 60

The power offset defined in (21) can be set via parameter dPCCHPowerOffset. If power
offset is set to high values this ensures a short synchronization time but may cause uplink
power overshoot (as described in Figure 15) and result in an increase of the uplink
interference. Otherwise if too low values are used, the uplink interference is minimized
but synchronization time may take longer thus negatively impacting the call setup delay.

Therefore optimal dPCCHPowerOffset parameter setting should ensure uplink power
overshoot minimization without negative impact on call set-up delay. This needs to be
verified in different RF scenarios as the initial uplink DPCCH power is calculated by the
UE based also on RSCP measurements.
Figure 15:Uplink Transmitted Power with and without overshoot.

Several values have been tested for this parameter with different UE, in different RF
conditions. A good compromise between overshooting and convergence time is achieved
with a value of 84dBm. This does not avoid completely the overshooting (over 15dB
were still measured in very good RF conditions) but some type of UE were not able to
establish a connection to the network with lower values. Further testing need to be done
to ensure 100dBm leads to an absence of overshoot and still good performances (Call
Success Rate is 100%).

The use of uplink DCH power control preamble is controlled using parameter
pCPreamble. The basic information for these parameters is summarized in Table 31.
& Range
UEGTFCS
Instance
Default
Value
Access
ULDPCHInfo
dPCCHPowerOffset Integer [dB]
-164, -162-6
All -84dB RNC-CLI
RW-EL

Subject:
Date:
June 2006


Page 61 of 61

pCPreamble Integer - 0..7
[Radio frames]
All 7
Table 31: Initial uplink power control related parameters.
6.3. OUTER LOOP POWER CONTROL FOR UPLINK DEDICATED CHANNELS
6.3.1. Uplink DCH Outer Loop Power Control Principle
The uplink Outer Loop Power Control is in charge of updating the SIR
target
in the NodeB
in order to provide the Block Error Rate performances corresponding to the requested
Quality of Service. The principle is depicted in Figure 16.
This control loop runs in the RNC with a speed up to 100 Hz and may use various
parameters to adjust the SIR
target
, such as:
The quality estimate of the data received at the NodeB,
The number of soft legs,
The interference/load level in the uplink channel,
Measurement reports from the UE.
The new SIR
target
is then sent to the NodeB over the IuB/Iur links and is used by the inner
loop function located in the NodeB as a reference against the SIR
est
,.
UE
BTS
DCH quality estimate,
CRC check reports
Inner
Loop
PC
RNC
Pmax, Pmin
= f(QoS, load,
cell size)
inner loop algo,
deltaTPC
up/down
data
NBAP
NBAP
DCCH
inner loop algo,
Measurements:
RSSI, SIR...
DCH FP
UL outer Loop PC
DCH FP
SIR target
update

Figure 16: Uplink outer loop power control principle
The method used to calculate the optimal SIR
target
value for providing the requested BLER
is not specified in the 3GPP standards, where only a set of minimal NodeB performances
for a set of given radio conditions is required (refer to [5]). The developed adaptation
algorithm is Lucent Technologies proprietary and this approach involves a close
interaction between the Radio Resource Management, the Call Admission Control, the
Uplink Overload Control and the Outer Loop Power Control functions.

Subject:
Date:
June 2006


Page 62 of 62

6.3.2. Inter-RNC soft handover
In case of Inter-RNC soft handover the uplink Outer Loop Power Control function is
located in the S-RNC, which will determine the SIR
target
for all the NodeBs involved in
the connection. Figure 17 shows the location of uplink Outer Loop Power Control
function within the different UTRAN Network Elements.
NodeB
NodeB
NodeB
NodeB
uplink
OLPC
Frame
selector
Iub
Iur
Iub
Iub
Iub
Drift RNC
Serving RNC
To
CN
UE
Iu

Figure 17: Location of the uplink Outer loop function.
The Outer Loop Power Control procedures can be separated into following steps:
1. Radio link quality estimation: Each NodeB involved in the call performs an
estimation of the radio link quality on the uplink DCH; the results are provided to
the S-RNC.
2. Frame selector function: the results of the radio link quality estimation are used by
the Frame Selector function to provide unique quality estimation values to the
outer loop power control algorithm.
3. Update of the SIR
target
: The outer loop power control algorithm uses the quality
estimation values as well as other parameters to adjust the SIR
target
if required.
4. Update of the SIR
target
at the NodeB(s): The new SIR
target
value is sent to the
involved NodeB(s) if required. Upon reception of this information, the NodeB
applies the new SIR
target
to the related uplink Inner Loop Power Control function.
The information on the radio link quality estimation (uplink direction) as well as on the
new SIR
target
(downlink direction) are transmitted using the DCH Framing Protocol.

Subject:
Date:
June 2006


Page 63 of 63

The DCH frames are transmitted every Transmission Time Interval whose values in
general can be equal to 10, 20, 40 or 80 ms depending on the different radio channel
coding.
6.3.3. Radio Link Quality Estimation
The radio link quality in the uplink can be estimated by the UTRAN using different
metrics described in the 3GPP standards [3]. In the following the most relevant ones are
presented and discussed.
6.3.3.1.BLER metric
The BLER metric is defined as the average error rate of transport blocks estimated on a
transport channel containing a CRC check.
The NodeBs send the result of the CRC check to the S-RNC over the I
ub
or the I
ur

interface, using the CRCI field of the DCH frames. For every transport block included in
the data frame a CRCI bit is present, irrespective of the presence of a transport block CRC
on the air interface. CRCI bit is set to 1 if CRC check failed and to 0 if either no CRC
was present or CRC check passed. The BLER estimation is done after the frame selection
by averaging the CRCI. Therefore this metric is computed at the S-RNC.
6.3.3.2.Quality Estimation metrics
The BLER is not an appropriate metric to evaluate the quality of transport channels with
high quality requirements (e.g. BLER
target
= 10
-3
).
Two alternative quality metrics are defined to refine estimate of the transport channel
quality: the Transport BER and the Physical Channel BER. These metrics are computed
at the NodeB and then forwarded to the S-RNC. More details can e found in [3].
6.3.3.2.1. Transport Channel BER metric
The Transport Channel BER is defined as the error rate of the DPDCH "bits" before
decoding. The "bits" before decoding should actually be called symbols to avoid
misunderstandings of the definitions. Using the above definition the "Transport Channel
BER" is the SER for a given DCH. The detailed definition of the Transport Channel BER
is provided in [3].
The "Transport Channel BER" is usually determined by decoding, re-coding and by
comparing the re-coded symbols with the received symbols after hard-decision. The
decoding and re-coding can be easily done for convolutional codes. At the point of time
where the most-likely trellis path is selected, the output bit and the according input
symbols are known. These symbols need to be compared with the input symbols to find
the input symbols being in error. By counting the number of wrong symbols the Transport
Channel BER can be determined.
There is no unique mapping between the SER (bits in error before decoding) and the BER
(bits in error after decoding). This mapping strongly depends on the radio environment
and can change dramatically due to changes in the delay profile or UE speed.
The Transport channel BER value is calculated within the range [0,1] and reported in the
unit TrCh_BER_LOG as defined in [6]. The mapping between the measured quantity
value and the reported value is included in Table 32. The defined range and resolution of

Subject:
Date:
June 2006


Page 64 of 64

the Transport Channel BER includes SER values of up to 100% while the expected value
of the SER can't exceed 50% in a long term average. Thus, the high SER are quite
unlikely to occur.
Reported value Measured quantity value
TrCh_BER_LOG_000 Transport Channel BER = 0
TrCh_BER_LOG_001 - < Log10(Transport Channel BER) < -2.06375
TrCh_BER_LOG_002 -2.06375 Log10(Transport Channel BER) < -2.055625

TrCh_BER_LOG_255 -0.008125 Log10(Transport Channel BER) 0
Table 32: Mapping of Transport Channel BER metric.
If the DCH FP includes transport blocks for the DCH, which was indicated as "selected"
in the QE-selector IE, this Transport Channel BER is also mapped onto the Quality
Estimation field of the DCH frame.
If no Transport channel BER is available the QE field is then filled with the Physical
channel BER.
The conversion of the transport channel BER in the QE is performed using the formula
(23). The reverse conversion is provided in (24).
Many contributions to the 3GPP standards process have that the mean Transport Channel
BER is not a reasonable measurement for estimating the BER or BLER after decoding.

= =
< +
(
=
0 BER for 0 QE
1 BER 0 for 255
0.008125
log10(BER)
INT QE

(23)
= =
< =
0 QE for 0 BER
255 QE 0 for 0.008125 *
10
255) - (QE
BER

(24)
6.3.3.2.2. Physical Channel BER metric
The Physical Channel BER is defined as the bit error rate on the DPCCH This
measurement has been included especially to provide quality information even if the
DPDCH does not contain any data. As for the Transport Channel BER, the Physical

Subject:
Date:
June 2006


Page 65 of 65

Channel BER is related to the SER. The SER on the DPCCH can be measured by using
the known pilot patterns, by decoding and re-coding of the TPC and TFCI information.
The detailed definition of the Physical Channel BER is provided in [3].
The Physical channel BER is measured in the range [0,1] and reported in the unit
TrCh_BER_LOG as defined in [6]. The mapping between the measured quantity value
and the reported value is included in Table 33.
Reported value Measured quantity value
PhCh_BER_LOG_000 Physical channel BER = 0
PhCh_BER_LOG_001 - < Log10(Physical Channel BER) < -2.06375
PhCh_BER_LOG_002 -2.06375 Log10(Physical Channel BER) < -2.055625

PhCh_BER_LOG_255 -0.008125 Log10(Physical Channel BER) 0
Table 33: Mapping of Physical Channel BER metric
If the IE QE-selector equals "non-selected" for all DCHs in the FP frame, then the QE is
the Physical channel BER.
In case no Transport channel BER is available, then the QE is the Physical channel BER.
The conversion of the physical channel BER in the QE is performed using the formula
(23). The reverse conversion is provided in (24).
6.3.3.2.3. Translation settings
The parameter qESelector, described in Table 34, defines the mode to be included in the
QE-selector IE in order to properly map the metrics in the QE field. The QE-selector IE is
conveyed in NBAP or RNSAP messages during Radio Link Setup or Radio Link
Reconfiguration procedure. Refer to [8] and [10] for more information.
Current default value indicates that Physical Channel BER is the metric used for radio
link quality estimation.

Subject:
Date:
June 2006


Page 66 of 66

name
Attribute Syntax
& Range
Parent Object
Instances
Default
Value
Access
All except 3 and 4 Y DedicatedTransport
Channel
qESelector Boolean
Y=selected,
N=non-selected
3 and 4 N
RNC-CLI
RW-EL
Table 34: Metric selection parameter.
6.3.4. Frame Selector Function
Each NodeB k involved in the call performs CRC checks on each transport block. The
resulting CRCI
k
and QE
k
are then sent to the S-RNC over the DCH-FP protocol. The
Frame Selector uses the CRCI
k
and QE
k
values to estimate a CRCI and QE which are
forwarded to the uplink outer loop power control algorithm.
6.3.5. Uplink Outer Loop Power Control Algorithm
6.3.5.1.Principle
The uplink outer loop power control algorithm uses the CRCI and QE values to evaluate
if there is a need to re-adjust the SIR
target
. The information flow of the outer loop power
control algorithm is depicted in Figure 18 assuming that the UE is in soft handover with
two cells.

In the RNC, the uplink outer loop function is made out one or more PHY entities. Each
PHY entity is associated to one transport bearer (i.e. one DCH FP pipe). In the case where
more than one DCH is associated to one transport bearer (e.g. coordinated transport
channels), one single PHY entity is in charge of that group of coordinated transport
channels. One example of coordinated transport channels is the case of the Voice Service
where the 3 voice sub-flows are associated to one single DCH Framing Protocol pipe.

Figure 19 gives an illustration of the uplink outer loop function for one call configured for
voice+64kPS+signalling configuration. Each PHY entity can request an update of the
uplink SIR
target
. This leads to the transmission of an uplink outer loop power control
command to every NodeB involved in the call to update the uplink SIR
target
, as well as the
re-synchronization of the SIR_calc values in the other PHY entities to that requested
value.

Subject:
Date:
June 2006


Page 67 of 67

Frame
Selector
CRCI, QE
OLPC
target SIR
ILPC
a

est. SIR
SIR
Estimator
Receiver
Decoder
ILPC
a

est. SIR
SIR
Estimator
Receiver
Decoder
target SIR target SIR
C
R
C
I
1
,

Q
E
1

ILPC
b

Transmitter
TX Pwr
TPCcmd
1
TPCcmd
2

UE
NodeB 2 NodeB 1
SRNC
C
R
C
I
2
,

Q
E
2

CRCI
1,
QE
1
CRCI
2,
QE
2

Figure 18:Information flow for Power Control based on CRC and QE
in soft handover scenario.

Subject:
Date:
June 2006


Page 68 of 68

PHY entity
for DCCH
PHY entity
for 64k PS bearer
PHY entity
for voice bearer
CRCI and QE information related to
DCCH transport blocks
CRCI and QE information related
to 64k PS transport blocks
CRCI and QE information related
to voice transport blocks
Request for update
of UL SIR target
Request for update
of UL SIR target
Request for update
of UL SIR target
SIRcalc
sync input
SIRcalc
sync input
SIRcalc
sync input
Generation of OUTER LOOP PC
frames
(as per 25.427)
UL SIR target commands sent
to BTS(s) involved in the call
UL outer loop function
sub flow
#1, #2
and #3

Figure 19: Uplink outer loop function for one call configured for voice+64kPS+signalling

Subject:
Date:
June 2006


Page 69 of 69

The parameters involved in the uplink outer loop power control algorithm are
summarized in the Table 35 and Table 37.

The parameters related to the SIR
target
initial value and limits are defined by translations
initialSIRTarget, minSIRTarget and maxSIRTarget on a per service combination basis
(e.g. PS data & signalling, standalone signalling, etc.) as they belong to the parent object
UEGTFCS. This means that a new set of these parameters is passed on to the TPU each
time a service is added or dropped from the call.
In general optimized settings of these parameters can be derived according to the
following rules:
InitialSIRTarget needs to be slightly above the average SIR target set point
MinSIRTarget needs to be equal or slightly below the minimum observed target.
MaxSIRTarget needs to be equal or slightly above the maximum observed target.
Furthermore the optimal initialSIRtarget value should minimize the convergence time,
hence the uplink power consumption & noise rise during the transient period, while the
optimal minSIRTarget and maxSIRTarget values should ensure that standard deviation of
measured uplink BLER around the defined target (e.g. 5%) is minimized as well as uplink
power interference.

Parent Object Parameter Name Attribute Syntax
& Range
UEGTFCS
Instances
Default
Value
Access
All Others +7dB
2 +5dB
3, 13, 14 +6dB
initialSIRTarget Real - [dB]
-8.2, -8.1, +17.3
47, 48, 49,
50, 51, 52,
53, 54
+8dB
All Except 3 0dB minSIRTarget Real - [dB]
-8.2, -8.1, +17.3 3 +4.9dB
All +8dB
2, 13, 14 +7dB
ULDPCHInfo
maxSIRTarget Real - [dB]
-8.2, -8.1, +17.3
31, 32, 33,
34, 35, 36
+9dB
RNC-CLI
RW-EL
Table 35: SIR target related parameters

Experimental results showed that the performances are degraded by using lower values
for qEThreshold. Therefore the decision on how to update the SIR_calc is based only on
the CRC check results.

Subject:
Date:
June 2006


Page 70 of 70

Optimal settings of parameters downSIRStep and upSIRStep are strictly related to the
uplink BLER target that has to be achieved on the considered uplink radio bearer. The
uplink BLER target on a transport channel is related to the value of the upSIRStep and
downSIRStep parameters by formula (25). During lab optimization different values of
downSIRStep and upSIRstep have been tested, verifying whether the standard deviation
of the uplink SIR target value at the NodeB is minimized, whilst maintaining the uplink
BLER performance at around 5% for PS calls.

|
|
.
|
\
|
=
yTarget BLERQualit 1
yTarget BLERQualit
p downSIRste upSIRstep

(25)

In general increasing upSIRStep values may improve the reactivity of the control loop in
highly variable radio environment, however that is to the detriment of its stability
resulting in SIR
target
overshoots.
6.3.6. Outer Loop Power Control Execution
When the uplink outer loop power control function generates a new SIR
target
value, it is
forwarded to the NodeB(s) involved in the connection to the UE. This information is sent
over I
ub
or I
ur
interface depending of the relation of the NodeB to the S-RNC.
The SIR
target
value is defined in dB with range [-8.217.3 dB] and step value of 0.1 dB.
Its transmitted values are mapped to the unit UL_SIR_TARGET as in Table 36.

SIR Transmitted values SIR Used values
UL_SIR_TARGET = 000 SIR Target = -8.2 dB

UL_SIR_TARGET = 254 SIR Target = 17.2 dB
UL_SIR_TARGET = 255 SIR Target = 17.3 dB
Table 36: Mapping of SIR
target
values.
The mapping of the SIR
target
value need no synchronization between the different
NodeB(s) involved in the radio connection. Upon reception of a DCH control frame with
an outer loop power control field, the NodeB applies (i.e. within 10 ms) the new SIR
target
value to the related uplink inner loop power control function.

& Range
Parent
Object
Instance
Default
Value
Access
1, 3, 4 255.0
All others 130.0
qEThreshold Real
0.0, 1,0.255.0
19, 20, 21, 22,
29
150.0
Dedicated
Transport
Channel
downSIRStep Real - dB 1 0.002
RNC-CLI
RW-EL

Subject:
Date:
June 2006


Page 71 of 71

2, 5, 27, 28,
29
0.001
3, 4 0
0.000,0.001 0.512
All others 0.01
All others 0.2
2 0.15
3, 4 0
upSIRStep Real - dB
0.000,0.001 0.512
27, 28, 29, 30 0.02
sIRAllowance Real - dB
0,0.1.10
All 0
1 0.4
2 0.025
3, 4 0
8 0.2
SIRDownAdditional Real - dB
0.000,0.001 0.512
All others 0.05
1, 6, 7, 9, 19 120
2, 3, 4, 27, 28,
29, 30
250
5 200

NbOfConsecutive
GoodFramesThreshold

IntType
[0,1 .255]
20, 21, 22 160

Table 37: SIR calculation related parameters.

Subject:
Date:
June 2006


Page 72 of 72

6.4. INNER LOOP POWER CONTROL FOR UPLINK DEDICATED CHANNELS
6.4.1. Uplink Inner Loop Power Control Principle
The Uplink Inner Loop Power Control is adjusting the transmit power of the UE in order
to achieve the SIR
target
provided by the outer loop power control function for a considered
uplink dedicated channel. Due to the speed of the inner loop (up to 1500 Hz), the only
elements involved are the UE and the NodeB. Figure 20 describes the principle of uplink
inner loop power control.

Node B UE
transmit power
=
f(TPC,deltaP,
Pmin, Pmax,
....)
compare with
SIR
target
SIR estimate
TPC
data
TPC = Power_up
or Power_down
c

Figure 20: Uplink inner loop power control principle.
The inner loop power control procedure can be separated in four basic steps:

1. The UE transmits the user data to the NodeB on an uplink DCH.
2. The NodeB performs the SIR estimation and compares the result with the SIR
target
.
3. If required, the NodeB will send the transmit power control commands to the UE
using the downlink DPCCH. The rate of TPC commands is up to 1500 Hz.
4. According to the transmit power control commands, the UE will either keep its
transmit power constant, or increase or decrease it by
TPC
[dB]. Irrespective of
the TPC commands the UE transmit power must remain within a range defined by
[P
min
;P
max
].
6.4.2. SIR Estimation and TPC Commands Generation
6.4.2.1.SIR Estimation
The algorithm implemented in the UMTS Releases from Lucent Technologies to
estimated the SIR
est
are based on explicit estimation where the signal energy and the noise
variance are estimated.

Subject:
Date:
June 2006


Page 73 of 73

6.4.2.2.Transmit Power Control commands
6.4.2.2.1. TPC Generation
Based on the SIR
est
, the NodeB generates TPC commands according to the following rule:

If SIR
est
>= SIR
target

Then
TPC command = "0" [requesting a power decrease from the UE]
Else
TPC command = "1" [requesting an increase of the power at the UE]

6.4.2.2.2. TPC bit pattern
Depending on the downlink slot format, the TPC command may be transmitted over the
air using one of the three TPC bit patterns described in Table 38 from [1]:

TPC Bit Pattern
N
TPC
= 2 N
TPC
= 4 N
TPC
= 8
Transmitter
power control
command
11
00
1111
0000
11111111
00000000
1
0
Table 38: TPC bit pattern
The relation between the slot format value and the N
TPC
values can be found in [1]. The
TPC commands are sent in TPC field of the downlink DPCCH.
6.4.2.2.3. TPC timing
According to the specifications, the UE is expected to change its uplink DPCH output
power at the beginning of the first uplink pilot field after the reception of the TPC
command, as illustrated in the Figure 21.
The SIR measurement periods illustrated are examples; other ways of measurement are
allowed to achieve accurate SIR estimation.
If there is not enough time for the UE to respond to a TPC, the action can be delayed until
the next slot at the beginning of the next pilot field.
To be noticed that this mechanism of uplink inner loop timing has only non-tunable
parameter. More information on the uplink power control timing can be found in [2].

Subject:
Date:
June 2006


Page 74 of 74

Data2 Data1 Data1
T
P
C
Data1
T
P
C
PILOT PILOT
PILOT
Response
To TPC
T
P
C
PILOT TFCI TPC
DL-UL timing
offset (1024 chips)
Slot (2560 chips)
PILOT PILOT Data2 Data1
T
P
C
PILOT PILOT TFCI TPC
Slot (2560 chips)
Propagation delay
UL SIR
measurement
In response
to TPC the UE changes
the power accordingly
DL DPCCH
at UTRAN
Propagation delay
DL DPCCH
at UE
UL DPCCH
at UTRAN
UL DPCCH
at UE
512 chips
TF
CI
TF
CI

Figure 21: Uplink inner loop timing
6.4.3. Uplink Inner Loop Power Control Algorithms
6.4.3.1.Overview
In soft handover situations the UE may receive up commands on some legs and down
command on others as depicted in Figure 22. In such case the UE should decrease its
transmit power, unless there is a consensus amongst the cells in the Active Set for
increasing it.

The softer handover case is an exception. In that case all the involved links are requested
to send back to the UE the same TPC command. The UE is informed via RRC Signalling
and can perform some further combining enhancing the detection of the correct TPC
word.
Upon creation of a new link the UE is informed, if this new link will send the same TPC
command as an existing one, by reading the TPC Combination Index IE. This is sent
embedded in the Downlink DPCH info RRC message as described in [7].
If the value is the same as the one of an existing link, it implies for the UE that the TPC
commands sent by the 2 radio links are identical and that it can apply the related
algorithm for softer combining case.
The RNC stores for each of the DCH that it is supporting the TPC combination index
value of each one of the supported links. Upon creation of a new link in softer handover,
the RNC sets the TPC combination index value of this new link to a value equal to the
other links in softer handover.

Subject:
Date:
June 2006


Page 75 of 75

Figure 22: TPC received commands in soft/softer handover scenario

In the case a UE is in soft(er) handover, upon reception of the TPC commands in a slot,
the UE derives a single TPC command, named TPC_cmd. There are two algorithm
supported by the UE to derive its TPC_cmd value from the TPC it receives from the
UTRAN. The RNC controls which one is used and configures the UE through RRC
signalling. The value of PowerControlAlgorithm is extracted from a related MIB profile,
which is manageable by OAM.

If PowerControlAlgorithm indicates algorithm1, then the layer 1 parameter
PCA takes the value 1 and algorithm 1 is supposed to be used.
If PowerControlAlgorithm indicates algorithm2, then the layer 1 parameter
PCA takes the value 2 and algorithm 2 is supposed to be used.

Note however that the behavior of the inner loop at the NodeB does not change upon the
algorithm. It always sends one TPC word per slot and the TPC value depends on the
comparison between SIR
estimated
and SIR
target
.

Therefore the description of the two algorithms is provided as information. Only the
method to configure the UE is to be specified from an UTRAN point of view.
6.4.3.2.Algorithm 1 for processing TPC commands
The first algorithm, algorithm1, is the basic inner loop algorithm running at a speed of
1500Hz. This algorithms works differently depending on which scenario the UE is
involved in.
6.4.3.2.1. TPC command received when UE not in soft/softer handover
When a UE is not in soft handover, only one TPC command, either 0 or 1, is received
in each slot. In this case, the value of TPC_cmd is derived as follows:

If TPC command = "0

Subject:
Date:
June 2006


Page 76 of 76

Then
TPC_cmd = "-1"
Else
TPC_cmd = "+1"
6.4.3.2.2. TPC commands received form links in softer handover
When the UE has the knowledge that some TPC commands received in a slot are the
same, e.g. when in softer handover, the UE combines these N TPC commands into one
single TPC_cmd
softer
.
6.4.3.2.3. TPC commands received from links in soft combining
In the case the UE receives different TPC commands from links in soft handover, it
conducts a soft symbol decision W
i
on each of the power control commands TPC
i
,
(i [1..N]), where N is the number of TPC commands from radio links of different radio
link sets, that may be the result of a first phase of TPC combination.
The UE derives a combined TPC command, TPC_cmd, as a function of all the N soft
symbol decisions W
i
:
TPC_cmd = (W
1
, W
2
, W
N
),
TPC_cmd can take the values +1 or -1 depending on the following:
If the N TPC
i
commands are random and uncorrelated, with equal probability of
being transmitted as "0" or "1", the probability that the output of is equal to +1
shall be greater than or equal to 1/(2
N
), and the probability that the output of is
equal to -1 shall be greater than or equal to 0.5. This rule means that, regardless
of the value of N, the probability of having power decrease is always greater than
the probability of having power increase. If N equals 2, for instance, then the
probability of power increase is equal to 1/2
N
=1/4=0.25 that is still less than 0.5.
If the TPC commands from all the radio link sets are reliably +1 then the output
of is equal to +1.
If a TPC command from any of the radio link sets is reliably 0 then the output of
is equal to 1.
6.4.3.3.Algorithm 2 for processing TPC commands
With the second algorithm, algorithm2, the inner loop speed is decreased to 300Hz.
This algorithm increases the reliability of the TPC commands and enables the emulation
of adjustment step sizes smaller than 1.0 dB in e.g. static radio conditions. Additionally, it
allows turning off the power control by transmitting an alternating series of TPC
commands.
The disadvantage of this algorithm is that due to the slower inner loop speed, the tracking
of fast fading channel can be affected.

Subject:
Date:
June 2006


Page 77 of 77

6.4.3.3.1. TPC command received for UE not in soft/softer handover
When a UE is not in soft handover, only one TPC command is received in each slot. In
this case, the UE processes the received TPC commands on a 5-slot cycle, where the sets
of 5 slots are aligned to the frame boundaries without any overlap between each set of 5
slots. Under such conditions, the inner loop power control runs at a rate of 300 Hz.

The value of TPC_cmd is derived as follows:
For the first 4 slots of a set, TPC_cmd = 0 regardless of the value of the TPC
command received in each of these slots.
For the fifth slot of a set, the UE uses hard decisions on each of the 5 received
TPC commands as follows:

If all 5 hard decisions in the set are 1 then TPC_cmd = +1 in the 5
th
slot
If all 5 hard decisions in the set are 0 then TPC_cmd = -1 in the 5
th
slot
Otherwise TPC_cmd = 0 in the 5
th
slot

6.4.3.3.2. TPC command received from links in softer handover
The same procedure as the one used for algorithm1 is applied in the case the TPC
commands are received from links in soft handover and a combined TPC_cmd
softer
is
generated.
6.4.3.3.3. TPC command received from links in soft handover
To combine the TPC commands from radio links of different radio link sets, the UE first
makes a hard decision on the value of each TPC
i
, where this parameter is defined as in
6.4.3.2.3.
The UE then follows this procedure for 5 consecutive slots, resulting in N hard decisions
for each of the 5 slots. During this step, the UE assumes that:
The sets of 5 slots are aligned on the frame boundaries and there is no overlap
between them.
The value of TPC_cmd is zero for the first 4 slots.
Finally, after the 5
th
slots, the UE determines the value of TPC_cmd for the 5
th
slot
according to Equation 26.

Subject:
Date:
June 2006


Page 78 of 78

A temporary TPC command, TPC_temp
i
, is determined for each of the N sets of 5 TPC
commands as follows

If all 5 hard decisions within a set are "1" then TPC_temp
i
= +1
If all 5 hard decisions within a set are "0" then TPC_temp
i
= -1
Otherwise TPC_temp
i
= 0

A combined TPC_cmd(5th slot) is calculated for the 5th slot following:

If
5 . 0 _
1
1
>
=
N
i
i
temp TPC
N

TPC_cmd(5
th
slot) is set to 1
If any of the TPC_temp
i
equal -1 TPC_cmd(5
th
slot) is set to -1
Otherwise TPC_cmd(5
th
slot) is set to 0

Equation 26: TPC_cmd calculation when in soft handover
The parameter describing the algorithm type is powerControlAlgorithm. It is recommended
to use the algorithm 1. The basic information is provided in Table 39.

Parent Object Parameter name Attribute Syntax &
Range
ULDPCHInfo powerControlAlgorithm Enumerated
(pCALGORITHM-1,
pCALGORITHM-2)
pCALGORITHM-1 RNC-CLI
RW-EL
Table 39: Uplink inner loop power control algorithm related parameter.
6.4.4. Uplink Inner Loop Power Control Execution
6.4.4.1.Principle
Upon derivation of a single TPC_cmd, the UE adjusts the transmit power of the uplink
DPCH(s) according to:

If TPC_cmd equals 1 then the transmit power of the uplink DPCCH and DPDCHs
shall be increased by
TPC
dB.
If TPC_cmd equals -1 then the transmit power of the uplink DPCCH and DPDCHs
shall be decreased by
TPC
dB
If TPC_cmd equals 0 then the transmit powers remain unchanged

Subject:
Date:
June 2006


Page 79 of 79

6.4.4.2.Transmit Power Step Size
The S-RNC sends to the UE the transmitter adjustment step size
TPC
via RRC
signalling.. The 3GPP standard specifies that the UE transmitter shall support TPC
commands adjustments steps
TPC
of 1dB, 2dB and 3dB. As the supported step size is UE
specific more information on this parameter and its accuracy should be searched in the
3GPP specification ([4]).
6.4.4.3.Transmit power limits Maximum transmit power
Because the performances of CDMA-based networks are limited by interference, it is
very important that each UE never transmits with a power above a maximum allowed
value. This value is set by the Maximum allowed uplink Tx power" transmitted from the
S-RNC to the UE via RRC signalling as described in [7].
In the case several radio links are involved in a connection, e.g. when in soft/softer
handover, the maximum transmit power for the UE should be calculated at the S-RNC as
follows:
Max_UL_Tx_Power = min(Max_UL_Tx_Power
1
, .. Max_UL_Tx_Power
N
)

Where N is the number of radio links involved in the call. This limit is calculated at call-
setup, at addition of a new link or at deletion of an existing one.
Upon reception of the relevant RRC message, the UE keeps its uplink transmit power
below the indicated power value. If the new calculated uplink transmitted power is above
an indicated maximum power value, the UE is expected to adjust the power to a level
below the limit.
6.4.4.4.Transmit power limits Minimum transmit power
The minimum transmit power cannot be set by the radio network and is specified in [4].
The minimum output power is defined as an averaged power in a time slot measured with
a filter that has a Root-Raised Cosine filter response with a roll off = 0.22 and a
bandwidth equal to the chip rate. The minimum output power shall be less than 50dBm.
Parameter tPCStepSizeUL defines the uplink power control step size. Its recommended
value is equal to step size of 1 dB.
The maximum allowed uplink transmit power in the UE is specified by parameter
MaxULDPCHPower. As a default setting, it is suggested to set the value to the maximum
power level achievable by the UE. This should ensure maximum uplink coverage. A
tuning of this value can be performed in the case of issues with uplink interference. The
maximum power value is defined via the UE Power Classes as described in Table 40.
More information for these parameters is included in Table 41.

Current default setting (i.e. 33dBm) ensures that the uplink power is limited by the UE
H/W capability thus no adjustment is required depending on the type of UE. Currently
available UEs belong to Power Class 3 or 4.

Subject:
Date:
June 2006


Page 80 of 80

Power Class Nominal maximum output power Tolerance
1 +33dBm +1/-3 dB
2 +27dBm +1/-3 dB
3 +24dBm +1/-3 dB
4 +21dBm 2 dB
Table 40: UE Power Classes.

& Range
UEGTFCS
Instance
Default
Value
Access
All 0 tPCStepSizeUL Integer
0 (1.0dB), 1 (2.0dB) 50, 54 1
ULDPCHInfo
maxULDPCHPower Integer - dBm
-50, ..,+33
All 33dBm
RNC-CLI
RW-EL
Table 41: Uplink power control step size and UE transmit power limit parameters.
6.4.5. Uplink DPCCH/DPDCH power gains
6.4.5.1.Effect of bit rate variability on uplink power control
On the uplink the DPDCH spreading factor and the amount of puncturing/repetition is a
function of the amount of data to be transmitted over a radio frame. Therefore if the
DPDCH power level is kept constant, signal energy per symbol will vary with the instant
data rate.
Considering as an example the service combination uplink 64kbps + signalling as
defined in [12]. Table 42 gives the DPDCH Energy per Symbol, assuming the UE
transmit power level is kept constant. The ratio of energy per symbol is calculated using
(27).

|
|
|
.
|
\
|
i
i i
i
i i
N RM
N RM
R
TFC reference for
TFC considered for
log 10

(27)
Where
The reference TFC is TFCI number 9
N
i
is the number of channel coded bit on DCH
i

RM
i
is the associated rate matching value (i.e. 140 for signalling, 160 for 64kbps)

At the same time, the uplink DPCCH is used by the uplink inner loop as reference
sequence on which the SIR
target
has to be achieved. Given the results in Table 42, if the
SIR
target
remains constant and if the power ratio between the uplink DPDCH and DPCCH
were to be constant, the SIR on the DPDCH (hence the QoS) would be a function of the
radio frame TFCI value, which is not acceptable.

Subject:
Date:
June 2006


Page 81 of 81

TFCI TFC
64kbps
bearer
TFC
signalling
bearer
Number of bits
/ radio frame
for 64kbps
Number of bits
/ radio frame
for signalling
SF DPDCH Energy
per symbol
0 TF0 TF0 0 0 No Tx No Tx
1 TF1 TF0 534 0 64 X + 6.3 dB
2 TF2 TF0 1062 0 32 X + 3.3 dB
3 TF3 TF0 1590 0 16 X + 1.5 dB
4 TF4 TF0 2118 0 16 X + 0.3 dB
5 TF0 TF1 0 129 256 X + 11.9 dB
6 TF1 TF1 534 129 32 X + 5.2 dB
7 TF2 TF1 1062 129 16 X + 2.7 dB
8 TF3 TF1 1590 129 16 X + 1.2 dB
9 TF4 TF1 2118 129 16 X
Table 42: DPCH energy per symbol figure per "uplink 64kbps+signalling"
In order to make the quality of service provided to a DPCH independent of the instant
data rate, the concept of power gain between the uplink DPDCH and DPCCH has been
introduced, allowing compensation for the fluctuations of processing gain inherent to rate
fluctuations. The mechanism is based on the fact that the uplink DPCCH and DPDCH(s)
are transmitted on different codes as illustrated in the Figure 23.

I
j
c
d,1
d
S
dpch,n

I+jQ
DPDCH

c
c
c
DPCCH

S
jQ

Figure 23: Uplink DPDCH and DPCCH transmission.

After channelization, the real-valued spread signals are weighted by power gain factors
c

for the DPCCH and
d
for all DPDCHs. At every instant in time, at least one of the values
c
and
d
has the amplitude of 1.0. The -values are quantized into 4 bit words as
described in Table 43.

Subject:
Date:
June 2006


Page 82 of 82

Signalling values for
c
and
d

Quantized amplitude ratios
c
and
d

15 1.0
14 14/15
13 13/15
12 12/15
11 11/15
10 10/15
9 9/15
8 8/15
7 7/15
6 6/15
5 5/15
4 4/15
3 3/15
2 2/15
1 1/15
0 Switch off
Table 43:Quantization of the gain factors.
6.4.5.2.Gain factors controlling procedures
6.4.5.2.1. Principle
In order to compensate for the DPDCH processing gain fluctuations, there is a set of
c

and
d
values defined per combination of the TFCS. They are transmitted by the RNC to
the UE via RRC signalling and to the NodeB via NBAP messages. Beside of the gain
factor values to be signaled in explicitly signalling case or to be taken as reference values
in computing case, a Reference TFC ID is included in order to calculate the gain factors
for the specific TFC.
There are two ways of controlling the gain factors of the DPCCH code and the DPDCH
codes for different TFCs in normal (non-compressed) frames:

c
and
d
are signaled for the TFC, or

c
and
d
are computed for the TFC, based on the signaled settings for a
reference TFC. Several reference TFCs may be signaled.

To be noticed that the update of the relative gain factors between DPDCH and DPCCH
occurs at a rate of once per radio frame and it is independent from the uplink inner loop
power control that updates every timeslot the DPCCH absolute power level with steps of
TPC
dB.
6.4.5.2.2. Signalling the gain factors
When the gain factors
c
and
d
are signaled for a certain TFC, the signaled values are
used directly for weighting of DPCCH and DPDCH(s). The gain factor values shall be

Subject:
Date:
June 2006


Page 83 of 83

populated with the values extracted from the OAM MIB table corresponding to the radio
bearer combination mapped onto the uplink DPDCH.

6.4.5.2.3. Computing the gain factors
Instead of being explicitly signaled, the gain factors
c
and
d
may be computed for
certain TFCs, based on the signaled settings for a reference TFC. An overview of this
method is provided as information in the next paragraphs.

First, the computed method introduces following variables:

c,ref
and
d,ref
denote the signaled gain factors for the reference TFC

c,j
and
d,j
denote the gain factors used for the j:th TFC
L
ref
denote the number of DPDCHs used for the reference TFC
L
,j
denote the number of DPDCHs used for the j:th TFC.

Then it defines the variable K
ref
and K
j
as defined in (28) and (29) and a nominal power
relation A
j
computed as in (30).

=
i
i i ref
N RM K

(28)
where:
RM
i
is the semi-static rate matching attribute for transport channel i,
N
i
is the number of bits output from the radio frame segmentation block for
transport channel i
the sum is taken over all the transport channels i in the reference TFC

=
i
i i j
N RM K

(29)
where the sum is taken over all the transport channels i in the j:th TFC.

ref
j
j
ref
ref c
ref d
j
K
K
L
L
A =
,
,

(30)

The gain factors for the j:th TFC are then computed as follows:

Subject:
Date:
June 2006


Page 84 of 84

If A
j
> 1 then
d,j
=1. and
c,j
is the largest quantized-value, for which the condition
c,j
1/A
j
holds
1

If A
j
1 then
d,j
is the smallest quantized-value, for which the condition
d,j
A
j

holds and
c,j
= 1.0

The RNC shall support the RRC signalling of
c
and
d
values based on the signaled
gain factors method.

In the case where the gain factors of all the combinations are not explicitly given, at least
one of the given values shall be used as a reference to derive the gain factor of the other
combinations according to the computed gain factors method.
6.4.5.3.Translations settings
The gain factors signalling method, the gain factors
c
and
d
are defined respectively by
the parameter gainFactorType, gainFactorBc and gainFactorBd. Information on the gain
factors related parameters is available in Table 44:
The prescribed method for the signalling of beta gain factor, in case of the computing
method, is the following:
explicitly declare the gain factor corresponding to the highest data rate of the
TFCS through the parameter referenceTFC and use this TFC as the reference
declare the other factor as computed using the above reference.
Due to the limitation of the UE
2
not supporting the computed method, the recommended
value for gainFactorType is sIGNALLED. With the latest UE versions
3
, supporting the
computed method, the parameter gainFactorType could also take the value
cOMPUTED.
Both parameters gainFactorBc and gainFactorBd can be defined with different values
depending on the selected service combination specified through different instances of
parent object UEGTFC. These parameter should ensure an optimal trade-off between low
BLER on both DPDCH and DPCCH and uplink power consumption.
As major performance issues were observed in soft/softer handover scenarios (due to
downlink outer loop power control not working, power imbalance, and soft/softer
handover delay) retune of gain factors has helped to mitigate the downlink DCH power
imbalance. This was achieved by increasing the gainFactorBc with respect to
gainFactorBd, i.e. from 8 and 15 to respectively 12 and 15. TPC detection success rate
has improved without causing higher uplink power consumption.

1
Since c,j may not be set to zero, if the above rounding results would equal zero, c,j shall be set to the
lowest quantized amplitude ratio of 1/15 as specified in Table 43.
2
This concerns mainly the Qualcomm TM5200, FM6208
3
This is for instance the case of the Qualcomm TM6200, WZ4220

Subject:
Date:
June 2006


Page 85 of 85

Parent
Object
& Range
Service
Combination
gainFactorBc Integer - 015
gainFactorBd Integer - 015
referenceTFC Integer
UEGTFC
gainFactorType Enumerated
(sIGNALLED,
cOMPUTED)
See default values in load
RNC-CLI
RW-EL
Table 44: Gain factors related parameters
In the case the UE supports it, the computed gain factors method can be used and the
gain factor set according to Table 45. Note that due to the high number of parameters to
be tuned and their impact on the system performances, it is highly not recommended to
modify the beta values. The default values provided in the loads should be kept and only
modified after testing in the Swindon Lab and verification in test environment.
Service profile c d Number of reference TFC
(as per SRD-UTRAN-RCC-1)
DCCH signalling alone 15 8 1 [i.e. (TF1)]
64 kbps PS data+DCCH signalling 8 15 9 [i.e. (TF4)(TF1)]
Voice+DCCH signalling 11 15 5 [i.e. (TF2)(TF1)(TF1)(TF1)]
Voice+64kbps PS data+DCCH signalling 5 15 29 [i.e. (TF2)(TF1)(TF1)(TF4)(TF1)]
64 kbps CS Data+DCCH signalling 8 15 3 [i.e. (TF1)(TF1)]
Table 45: Default gain factors
Table 46: Gain factor tunings

For reference TFC:
UEGTFC:gainFactorType set to signalled
UEGTFC:isReferenceTFC set to true
UEGTFC:ReferenceTFC and TFCNumber set to 0
UEGTFC:Bc and Bd set as per table Table 45 (depends on service combination)

For other TFCs:
UEGTFC:gainFactorType set to computed
UEGTFC:isReferenceTFC set to false
UEGTFC:ReferenceTFC and TFCNumber set to 0
UEGTFC:Bc and Bd set to any value

Subject:
Date:
June 2006


Page 86 of 86

6.4.6. Impact of Compressed Mode on uplink transmit power control

Uplink power control remains essentially the same as in the normal case, using the same
UTRAN supplied parameters for Power Control Algorithm and step size (
TPC
), but with
additional mechanism designed to help the recovery of the interrupted control loop after
each transmission gap.

Due to the transmission gaps in compressed frames for both uplink DPDCH(s) and
DPCCH, there may be missing TPC commands in the downlink. If no downlink TPC
command is transmitted, the corresponding TPC_cmd derived by the UE is set to zero.

6.4.6.1.Impact at the UE
6.4.6.1.1. Impact of the change of number of pilot bits
Compressed and non-compressed frames in the uplink DPCCH may have a different
number of pilot bits per slot due to change in slot format. A change in the transmit power
of the uplink DPCCH is needed in order to compensate for the change in the total pilot
energy. Therefore at the start of each slot the UE derives the value of a power offset
PILOT. If the number of pilot bits per slot in the uplink DPCCH is different from its
value in the most recently transmitted slot, PILOT [dB] is given in (31)

|
.
|
\
|
=
curr PILOT
prev PILOT
PILOT
N
N
Log
,
,
10 * 10
(31)

where
N
PILOT,prev
is the number of pilot bits in the most recently transmitted slot
N
PILOT,,curr
is the number of pilot bits in the current slot.

Otherwise, including during transmission gaps in the downlink, PILOT is set to zero.

6.4.6.1.2. Impact of compressed mode on the uplink power control step size
Unless otherwise specified, in every slot during compressed mode the UE is expected to
adjust the transmit power of the uplink DPCCH with a step of DPCCH [dB] which is
given in (32).

PILOT cmd TPC TPC DPCCH + = _ *
(32)

Subject:
Date:
June 2006


Page 87 of 87

6.4.6.1.3. Initial transmit power after an Uplink transmit gap
At the start of the first slot after an uplink or downlink transmission gap the UE applies a
change in the transmit power of the uplink DPCCH by an amount DPCCH [dB], with
respect to the uplink DPCCH power in the most recently transmitted uplink slot according
to (33).

PILOT RESUME DPCCH + = (33)

The value of RESUME [dB] is determined by the UE according to the Initial Transmit
Power mode. The ITP is a UE specific parameter, which is signaled by the network with
the other compressed mode parameters. The different modes are summarized in Table 47.

Initial Transmit Power mode Description
0
gap TPC
cmd TPC RESUME _ * =
1
last
RESUME =
Table 47 - Initial Transmit Power modes during compressed mode

In the case of a transmission gap in the uplink, TPC_cmd
gap
is the value of TPC_cmd
derived in the first slot of the uplink transmission gap, if a downlink TPC_command is
transmitted in that slot. Otherwise TPC_cmd
gap
equals zero.

last
is the most recently computed value of
i
, which is updated according to the following
recursive relations, which is executed in all slots in which both the uplink DPCCH and a
downlink TPC command are transmitted, and in the first slot of an uplink transmission
gap if a downlink TPC command is transmitted in that slot.

sc TPC i i i
k cmd TPC * * _ * 96875 . 0 * 9375 . 0
1
=
(34)

Where:
TPC_cmd
i
is the power control command derived by the UE in that slot.
ksc = 0 if additional scaling is applied in the current slot and the previous slot and
ksc =1otherwise.

i-1
is the value of
i
computed for the previous slot. The value of
i-1
is initialized
to 0
o when the uplink DPCCH is activated,
o at the end of the first slot after each uplink transmission gap,
o at the end of the first slot after each downlink transmission gap.
The value of
i
is set to zero at the end of the first slot after each uplink
transmission gap.

Subject:
Date:
June 2006


Page 88 of 88

6.4.6.1.4. Recovery period on the UE side
After a transmission gap in either the uplink or the downlink, the period following
resumption of simultaneous uplink and downlink DPCCH transmission is called a
recovery period.

RPL is the recovery period length and is expressed as a number of slots.
RPL is equal to the minimum value out of the transmission gap length and 7 slots.

If a transmission gap is scheduled to start before RPL slots have elapsed, then the
recovery period ends at the start of the gap, and the value of RPL is reduced accordingly.

During the recovery period, 2 modes are possible for the power control algorithm. The
Recovery Period Power control mode (RPP) is signaled with the other compressed mode
parameters.

The different modes are summarized in Table 48.

Recovery Period power
control mode
Description
0
Transmit power control is applied using the algorithm
determined by the value of PCA, with step size TPC
1 Transmit power control is applied using algorithm 1 with
step size
RP-TPC
during RPL slots after each
transmission gap
Table 48 Recovery Period Power Control modes during compressed mode

For RPP mode 0, the step size is not changed during the recovery period and ordinary
transmit power control is applied, using the algorithm for processing TPC commands
determined by the value of PCA. Default PCA is PCA-1. Mode 0 is the default RPP
mode.

For RPP mode 1, during RPL slots after each transmission gap, power control algorithm 1
is applied with a step size
RP-TPC
. The change in uplink DPCCH transmit power (except
for the first slot after the transmission gap) is given in (35).

PILOT cmd TPC TPC RP DPCCH + = _ *
(35)

RP-TPC
[dB] is called the recovery power control step size.
If PCA has the value 1, RP-TPC is equal to the minimum value of 3dB and 2
TPC.
If PCA has the value 2,RP-TPC is equal to 1dB.

Subject:
Date:
June 2006


Page 89 of 89

After the recovery period, ordinary transmit power control resumes using the algorithm
specified by the value of PCA and with step size TPC.

If PCA has the value 2, the sets of slots over which the TPC commands are processed
shall remain aligned to the frame boundaries in the compressed frame.

Hence, for both RPP mode 0 and RPP mode 1, if the transmission gap or the recovery
period results in any incomplete sets of TPC commands, TPC_cmd shall be zero for those
sets of slots, which are incomplete.

6.4.6.2.Impact on the uplink SIR target
When compressed mode is active the target SIR becomes a function of the presence, in
the current radio frame, of uplink compressed mode gap and of the number of pilots bits.

When compressed mode is active, the Node B shall calculate the uplink SIR target
SIR
cm_target
as defined in (36).

coding SIR coding SIR SIR SIR SIR
PILOT et t et t cm
_ 2 _ 1
arg arg _
+ + + =
(36)

Where
SIRtarget is the uplink SIR target value sent by the RNC
SIR1_coding and SIR2_coding shall be computed using the uplink parameters
DeltaSIR1, DeltaSIR2, DeltaSIRafter1, DeltaSIRafter2 sent by the SRNC through
NBAP/RANSAP signaling, according to (37) and (38).

SIR1_coding = DeltaSIR1 if the start of the first transmission
gap in the transmission gap pattern is within the current uplink
frame.
SIR1_coding = DeltaSIRafter1 if the current frame just
follows a frame containing the start of the first transmission gap
in the transmission gap pattern.
SIR1_coding = 0 dB in all other cases
(37)

SIR2_coding = DeltaSIR2 if the start of the second
transmission gap in the transmission gap pattern is within the
current uplink frame.
SIR2_coding = DeltaSIRafter2 if the current frame just
follows a frame containing the start of the second transmission
gap in the transmission gap pattern.
SIR2_coding = 0 dB in all other cases.
(38)

|
.
|
\
|
=
frame curr PILOT
N PILOT
PILOT
N
N
Log SIR
_ ,
,
10 * 10
(39)

Subject:
Date:
June 2006


Page 90 of 90

SIR
PILOT
shall be calculated as defined in (39), where:
Npilot,N is the number of pilot bits per slot in a normal uplink frame without a
transmission gap.
Npilot,curr_frame is the number of pilot bits per slot in the current uplink frame.

In case several compressed mode patterns are used simultaneously, SIR1_coding and
SIR2_coding offsets shall be computed for each compressed mode pattern and all
SIR1_coding and SIR2_coding offsets shall be summed together.

The corresponding translation settings are described in Table 49.

& Range
UEGTFCS
Instance
Default
Value
Access
deltaSIR1
Integer [dB]
0, 0.01, 3
0
deltaSIRAfter1
Integer [dB]
0, 0.01, 3
0
deltaSIR2
Integer [dB]
0, 0.01, 3
0
Transmission
GapPattern
deltaSIRAfter2
Integer [dB]
0, 0.01, 3
All
0
RNC-CLI
RW

7. Power control for Uplink Common Channels
The only uplink common channel supported in U03.03 is the Random Access Channel.
The power control technique used on this channel is a mix of open loop and a crude
version of inner loop power control based on preamble power ramp up technique. Since
this topic belongs to procedures in the RACH channel it is described in details in the TAN
Access Procedures [15].

Subject:
Date:
June 2006


Page 91 of 91

8. Power Overload Control
As sector/carrier power utilization is approaching the upper limit of sector/carrier power,
the demanded power on the downlink may exceed the rated power of the amplifier. In
order to maintain the signal quality of the existing calls and to protect the amplifier from
overheating and shutting down, the UTRAN act on existing calls, new calls and/or
incoming handover. The overload control can be basically separated into following
categories:

Power Overshoot Control (downlink only):
This is a fast process with a reaction time of the order of a radio frame. It is based
on the estimation of the overall transmit power on a given sector/carrier. If the
measured value is higher than the rated power of the amplifier, various quick
actions can be taken in order to reduce the aggregate transmit power of the sector.
These actions consist of procedures as gain clipping, Aggregate Overload Control
(AOC) and Modified Aggregate Overload Control (MAOC).
Amplifier protection (downlink only):
These strategies are based on the measurement of the temperature of the power
amplifier. If a given critical temperature is reached, the amplifier enters a cool-
down state involving a given set of transmit power limiting actions, such as power
overshoot and call admission control. The sector transmitter resumes to its normal
behavior when it switches back to the normal state.
Load control algorithms (uplink and downlink):
This is a slow process consisting in three main different algorithms, i.e. Call
Admission Control, Dynamic Bearer Control and Congestion Control that
respectively deny access to new incoming calls, deny RAB establishment or take
actions on the already connected users. On the downlink the load evaluation is
based on the downlink aggregate transmitted power.

8.1. UPLINK OVERLOAD CONTROL
Different procedures are implemented to control the overload in uplink. For more
information refer to the Translation Application Note dealing with the Load Control
Algorithms [13] as well as to the Access Procedures [15].

Customer UMTS RF TAN U03.03 Power Control

Enviado por

Dados do documento

Descrição original:

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Customer UMTS RF TAN U03.03 Power Control

Enviado por

Direitos autorais:

Formatos disponíveis

Lucent Technologies Proprietary

This document contains proprietary information of

Você também pode gostar