UMTS Optimization

Download as xls, pdf, or txt
Download as xls, pdf, or txt
You are on page 1of 79

CS & PS

Access Failure

ES
SI
BI
LI
TY

OFR KPI

RRC Failure + RAB Failure


RRC Failure
RAB Failure
kpi

MO
BI

LIT

AC
C

CSSR PS
HSDPA R99 RECONFIG SR
RB CS SR
RRC CS SR
RRC SR
RRC PS SR
RRC SIGN SR
DCR AMR E2E
RETAINABILITY
DCR PS E2E
ASU CS SR
ASU PS SR
CCO PS SR
HHO IF OUTGOING CS SR
HHO IF PS SR
HO ITRAF HSD SR
IRAT HO AMR MULTIRAB SR
IRAT HO AMR RATIO
ASU SR
AMR ERLANG
TRAFFIC PS HSD TRAFFIC RATIO
PS HSU TRAFFIC RATIO

AVG
99.88
99.53
99.78
99.86
99.87
99.88
99.86
0.35
2.44
99.96
99.89
94.75
99.63
99.28
97.77
98.80
10.57
14815.56
97.39
65.13

Accessibility Check List


S. No
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

Checklist in the order of Priority


RBS & RNC Hardware\Software Alarms
VSWR Issues
RTWP Issues
Missing Neighbors
Faulty Antennas
Outages
Taking unusual traffic due to issues\outages on neighboring cells
Over-propagating
MP Load issues
Admission Control Blocking ( Due to Adm )
Failures after Admission (pmNoFailedAfterAdm)
License Issues
Node Blocking
TN Blocking
Congestion on Control Plane (Iub, Iu, Iur)
Congestion on Iub User Plane
Failures due to RNC Issues
Core Transport Network Congestion
Failures due to UE Capability Issues
Failures due to Radio Reasons
Other Parameters to improve Accessibility
Lack Of HW_CE CONGESTION ANALYSIS ( UL & DL )
High UL Noises

Retainability Analysis

Retainability Analysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

MOBILITY

Missing Neighbros

DropMissingNcell

Prop Delay & WNCS


One way Neighbros
IntraFreq Ngbr count=31?
Alarms
Alarms Ngbr Cell
CPICH Power

DropULSyncSpch

CPICH Delta
UL RSSI
2G neighbors if IRAT enabled?
Terrain profile
electricalUlDelay
Alarms
Alarms Ngbr Cell
UL RSSI
CPICH Power

DropSoHoSPch

CPICH Delta
Prop Delay
IUB congestion
T1 down time
T1 error
See if Accesibility issue
Alarms
Alarms Ngbr Cell
SC Collision
CPICH Power

DropOtherSpch

CPICH Delta
UL RSSI
IUB congestion
T1 down time
T1 error

re + RAB Failure
RRC Access Failure + RRC Setup Failure
RAB Access Failure + RAB Setup Failure
kpi
CSSR_PS_M
HSDPA_R99 reconfig_SR_M
RB_CS_SR_M
RRC_SSR_CS_M
RRC_SSR_M
RRC_SSR_PS_M
RRC_SSR_Signalling_M
DCR_AMR_E2E_M
DCR_PS_E2E_M
ASU CS_SR_M
ASU PS_SR_M
CCO_PS_SR_M
HHOinterfreq_OUT_CS_SR_M
HHOinterfreq_PS_SR_M
Intrafreq HO HSD SR_M
IRATHO_AMR_MultiRab_SR_M
IRATHO_AMR_Ratio_M
ASU_SR_M
AMR_Erlang_M
Ratio PS_HSD_Mbyte vs ALL_Mbyte
Ratio PS_HSU_Mbyte vs ALL_Mbyte

Source of Information
OSS, NodeB, RNC (MOSHELL)
MOSHELL
MOSHELL, PM Counters
WNCS
FFAX
PM Counters
PM Counters
PM Counters
PM Counters
PM Counters
PM Counters
Alarms, PM Counters
PM Counters
PM Counters
PM Counters
PM Counters
PM Counters
PM Counters
PM Counters, GPEH, UETR Traces
WMRR
PM Counters
PM Counters
PM Counters

=100*((([ReqPsSucc])/([ReqPs]KPI Formula
[pmNoLoadSharingRrcConnPs]))*(([pmNoRabEstabl
=100*(([pmNoRabEstablishSuccess].[Speech]+
ishSuccess].[PacketInteractive]/
=
100*[pmHsToDchSuccess]/[pmHsToDchAttempt]
[pmNoRabEstablishSuccess].[Cs64])/
[pmNoRabEstablishAttempt].[PacketInteractive])))
([pmNoRabEstablishAttempt].[Speech]+
=100*([ReqCsSucc]/([ReqCs][pmNoRabEstablishAttempt].[Cs64]+
[pmNoLoadSharingRrcConnCs]))
[pmNoDirRetryAtt]))
#N/A
=100*(([ReqSuccess]-[ReqCsSucc]-[ReqPsSucc]=100*([ReqPsSucc]/([ReqPs]=100*([Query
1 with TP Ericsson RNC PM].
[SuccessIratCellResel])/([Req]-[ReqCs]-[ReqPs][pmNoLoadSharingRrcConnPs]))
[pmNoSystemRabReleasePacket]-[Query
1 with TP
[AttIratCellResel]-[pmNoLoadSharingRrcConn]+
=100*([pmNoSystemRabReleaseSpeech])/
Ericsson
RNC PM].
[pmNoLoadSharingRrcConnCs]+
([pmNoNormalRabReleaseSpeech]+
[pmNoSystemRabReleasePacketUra])/
[pmNoLoadSharingRrcConnPs]))
[pmNoSystemRabReleaseSpeech])
([pmNoNormalRabReleasePacket]=100*([pmRlAddSuccessBestCellSpeech]/
=100*(([pmRlAddSuccessBestCellPacketLow]+
[pmNoNormalRabReleasePacketUra]+[Query
1
[pmRlAddAttemptsBestCellSpeech])
[pmRlAddSuccessBestCellPacketHigh])/
with
TP Ericsson RNC PM].
([pmRlAddAttemptsBestCellPacketLow]+
=100*[pmNoOutIratCcSuccess]/
[pmNoSystemRabReleasePa
=100*([InterFreqHoCsSpeech12]+
[pmRlAddAttemptsBestCellPacketHigh]))
=100*([InterFreqHoPsInteractiveLess64]+
[pmNoOutIratCcAtt]
[InterFreqHoCsConversational])/([CsSpeech12]+
[InterFreqHoPsInteractiveGreater64]+
[CsConversational])
[pmSuccNonBlindIfhoPsIntEul])/
([PsInteractiveLess64]+[PsInteractiveGreater64]+
=100*[pmNoHsCcSuccess]/[pmNoHsCcAttempt]
=100*([pmNoSuccessOutIratHoSpeech])/
[pmAttNonBlindIfhoPsIntEul])
=100*([pmRlAddSuccessBestCellSpeech]+
=100*([pmNoSuccessOutIratHoSpeech])/
([pmNoAttOutIratHoSpeech]+[SpeechGsmFailure])
=([pmSumBestCs12Establish]+
[pmRlAddSuccessBestCellPacketLow]+
([pmNoNormalRabReleaseSpeech]+
[pmSumBestAmr4750RabEstablish]+
[pmRlAddSuccessBestCellPacketHigh])/
[pmNoSystemRabReleaseSpeech])
[pmSumBestAmr5900RabEstablish]+
([pmRlAddAttemptsBestCellSpeech]+
[pmSumBestAmr7950RabEstablish]+
[pmRlAddAttemptsBestCellPacketLow]+
[pmSumBestAmr12200RabEstablish]+
=
100*[pmDlTrafficVolumePsIntHs]/
[pmRlAddAttemptsBestCellPacketHigh])
[pmSumBestAmrNbMmRabEstablish]+
= 100*[pmUlTrafficVolumePsIntEul]/
[pmSumBestAmrWbRabEstablish])/720

Link For Details


HW Issue

MP Load
Due to Adm
Failures after Admission
License Issues
Node Blocking
TN Blocking
Congestion on Control Plane
Congestion on Iub User Plane
RNC Issues
Core Network Cong
UE Issue
Radio Reasons
Other Parameter
Lack of UL HW_ CE UL
High RSSI_UL Noise

ghbros

& WNCS

eighbros

gbr count=31?

r Cell

er

rs if IRAT enabled?

le

Delay

r Cell

er

tion

me

sibility issue

r Cell

er

tion

me

Missing NBR

For Details

W Issue

Lack of UL HW_ CE DL

WNCS

Back to Home_1
Admission Control Blocking ( Due
to Adm )
Counter Pegged

pmNoRrcReqDeniedAdmDlPwr

pmNoRrcReqDeniedAdmDlChnlCode

pmNoRrcReqDeniedAdmDlHw

pmNoRrcReqDeniedAdmUlHw

pmNoRejRrcConnMpLoadC
pmNoFailedRrcConnectReqHw
pmNoFailedRrcConnectReqCsHw
pmNoFailedRrcConnectReqPsHw

PmNoFailedRabEstAttemptLackDlPwr

pmNoFailedRabEstAttemptLackDlChnlCo

pmNoFailedRabEstAttemptExceedConnLimit
pmNoFailedRabEstAttemptLackDlAse

pmNoFailedRabEstAttemptLackUlAse

pmNoFailedRabEstAttemptLackDlHw

pmNoFailedRabEstAttemptLackDlHwBest

pmNoFailedRabEstAttemptLackUlHw

pmNoFailedRabEstAttemptLackUlHwBest

pmNoFailedRabEstAttemptExceedConnLi

Failures after Admission (pmNoFailedAfterAdm)


Back to Home_2

pmNoFailedAfterAdm

Iub Congestion
Back to Home_3
Background

Back to Home_3
> Iub congestion is a common reason for high number
of failures after admission events
> Depending on the volume of traffic per service or
RAB, certain QoS could be congested at the AAL2.
> All RABs excluding HSDPA are configured to use Class
A or Class B being these two mostly impacted by
congestion
> Most of the time Class B is the first QoS to get
congested leading to failures after admission events
> Iub congestion could be caused by T1 issues
> Missconfigurations of AAL2 profile at the Node B, RNC
or RXI/MSN can lead to Iub congestion

Core Transport Network Congestion


Back to Home_4
Background
> Accessibility issues are observed on all sites at the
RNC without major issues with Iub congestion
(especially at peak hour)
> Congestion will be observed at
Iu-CS (RNC<-->MGW) or Iu-PS (RNC<-->SGSN) links

KPIs to monitor TN blocking on the Iub link, when I


Back to Home_5

Outgoing AAL2 Setup Failure Rate for QoS class X = 100 * (pmUnSuccOutCo
pmSuccOutConnsRemoteQoSClassX)

Incoming AAL2 Setup Failure Rate for QoS class X = 100 * (pmUnSuccInConn
pmSuccInConnsRemoteQoSClassX)

Total AAL2 Setup Failure Rate for QoS class X = 100 * (pmUnSuccOutConnsLo
(pmUnSuccOutConnsLocalQoSClassX + pmUnSuccOutConnsRemoteQoSClassX + p

Failures due to Node Blocking and Transport Netwo


Back to Home_6
Counter Pegged
Counters 85-93 in Counters sheet

Counters 69-84 in Counters sheet

Counters 69-84 in Counters sheet

Congestion on Control Plane (Iub, Iu, Iur)


Back to Home_7

The purpose of this section is to help predict when network expansion is needed by
All counters should be active continuously to monitor congestion trends. If the num
reconfiguration of the signaling link or expansion of the network should be conside
For Iub over ATM specifically, reconfiguration of the signaling link, usually configure
detected.
For Iub in general, expansion of the network could be done by increasing the numb
Monitor counters from 98-101 in counters sheet.

By monitoring the number of discarded AAL2 CPS packets per AAL2 path due to qu
QoS separation on ATM layer is used, then an AAL2 path can be configured to carry
If QoS separation on AAL2 layer is used, then each AAL2 path can be configured to
to carry Class A, B and C traffic. This would result in Class A and B traffic being cons
When QoS separation on AAL2 layer is used it is not possible to monitor the conges
the impact of AAL2 queue congestion on the different services.
Not all ET-boards support AAL2 QoS separation.
Configuration of an AAL2 Path to traffic of certain QoS class(es) is done using MO cl
Monitor counters 102-103 from the counters sheet.
Discarded CPS Packet Rate = 100 * pmDiscardedEgressCpsPackets / (pmEgress

Congestion on Iub User Plane


Back to Home_8

Monitor counters from 104 - 118 from counters sheet


HSDPA Frame Loss Rate on the path from SRNC to RBS = 100 * pmHsDataFr
EUL Frame Loss Rate on the path from RBS to SRNC = 100 * pmEdchDataFra
Frame Loss Rate over Iub for DCH Traffic = 100 * pmDchFramesTooLate / (pmD

Failures due to Radio Reasons


Back to Home_9
These failures are mostly RF related and can be split between:
Bad Coverage (Low RSCP)
Interference (Good RSCP & Low Ec/No)
Route cause analysis will in most cases require WMRR/drive test analysis.
Also verify that the uplink RTWP value for these sites is within acceptable range to

Failures due to UE Issues


Back to Home_10

The cause for this counter can be coverage even though L1 synchronization has be
coverage related. If this is an RNC wide problem it needs to be determine if it is a s
Also verify that the uplink RTWP value for these sites is within acceptable range to

Failures due to RNC Issues


Back to Home_11

If the counter is showing a step change it might be related to a new software load t
particular sites are showing a high count for this counter investigate if there is any
availability and BER for T1)

Failures due to Hardware Issues


Back to Home_12
Issue
RSSI (UL Interference)
VSWR
Errors/Crashes
T1 Errors
Alarms
Cell Unavailability

Other Parameters to improve Accessibility


Back to Home_13
Parameter

minPwrRl (dB)

minPwrMax (dB)

maxFach1Power (dB)

spreadingFactor

pO3

License Issues
Back to Home_14
Parameter

pmNoFailedRrcConnectReqCsHw

(CS RRC denied - Insufficient


Licensed Capacity)

Lack of UL HW_ CE UL
Back to Home_15
Parameter

Back to Home_15

RRC : pmNoFailedRabEstAttemptLackUlHwBest RAB :


pmNoFailedRabEstAttemptLackUlHW

Lack of DL HW_CE DL
Back to Home_16
Parameter

RRC : pmNoFailedRabEstAttemptLackDlHwBest
RAB : pmNoFailedRabEstAttemptLackDlHW

HIGH PROCESSOR LOAD ( Mp LOAD )


Back to Home_17
pmSumMeasuredLoad / pmSamplesMeasuredLoad

pmprocessorload

High UL Noises
Back to Home_18
Parameter

Back to Home_18

RSSI : pmAverageRssi
[Low RSSI<-105,High RSSI>-95]

Back to Home_19

Missing Neighbor
Parameter

Missing Neighbor

Back to Home_19

DropULSyncSpch
Parameter

Uplink Sync Drop :Investigating the UE TxPwr by WMRR.

Back to Home_19

IRAT Drop :
Parameter

Parameter (usedFreqThresh2dRscp )value change (-112


to -109)

Troubleshooting and Optimization

> OCNS is deactivated on the cell


> Feeder length attenuation & TMA parameters are set correctly
> Check for neighboring cell outages/malfunctions
> Check if the site is over-propagating
> pwrAdm, pwrOffset parameters are consistent with baseline values
> maximumTransmissionPower alligned to maxDlPowerCapability
> ratio between primaryCpichPower and
maximumTransmissionPower
> Explore IFLS with other Carrier
> Balance traffic between cells using QOffset1sn and QOffset2sn
> Enable or Increase Directed Retry to GSM
> Reduce sf16Adm/sf8Adm
> Consider changing to 60W RU
> Add another Carrier
> Check for neighboring cell outages/malfunctions
> Check if the site is over-propagating
> Check dlCodeAdm values and see if adjustment possible
> Reduce 384k Users on DL (sf8Adm)
> Reduce HS-PDSCH Codes
> Reduce HS-SCCH Codes

> Check for neighboring cell outages/malfunctions


> Check if the site is over-propagating
> Check Locked TX Boards
> Check hanging CE (restart)
> Check HS resource allocation
> Check dlHwAdm value and see if adjustment possible
> Adjust channel down-switch from DCH to FACH (increase downswitch
attempts, decrease upswitch attempts, results in better usage of
resources)
> Replace TX Boards
> Increase CE licensing
> Add DUW
> Check for neighboring cell outages/malfunctions
> Check if the site is over-propagating
> Check Locked RAX Boards
> Check hanging CE (restart)
> Check EUL resource allocation
> Check ulHwAdm value and see if adjustment possible
> Adjust channel down-switch from DCH to FACH (increase downswitch
attempts, decrease upswitch attempts, results in better usage of
resources)
> Replace RAX Boards
> Increase CE licensing
> Add DUW
>
>
>
>

Check MP load for Main Racks / Extension Sub Racks High??


Check for Core Network (Call servers / SGSN ) outages
Re-module Sites
Re-home Sites

> Verify/correct individual licensed capacity levels for the node


> Check CE usage trends
> Verify/correct individual licensed capacity levels for the node
> Check CE usage trends
> Verify/correct individual licensed capacity levels for the node
> Check CE usage trends

> OCNS is deactivated on the cell


> Feeder length attenuation & TMA parameters are set correctly
> Check for neighboring cell outages/malfunctions
> Check if the site is over-propagating
> pwrAdm, pwrOffset parameters are consistent with baseline values
> maximumTransmissionPower alligned to maxDlPowerCapability
> ratio between primaryCpichPower and
maximumTransmissionPower
> Explore IFLS with other Carrier
> Balance traffic between cells using QOffset1sn and QOffset2sn
> Enable or Increase Directed Retry to GSM
> Reduce sf16Adm/sf8Adm
> Consider changing to 60W RU
> Add another Carrier

>
>
>
>
>
>

Check for neighboring cell outages/malfunctions


Check if the site is over-propagating
Check dlCodeAdm values and see if adjustment possible
Reduce 384k Users on DL (sf8Adm)
Reduce HS-PDSCH Codes
Reduce HS-SCCH Codes

dConnLimit
>
>
>
>

Verify settings for aseDlAdm


Check for traffic load on the cell
Check if the site is over-propagating
Optimize radio network to reduce interference / pilot pollution

>
>
>
>

Verify settings for aseUlAdm


Check for traffic load on the cell
Check if the site is over-propagating
Optimize radio network to reduce interference / pilot pollution

> Check for neighboring cell outages/malfunctions


> Check if the site is over-propagating
> Check Locked TX Boards
> Check hanging CE (restart)
> Check HS resource allocation
> Check dlHwAdm value and see if adjustment possible
> Adjust channel down-switch from DCH to FACH (increase downswitch attempts,
decrease upswitch attempts, results in better usage of resources)
> Replace TX Boards
> Increase CE licensing
> Add DUW

> Check Locked TX Boards


> Check hanging CE (restart)
> Check HS resource allocation
> Check dlHwAdm value and see if adjustment possible
> Adjust channel down-switch from DCH to FACH (increase downswitch attempts,
decrease upswitch attempts, results in better usage of resources)
> Replace TX Boards
> Increase CE licensing
> Add DUW
> Check for neighboring cell outages/malfunctions
> Check if the site is over-propagating
> Check Locked RAX Boards
> Check hanging CE (restart)
> Check EUL resource allocation
> Check ulHwAdm value and see if adjustment possible
> Adjust channel down-switch from DCH to FACH (increase downswitch attempts,
decrease upswitch attempts, results in better usage of resources)
> Replace RAX Boards
> Increase CE licensing
> Add DUW
> Verify Admission settings / Usage trends for different SF RAB
> Fine tune parameter settings to maintain balance between accessibility &
throughput

NoFailedAfterAdm)
The counter pmNoFailedAfterAdm is generally pegged for the reasons stated
below :
> NBAP RL setup failure (RAX or TXB congestion)
> AAL2 setup failure (due to congestion or miss configuration)
> Channel Code Allocation Failure (code not available)

Action:
Check
Check
Check
Check

and
and
and
and

remove TAX/RAX Board congestion


upgrade IUB link in case of congestion
clear Hardware issues
clear Iub Link outages.

Indicators

> Check for congestion at the Iub link


> Check for T1 issues and history of alarms of T1s (for intermittent T1 alarms)

estion
Indicators
> Signalling bearer rab (SRB) issues for either CS or PS would increment on
these cases.
> SRB failures for CS: 1-(pmNoIuSigEstablishSuccessCs /
pmNoIuSigEstablishAttemptCs).
> SRB failures for PS: 1-(pmNoIuSigEstablishSuccessPs /
pmNoIuSigEstablishAttemptPs)
> Congestion will be observed at Aal2 access point (Aal2Ap) entity starting with
g (note: Aal2Ap entities starting with b correspond to node b, starting with
r corresponds for Iur links,etc)

on the Iub link, when Iub is over ATM

class X = 100 * (pmUnSuccOutConnsLocalQoSClassX + pmUnSuccOutConnsRemoteQoSClassX) / (pmUnSuccO

class X = 100 * (pmUnSuccInConnsLocalQoSClassX + pmUnSuccInConnsRemoteQoSClassX) / (pmUnSuccInCon

s X = 100 * (pmUnSuccOutConnsLocalQoSClassX + pmUnSuccOutConnsRemoteQoSClassX + pmUnSuccInConn


uccOutConnsRemoteQoSClassX + pmUnSuccInConnsLocalQoSClassX + pmUnSuccInConnsRemoteQoSClassX + p

g and Transport Network Blocking


Troubleshooting and Optimization
> Check for node configuration error, node limitations or transport network layer
service unavailability (T1 related issues which can be checked from T1 error
counters) to identify the main cause for these failures
> Check for congestion on user plane (AAL2)/ control plane (UniSaal /SCTP) for
IuB resources.
> Expand TN bandwidth
> Check T1 Errors
> Check Locked T1s
> Check Iub/TD (Vc/Vp) Mismatches
> Share Traffic with Nbr or Escalate for expansion

> Check for congestion on user plane (AAL2)/ control plane (UniSaal /SCTP) for
IuB resources.
> Expand TN bandwidth
> Check T1 Errors
> Check Locked T1s
> Check Iub/TD (Vc/Vp) Mismatches
> Share Traffic with Nbr or Escalate for expansion
> Reduce CPICH or Tilt to take less traffic
> Increase Directed Retry to Offload to GSM
> Check IuCs/IuPS

(Iub, Iu, Iur)

en network expansion is needed by monitoring congestion on the control plane of the Iub (NBAP and Q.2630), Iu
onitor congestion trends. If the number of congestion occurrences and discarded messages tends to be consiste
n of the network should be considered.
the signaling link, usually configured as UBR+, can be done by increasing the MCR for the VC carrying NBAP-C,

uld be done by increasing the number of links to an RBS, changing the physical links to increase capacity or add

Congestion on Iub per AAL2 Path


S packets per AAL2 path due to queue overflow, congestion per AAL2 Path can be ascertained. If the discard ra
AL2 path can be configured to carry traffic of one QoS class only. A high CPS packet discard rate for this AAL2 pa
ch AAL2 path can be configured to carry traffic of more than one class but the different classes can be prioritize
lt in Class A and B traffic being considered as higher priority and Class C being considered as lower priority.
not possible to monitor the congestion impact per class. Consequently, the following counters cannot be used
fferent services.

n QoS class(es) is done using MO class Aal2PathVccTp with attribute SupportedQoSClasses.


eet.
rdedEgressCpsPackets / (pmEgressCpsPackets + pmDiscardedEgressCpsPackets)

sheet
SRNC to RBS = 100 * pmHsDataFramesLostSpiXX / (pmHsDataFramesLostSpiXX + pmHsDataFramesReceivedS
S to SRNC = 100 * pmEdchDataFramesLost / (pmEdchDataFramesLost + pmEdchDataFramesReceived)
100 * pmDchFramesTooLate / (pmDchFramesReceived * 1000)

split between:

WMRR/drive test analysis.


sites is within acceptable range to isolate the problem due to uplink issues.

n though L1 synchronization has been achieved it can still be that a sudden loss in coverage was experienced. I
it needs to be determine if it is a specific Ue type that is causing the problem, this will require advance analysis
sites is within acceptable range to isolate the problem due to uplink issues.

be related to a new software load that was loaded on the RNC or core network side. To trouble shoot these prob
counter investigate if there is any common point of failure for these cells ( Same RNC, same Transmission path

Troubleshooting and Optimization


> Check Attenuation/Losses
> TMA Check
> Verify RFDS
> External Interference
> Restart Board
> Cold Restart Node
> Verify SW on node
Escalate
> Swap Cards/Boards
> Replace Cards/Boards

Accessibility
Description
If the power of a radio link is very low, it is very sensitive to various interference.
To avoid the Power Control function (more information can be found in Power
Control) decreasing the power too much due to temporary good radio conditions,
a minimum DL transmitted code power value is configured. Parameter minPwrRl
(dB) sets this value per cell and it is relative to primary CPICH power in a cell.

minPwrMax is a utrancell parameter that defines the maximum power per Radio
Link relative to CPICH power where Radio Link bit rates are equal to or below
minimumRate.
Connections that fall into this range are:
Signaling Radio Bearers (3.7 Kbps)
Standalone Signaling Radio Bearers (14.8 Kbps)
Conversational Circuit Switch Speech AMR 12.2 Kbps (15.9 Kbps)
Maximum DL transmitted code power must not be set to a lower value than the
minPwrRl.

defines the maximum power used for transmitting the first FACH channel, relative
to the primaryCpichPower value.
The first FACH is used for logical channels BCCH, CCCH, and DCCH control
signaling.
Minimum spreading factor to use for PRACH.
Value mapping:
32: TTI = 10 ms.
64: TTI = 20 ms.
Power offset for the pilot bits.
Value mapping:
0: 0.00 dB
1: 0.25 dB
...
24: 6.00 dB
Unit: 0.25 dB

Description

# CS calls denied by admission control due to insufficient licensed


capacity in the RBS.

Troubleshooting and Optimization

ADMISSION REJECTION : CE Usage > Available CEs in DUW


Action:
Check availableRbsChannelElementsUplink &
licenseCapacityRbsChannelElementsUplink
Rejection in only one DUW (out of available two DUWs) - Tune the
ulLicFractBbPool2
Recommend for CE addtion.
Reduce the values for sf4AdmUl, sf8AdmUl, sf16AdmUl,
Switch off the 2ms TTI for EUL eulServingCellUsersAdmTti2
Send more traffic to GSM if GSM is not congested Change
Directed Retry to GSM loadSharingGsmThreshold
Increase UlHwAdm

Troubleshooting and Optimization


ADMISSION REJECTION : CE Usage > Available CEs in DUW
Action:
Check availableRbsChannelElementsDownlink &
licenseCapacityRbsChannelElementsDownlink
Rejection in only one DUW (out of available two DUWs) - Tune the
dlLicFractBbPool2
Recommend for CE addtion.
Reduce the values for sf4Adm, sf8Adm, sf16Adm,
Send more traffic to GSM if GSM is not congested Change
Directed Retry to GSM loadSharingGsmThreshold
Increase DlHwAdm

LOAD )
Action
Terminate MRR Recording
Turn off unnecessary traces
change reportingRange1a from 6 to 4,
reportingRange1b from 10 to 8, and
timeToTrigger1c from 11 to 12
Incase of IUB congestion, Check and upgrade IUB

Troubleshooting and Optimization

Defective TMA and antenna


Faulty Hardware
Increase in traffic mainly EUL traffic

Troubleshooting and Optimization


Check neighbor cell availability.
Add suggested Neighbors cell with WNCS/GPEH/Drive Test data.
Analyze existing neighbors with 3G-3G cell relation report.
Analyze cell coverage by RF Planning tool /Drive Test data Plots.
Check Propagation Delay to detect & down tilt the overshooting cells.

Troubleshooting and Optimization


Uplink Sync Drop :Investigating the UE TxPwr by WMRR.
Analyze cell coverage by Propagation delay counter.
Check UL-RSSI issues.
Parameter tuning like rlFailureT & nOutSyncInd as mentioned in the Slide No: 15.

Troubleshooting and Optimization


Neighbor list optimization.
GSM coverage area verification for IRAT neighbors.
Consistency check for external GSM cell (BSIC, BCCH, LAC, CI)
Selection Priority the neighbor list based on distance and attempt.
Check parameter setting : hoType, qRxLevMin, usedFreqThresh2dEcno,
usedFreqThresh2dRscp

Parameters that can be tuned

> Increase pwrAdm (75, 80, 85, 90 & 95) and reduce pwrOffset (15,10,5 &
considering the Total Non HS Power Utilization and SHO factor
> Fine tune QOffset1sn and QOffset2sn to balance traffic
> Reduce sf16Adm (16) /sf8 (8) Adm
> Allign maximumTransmissionPower with maxDlPowerCapability
> loadSharingGsmFraction (100)---->specifies the percentage of Directed R
candidates to be diverted to GSM while the cell load is above the specified load
threshold
> loadSharingGsmThreshold (75)--->specifies the minimum load at which o
loading to GSM begins: ex 80% of pwrAdm
>
>
>
>
>
>
>
>
>
>
>

dlCodeAdm----> (80, 90 & 95)


sf8Adm----sf16Adm
sf32Adm
sf16gAdm
sf4AdmUl
sf8AdmUl
sf16AdmUl
sf8gAdmUl
hsdpaUsersAdm
eulServingCellUsersAdm

> dlHwAdm ----> (95)


> dlLicFractBbPool2

> ulHwAdm -----> (95)


> ulLicFractBbPool2

>
>
>
>
>
>

Increase pwrAdm and reduce pwrOffset considering the Total Non HS Power Utilization and SHO
Fine tune QOffset1sn and QOffset2n to balance traffic
Reduce sf16Adm/sf8Adm
Allign maximumTransmissionPower with maxDlPowerCapability
loadSharingGsmFraction
loadSharingGsmThreshold

>
>
>
>
>
>
>
>
>
>
>

dlCodeAdm
sf8Adm
sf16Adm
sf32Adm
sf16gAdm
sf4AdmUl
sf8AdmUl
sf16AdmUl
sf8gAdmUl
hsdpaUsersAdm
eulServingCellUsersAdm

> aseDlAdm

> aseUlAdm

> dlHwAdm
> dlLicFractBbPool2

> ulHwAdm
> ulLicFractBbPool2

Description : Number of Radio Resource Control (RRC) or Radio Access Bearer (RAB) establishment req
failed after being admitted by admission control.
REASONS:Failures due to RAXB congestion
Failures due to TXB congestion
DL Channel code alloc Failure
and RAB establishment failures which occur after admission control can be due to
Iub transport Network blocking
Node blocking
NBAP RL setup failure (RAX or TXB congestion)
Channel Code Allocation Failure (code not available)
Core transport network congestion or failures
timeout in the UE, RNC or RBS
invalid parameter settings

Solutions

> Enable Directed retry (short term solution)


> Correct possible AAL2 miss configuration at Node B, RNC and RXI or MSN (AAL2 profile must match
three entities)
> Order new T1s (long term solution)
> Change AAL2 QoS configuration depending on services request volume (CS voice, R99 data, HSDPA

Solutions
> This issue will require support from the client in order to determine high utilisation over Iu-CS and I
links

oteQoSClassX) / (pmUnSuccOutConnsLocalQoSClassX + pmUnSuccOutConnsRemoteQoSClassX +

oSClassX) / (pmUnSuccInConnsLocalQoSClassX + pmUnSuccInConnsRemoteQoSClassX +

SClassX + pmUnSuccInConnsLocalQoSClassX + pmUnSuccInConnsRemoteQoSClassX) /


nConnsRemoteQoSClassX + pmSuccOutConnsRemoteQoSClassX + pmSuccInConnsRemoteQoSClassX)

Parameters that can be tuned

the Iub (NBAP and Q.2630), Iur (RNSAP and Q.2630) or Iu (RANAP and Q.2630) link.
essages tends to be consistently high for the control plane, not just high during short peak periods, then either
for the VC carrying NBAP-C, NBAP-D and/or Q.2630 traffic, depending on which UniSaalTp the congestions are

s to increase capacity or adding a new RBS.

r AAL2 Path
ascertained. If the discard rate is consistently high it indicates an under dimensioning problem with the AAL2 pa
discard rate for this AAL2 path indicates congestion for that traffic class.
erent classes can be prioritized into two levels. For example, one configurable option is to have an AAL2 Path con
sidered as lower priority.
ng counters cannot be used for tuning of different traffic types and additional counters should be monitored to o

SClasses.

pmHsDataFramesReceivedSpiXX)
ataFramesReceived)

coverage was experienced. If only one site is showing a high count for this counter the most likely cause is that
will require advance analysis with the help of GPEH, UETR & CTR

e. To trouble shoot these problem it will require in-depth analysis on the RNC which should be done by Tier 2. If
RNC, same Transmission path/ hardware) Also investigate if the T1 for these sites are error free (Look at the cell

Recommendation

Increase minPwrRl from the default value in case DL power is not limiting (e.g -150)

can tune it to -100.


Rule of Thumb
minPwrRl = max DL Tx Power primaryCpichPower 25

Increase minPwrMax from the default value in case DL power is not limiting (e.g 30)
Increase maxFach1Power from the default value in case DL power is not limiting (e.g 38)

Change spreadingFactor for RACH from 64 to 32

Note that setting spreadingFactor = 32 may reduce RACH coverage. If this is seen it is recommende
tune parameter powerOffsetPpm, which sets the power level for the RACH message based on the re
pre-amble power

The parameter pO3 sets the power offset between pilot bits and data bits in the downlink. Previous fi
have shown very positive effects on used A-DCH power in the downlink. The A-DCH power can be sub
in cells with many HSDPA users . The change improves HSDPA performance since less non-HS power i
in the downlink. The reduced power consumption also has a positive effect on downlink interference,
downlink interference is lowered.
Change pO3 to 0 (0 dB)

Recommendation

Incremented by one when an RRC connection request with cause Originating Conversational C
Terminating Conversational call, or emergency call is denied by admission control due to insu
licensed capacity in the RBS.

Parameters that can be tuned

Parameters that can be tuned


SPECIAL CHECK (for Hybrid & IP Nodes)
DlCreditUsage != AvgDLCEUsage
RECOMMENDATION: IUB Reset

Parameters that can be tuned

Action:
Reduce OwnUuload and RotCoverage (if EUL throughput is currently good,
otherewise we will be reducing EUL thp to an unacceptable level) to 80 and 100
respectively
Reduce aseUlAdm
Set Sf4AdminUl to 0, set sf8AdmUl to 0 (if (not already done)
reduce compModeAdm in outdoor cells
Consider increasing CPICH of indoor sites

Parameters that can be tuned


WNCS Analysis,Mapinfo,Mcom

Parameters that can be tuned

PARAMETERS FOR TUNING


To Reduce Drops due to UlSynch, can tune the parameters " rlFailureT " and nOutSyncInd "
For eg: A Cell is tuned with rlFailureT ( from 10 to 50) and nOutSyncInd (from 10 to 50) has shown a 2
improvement in Speech Drop Rate

Parameters that can be tuned

PARAMETERS FOR TUNING (UsedFreqThresh2dRscp)


GPEH analysis shown IRAT Drops were happening in Weaker Coverage Area.
In site VIEUX_LILLE_NORD_TEMP_U11 having high number of drop. After the implementation of the CR
31st on Oct the drop rate is reduced.

ACTion plan

Check and align the Power parameters & DL attenuation parameters.


Increase the value for pwrAdm + pwrOffset and could be less than 100%
Optimize the parametrs minPwrRl, minPwrMax, interPwrMax, maxPwrMax, maximumTransm
If 2nd carrier cell is not congested then share traffic by Qoffset (Idle mode)
If 2nd carrier is congestion also, then check neighbor cells and try to off load traffic by reduc
adjust Qoffset values
Do physical optimization incase of overshooting

FAILURE Counters:
RRC : pmNoRrcReqDeniedAdmDlChnlCode
RAB : pmNoFailedRabEstAttemptLackDlChnlCode
CRITERIA
ADMISSION REJECTION : code > dlcodeadm
Action:
Increase dlCodeAdm from 90 to as high as 95
Decrease pwrLoadThresholdDlSpeech.amr12200 from 50 to possibly as low as 0
Reduce NumHsScchCode (3 -----> 1)
Reduce NumHsPdschCode (1 reduce it if value is more than 1)
Reduce the values for sf4Adm, sf8Adm, sf16Adm to free some codes for other uses Carrier ad
SPECIAL CHECK (for Hybrid & IP Nodes)
UlCreditUsage != AvgULCEUsage
ACTION: IUB Reset

ers.
an 100%
xPwrMax, maximumTransmissionPower.
dle mode)
y to off load traffic by reducing cpich (-1 db) and

bly as low as 0

n 1)
des for other uses Carrier addition

pathlossThresholdEulTti2 & threshEulTti2Ecno parame


Accessibility where inter-site distance is high and the
conditions.

ulTti2Ecno parameters can be tuned to improve EUL and HS


e is high and the main reason for rejections is poor radio

EUL and HS
oor radio

You might also like