Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Feature & Strategic Optimisation Activities

Download as pdf or txt
Download as pdf or txt
You are on page 1of 20

Security Level:

Feature & Strategic


optimisation activiti
es

www.huawei.com
Feature Dashboard – Huawei & Ericsson

Feature Alignment Feature Trial Total Feature Progress Progress timeline

Feature Audit
Total Feature Progress

100% 85% 95%


100% 96%
Week W06 W37 Dec’18

Feature Matrix
Highlight improvement
Huawei Ericsson
System Category 100%
Done SOW Done SOW 100% 27%
2G RF Quality 1 1 0 0 100%
Resource efficiency 0 0 1 1
Data Experience 25 25 13 14
Quality enhancement 47 49 40 41
3G Voice Experience 12 12 0 0 96%

RF Quality 6 6 7 7
3G Huawei | HSDPA Iub Flow Control Enhancement
Traffic Management Layering 4 4 19 22
Next Feature Activity Plan :
Resource efficiency 0 0 5 5 Trial Activity
Data Experience 10 11 3 3 • 3G | Admission Block Redirection (E)
Voice Experience 39 43 27 28 • 4G | Video Rate Adaptation (H), Dynamic PUCCH (E)
4G 95%
RF Quality 2 2 7 7 Roll Out
• NPS Proven strategy Batch-3 & PHICH Spreading Group
Traffic Management Layering 6 6 9 9
Strategic Optimisation Dashboard

Strategic Optimisation Progress timeline

Huawei Ericsson Strategic activity timeline


System Category Overal Progres
Done SOW Done SOW 2018 92% Dec’18
Voice Experience 0 0 1 1 W36

2G RF Quality 5 5 18 18 100%
Highlight improvement
Traffic Management Layering 0 0 3 3
18% HS to URA
Resource efficiency 2 2 6 6
measured
Data Experience 15 15 10 12

Voice Experience 14 14 23 26
3G 94%
Quality enhancement 0 0 1 1

RF Quality 4 4 4 4 3G Ericsson | HS/DCH to URA Activation

Traffic Management Layering 0 0 2 2 Next Strategic Optimisation

Resource efficiency 3 3 1 2 Trial Activity


• 3G | Dynamic Code Allocation NodeB & RNC (H), E2F State Transition (H),
Data Experience 24 28 12 15 Downswitch threshold (E), Power Allocation (E)
• 4G | BLER Improvement (H), Layering Step-3 (H), Layering Step-3 (E)
4G Voice Experience 0 0 1 1 85%
Roll Out
RF Quality 0 0 1 1 • NPS Proven strategy Batch-3

Traffic Management Layering 1 1 2 2


2G Ericsson | Strategic: QOS Setting and DSCP bit Mappings

Background › Rate of DL Framloss is


› Current QOS Setting is not match with Improved +17.04%
from 1.27% to 1.05%
Transmission Guideline (MOP Project
› CS Traffic Increased
QoS). RSL config in DSCP 51.
+7.02% from 7549.48
Benefit Erl to 7019.55 Erl
› Setting a proper config RSL to DSCP
40, schedulers change from priority
high to strict-high, bandwidth percent
4 to priority police rate percent 10
peak-rate percent 20.
Parameter
LBG, NWIE_BSC, PSTU, dscp2pcp40
Execution › Drop Call Number is
Decreased -5.37%
› 41 BSC, 755 Site. from 2289 to 2166.
› Drop Call Rate is
Maintainde from 1.50%
to 1.54%.

Site List Qos


Setting W37
2G Ericsson | Strategic: Speech Quality Improvement (1/2)

Background › SQI Good is improved


› AMRFRSUPPORT : Indicates if AMR 4.93% from 81.89% to
86.48%.
FR is turned ON or OFF and also
› SQI Bad and Accepted is
which full rate codec set that shall be
decreased 4.93% from
used within the BSC.
18.45% to 13.52%.
Benefit
› Speech Quality improvement.
Parameter
AMRFRSUPPORT set from “5” (support
4 Codec Mode 4.75 kbps, 5.9 kbps, 7.4
kbps, 12.2 kbps) to “0” (not support).

Execution › AMR FR stop working after


AMRSUPPORT set to “0”
BSC Before After (not support).
› Other KPI such as Speech
BJKT28 AMRFRSUPPORT-5 AMRFRSUPPORT-0
AMRFRSUPPORT-5 AMRFRSUPPORT-0
Drop, Speech Access and
BJKT29
BJKT31 AMRFRSUPPORT-5 AMRFRSUPPORT-0 Traffic maintain
BJKT32 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT33 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT34 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT35 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT36 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT37 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BJKT38 AMRFRSUPPORT-5 AMRFRSUPPORT-0
2G Ericsson | Strategic: Speech Quality Improvement (2/2)

Background › SQI Good is improved


› AMRFRSUPPORT : Indicates if AMR 4.09% from 78.68% to
82.78%.
FR is turned ON or OFF and also
› SQI Bad and Accepted is
which full rate codec set that shall be
decreased 4.09% from
used within the BSC.
21.32% to 17.22%.
Benefit
› Speech Quality improvement.
Parameter
AMRFRSUPPORT set from “5” (support
4 Codec Mode 4.75 kbps, 5.9 kbps, 7.4
kbps, 12.2 kbps) to “0” (not support).

Execution › AMR FR stop working after


AMRSUPPORT set to “0”
BSC Before After (not support).
› Other KPI such as Speech
BKRW02 AMRFRSUPPORT-5 AMRFRSUPPORT-0
Drop, Speech Access and
BLBA03 AMRFRSUPPORT-5 AMRFRSUPPORT-0
AMRFRSUPPORT-5 AMRFRSUPPORT-0 Traffic maintain
BLBA04
BLBA2 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BLBK1 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSKB03 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSKB2 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSRG03 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSRG2 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSRP1 AMRFRSUPPORT-5 AMRFRSUPPORT-0
BSRP2 AMRFRSUPPORT-5 AMRFRSUPPORT-0
2G Huawei | Strategic: Access and HO Parameter tuning

Background › Perceived Drop rate


improved 7.9% |
› Make sure user on best server
1.4% to 1.29%
during idle and dedicated mode.
› Handover SR
› Proper access parameter based
improved from 96.8%
on current traffic load.
to 97.00%
Benefit
0.21%
› Improved perceived drop rate 7.9%
› Improve handover success rate
Script:
› SET
GCELLBASICPARA:IDTYPE=BY
NAME,CELLNAME=“xx",LAYER=
x,RXMIN=xx;{BSCxx} › TCH Traffic increased
› SET 1.23% from 178 KErl
GCELLIDLEBASIC:IDTYPE=BYN to 180 KErl
AME,CELLNAME=“xx",PI=YES,C › Good VQI improved
RO=x;{BSCxx} 1.42%, 72.47% to
› SET 73.49%.
GCELLHOBASIC:IDTYPE=BYNA
ME,CELLNAME=“xx",HOCTRLS 1.42%
WITCH=HOALGORITHM1,ULED
GETHRES=x,DLEDGETHRES=x,
HOTHRES=x,LEVHOHYST=x,HO
CDCMINUPPWR=x,HOCDCMIND
WPWR=x;{BBDG12}
2G Ericsson | Strategic: TRX Downgrade and GFR (Outer Jabo)

Activity : › Perceived Drop Rate is


improved DCS1800 from
TRX downgrade Batch1 07-September 1.00% to 0.79%, GSM900
2018 23:00 WIB. from 2.00% to 1.63%.
TRX downgrade Batch2 11-September
2018 23:00 WIB.
GFR 16-17 September 2018 23:00-
03:00 WIB.

Comparison 10 and 17-September


2018 :

Result Other KPI


• Assignment Success Rate is
Maintained. › DL RxQual 0-5 DCS1800 is
• Perceived Congestion is Maintained, Maintained.
GSM900 improved from 0.29% to › UL RxQual 0-5 DCS1800 is
0.03%. Maintained.
• SDCCH Congestion Rate is improved
DCS1800 from 0.65% to 0.06%,
GSM900 from 1.72% to 0.31%.
• DL TBF Faiure Rate is improved
DCS1800 from 0.98% to 0.52%,
GSM900 degraded from 0.86% to
1.41%.
• SDCCH Drop Rate is Maintained.
• HOSR is Maintained.
3G Ericsson | Strategic: IFHO Improvement - CBNIFHO
Background
› PowerControl_cNbIFHO :
› Packet for RNJKT36
Factor used to take into account
Weekend to
the non-blind inter-frequency
Weekend Improve
handover margin.
from Peak Drop
› used to set the initial power at a
1.25% to stable at 1%
more conservative level to take
into account the inter-frequency
HO margin.
Benefit
› increase the reliability of the
inter-frequency handover.
› Better Inter Frequency HO due
to initial higher power allocated › Number of IFHO
Parameter Failure Improved 50%
from ~1500 to ~1000
PowerControl_cNbIFHO 15 to Drop Hourly
100
Execution
› Trial 1 RNC RNJKT36

Propose to Roll Out


3G Ericsson | Strategic: HS/DCH to URA Activation

Background › Retainbility for PS


› RabHandling_state128_128Support Drop is Improve –
ed : When set to TRUE, the Fast 18.10% from 1.16 %
Dormancy Handling feature will not Kbps to 0.95 %.
use the direct transition from
Cell_DCH to URA_PCH at reception
of a Signaling Connection Release
Indication with cause value "UE
requested PS data session end".
Instead the UE will be switched to
Cell_FACH.
Benefit
› Allow users transition from HS/DCH
directly to URA
› Data Traffic whole
Parameter Day Is Increase
RabHandling_state128_128Supported 1.76% from 4981 GB
TRUE  FALSE to 5068 GB.
› HS2Ura SR (98.56%)
Execution › DCH2Ura SR
(99.88%)
› 3 RNC RNBGR01 RNJKT26 &
RNTGR05

Step 2 FastDormancyHandling to
URA_PCH
3G Ericsson | Strategic: IuR Network Initiated CRE

Background › IurSpeechDrop is
› Iur_nwInitCallReestEnabled : Improve 10.75% from
Enables or disables Network Initiated 5.21% to 4.65%
Call Re-establishment support over › Attempt/Request
Iur. occur after activation
is enable
Benefit
› Speech Drop improvement on RNC
Border
Parameter
IurLink_nwInitCallReestEnabled

FALSE  TRUE

Execution › RRC Success rate


increase from 99% to
› 1 RNC RNJKT29
99.1%
› Overall speech drop
maintain
Propose Roll Out
3G Huawei | Feature: HSDPA Iub Flow Control Enhancement
Background › HSDPA User
100% Throughput improved
› In the original flow control detection feature, when the
packet loss rate is high, this feature determines that 100% from 150kbps
congestion occurs. Therefore, if severe non- to 300kbps in busiest
congestion-caused packet loss occurs, the available hours.
bandwidth on logical ports may be decreased to a › Total Payload
large extent, therefore affecting the data rate or even increased by 30%.
causing TRB to reset. HSPA flow control
enhancement is introduced in RAN17.1. This feature
can correctly determine the Iub congestion state in
non-congestion packet loss scenarios and adjust the
available Iub bandwidth.
Benefit
› After flow control is enhanced, whether a transport
network is congested when packet loss occurs can be
correctly determined. In this way, the available
bandwidth can be correctly adjusted, achieving the
maximum Iub bandwidth usage while ensuring that Iub
transmission is not congested. Improving Data
Experience.
Parameter 27%
› ADD GPM: GPMID=0, DIR=DOWN, DDM=FPDT,
DETECTOBJECT=RSCGRP, SN=xx,
SBT=BASE_BOARD, PT=ETH, PN=xx,
RSCGRPID=DEFAULTPORT;
› SET DLFLOWCTRLPARA: SN=xx,
SBT=BASE_BOARD, BEAR=IP, PT=ETH, PN=xx,
SWITCH=BW_SHAPING_ONOFF_TOGGLE,
DLFLOWCTRLENHSW=ON;
Execution
• Top 8 NodeB with severe Iub Congestion › RTT reduced by 27%
› Webpage Throughput by 31%.
Propose National Roll-out
3G Huawei | Feature: Sleeping Cell Detection and Recovery
› After activation sleeping
Background
cell still have issue.
› Sleeping Cell Detection and Recovery Payload & RRCAt › Root cause finding and
Feature Trial 100 enhance troubleshoot
Benefit 90 required
› With the Sleeping Cell Detection and 80
Recovery feature, operators can detect 70
sleeping cells and restore services of 60
sleeping cells, thereby reducing the impact 50
of sleeping cells on networks and improving 40
O&M efficiency as well as user experience. 30
20
Parameter
10
1. SET NODEBALGPARA: 0
SLEEPINGCELLDETECTOPTION=ULCOMPH
YCHANNELDETECT- Cell Availability(%) TotalPayloadGB
1&ULDEDPHYCHANNELDETECT- RRCat Sum of VS.RRC.AtConnEstab.Sum
1&DLCOMPHYCHANNELDETECT-
1&DLDEDPHYCHANNELDETECT-
› ALM-28209 Cell No
1&IUBDATADETECT-1&IUBSIGNALDETECT- Traffic appear after
1;SET feature enable to
USRBRATECOVSEL:CsF2DSrbRateCovSelSwi identify cell no traffic
tch=ON,CsF2DSrbRateGoodEcNoThd=0,SrbRa › Reset self-recovery
teForUENoEcNo=DCH_13.6K_SIGNALLING;
function working on
2. SET NODEBALGPARA: this site.
SLEEPINGCELLALARMSW=ON;
3. SET ULOCELLNOACCESSPARA:
ULOCELLID=1,NOUETIMER=0,
NOFSTRLTIMER=0,
AUTORCVRMTHD=BBRESOURCERESET;
Next Plan

Propose national Roll Out


3G Huawei | Strategic: CS Service Setup-triggered F2D State Transition
Background › KQI Call Setup
› Trial RAB Feature CS Service Setup-triggered F2D Call Setup Success Rate(%) 0.52%
Success Rate improve
State Transition. F2D Coverage-based Adaptive 97 3 4 12.82% from 96.1% to
SRB Rate Adjust SW. When the feature is on the 1 2 96.6
96.6% on Busy hour
Ec/No when UEs perform CELL_FACH-to- 96.596.1 96.2
CELL_DCH state transitions prior to setting up CS
96
services. 95.5
95.5
Benefit
95
› Improvement CS Accessibility & delay time
94.5
Parameter
94
1. SET USRBRATECOVSEL:RRCSETSRBRATEBADECNOTHD=-
8,CsP2DSrbRateBadEcNoThd-8, CsF2DSrbRateBadEcNoThd=- 93.5
8;SET USRBRATECOVSEL:CsF2DSrbRateBadEcNoThd=-20;
SET USRBRATECOVSEL:CsF2DSrbRateBadEcNoThd=-8; 93
2. SET USRBRATECOVSEL:RRCSETSRBRATEBADECNOTHD=-
ug ug ug ug S ep S ep S ep S ep S ep S ep S ep S ep S ep ep ep ep ep Se
p ep ep ep ep
20,CsP2DSrbRateBadEcNoThd-20, -A -A -A -A 1- 2- 3- 4- 5- 6- 7- 8- 9- -S -S -S -S - -S -S -S -S
CsF2DSrbRateBadEcNoThd=-20; 28 29 30 31 10 11 12 13 14 15 16 17 18
3. SET USRBRATECOVSEL: SRBRATEFORUENOECNO=OFF;
SET URRCESTCAUSE:RRCCAUSE=ORIGCONVCALLEST, › E2E Call
SIGCHTYPE=DCH_3.4K_SIGNALLING; SET
Connection Delay
URRCESTCAUSE:RRCCAUSE=TERMCONVCALLEST,
SIGCHTYPE=DCH_3.4K_SIGNALLING;
E2E Call Connection Delay (ms) improve 1.4% (~69
5800 1.4%
4. SET URRCESTCAUSE:RRCCAUSE=ORIGCONVCALLEST, ms) from 5061 ms
SIGCHTYPE=DCH_13.6K_SIGNALLING; SET
URRCESTCAUSE:RRCCAUSE=TERMCONVCALLEST,
5600 to 4992.5 ms
SIGCHTYPE=DCH_13.6K_SIGNALLING; SET
USRBRATECOVSEL:
5400
SRBRATEFORUENOECNO=DCH_13.6K_SIGNALLING;
5172
52005061 5052
SET USRBRATECOVSEL: RRCSETSRBRATEBADECNOTHD=-
20,CSP2DSRBRATEBADECNOTHD=- 5000
20,CSF2DSRBRATEBADECNOTHD=-20; 4992.5
4800
Execution
4600
› Trial done in RNMKS02
4400
Next Plan g g g g p p p p p p p p p p p p p p p p p p
-Au -A u -Au -Au 1-Se 2-Se 3-Se 4-Se 5-Se 6-Se 7-Se 8-Se 9-Se 0-Se 1-Se 2-Se 3-Se 4-Se 5-Se 6-Se 7-Se 8-Se
Propose national Roll Out use setting no 4 28 29 30 31 1 1 1 1 1 1 1 1 1

Trial in RNMKS02
4G Ericsson | Strategic: filterCoefficient

Background › Redirect to 3G
› Measurements of RSRP or RSRQ are improve 26.5% from
performed on the serving and detected 129183 to 94926.
neighboring cells. The UE averages a
number of measurements to filter the
impact of large scale fast fading.
26.5%
Different filter values can be defined for
RSRP and RSRQ. Filter parameters
filterCoefficientEUtraRsrp can be used
to average out the impact of fast fading.

Benefit
› User Data Fallback 4G to 2G/3G
expected decrease
› Data Fallback 4G to
2G/3G improve 25%
Parameter from 9.2% to 6.9%.
› filterCoefficientEUtraRsrp = 8 25%

Execution
› 1 Kabupaten done execute (Kota
Yogyakarta)

Waiting result from Jabo before Roll Out


4G Ericsson | Strategic : Trial a1a2SearchThresholdRsrp
Background
› a1a2SearchThresholdRsrp :
RSRP threshold for the Event
457.76% -4.90%
A1 and Event A2
measurements detecting the
search zone
› a1a2SearchThresholdRsrp
adjustment from -116 dBm to 4G payload reduce around 3 TB
-116 dbm to know to makes TRIAL
easier 4G release to 3G
Period
Benefit/Impact
› User Fallback 4G to 3G will
increase
Parameter
Set ReportConfigSearch
a1a2SearchThresholdRsrp from -118 to
-105
Execution
23 sites of Kab Tangerang area
Result
› 4G user fallback to 3G
increase 457%
› 4G Payload decrease 5% (3 3G payload increase around 2 TB
TB) and 3G Payload increase
6% (2 TB)
› Suggest not to set aggressive
threshold to balance user
throughput experience which
impacting to payload overall
4G Ericsson | Strategic : Trial a1a2SearchThresholdRsrp
Background LTE Handset sensitivity
› a1a2SearchThresholdRsrp : Phone 5 Bars 4 Bars 3 Bars 2 Bars 1 Bar 0 Bar No Signal
RSRP threshold for the Event
Samsung   > -95 -96, -101 -102, -110 -111, -116 -117, -124 < -124
For user perception we propose
A1 and Event A2
measurements detecting the iPhone > -88 -89, -96 -97, -105 -106, -112 -113, -124   < -124 set to <= -116 dBm
search zone Source: Qualcomm test

› a1a2SearchThresholdRsrp User strength signal on Samsung having low perception comparing Iphone
adjustment from -116 dBm to
-116 dbm to know to makes
easier 4G release to 3G 4G Signal level -112 3G Signal level > -95
to -124 (0-1 Bar) to -124 (3-4 Bars)
Benefit/Impact
› User Fallback 4G to 3G will
increase
Parameter
Set ReportConfigSearch
a1a2SearchThresholdRsrp from -118 to
-105
Execution
23 sites of Kab Tangerang area
Result
› 3G Serving increase from 6%
to 19%
› Need to check throughput
3G 6% 3G 19%
comparison with 4G (0-1 Bar) 4G 94% 4G 81%
and 3G (3-4) Bars on Samsung
phone
4G Huawei | Strategic: CSFB Optimization
› MT Fallback delay
Background improved 1% |
› Add Utrannfreq,MOD 1195 ms to 1186
CSFALLBACKHO, MOD 10ms ms
CSFallbackHO, MOD › MO Fallback delay
Enodebalgoswitch improved 1% |
Benefit 1387 ms to 1378
ms
› CSFB improvement
› Fallback Delay improvement
Parameter
- Add UTRANNFREQ
- UtranCsfbSwitch-
1,HOMODESWITCH=UtranRedir
ectSwitch-1&BlindHoSwitch-1 › 4G Payload
- BlindHoA1A2ThdRsrp=-121, maintained
BlindHoA1A2ThdRsrq=-28, › Session Setup SR
HoModeSwitch=BlindHoSwitch-1 maintained

Execution
› Rollout West Java
4G Huawei | Feature: Ue MRO

Background
› eRAB drop improved
 Ping-pong HO between intra-freq cell can from 0.90% to 0.59%
› Payload Maintain
cause bad experience to user. To overcome
ping pong HO Ue level MRO can be applied
to UE so that each use set with different CIO
value
Benefit
 Ping-Pong HO reduced,
 HO success rate improved

Parameter
MOD CELLALGOSWITCH:LocalCellId=xx, › Ping pong ratio
MroSwitch=UEMroSwitch-1; improved from 1.19%
MOD MRO: PingpongTimeThd=2, to 0.97%
PingpongRatioThd=2; › Payload Maintain

Execution
› 50 eNodeB in East Java

Next Plan
› National Rollout

You might also like