VoLTE Performance Analysis using SQI and Performance improvement

through Parameter Optimization

By:

Shehraz Shaikh/s00363887

Keywords: P3, Benchmark, LTE, Optimization, Parameter, Feature, VoLTE, SQI

Abstract:

SQI trial was performed in Country A Operator V’s network with specific targets for
optimization based on KQI information. Only the SQI platform was deployed without
SEQ, no probes were deployed as it was a small scale trial project for SQI alone, hence
no CN side analysis could be performed.

SQI platform was used for network insight and evaluation of webpage, streaming,
speed test, VoLTE, etc., services. This document describes the network insight and
evaluation of VoLTE service based on SQI data input along with optimization and
their results performed for VoLTE service improvement.

By: Shehraz Shaikh VoLTE Parameter Optimization ID: S00363887

Contents
Introduction/Situation .................................................................................................................... 3
1. Description of Problems/Challenges ....................................................................................... 4
2. Analysis, Strategies and Actions........................................................................................................ 5
2.1 Large VoLTE setup delay ............................................................................................................... 5
2.1.1 Analysis/Strategy ...................................................................................................................... 5
2.1.2 Action ........................................................................................................................................ 6
2.1.3 Results ....................................................................................................................................... 7
2.2 VoLTE Call Setup/Drop Rate/Others ............................................................................................. 8
2.2.1 Analysis/Strategy ...................................................................................................................... 8
2.2.2 Actions....................................................................................................................................... 9
2.2.3 Results ..................................................................................................................................... 11
2.3 VoLTE layering strategy/service steering .................................................................................... 13
2.3.1 Analysis/Strategy .................................................................................................................... 13
2.3.2 Actions..................................................................................................................................... 14
2.3.3 Results ..................................................................................................................................... 16
3. Results/Suggestion.......................................................................................................................... 16

Huawei Confidential 2|Page

Introduction/Situation

With the SQI trial, Operator V requested for deep analysis on VoLTE service for their network and provide
optimization techniques and strategies to improve the key KPIs related to VoLTE for P3 benchmark score.

Huawei used the SQI trial project as a model to analyze the network and provide solutions that can boost
the VoLTE experience for operator V users and thereby show the expertise of the SQI product.

SQI solution had the following focus points for deep analysis:
a. Optimization target set to KQI improvement rather than traditional KPI methodology.
b. E2E root cause analysis, including IP/SP/Device.
c. PM/MR/CHR/Probe data, multi data correlative analysis, precise demarcation as opposed to only
performance data.
d. Focus is service and end-user.

SQI trial team worked on a parallel with P3 Optimization Project Team and the structure was as below,

This report describes the evaluation and recommendations provided to operator V based on SQI and some
of the approved trials which were then rolled out as a part of P3 Optimization by operator V.

Huawei Confidential 3|Page

 1. Description of Problems/Challenges

Service quality and user experience are the focus items in any benchmarking tests and these areas have
previously been identified as weak for operator V, especially for VoLTE services. With the inclusion on
VoLTE in P3 tests for year 2017, short comings in performance for VoLTE had to be optimized, which led
to this exercise.

Business challenge for any operator is manage the User experience. This is where SQI provided a deep
insight.

Based on SQI platform, following analysis overview was presented to operator V to highlight the issues
in the network for VoLTE Service.

Huawei Confidential 4|Page

VoLTE Analysis Highlights:

VoLTE KQIs were broken down and following are the highlights of operator V’s short comings:

• Large call setup duration

• Call setup failures/Drops due to coverage/quality (NCCI/ACP project was setup for overall RF
tuning)
• Poor IMS and PDN connection SR (Operator V CN team engaged to follow up)
• Poor SRVCC performance

2. Analysis, Strategies and Actions

2.1 Large VoLTE setup delay

2.1.1 Analysis/Strategy
A breakdown analysis was performed and certain actions were undertaken to improve call setup delay.
Most delay was noted to be between invite and 183 and this was the key area for improvement. Good
improvement in Call setup delay was noted with optimization.

Huawei Confidential 5|Page

Further analysis based on comparison between operators showed a large difference which was due to
pre-conditioning being switched OFF for the other operator. Operator V considered Huawei’s
recommendation to switch OFF pre-conditioning, however testing with CN could not be completed in
time and this has been set to be completed in the following quarter after P3 final benchmark.

2.1.2 Action
With invite to 183 being the worst contributor detailed call traces for MO and MT UEs were checked and
it was drilled down to the delay contribution to be from VoLTE idle paging on MT side. The delay in
paging was the main contributor. Following parameter was modified to improve the delay.

Huawei Confidential 6|Page

 2.1.3 Results
Based on the above analysis, Default paging cycle was optimized from 128RF to 64RF and then
further to 32RF. Excellent results were observed from these changed and operator V rolled out the
parameter before P3 final benchmark.

128RF to 64RF:

64RF to 32RF:

Huawei Confidential 7|Page

 2.2 VoLTE Call Setup/Drop Rate/Others

2.2.1 Analysis/Strategy
Breakdown of the Call setup failures were analyzed and few issues were detected. Along with that a set
of parameters were implemented to improve the QCI packet loss, packet delay, improve HO SR, drop
rate, MOS, reduced retransmissions.

RF reasons was noted to be the biggest contributor for most KPIs and a team was setup to work on ACP
tool using live MR data for accurate RF optimization.

A set of parameters/switches were optimized to improve performance at network level with operator V
rolling them out based on results from DT and performance data.

Brief description of issues is provided below for reference.

Huawei Confidential 8|Page

 2.2.2 Actions
Following parameter were implemented to improve performance related to packet loss/delay,
scheduling delays, RRC re-establishments, cross procedures. Detailed parameter description provided
below.

Huawei Confidential 9|Page

Huawei Confidential 10 | P a g e
2.2.3 Results
 VoLTE MOS/VQI improved as observed from both DT and KPIs also and improvement in VoLTE
packet error rate noted as expected.
 Overall Radio drop rate and VoLTE drop rate including MME drops improved with the RRC Re-
establishment parameter changes.
 Improvements in RRC Re-establishment rate and HOSR noted with the optimization.
 DTx ratio improved as expected for PUCCH Power Control.
 PUCCH (initial) interference KPI increased as expected with power control parameter change.

Huawei Confidential 11 | P a g e
Huawei Confidential 12 | P a g e
 2.3 VoLTE layering strategy/service steering

2.3.1 Analysis/Strategy

RSRP/SINR Comparison shows that L850 coverage and quality is better than L1800 and the also the
number of L850 sites in operator V network is larger than L1800.

Based on the above information, proposal to perform service steering during QCI 1 setup was proposed
to Operator V.

Huawei Confidential 13 | P a g e
  VoLTE (QCI1) service steering solution:
 QCI1 on L18A/L18B/L21 can be transferred to L08A by “Service-based Inter-Frequency
Handover”.
 QCI1 released and enter Idle mode, then can:
 Cell Reselect to L18A/L18B/L21 based on frequency priority.
 MLB (RRC Release Mode) can be used to balance users between carriers.
 QCI1 released and still in connect mode, then can:
 MLB (Connected mode) can be used to shift load to high band.
 Frequency based HO can be used to shift load to high band.

2.3.2 Actions
Parameter changes were implemented and following were the observations.

 MOD ENODEBALGOSWITCH: HoAlgoSwitch=ServiceBasedInterFreqHoSwitch-1;

 MOD CELLALGOSWITCH: LocalCellId=1, HoAllowedSwitch=SrvBasedInterFreqHoSw-1;
 MOD INTERFREQHOGROUP: LocalCellId=1, InterFreqHoGroupId=1, InterFreqLoadBasedHoA4ThdRsrp=-
112;
 MOD
SERVICEIFHOCFGGROUP:CNOPERATORID=0,SERVICEIFHOCFGGROUPID=1,INTERFREQHOSTATE=PERMIT_H
O,DLEARFCN=L850_EARFCN;
 MOD
SERVICEIFDLEARFCNGRP:CNOPERATORID=0,SERVICEIFHOCFGGROUPID=1,DLEARFCNINDEX=0,DLEARFCN=
L850_EARFCN,SERVICEHOFREQPRIORITY=7;
 MOD CNOPERATORSTANDARDQCI:CNOPERATORID=0,QCI=QCI1,SERVICEIFHOCFGGROUPID=1;
 MOD CELLSTANDARDQCI: LocalCellId=1, Qci=QCI1, QciPriorityForHo=1;

Huawei Confidential 14 | P a g e
Based on testing, the behavior was as expected. Call were being handover to L850 when QCI 1 was
setup. Behavior when call was completed or any handover occurred were within expectations.

VoLTE and CA users behavior was as follows,

 User is performing CA with L1800A as Pcell and L1800B or L850 as Scell in 2CC or L1800B and L850 in 3CC.
 VoLTE call is initiated and QCI 1 is assigned.
 eNB sends the RRC Conn ReCfg to the UE with L08A information and the UE measures the L08A cell and
reports Event A4.
 eNB sends the mobility information to handover to L08A and includes the IE ScellRemoveList to remove the
Scell configuration.
 UE is now performing VoLTE call in L08A.
 Since QCI 1 is already assigned and all Scell configuration with the L08A being the Pcell are measurement
based, no CA configuration can be done until the QCI 1 is released.

Huawei Confidential 15 | P a g e
 2.3.3 Results
MOS was noted to have improved with a slight impact on setup delay in scenarios where gap mode is
still ongoing (in very poor coverage areas only) while the IMS SIP Invite for Ringing is expected. Since the
coverage area and indoor penetration for L850 is larger, Operator V rolled out the parameters to
enhance the VoLTE experience.

KPIs confirmed that the VoLTE traffic layers have shifted while maintaining the drop rates.

3. Results/Suggestion

Inputs from SQI, P3 drives and performance data were used in conjunction to perform VoLTE optimization
resulting in good performance improvement. All parameters were rolled out by operator V and based on internal
P3 DT performed, results were noted to have improved.

Operator V and Huawei partnered again in 2017 to optimize the network for P3 benchmarks and introduction of
SQI trial platform gave a more detailed insight on to the network. The approach of building a strong base and

Huawei Confidential 16 | P a g e
adding on top enhancements translated to a consistent performance in all test scenarios in every clutter area of
the network, proving that the tried and tested bottom top method is bound to provide results. The main target
for Operator V is to achieve best score in major where it has most of its customers.

The above details of parameter optimization activities on top of detailed re-planning, layering strategies,
transmission expansions, capacity expansions accelerated the improvements in Operator V network in
order to achieve the results. However, each network is unique and it has its own challenges and
difficulties. In order to optimize network for P3 purposes, the best starting point is to understand area
where it is possible to gain points based on gap analysis and improve rather than blindly start optimizing
everything. The approach has to be holistic and not narrow.

P3 tests user experience and thereby P3 optimization is different than improving network KPI and not
necessary that activities above improve network KPI. Customer expectations have to be set and
managed to avoid unnecessary rework.

Finally, it is team work that pays off and both internal and external parties need to work in sync towards
the target.

Huawei Confidential 17 | P a g e