Drivetest Session I & II

May 11th & 18th, 2006

 Table of content
 Introduction :
 Tuning Flow Diagram
 Responsibility
 Preparation & Drivetest
 Analysis Process
 Reports
 Introduction
• Drivetest / Walktest :
 to measure downlink signals from Node B to UE

 to test the network environment to achieve predefined KPIs

• Objective :
 to get the image of Cluster Tuning Process

 to understand the Responsibility of each phase in the whole

Cluster Tuning activities

 to understand how to achieve high quality of cluster tuning

 Tuning Flow Diagram (1/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work: Generating pre-simulation project

and preparing drive test
(vehicle, equipment & site ready, etc)
•Output: Drive test team ready
Route map
Site information
Site location map
Pre-simulation of cluster
Problem Area Sheet
Drive Data Sheet
•Owner: Navigator, Field test engineer,
Data analyst
•Period: <= 1 day
 Tuning Flow Diagram (2/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work: Performing measurement along the

defined drive route and confirming
Problem Area Sheet.
•Output: Measurement raw data (2 CD-Roms)
Problem Area Sheet
Drive Data Sheet
•Owner: Navigator, Field test engineer
•Period: 2 days
 Tuning Flow Diagram (3/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work: Data processing of raw data for further

analysis.
Finding a draft solution from drive test
data.
Simulation and verification of tuning
solution
using NetActPlanner.
•Output: Tuning solution
Plots of NetActPlanner simulations
Plots of measurement data
Improvement calculation of when tuning
solution is applied
All files related to the above items
•Owner: Data Processor, Data analyst
•Period: 2 days
 Tuning Flow Diagram (4/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work: Reporting measurement result of the

cluster and verification of improvement
•Output: Cluster tuning report (2 copies)
Deliverables defined in Report format
•Owner: Data analyst
•Period: 1-2 days
•After Report is made, it should be discussed with
Nokia NWP
 Tuning Flow Diagram (5/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work1: Report and all deliverables are checked

by customer and discussion for approval of
tuning
solution
•Output: Customer’s approval
•Owner: Nokia NWP team, data analyst, sub-
contractor’s coordinator
•Period: 2 days (Submission + Approval)
•Work2: Request for BTS implementation change
to Nokia’s implementation team and
customer
•Output: Change request for Nokia and Customer
•Owner: Nokia RF team, Customer
•Period: Immediately triggered after approval
 Tuning Flow Diagram (6/6)
• Cluster tuning flow diagram

Preparatio Drive Test Analysis Report Approval Implement

n

•Work: BTS implementation change by Nokia’s

implementation team and customer
•Output: Change of implementation
•Owner: Nokia Implementation team, Customer
•Period: 1 day
 Responsibility (1/2)
Drive the car according to drive test route been plan
Driver Control the car speed at adequate speed (20-40 km/h)
Ensure the car is following right route

Navigator To Prepare drive plan and route map

To Discuss and agree with other drive team navigators on
which area is
their responsibility to avoid unnecessary additional drive test.
To Assist Field test engineer to located the problem area on the
map
and check the area condition.
To Check out the actual testing car’s position
Ensure the car is following right route
To Study the pre-analysis result from the simulation team
The road on 2nd drive test is following 1st drive test route (if
Field Test Engineermismatch,
To Perform explanation is needed.)
the Field testing
To Study the pre-analysis result from the simulation team
To Setup the tools in the car and ensure in the good condition
To Observe and identify the problem area during the drive test.
Fill in the data in the problem sheet.
To Write down your observation during the drive test
To “Save” the drive test data in file name according Nokia definition
To Copy the data into CD Rom and submit to Window person
To inform the problematic status to Window person
To Fill in the data sheet
 Responsibility (2/2)
To Perform data process & Print out plots and graphs for Data Analysis
Data Processor To Copy / plot the common picture to the report
To Take care all files and data related to analysis and report
To Prepare the data which requested by analysis engineer

Data Analyst To Analyze the data and carry out Simulation
Reporting. Submit the report to Nokia before see the customer
Explain to customer about the Solution (together with Nokia engineer)
Improve the report according to customer request
Ensure the changes are implemented to the network
Instruct Drive test team to start the 2nd drive without any delay

Window Person Confirm the site status with Nokia NMS engineers
Located in Nokia office
Ensure the RAN tuning activity is carried out according to the schedule
Participate the progress meeting if necessary
Main interface between Nokia and subcons
 Table of content
 Introduction :
 Tuning Flow Diagram
 Responsability
 Preparation & Drivetest
 Analysis Process
 Reports
 Drivetest Preparation

• Preparation item list :

• Road Map which specified the road that should be measured.
• Route Map which specified the actual measurement route.
• Site information of cluster
• Site location Map
• Pre-simulation results (NetAct if any)
• Problem Area Sheet and Drive Data Sheet for each cluster
• Test equipment
• Measurement Vehicle
• Set up the Measurement Vehicle and test equipments
• Sufficient Drive test teams
 Drivetest Preparation
• Map
• Road Map
• Nokia prepares Road Map agreed with Telkomsel.
• Road map shows road that should be measured during drive test.
• The Road Map is based on cluster categories as shown in the table.

Sub-cluster　Categories Description
BTS　Sites Max DT Route [km]
Urban 6-10 160 km
Categories－1
Sub-urban 1-5 200 km
Urban 11-15 320 km
Categories－2
Sub-urban 6-10 400 km
Urban 16-20 450 km
Categories－3
Sub-urban 11-15 600 km
 Drivetest Preparation
• Route Map
• Navigator plans Route Map according to Road Map
• Every road in the Road Map should be included but drive route in
route map should not be overlapped each other if possible.
• If highway exists in cluster, the measurement in highway should
basically be performed and stored in separated data file.
• Problem areas should be marked in one of Route maps
• Navigator handles 2 Route map – one is for navigation and the other
is for announcing field test engineer the problem area during drive test.

No good

Modify
 Drivetest Route

Nokia trial sites

 Drivetest Preparation
• Site information of each cluster
• Navigator prepares site information, obtained from Nokia, for the
field test engineer.
• Contents are Site ID, Site Name, Scrambling Code and Ant Info.
• Site location can be confirmed by Site Location Map
Primary
Scrambling
WBTS Name Site ID WBTS Cell Cell ID Azimuth E Tilt M Tilt Antenna Type Long Lat RNC ID LAC SAC RAC URA Code UARFCN
EXCHANGEHSEMW 1501 EXCHANGEHSEMW1 15011 20 4 2 Kathrein 742 215 106.832 -6.23306 301 116 15011 93 116 8 10688
EXCHANGEHSEMW2 15012 160 4 2 Kathrein 742 215 106.832 -6.23306 301 116 15012 93 116 16 10688
EXCHANGEHSEMW3 15013 270 4 2 Kathrein 742 215 106.832 -6.23306 301 116 15013 93 116 24 10688
DRJLISTRIKMW 1502 DRJLISTRIKMW1 15021 40 4 2 Kathrein 742 215 106.83359 -6.22625 301 116 15021 93 116 9 10688
DRJLISTRIKMW2 15022 200 4 2 Kathrein 742 215 106.83359 -6.22625 301 116 15022 93 116 17 10688
DRJLISTRIKMW3 15023 320 4 2 Kathrein 742 215 106.83359 -6.22625 301 116 15023 93 116 25 10688
UPPINDOMW 1503 UPPINDOMW1 15031 40 4 2 Kathrein 742 215 106.832 -6.21917 301 116 15031 93 116 10 10688
UPPINDOMW2 15032 160 4 2 Kathrein 742 215 106.832 -6.21917 301 116 15032 93 116 18 10688
UPPINDOMW3 15033 260 4 2 Kathrein 742 215 106.832 -6.21917 301 116 15033 93 116 26 10688
GDGMIPMW 1504 GDGMIPMW1 15041 60 4 2 Kathrein 742 215 106.828639 -6.213722 301 116 15041 93 116 11 10688
GDGMIPMW2 15042 180 4 2 Kathrein 742 215 106.828639 -6.213722 301 116 15042 93 116 19 10688
GDGMIPMW3 15043 260 4 2 Kathrein 742 215 106.828639 -6.213722 301 116 15043 93 116 27 10688
GEDUNGLINAMW 1505 GEDUNGLINAMW1 15051 10 4 2 Kathrein 742 215 106.82942 -6.20906 301 116 15051 93 116 12 10688
GEDUNGLINAMW2 15052 90 4 2 Kathrein 742 215 106.82942 -6.20906 301 116 15052 93 116 20 10688
GEDUNGLINAMW3 15053 270 4 2 Kathrein 742 215 106.82942 -6.20906 301 116 15053 93 116 28 10688
KENDALMW 1506 KENDALMW1 15061 90 4 2 Kathrein 742 215 106.82675 -6.202861 301 116 15061 93 116 13 10688
KENDALMW2 15062 270 4 2 Kathrein 742 215 106.82675 -6.202861 301 116 15062 93 116 21 10688
KENDALMW3 15063 310 4 2 Kathrein 742 215 106.82675 -6.202861 301 116 15063 93 116 29 10688
RKPECENONGANMW 1507 RKPECENONGANMW1 15071 50 0 0 Kathrein 742 215 106.8268 -6.16625 301 116 15071 93 116 14 10688
RKPECENONGANMW2 15072 150 0 0 Kathrein 742 215 106.8268 -6.16625 301 116 15072 93 116 22 10688
RKPECENONGANMW3 15073 240 0 0 Kathrein 742 215 106.8268 -6.16625 301 116 15073 93 116 30 10688
TANAHABANGMW 1508 TANAHABANGMW1 15081 40 0 0 Kathrein 742 215 106.81985 -6.176 301 116 15081 93 116 15 10688
TANAHABANGMW2 15082 160 0 0 Kathrein 742 215 106.81985 -6.176 301 116 15082 93 116 23 10688
TANAHABANGMW3 15083 280 0 0 Kathrein 742 215 106.81985 -6.176 301 116 15083 93 116 31 10688
 Drivetest Preparation
• Site Location Map (Navigator prepares using MapInfo)
•Field test engineer makes sure site location, scrambling codes of cells using this
map.
•This map helps finding SC of which RSCP and Ec/No is measured during drive
test.
 Drivetest Preparation

• Pre-Simulation Results
• Default propagation model (Okumura Hata model) is used for pre-
simulation
• Problem Candidate Areas found through Pre-simulation result are
marked in one Route map and Problem Area Sheet – This work is
done by Data Analyst.
• Plot of Simulated Best Server RSCP:
Identifies poor best server’s RSCP coverage areas. The threshold
level is –92dBm.
• Plot of Simulated Best Server Ec/Io:
Identifies poor best server’s Ec/No coverage areas. The
threshold level is –9 dB.
• Plot of Simulated 4th Best Server RSCP:
Identifies excessive no. of SC – pilot pollution - from RSCP
point of view. The threshold is –92 dBm as well.
 Drivetest Preparation

• Problem Area sheet

• Data Analyst should make the Problem Area Sheet to inform
Field test engineer of the points where problem is likely to
take place before drive test.
• The Field test engineer should observe the problem area
with attention during measurement and report the result and
pre-investigation to analysis engineer.
• Plus, Field test engineer should inform Data Analyst of the
location or environment that he could observe.
• The problem areas in this sheet are based on Pre-simulation
result.
 Drivetest Preparation
Ec/No (from UE)
The connection between Siemens RAN and Nokia Core was implemented successfully. Iur
was not still in place, therefore no Inter-vendor (Nokia-Siemens) soft handover was possible

Active set update not possible, UE

Ec/No is not reliable with the real
Ec/No from Scanner
Drivetest Preparation
RSCP (from UE)

Active set update not possible, UE

RSCP is not reliable with the real
RSCP from Scanner
 Drivetest Preparation
• Problem Area Sheet format 1st drive test 2nd drive test
 Drivetest Preparation
Ｄｒｉｖｅ　Ｄａｔａ　Ｓｈｅｅｔ

• Drive Data Sheet ＊地域(Region)

( Cluster Drive Test )

東北(Tohoku) ： H ／ 関西(Kansai) ： K ／ 中国(Cyugoku) ： N ／ 四国 (Shikoku) ： S 07/Nov/01 ver 1.0B

地域名(Region Name) Cluster Name

• Used for confirmation of drive team 測定日(Ｄａｔｅ) 　 　　／　　　／

測定の種類 (measurement type)
～ 　 　　／　　　／
１ｓｔ　・　２ｎｄ　・　３rd　・　Moniter　・　ｅｔｃ

information, status of drive testing, Tool　No　（ ＩＤ ）

Leader　Name
Team　Name
Assistant　Name
Ｃａｒ　Number
Driver　Name

＜ 使用機材の確認 ( Confirmation of machine parts ) ＞

equipments, sites, and measurement GPS and Gyro ・　GPS－ANT
Before
□
After
□ Personal ・　PC
Before
□
After
□

file. ・　Gyro or Converter

・　ＲＳ２３２Ｃ　Cable
□
□
□
□
Computer ・　POW　Cable
・　POW　Adapter
□
□
□
□
Before After Before After
Anritsu　Tool ・　ML8720B □ □ etc. ・　DC-AC　Converter □ □

• The Field test engineer needs to bring

・　ＲＳ２３２Ｃ　Cable □ □ ・　socket □ □
・　POW　Cable □ □ ・　Map　Two pieces □ □
・　POW　Adapter □ □ ・　Problem　Area　Sheet □ □
・　RX－ANT □ □ □ □

this sheet for every cluster test. ・　ATT　X ｄB

・　PC card
□
□
□
□
□
□
□
□

This is the basic information of drive test ＜　局の状態確認 ( Ｓｉｔｅ　Ｓｔａｔｕｓ )　＞

condition. All information is entered before drive 測定対象のCluster

( Cluster for measurement )

test start. 局状態(Ｓｉｔｅ Ｓｔａｔｕｓ) ：

局番号(Ｓｉｔｅ ＩＤ) 局名(Site Name) 備考(note)
局状態(Ｓｉｔｅ Ｓｔａｔｕｓ) ：
局番号(Ｓｉｔｅ ＩＤ) 局名(Site Name) 備考(note)

Parts checking box. There are two check boxes

of Before and After on each parts. The
existence of every parts must be checked
before and after drive test.
This is the site status Table. It is entered in case ＜　データ・ファイルの管理 ( Management of Data File )　＞

that measurement engineer find the site ファイル名

(File Name)
開始時間 開始時の距離 停止時間 停止時の距離 停止した理由
(Start Time) (Start mileage) (Stop Time) (Stop mileage) (standstill reason)

problem during drive test. 1

2
　

Data file management Table. File name, Start 4

5

Time and Start mileage are entered when 6

7
logging is started. Stop Time, Stop mileage and 8

reason of Stopping are entered when logging is 9

10

stopped. Odometer value of car is used for 11

12
mileage. 13
14
 Drivetest Preparation

• Test equipments :
• Nemo SAM & TOM dongles & software
• Nemo handy (phone 6630, 6650, etc), with USB cable
• GPS + car roof antenna, preferably Garmin III+
• Scanner, preferably Anritsu (high accuracy).
• Laptop
• Inverter + 3m cable
• Dual (or triple ?) serial port (PCMCIA/USB)
• Handlers Licenses (GPS, Handphone, scanner)
• Extension Cable
• Case
 Drivetest Tool Configuration
• Scanner, AMR & Data call test connection configuration of equipments

CS
Call

PS
call

* Example of drive test tool set up

 Cluster Drivetest
Start

Confirmation of site status

Equipment check
Can Can’t
Whether measurement
Logging Start can be continued or
not.(In office)

Logging

Logging stop and inform

No the status to the Office
No
No

Problem Area? The equipment or site

problem is found Yes
Is it necessary to
Yes
stop Logging? Measurement suspended
Yes
Fill in the Problem Area sheet
Logging Stop
Report to Office

No Yes Drive Test

Are all schedules Report to Office
completion? Finish
 Cluster Drivetest
• Window person
• Has to check if all sites are ready for drive test
• Window person handles information related to drive test between
Field test engineer and Nokia RF and needs to stay in Nokia office.
• Prepare the report on the drive test status and checklist confirmed
during the drive test.

(in the car) (in Nokia

Driver office) Tools intellectual

NMS Person
Navigator Measurement Engineer Window person
RNC Engineer

Etc
Driver
Driver
Driver
Driver

Navigator MeasurementEngineer
Engineer
Navigator Measurement
Navigator Measurement Engineer
Navigator Measurement Engineer
 Cluster Drivetest
• Report of Drive Test Status (for Window person)
 Cluster Drivetest
Drive test team SHOULD BRING the following items :

• Two Route maps for navigation and marking problem areas

• Site information sheet
• Site location map
• Problem Area sheet
• Drive Data sheet
• Measurement equipment
 Cluster Drivetest
The confirmation check list before drive test :
• Site status in the Cluster
 Nokia RF must confirm the site status in planned cluster from NMS
every morning (before Drive test) and inform Window person of
whether drive test can start or not.
• Route maps, equipments, and Vehicle
 Drive test team must confirm whether the drive route(Map, Cluster) is
the same with assignment from coordinator, and equipment number,
vehicle number in the Drive Data Sheet.
• Existence of equipment parts.
 Drive test team must confirm if all equipments are well connected and
stabilized and remark it on the Drive Data sheet.
• Measurement equipment is working
 Drive test team must confirm to behave the measurement equipment
normally (RF, GPS ,etc): test measurement nearby office.
 Cluster Drivetest
• Notes at the time of Logging Start
• The Required matters are filled in the Drive Data Sheet
• File name(Detail is after page), Start time, Value of odometer
• The Start point is marked in the route map
• Confirmation of GPS reception
 Cluster Drivetest
• Notes during Logging
• Confirmation of equipment being working and site normality
• Confirmed that the Logging is performed normally by RF,GPS and PC.
• Confirmed that the signal of neighboring site are received by checking the SC and its
RSCP & Ec/No
• Field test engineer must stop the Logging and inform the status to
Window person in case of followings:
• The measurement can’t be performed due to equipment problem after trying to improve
• Site problem is found (Cell or Site down, etc)
• In the office side (Window person), decide whether measurement
can be continued or not and inform Field test engineer
• Field test engineer may pause measurement while stopping on
traffic signal and then start measurement when moving again.
 Cluster Drivetest
• Notes during Logging
• In case that Field test engineer come near the problem area
according to the Problem Area Sheet
• To describe types of problems and environments
• In case that Field test engineer finds a new problem area
• The criteria of problem area are:
• Poor Best Server’s RSCP coverage: best server’s RSCP is less
than –92 dBm
• Poor Best Server’s Ec/No coverage: best server’s Ec/No is less than
–9 dB
• Excessive No. of SC (RSCP) : No. of SC >= 4 & Threshold of RSCP
>= –92dBm
• If Field test engineer finds a new problem area, he will enter the
necessary matters (description – problem property, values) in
Problem Area Sheet and the position marked in the Route map.
 Cluster Drivetest
• Notes at the time of Logging Stop
• Condition of the Logging Stop
• All planed tests are finished
• Field test engineer is directed to stop Logging
• A break is taken (i.e Lunch)
• Logging file is updated (ex. every one hour)
• Notes
• The stop point is marked in the route map
• The Required matters are filled in the Drive Data Sheet
• Stop time, reason of stopping, value of odometer
 Cluster Drivetest
• Notes at the time of measurement finished
• Confirmation whether planned route is completed or not
• Confirmation whether data logging was performed normally or
not by using Replay function of logging tool
• The first, the data logged in measurement PC is confirmed.
• The data logged in PC card in Anritsu measurement tool is
confirmed if the data logged in measurement PC have some
problem.
• Informing the office of the finish of measurement
 Cluster Drivetest
• Data Management
• Definition of the File Name
• The measurement data files must be named as follows figure.
• In case that “Auto Saving” setting in Anritsu measurement tool is set to
“Auto Increment File Name”, the file is named automatically with only one
increment of File No from second time if file is named manually with File
No “-001” at the first time.

• However, they have to

Region Date Cluster Name Drive team File No
name the file with the Name xx xx xx No ―
―
continuous file number
manually if they turn off the
power so that file name
K yymmdd_ A 1 C - 002
setting is reset to default one. Region Name; Jakarta : JKT Cluster Name; Cluster ID
Bogor : BGR
Bandung : BDG Block No,; Number of Cluster division
(Cluster is divided by 1day drive unit)
Date; Measurement Date
File No,; File Serial Number
 Cluster Drivetest
• Data Management
• Drive Data Management
• Data transfer to Nokia is performed using CD-ROM
• The CD-R should be made after whole drive test of the cluster is
completed, and two CD-ROMs are needed.

(in the car) (in the office)

Database Server

Card
Reader
NOKIA
ML8720B software CD-R
or

NOKIA
software Local Network

and Logged in this PC H/D

PC-Card This organization is different according to the
network structure and the peripherals in the each
office.
 Table of content
 Introduction :
 Tuning Flow Diagram
 Responsability
 Preparation & Drivetest
 Analysis Process
 Reports
 Measurement Data Analysis
• Role of Analysis
 To extract the problematic area or/and event
 To make tuning solution against the problems
 Simulation using NetActPlanner in order to verify the tuning solution if the
problem is related to poor RSCP or Ec/No coverage
• Analysis process consists of
 Most of data processing can be done using Nemo SAM / Macro (ACTIX is
supplementary)
 Field Data Analysis :
 Call setup failure analysis
 Call drop analysis
 BLER analysis
 MOS analysis (Swissqual)
 Simulation Analysis: Needed for verifying RSCP and Ec/No coverage
optimization
• Data Analysis and Simulation processes are combined for coverage
optimization
 A draft tuning solutions are verified using NetActPlanner
 Simulations to confirm the effectiveness of the tuning solutions(RSCP & Ec/No)
 Result before applying the tuning solutions
 Result after applying the tuning solutions
 However, the simulation result will not be presented in the report. (Keep it as
BACKUP)
 Analysis Process Flow Diagram
Field Data Project

Data Processing Model tuning

Processed Data Prob. area sheet Propagation model Other tuned model

Field Data Analysis branch

Problem Check Other simulation Simulation

Simulation branch
Problematic area Simulation result
w/o solution

Analysis Re- simulation

Legend

Input Draft solutions Simulation result

with solution
Step
Comparison
data Feedback

Output Solution
 Data Processing & Analysis
Plot of Measured Best
Processing Server RSCP

Plot of Each SC RSCP)

Plot of Measured Best
Material For Report Server Ec/No
Plot of Each SC Ec/No
Material for only Plot of Measured No. of
analysis SC (RSCP)

(Macro (MapInfo Plot of Measured Best

) ) SC
Making MapInfo Making Map Problem Area Map
Input file
Raw Data Change
Format
Making Exporting and
Graph making graph

Graph of RSCP and Ec/No

in problem area
Summary average
value
Histogram and CDF of RSCP

Distribution of RSCP and Ec/No of Cluster Histogram and CDF of Ec/No

Distribution of RSCP and Ec/No of problem area Histogram and CDF of No. of SC
 Data Processing & Analysis
• Required material for making Cluster / Area Tuning Report
Item Detail
RF Performance Analysis (note: Problem Area Map
each UE will have own CPICH Best Server CPICH RSCP
RSCP and Ec/No data for the Best Server CPICH Ec/No
analysis) Cumulative Distribution of RSCP
Cumulative Distribution of Ec/No
Distribution of RSCP and Ec/No
Voice AMR

Call Set-up Failure Ratio

Call Drop Ratio
HO Ratio
AMR Measurement Distribution of MOS Evaluation
BLER
MOS
Call Drop Point Analysis Call drop point indicated on map
PS

Throughput Cumulative Distribution

Average Download time
Call Set-up Failure Ratio
Call Drop Ratio
PS Measurement FTP Throughput
BLER
Coverage Map Throughput
UDI

Average Download time

Call Set-up Failure Ratio
Call Drop Ratio
FTP Throughput
Coverage Map BLER
Throughput
 Data Processing & Analysis
• Plot of Measured Best Server RSCP
• Plotted along the drive route on the map
• Shows measured best server’s CPICH RSCP level so that Data Analyst
can see the Coverage problem area
• Plot of Measured Best Server Ec/No
• Plotted along the drive route on the map
• Shows measured best server’s CPICH Ec/No level so that Data Analyst
can see the Dominance problem area (Low Best Server Ec/No)
 Data Processing & Analysis
• Problem Area Map
• Shows all problem areas in the Cluster / Area
• MapInfo is used to draw polygons of problem areas on the map.
• This map is used for producing Graph of RSCP and Ec/No in problem
area, which is explained later section
• Minimize the size of polygon for easy area identification
• Several polygons that have same solution/causes of problem could be
combined on one problem sheet.
• Case by case. Analysis is needed in order to identify whether it is real
problem area or not => then to be put as problem area or not could be
decided. Some notes could be taken for backup explanation.
 Data Processing & Analysis
• Histogram and CDF of RSCP
• Shows the percentage of samples at different received RSCP level for whole
drive data
• Histogram and CDF of Ec/No
• Shows the percentage of samples at different received Ec/No level for whole
drive data

Measured Ec/No RSCP distribution

100 100
25.0 % 100%
90 90
95%
90% 80 80
85%
20.0 % 80% 70 70
75%
70% 60 60

% cdf
%pdf
65%
15.0 % 60% 50 50
55%
40 40
50%
45% 30 30
10.0 % 40%
35% 20 20
30%
25% 10 10
5.0 % 20%
15% 0 -32 0
-37
-42
-47
-52
-57
-62
-67
-72
-77
-82
-87
-92
-97
-102
-107
-112
-117
10%
5% RSCP in dBm
0.0 % 0%
First drive pdf(%) Second drive pdf(%)
-16 -15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 Third Drive pdf (%)(if any) First drive cdf (%)
Second drive cdf (%) Third Drive cdf (%)(if any)
 Data Processing & Analysis
• Distribution of RSCP and Ec/No
• Distribution of RSCP and Ec/No for Problem Area: Dotted distribution of
measured CPICH RSCP and Ec/No profiles for problem area.
• This plot is very useful to compare the network improvement achieved in
2nd drive test with that of 1st drive test.
A5-1 Distribution of Best Server RSCP and Ec/No (First Drive)

-30

-25

12.61% 14.51%

-20

-15
Ec/No (dB)

-10
-9.7

-5

71.55% 1.34%
0

-92
5
-30 -40 -50 -60 -70 -80 -90 -100 -110 -120 -130

RSCP (dBm)
 Data Processing & Analysis : AMR call
• Call set-up failure ratio
• This includes RRC connection set-up failure and RAB connection set-up failure .
• The ratio is calculated according to the whole drive route as a reference purpose
only.It is not the indication of a cell-based performance.

• Call drop ratio

• This includes RRC connection drop and RAB connection drop.
• The ratio is obtained from the data of the whole drive route , not the cell-based
performance indication and thus shall be treated like nothing but a reference.
 Data Processing & Analysis : AMR call
• Distribution of MOS evaluation
• MOS (mean of score) is collected along the test route with the 3rd party software
at the UE/PC end.
• Score1:Bad; Score2:Poor; Score3:Fair; Score4:Good; Score5: Excellent.
• These values express how close the voice quality is to real-life communication.
• The MOS distribution is the collection of portion of scores along the whole drive
test route.

MOS distribution
1%
2%

21% 12%

64%

1:Bad 2:Poor 3:Fair 4:Good 5:Excellent

 Data Processing & Analysis : AMR call
• BLER
• Block error ratio:BLER is measured from the CRC – Transport Block (TB) is a
basic unit that is exchanged between Layer 1 and Layer 2 (MAC). Layer 1 adds
CRC for each TB and BLER is calculated from the CRC.
• The BLER is plotted along the whole drive test route as a sort of quality
indication for the whole route and can be an aid to other problem events’
analysis.
• BLER performance will also be expressed as a statistical figure which mean
value is shown in it.
Mean
Frequency
10.00%

9.00%

8.00%

7.00%
Freq of BLER

6.00%

5.00%

4.00%

3.00%

2.00%

1.00%

0.00%
0.00% 0.25% 0.50% 0.75% 1.00% 1.25% 1.50% 1.75%

BLER
 Data Processing & Analysis : AMR call
• Call drop point indicated on map
• Showing the problem events on the map for a clear indication and to
facilitate the problem analysis.
 Data Processing & Analysis : PS call
• Throughput cumulative distribution
• For the PS call session test , throughput is used as an important index for the
session quality.And the distribution can be used to be a reference for the whole
drive test route.
Measured Throughput

14.00% 100%
95%
90%
12.00% 85%
80%
75%
10.00% 70%
65%
60%
8.00%
55%
50%
45%
6.00%
40%
35%
4.00% 30%
25%
20%
2.00% 15%
10%
5%
0.00% 0%
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900

Example of HSDPA throughput

 Data Processing & Analysis : PS call
• Average download time
• The PS call is tested with the help of a FTP server locates at the core
network.The basic idea is to download a certain content from the FTP server.The
content is to be kept uniform for the whole drive test.Thus we could have a
download time for every session.The timing issue can be used as an aid to the PS
session quality evaluation.
• Call set-up failure ratio
• The proportion of the session set-up failure along the whole drive route.Again , it
is simply for a reference purpose only(or bench-marking).
• Call drop ratio
• The ratio of the session drops along the test route.
• FTP throughput
• The data rate in the transmission pipe. The better the transmission media quality ,
the higher the throughput.
• BLER map and statistical distribution
• Plot the PS call BLER along the drive test route.
• Indication of the quality and a good aid to the problem event analysis.
• BLER performance will also be expressed as a statistical figure which mean value
is shown in it.
• Throughput map and statistical distribution
• Plot the throughput collected along the drive route and the statistical distribution
chart together with the BLER map can provide good assistance to the problem
event analysis.
 Data Processing & Analysis : UDI call
• Throughput cumulative distribution
• For the UDI call session test , throughput is also used as an important index for
the session quality. And the distribution can be used to be a reference for the
whole drive test route.
 Data Processing & Analysis : UDI call
• Average download time
• The UDI call is tested with the help of a server locates at the core network.The
basic idea is to download a certain content(probably video streaming) from the
server.The content is to be kept uniform for the whole drive test.Thus we could
have a download time for every session.The timing issue can be used as an aid to
the PS session quality evaluation.
• Call set-up failure ratio
• The proportion of the session set-up failure along the whole drive route.Again , it
is simply for a reference purpose only(or bench-marking).
• Call drop ratio
• The ratio of the session drops along the test route.
• BLER map and statistical distribution
• Plot the UDI call BLER along the drive test route.
• Indication of the quality and a good aid to the problem event analysis.
• BLER performance will also be expressed as a statistical figure which mean value
is shown in it.
• Throughput map and statistical distribution
• Plot the throughput collected along the drive route and together with the BLER
map can provide good assistance to the problem event analysis.
• Throughput performance will also be expressed as a statistical figure for a
reference.
 Data Processing & Analysis

• Additional materials required for Data Analysis :

• Plot of Each Scrambling Code RSCP from cluster tuning result

• Plot of Each Scrambling Code Ec/No from cluster tuning result
• Graph of RSCP and Ec/No in problem area if the cause of the problem is poor
coverage of RSCP or Ec/No
• Real Network’s RAN parameter settings including Neighbour list of serving cells
• Signaling message logs to help tracing the cause of problem
 Data Processing & Analysis
• Graph of RSCP and Ec/No in problem area
• Special process of making “Graph of RSCP and Ec/No vs Data No, in Problem
Area”.
• The only data included in each problem area polygon are exported by using
MapInfo function after Problem Area map is made.
• This graph is very useful to analyze the problem area and to find a draft solution
• RSSI is also plotted together in this graph

Plot of RSCP and Ec/No vs Data No. in problem Area

0 0
1 002
Data No,
-5
-20
1200 -10

1600 Exporting -40 -15

1 098

RSCP and RSSI (dBm)

and -60
-20

Ec/No (dB)
Problem 1400 Making -25

Area 1 008 graph

-80
-30

-100 -35

-40
-120
1300 -45

1 037
-140 -50
1200

1300
1400

1600
Data No.

ANT1_RSSI SC97 RSCP SC106 RSCP SC103 RSCP SC69 RSCP

SC174 RSCP SC180 RSCP SC97 Ec/No SC106 Ec/No SC103 Ec/No
SC69 Ec/No SC174 Ec/No SC180 Ec/No
 Field Data Analysis Process
• Cluster /Area tuning is event-based optimization so that if there
is problematic event, the cause of the event should be traced
and solved according to the following optimization procedures :

 Call Setup Failure Optimization

 Call Drop Optimization
 BLER Optimization
 MOS Optimization
 Throughput Optimization
 Neighbour list Optimization
 CPICH RSCP and Ec/No Coverage Optimization
 ISHO Optimization

• All of these optimizations will be discussed in Optimization

Process & Procedures
 Field Data Analysis Process
• There are two problem categories: Problem Area and Problem
Event.
• Problem Event
• There are problematic cases like Call Setup failure or Call Drops
during the drive test and the reasons to this might be different
from case to case so the optimisation methods applied to this
would be different accordingly.Thus it requires detailed analysis
on each happening and the concept is termed as “problem
event”. Example: Call set-up failure, call drop.
• Problem Area
• For some performance deteriorations, they are not coming as an
outstanding case but rather as a form of an “area” which are
inter-related. Thus we need to analyse it as a whole. This concept
is termed as “problem area”. Example:CPICH RSCP, CPICH
Ec/No, BLER , MOS, Throughput
 Field Data Analysis Process
• UE is different from Anritsu RF Scanner in measurement:
Anritsu Scanner measures all SC’s signals but UE only
measures the signals of cells that the system informs through
BCH (neighbour list)
• Therefore if an area shows poor CPICH RSCP coverage, it
might be real poor coverage problem but there is possibility
that there is a site covering the area with high enough RSCP
level but the site is missing in the neighbour list of serving cell.
• This phenomenon can mislead the analyst to propose incorrect
solution or even cause serious side effect to other area so that
we have to use Best server’s CPICH RSCP and Ec/No plots
and Each SC’s CPICH RSCP and Ec/No plots from Cluster
Tuning to see if there is a dominant cell, which is missing in the
neighbour list of serving cell.
 Table of content
 Introduction :
 Tuning Flow Diagram
 Responsability
 Preparation & Drivetest
 Analysis Process
 Reports
Procedure of Reporting to Nokia RF / NWP

Prepare Report
according to defined
format

Submit Draft to Nokia

by Email

No
Visit customer
Agree? with Partner and
Submit the tuning
Yes report

Submit report to
Customer
 Structure of Report Format
•Cluster ID, map, & site information in the cluster
•How does the cluster look like?
•Transmitting sites in neighbor cluster
Cluster •Total length of drive route
Information •Comments on anything special about environment
Introduction and measurement condition
of Cluster Measurement •Model name, serial no.
Area Tool Information •Measurement mode, option, and data proc. method

•Measurement raw data name

Contents of Plots
•Plots of drive test route
•Plots of Best CPICH RSCP, Ec/No, no. of SC, SC,
Measurement best CPICH SC, RSSI, and distribution profile of
best RSCP
Histograms •Attachment for verification of solution
•Histogram & CDF of CPICH RSCP and Ec/No
•Histogram of no. of cells per location
Measurement List of
Analysis problematic •Describe problematic areas on drive route in map
Areas
Solution Report •Problematic area ID, cluster ID, and site list in the
area
•Description of problems, solutions, and verification
Appendix & of solution
Site Info •Verification should be done in both theoretical
Attachments method and simulation.
Two different solutions for a problem area
 Table of Deliverables (1/2)
Item Detail
RF Performance Analysis (note: Problem Area Map
each UE will have own CPICH Best Server CPICH RSCP
RSCP and Ec/No data for the Best Server CPICH Ec/No
analysis) Cumulative Distribution of RSCP
Cumulative Distribution of Ec/No
Distribution of RSCP and Ec/No
Voice AMR

Call Set-up Failure Ratio

Call Drop Ratio
HO Ratio
AMR Measurement Distribution of MOS Evaluation
BLER
MOS (need to check how to score MOS)
Call Drop Point Analysis Call drop point indicated on map
PS

Throughput Cumulative Distribution

Average Download time
Call Set-up Failure Ratio
Call Drop Ratio
PS Measurement FTP Throughput
BLER
Coverage Map Throughput
 Table of Deliverables (2/2)

CS UDI

UDI Measurement Throughput Cumulative Distribution

Average Download time
Call Set-up Failure Ratio
Call Drop Ratio
Coverage Map BLER
Throughput
 Solution Report Form (1/2)
Site list involved
in Problem Area

Description of
surroundings of
problem area

Cause of
problem area

Plot of measured data of problem area, which

is highlighted by circle or polygon.
If the problem area has more than one problem
at the same time, all plots related to the problems
should be plotted together.
 Solution Report Form (2/2)
At least Two
different tuning
solutions are
needed
Tuning solution

Describing the
expected
Good and adverse
Effects when the
above
method is applied
Suggestion is made Explanation about the improvement when the solution is
here when the applied.
problems To verify the improvement, basically NetActPlanner
could not be solved simulation is
referred and sometimes theoretical calculation is also
needed.
 Important Concerns
• How is the polygon is defined
• Executive summary should be there and match with the
adjective
• Problem could not be solved should be analyzed and
explained in the report
• Action required toward the failure , Additional drive test,
optimize during the Area tuning or re try with alternative
solution
• Does the result match with our expectation
• Clear relationship between the overall performance
improvement and the improvement in each problem area
should be established
• In the analysis, the appendix, figure used in analysis work
should indicated in the explanation part.
Change Request Form (1/2)
Get approval from customer on the proposed solution
 Change Request Form (2/2)
•In AREA tuning, we recommend to change RAN parameter as well according
the cause of the problems

RAN Parameters Modification

Problem BTS Name, ID Sector Cell ID Ran Parameter to Change contents
Even ID change
A1-1 JPBTS, 1001 A 8008 Window_Addition From 4dB to 3dB
Implementation Change Procedure