DESIGN AND IMPLEMENTATION OF A TOOLKIT FOR THE PROCESSING OF FILES ISSUED FROM RADIO MEASUREMENTS AND THE AUTOMATIC ALLOCATION OF CELL Ids TO NEW

List of figures and tables List of figures and tables6 Design and Implementation of a Toolkit for the Processing of files issued from Radio Measurements and the Automatic Allocation of

Trang 1

DEPARTEMENT DE GENIES ELECTRIQUE

ET TELECOMMUNICATIONS

DEPARTEMENT OF ELECTRICAL AND TELECOMMUNICATIONS

DESIGN AND IMPLEMENTATION OF A TOOLKIT

FOR THE PROCESSING OF FILES ISSUED FROM

RADIO MEASUREMENTS AND THE AUTOMATIC

ALLOCATION OF CELL Ids TO NEW CELLS

End of Course dissertation written by

CHEFUH DIVINE NGWA

In partial fulfilment of the requirements for a

MASTERS DEGREE IN ENGINEERING SCIENCES

In Telecommunications Engineering

Under the Supervision of

Defended before the Jury composed of:

President: Pr TIEDEU Alain Professor (E.N.S.P)

Superviser : Pr TONYE Emmanuel Professor (E.N.S.P)

Members : Dr Bell BITJOKA Georges Senior Lecturer (E.N.S.P)

Dr NGOHE Paul Salomon Senior Lecturer (E.N.S.P)

Mr MASSOGUE Vincent Radio Engineer ( Orange

Cameroun SA)

2010-2011 Academic year Defended on the 25 th of June 2011

Trang 2

Dedication

2

Design and Implementation of a Toolkit for the Processing of files issued from Radio Measurements and the Automatic Allocation of Cell Ids to New Cells

End of course Dissertation written by: CHEFUH DIVINE NGWA – June 2011

DEDICATION

To my lord JESUS CHRIST

I thank you lord for your constant intervention in all aspects of my life I thank you especially

for your help during the period of my end of course dissertation Father you gave me all the

strength and support I needed You made available to me all the resources I asked for, be it

human or material You nourished my spirit and that gave me the strength and courage to

breakthrough all difficulties Lord, yours I am and yours I want to remain Do with me

whatever you wish

To my loving parents

My father CHEFUH James MUYAH and mother CHEFUH Patricia LUM I thank you

very much for your love and care I know you have always wanted to give me the best in all

situations May the almighty reward all your efforts and grant your greatest desires

Trang 3

ACKNOWLEDGEMENTS

Thanks to everyone who assisted me to realise this work

They include:

My Academic Supervisor, Pr Emmanuel TONYE for all the advice and

encouragement during this internship and during my stay at ENSP Also, to all the

other Lecturers of ENSP

Pr Alain TIEDEU, for accepting to preside over my jury

Dr Bell BITJOKA Georges of Electrical and Telecommunications department, for

accepting to examine my work

Dr Paul Salomon NGOHE of the department of Mathematics and Physical sciences,

for accepting to examine my work

My professional Supervisor, Mr Vincent MASSOGUE (radio engineer OCM) for

all the help and contributions he made to make this work better

The chief of service of the Littoral-West radio department, Mrs NGO BIBOUM

Clémence for her constant encouragements and motherly care

The Littoral-West radio engineers of Orange CAMEROUN SA, Mr Anicet

KEMAYOU, Mr André YOKO, Mr Yves NYEMB, Mr Charly YAMB, all the

technicians Mr AMADIANG Charly, Mr Eric MINE, Mr Gerald HIOM and

Mr Christian MEKOUNDI for accepting me in their mist

The NSS Engineering department especially Mr Freddy OYONO for the training I

received from them during my 2010 internship in Orange Cameroun SA

All my classmates of GTEL promo 2011

AYUNI SENGUM FAI, for being my friend, study mate…, for everything

All my friends who have continually encouraged me, especially NJUMBE Divine

All my Aunties and Uncles for the wonderful support they gave me

My Brothers and sisters BISI, EMI, VAL, VEVE, NENE and JAY-JAY

My grandparents and all other relatives

Trang 4

List of abbreviations

4

LIST OF ABBREVIATIONS

BSC Base Station Controller

BSSAP Base Station Subsystem Application Part

BTS Base Transceiver Station

CEPT Conference of European Posts and Telegraphs

CGI Cell Global Identity

DCS Digital Cellular System

EIR Equipment Identity Register

ETSI European Telecommunications Standards Institute

IMEI International Mobile Equipment Identity

IMSI International Mobile Subscriber Identity

ISDN Integrated Services Digital Network

LAI Location Area Identification

MSC Mobile Station Controller

MSIN Mobile Station Identification Number

MSISDN Mobile Station ISDN Number

MSRN Mobile Station Roaming Number

OMC Operations and Maintenance Centre

SIM Subscriber Identification Module

TDMA Time Division Multiplexing

TEMS Test Mobile System

TMSI Temporal Mobile Subscriber Identity

TRAU Transcoder and Rate Adaptation Unit

VLR Visitor Location Register

VSAT Very Small Aperture Terminal

WIMAX Worldwide Interoperability for Microwave Access

Trang 5

In a cellular network, a good frequency

plan is a fundamental factor to reduce

interferences Several applications have

been designed to assist in the realisation of

frequency plans, like the software ATOLL

This software relies on prediction

calculations to characterise interferences

between cells Prediction calculations have

limitations since they rely on geographical

databases which are not accurate in some

regions Thus, real field data is needed to

complete that which is gotten from

prediction calculations

To supply field data into the frequency

allocation process the approach here is to

use log files issued from drive tests (radio

measurements) to generate interference

matrices which are injected into Frequency

Planning Modules

In this project, a toolkit was designed that

permits the generation of all ATOLL

compatible interference matrices These

matrices were imported into ATOLL for

use in frequency planning, and they also

served as basis for interference analyses

between couples of cells

Another aspect realised in this project is

the automatic allocation of cell identities to

new cells

Dans un réseau cellulaire, un bon plan de fréquence est nécessaire pour la réduction des interférences Plusieurs logiciels aident

à la réalisation des plans de fréquence telle qu’ATOLL Ce logiciel s’appui sur les calculs de prédictions pour caractériser les interférences entre les cellules Ces calculs

de prédictions sont basés sur les données géographiques qui ne sont pas précises dans certaines régions Alors, il est nécessaire de compléter ces données de prédictions avec celles du terrain

Pour fournir les données du terrain au processus d’allocation de fréquences, l’approche consiste à utiliser les fichiers logs issus des mesures radio Afin de générer des matrices d’interférences et les injecter aux Modules de Planification de Fréquences

Dans le cadre de ce mémoire, un outil a été conçu pour permettre la génération des matrices d’interférence compatible à ATOLL Ces matrices ont été importées dans ATOLL ó elles seront utilisées pour

la planification de fréquences, et constitueront les données de base d’analyse d’interférences

Un autre volet de ce travail est l’allocation automatique des « CELL ID » aux

nouvelles cellules

Key Words: Interference matrix, Cell id, Drive tests, Frequency planning, QoS

Trang 6

List of figures and tables List of figures and tables

6

LIST OF FIGURES

Figure 1: Architecture of Generic GSM Network 14

Figure 2: The Orange Cameroon Network (May 2011) 16

Figure 3: Fields of LAI 18

Figure 4: Fields of CGI 18

Figure 5: Other transmitters emitting at the same frequency band as serving transmitter 22

Figure 6: Other transmitters emitting at the adjacent frequency bands as serving transmitter 22

Figure 7: Frequency reuse pattern 23

Figure 8: Inputs into the AFP module of ATOLL 24

Figure 9: Some ATOLL modules 32

Figure 10: Procedure for interference matrix generation 34

Figure 11: Sample im0 file format interference matrix 36

Figure 12: Sample clc file format interference matrix 37

Figure 13: Sample dct file format interference matrix 39

Figure 16: Flowchart for generation of the im0 file format interference matrix 45

Figure 17: Flowchart for generation of the im2 file format interference matrix 46

Figure 18: Correlation analyses show highly reliable drive test based data 49

Figure 19: Correlation analyses show highly reliable drive test based data 50

Figure 20: Factors to consider during the allocation of new CI values 51

Figure 21: Steps to put in place mechanism for CI allocation 53

Figure 22: Sample BSS Configuration files in the directory ACME of the OMC-R 54

Figure 23: Header of a sample BSSConf file (BSSConf.omcrdla1.20110510090500) 54

Figure 24: CELL_SECTION of sample BSSConf file 55

Figure 25: Illustration of mechanism for CI attribution as in OCM 56

Figure 26: Flowchart for automatic allocation of CI values 58

Figure 27: Platform architecture 59

Figure 28: Database conception 60

Figure 29: Final database tables 60

Figure 30: Toolkit arborescence 63

Figure 31: Authentication page 64

Figure 32: Application Home Page 65

Figure 33: Generation of an Interference matrix 66

Figure 34: Sample of im0 interference matrix generated from IM_CI Tool 67

Figure 35: Upload new log file to server 67

Figure 36: Combining multiple log files 68

Figure 37: Loading interference matrix for comparison 69

Figure 38: Example Results of evaluation 69

Figure 39: Interactive interference analyses 70

Figure 40: C/I histograms for server-interferer couple 71

Figure 41: Channel information for frequency analyses 71

Figure 42: Input parameters for CI allocation 72

Figure 43: Cell Ids generated for 3 cells of tri-sectorial site 73

Figure 44: Mailing list 73

Trang 7

LIST OF TABLES

Table 1: GSM key Interfaces and protocols 17

Table 2: logic channels 19

Table 3: Different RxLev ranges 20

Table 4: Correspondences between BER and signal quality (RXQUAL) 21

Table 5: Channel protection ratios 23

Table 6: Comparison with NGAN TAFAM 30

Table 7: Comparison with NDEFO TAKA 31

Table 8: columns definition for im0 files format interference matrices 35

Table 9: Columns definition for clc files format interference matrices 37

Table 10: Columns definition for dct files 38

Table 11: Columns definition for im1 files format interference matrices 40

Table 12:Columns definition for im2 files format interference matrices 41

Table 13: List of log-file information elements for IM generation and interference analyses 44

Table 14: Interference matrices with high correlation 48

Table 15: Interference matrices with low correlation 49

Table 16: CI value ranges for GSM band per zone as used in Orange Cameroon 51

Table 17: Ranges of CI values for DCS band per zone as used in Orange Cameroon 52

Table 18: Example of CI values for site with 3 cells 52

Table 19: Programming Languages 61

Table 20: Programming tools 61

Table 21: Template for exporting TEMS log file 76

Trang 8

Table of Contents Table of Contents

8

1.1.1 General Presentation of GSM Network 13

1.1.2 The GSM Radio Interface _ 19

2.1 State of the art 30

2.2 Generation of Interference Matrices _ 34

2.2.1 Study of Formats of ATOLL Interference Matrices _ 35

2.2.2 Study of Log-files from Drive Tests 42

2.2.3 Cross-compatibility between ATOLL Interference Matrices and TEMS Log-files 43

2.2.4 Transformation of Exported Log-file fields to Interference Matrices _ 44

2.2.5 Evaluation of Generated matrices (Correlation Analyses) _ 47

2.3 Automatic Allocation of Cell Identities _ 51

2.3.1 Obtaining list of existing Cell Ids _ 53

2.3.2 Constraints for site being created 56

2.3.3 New mechanism for Cell Id allocation _ 56

Trang 9

3.1 Results from the application _ 63

Conclusion AND PERSPECTIVES _ 74

Bibliography and Webography _ 75

Annex _ 76

Trang 10

Introduction Introduction

10

INTRODUCTION

One of the major worries of a mobile phone operator is to insure a good quality of service to

its customers This quality of service is principally linked to 2 parameters which are: good

signal coverage and good signal quality But then, interferences arise due to frequency reuse

These interferences degrade the signal quality, and hence cause a drop in the quality of

service To reduce this, it was required to Design and Implement a toolkit for the

processing of files issued from radio measurements and the automatic allocation of Cell

Ids to new cells

This project tackles two aspects, first supplying data issued from radio measurements into the

frequency allocation process This will permit the amelioration of frequency plans and thus

reduce interferences The radio measurement data could also serve as input for an interactive

interference analyses between couples of cells with the goal of detecting and reducing

interferences Reducing interferences will improve on the QoS

Secondly to put in place a mechanism for an automatic allocation of cell ids to new cells

integrated into the operators’ network

This document presents details of all that was done to realise the project

The first chapter of the document, Context and Problem statement, describes relevant

concepts to the project for a clear understanding Here also, the problem at hand is described

with all its details

The second chapter, methodology, is where the approach adopted to resolve the problem is

described clearly

The third chapter, results and comments, shows screen shots of the developed toolkit and

comments on the screen shots

Trang 11

CHAPTER 1: CONTEXT AND PROBLEM

STATEMENT

The context starts with a brief description of the GSM norm: its origins, architecture, main components and the radio interface It also mentions some notions on interferences

It then goes on to give information on frequency planning and drive tests, giving tools used for these in the company Orange Cameroon

The Problem Statement outlines the Project specifications made by Orange Cameroon

Trang 12

Context

12

OUTLINE

CONTEXT

• General Presentation of GSM Network

Origin of GSM and Definition of the Norm The Network

Architecture

Generic Architecture Orange Cameroon Network Interfaces and Protocols

Cell identification in GSM networks

• The GSM Radio Interface

TDMA Structures Logic Channels Radio Interface Performances

• Drive Tests (Radio Measurements)

PROBLEM STATEMENT

• Problem description

• Objectives to be attained

Trang 13

1.1.1 General Presentation of GSM Network

1.1.1.1 Origin of GSM and Definition of the Norm

Origin of GSM:

During the early 1980s, analogue cellular telephone systems were experiencing rapid growth

in Europe, particularly in Scandinavia, the United Kingdom, France and Germany Each

country developed its own system, which was incompatible with everyone else's in

equipment and operation This was undesirable; not only was the mobile equipment limited

to operation within national boundaries, but there was also a very limited market for each

type of equipment

The Europeans realized this early on, and in 1982 the Conference of European Posts and

Telegraphs (CEPT) formed a study group called the Groupe Spécial Mobile (GSM) to study

and develop a pan-European public land mobile system The proposed system had to meet

certain criteria, among which were:

• good subjective speech quality,

• low terminal and service cost,

• support for international roaming,

• ability to support handheld terminals,

• support for range of new services and facilities,

• spectral efficiency, and

• ISDN compatibility

Definition of GSM Norm:

GSM (Global System for Mobile Communication) is the first cell phone norm to be made of

an entirely digital system This norm has two bands, each with a bandwidth of 25MHz: the

890-915 MHz band for the transmission from mobile stations to the network and

916-930MHz Band for transmission from the network to mobile stations

This norm also has a few variants, such as the DCS 1800 (Digital Cellular System), DCS

1900 and PCS 1900 (Personal Communication System) These are identical to the GSM

norm, and differ only in terms of the frequencies used in their exploitation A network can

conveniently work with some of its equipment emitting and receiving on the traditional GSM

Trang 14

Design and Implementation of a Toolkit for the Processing of files issued from Radio Measurements and the Automatic Allocatio

End of course Dissertation written by: CHEFUH DIVINE NGWA

band and other equipment using the

equipments inter-functioning with each other

1.1.1.1 The Network

Architecture

Generic Architecture

Apart from connecting subscribers to a network, the GSM, like all other mobile norms has

two other major challenges:

• It must permit subscribers to be able to continue communication while moving around

without having any rupture or break in communication, this is the concept of

HANDOVER

• It must also permit the subscribers to be able to call or receive calls and all other services

offered by the network from any point within the area covered by the network: this is the

concept of ROAMING

Figure 1: Architecture of Generic GSM Network

As seen from the above diagram, the GSM network can be split into four main entities:

Mobile Station (MS), Base Station Subsystem (BSS), Network Subsystem (NSS) and

band and other equipment using the bands of one or more of its variants, with these different

functioning with each other

The Network

Architecture

Generic Architecture

It must permit subscribers to be able to continue communication while moving around

It must also permit the subscribers to be able to call or receive calls and all other services

f Generic GSM Network

bands of one or more of its variants, with these different

It must permit subscribers to be able to continue communication while moving around

It must also permit the subscribers to be able to call or receive calls and all other services

Trang 15

included in the BSS or considered as a separate entity

• The MS (Mobile Station):

This is the mobile part of the network An MS is any physical equipment capable of

communicating with the network It is characterised by two main entities:

• The International Mobile Equipment Identity (IMEI) number, which is put in the memory

of the phone during its fabrication

• The SIM (Subscriber Identification Module) card: This is an electronic card containing

personal information of the subscriber (serial number, Pin code, PUK code, received

SMS, directory of phone numbers, IMSI, TMSI, etc) It provides personal mobility, so

that the subscriber can have access to all subscribed services regardless of the terminal

being used or location of that terminal

• The BSS (Base Station Subsystem):

It is also known as access network It assures radio-electric transmissions and manages the

radio resources The BSS is comprised of the BTS (Base Transceiver Station) and the BSC

(Base Station Controller)

• The NSS (Network Subsystem):

The NSS is the part of the GSM system that assures establishment of calls and mobility It

contains the switches and network databases It is made up of the following entities: the HLR,

VLR, AUC, and MSC

• OSS (Operation Subsystem):

This permits the network operator to administer his network The Operation and Maintenance

Centre (OMC) of the OSS is made of two parts: OMC_S (Operation and Maintenance Centre

Switching part) which supervises detects and corrects abnormalities of the NSS.OMC-R

(Operations and Maintenance Radio part) The OMC-R exploits and maintains the radio

sub-system

Trang 16

Orange Cameroon Network

Figure

Interfaces and Protocols

The table below gives a summary of key

protocols used

DOUALA YAOUNDE WEST/NORTH W LITTORAL/SOUTH W

CENTRE/EAST NORTH/ADAMAOUA/ EXTREME NOR

Orange Cameroon Network

Figure 2: The Orange Cameroon Network (May 2011)

and Protocols:

The table below gives a summary of key interfaces in GSM and their descriptions plus

More than 730 GSM sites

Number of GSM sites Number of

interfaces in GSM and their descriptions plus

More than 730 GSM sites

01 Mobile BTS

46 WIMAX Sites

30 VSAT

02 Principal technical Platforms

04 Technique Centres Yaoundé-Garoua-Bafoussam) One High Bit rate Transmission loop (Douala- Yaoundé -Bafoussam)

(Douala-Number of WIMAX sites

Trang 17

Name Position Role Protocol Support

- Voice at 16 Kbits/s per communication

- signalling for traffic management

- signalling for BTS maintenance and exploitation

LAPD 2 Mbits/s links

Ater BSC – TC Transport of :

-Voice at 16 Kbits/s used for communication

-on some equipment signalling for management of the TRAU

(SS7 basic) + BSSAP (BSSAP = BSSMAP+DTA P)

2 Mbits/s links

A TC - MSC Transport of :

-Voice at 64 Kbits/ used for communication

(SS7 basic) + BSSAP

2 Mbits/s links

B MSC - VLR Signalling for Mobile Application Part

(MAP)

(SS7 basic) + MAP

2 Mbits/s links

C MSC - HLR Signalling for Mobile Application Part

(MAP)

(SS7 basic) + MAP

2 Mbits/s links

D VLR - HLR Signalling for Mobile Application Part

(MAP)

(SS7 basic) + MAP

2 Mbits/s links

E MSC-MSC Transport of :

- Signalling for MAP

- Voiceat 16 Kbits/s used for communication

(SS7 basic) + MAP

2 Mbits/s links

F MSC - EIR Signalling for Mobile Application Part

(MAP)

(SS7 basic) + MAP

2 Mbits/s links

G VLR - VLR Signalling for Mobile Application Part

(MAP)

(SS7 basic) + MAP

2 Mbits/s links

Trang 18

Context

18

1.1.1.2 Cell Identification in GSM Networks

LAI (Location Area Identity)

A GSM network is divided into Location Areas Each location area is a collection of cells A

location area is identified by a Location Area Code (LAC) The LAC is freely attributed by

the network operator

CGI (Cell Global Identity)

Within a location area, a cell is identified by the Cell Global Identity (CGI) This is obtained

by doing a concatenation of the location area identity and a unique cell identifier called the

Cell Identity (CI) I.e CGI = LAI + CI

In Orange Cameroon, all cells of the network have a unique Cell Identity value

Figure 3: Fields of LAI

Figure 4: Fields of CGI

Trang 19

1.1.2.1 TDMA Structures

Each of the frequency channels in GSM is divided into timeslots of approximately 577µs

duration These timeslots are grouped together in sets of 8 consecutive timeslots as one

frame These frames are then grouped together in one of two ways as multiframes:

A 26-frame multiframe with duration of 120ms, comprising 26 TDMA frames; this

multiframe is used to carry traffic channels and their associated control channels

A 51-multiframe with duration of approximately 235.4ms, comprising 51 TDMA frames; this

multiframe is exclusively for control channels

1.1.2.2 Logic Channels

These are a set of slots in a multiframe permitting the transportation of either control or

signalling information with a given periodicity

Logic channels have the advantages that:-

• They permit better usage of radio resources,

• Limit scrutinizing efforts of equipments

Logic channels can be classified into 2 big classes: dedicated channels, which are generally

duplex channels and non dedicated channels which are generally simplex channels The table

below shows a list of all major logic channels and their primary functions:

Table 2: logic channels

Broadcast Channels (BCH),

Downlink

Frequency Correction Channel (FCCH)

Blocking on carrier frequency

Synchronisation Channel (SCH)

Synchronisation and Identification Broadcast control channel

(BCCH)

Broadcast of System information

Common Control Channels

(CCCH), Uplink or

Downlink

Paging Channel (PCH) (downlink)

Call of mobile

Random Access Channel (RACH) (uplink)

Random access of mobile

Access Grant Channel (AGCH) (downlink)

Allocation of resources

Cell Broadcast Channel (CBCH)

Diffusion of short messages

Dedicated Control Channels,

Uplink and Downlink

Stand-Alone Dedicated Control Channel (SDCCH) (uplink and downlink)

Signalling

Slow Associated Control Channel (SACCH) (uplink and down link)

Link supervision

Trang 20

Context

20

Fast Associated Control Channel (FACCH) (uplink and downlink)

Handover execution

Traffic Channels (TCH) Traffic Channel for coded

Speech (TCH/FS and TCH/HS) (uplink and downlink)

Full rate and Half rate voice

Traffic Channel for data (user rate), 9.6kbit/s, 4.8kbit/s, less than 2.4kbit/s

User data

1.1.2.3 Radio Interface Performances (Notions of RXLEV and

RXQUAL)

To appreciate the quality of a radio link, there are 2 major parameters that are used These

parameters are the RXLEV and the RXQUAL They are both measured at the level of the

BTS and at the level of the mobile to appreciate both the uplink and downlink respectively

RXLEV:

This is the signal strength received at the mobile from the BTS or inversely

It is measured over 64 different levels, from 0 to 63 These 64 levels correspond to signal

strengths ranging from -110dBm to -47dBm, with steps of 1dBm The table below

illustrates this

Table 3: Different RxLev ranges

RXLEV Deep Indoor -47 dBm to -64 dBm

This parameter is used to measure the quality of the received signal It is obtained by doing a

quantisation of the Bit Error Rate (BER) according to the correspondence defined by the table

below (3GPP TS 45.008, section 8.2.4)

Trang 21

Table 4: Correspondences between BER and signal quality (RXQUAL)

RxQual BER, Ranges Representative Values Interpretation

3 0.8% ≤ BER ≤1.6% 1,13% Average quality

4 1.6% ≤ BER ≤ 3.2% 2,26% Average quality

6 6.4% ≤ BER ≤ 12.8% 9,05% Bad quality

1.1.3 Notions on Interferences

1.1.3.1 Definition

Interference is anything that alters, modifies or disrupts a signal as it travels along a channel

between a source and a receiver

In radio-mobile networks, there are 2 major kinds of interferences:

• Co-channel Interference

• Adjacent channel interference

Co-channel Interference:

It occurs when for the transmitter to which the mobile is accorded (called the serving

transmitter), emitting in a defined frequency band, there are other transmitters emitting in the

same frequency band such that:

C/I = Carrier over Interference ratio,

C = Signal strength of serving transmitter,

Ij = Signal Strength of number j interferer

Trang 22

Figure 5: Other transmitters emitting at the same frequency band as serving transmitter

Adjacent channel Interference

transmitter), emitting in a defined frequency band, there are other transmitters emitting in

adjacent frequency bands such that:

Ij = Signal Strength of number j interferer

Figure 6: Other transmitters emitting at the adjacent frequency bands as serving transmitter

: Other transmitters emitting at the same frequency band as serving transmitter

Adjacent channel Interference:

adjacent frequency bands such that:

9

= Signal Strength of number j interferer

: Other transmitters emitting at the adjacent frequency bands as serving transmitter

: Other transmitters emitting at the same frequency band as serving transmitter

(2)

: Other transmitters emitting at the adjacent frequency bands as serving transmitter

Trang 23

The GSM norm (ETSI 05.05, ETSI = European Telecommunications Standards Institute)

defines protection ratios for a channel given by the following table:

An operator fixes a quality margin for practical uses This quality margin is added to the

co-channel protection ratio to get the practical margin (as in the table above) The value for the

quality margin is 3dB for Orange Cameroon The Adjacent Channel Suppression (ACS)

ratio is defined as I a / I c

1.1.4 Frequency Planning

1.1.4.1 Frequency Reuse Pattern

In the GSM network like in other radio-mobile networks, there are predefined frequency

bands for usage An operator cannot acquire the entire band Only a fixed number of channels

are attributed to the operator by the Telecommunications Regulatory Board For example

Orange Cameroon has 40 channels (ARFCN 85 to ARFCN 124) in the GSM 900 band and 60

channels (ARFCN 512 to ARFCN 571) in the DCS 1800 band

These limitations in frequency imply that the operator must put in place a frequency plan

(define a reuse pattern) that will take into consideration co-channel and adjacent-channel

interferences

Figure 7: Frequency reuse pattern.

Practical 12dB

2 nd Adjacent channel C/Ia2 -41dB 50dB -38dB

3 rd Adjacent channel C/Ia3 -49dB 58dB -46dB

Table 5: Channel protection ratios

Trang 24

Context

24

D is called the reuse distance It is calculated as:

Where R is the radius of a cell and N the number of cells in a cluster (in the above example

N=12)

1.1.4.2 Frequency Planning in OCM (use of ATOLL)

The frequency plan needs to be constantly ameliorated to take into account newly integrated

network cells This is the optimisation of the frequency plan

There exist several tools that permit the design and optimisation of frequency plans Among

these is ATOLL which is used in Orange Cameroon

Using ATOLL, there are 3 methods of frequency attribution to cells:

• Assigning frequencies Manually,

• Automatic Frequency Planning (AFP),

• Interactive Frequency Planning (IFP)

Among these frequency assignment methods, those considered here are the AFP and IFP

The Automatic frequency planning module of ATOLL has multiple input parameters as

shown on the figure below:

Interference

Matrices

Current Frequency Plan

Network Information

Separation

Constraints

Other Constraints

Automatic Frequency Planning (AFP)

Cell Traffic Input

Figure 8: Inputs into the AFP module of ATOLL

Trang 25

Among the multiple inputs into the AFP module of ATOLL is the interference matrix

Definition: In Atoll, the interference matrix is an array that stores the probability of

interference between a couple of cells (a victim cell and an interfering cell)

1.1.5 Drive Tests (Radio Measurements)

Drive testing is a method of measuring and assessing the coverage, capacity and Quality of

Service (QoS) of a mobile radio network

The technique consists of using a motor vehicle containing mobile radio network air

interface measurement equipment that can detect and record a wide variety of the physical

and virtual parameters of mobile cellular service in a given geographical area

By measuring what a wireless network subscriber would experience in any specific area,

wireless carriers can make directed changes to their networks that provide better coverage

and service to their customers

Drive testing requires a mobile vehicle outfitted with drive testing measurement equipment

These equipments are usually highly specialized electronic devices that interface to mobile

handsets This ensures measurements are realistic and comparable to actual user experiences

Drive test equipment typically collects data relating to the network itself, services running on

the network such as voice or data services, radio frequency scanner information and GPS

information to provide location logging

The dataset collected during drive testing field measurements can include information such

as: signal intensity, signal quality, interference, dropped calls, blocked calls, anomalous

events, call statistics, service level statistics, QoS information, handover information,

neighbouring cell information and GPS location co-ordinates

Drive tests are carried out in Orange Cameroon using the software TEMS Investigation that

will be described briefly in the next chapter of this document

Trang 26

Problem Statement

26

1.2.1 Problem description

The problem faced by the company Orange Cameroun had two major axes

• On one hand, to use field data gotten from drive tests for the generation of

interference matrices that will serve:

In the supply of AFP (Automatic Frequency Planning) and IFP (Interactive

Frequency Planning) modules of ATOLL (radio frequency design tool); these ATOLL modules will use the matrices combined with others issued from prediction calculations to produce better frequency plans This in view of minimising co-channel and adjacent channel interferences between transmitters in the network

As data that will be used to do an interactive interference analyses between

couples of cells (victim cell and interfering cell) This in view of determining

if a given cell is suffering from co-channel or adjacent-channel interferences

If it’s the case, to determine exactly which cells are interfering

• On the other hand, to put in place a mechanism for allocation of cell identities to

new cells integrated into the network The access network of Orange Cameroon is

in a rapid extension phase and CI values are limited (to 65535) Thus the proposed

mechanism should be:

Automatic

Centralised and

Not wasteful

1.2.2 Objectives to be attained

Based on the problem posed, some objectives were set The major objective was to design

and implement a toolkit whose functionalities include the following:

• Obtain log files gotten from the tool TEMS Investigation (a tool used to conduct drive

tests) These log files contain information that describes the state of the network

(precisely for the area covered during the drive tests)

• Use these log files to generate ATOLL compatible interference matrices

Trang 27

matrix

• Use the interference matrices in an in-built module of the toolkit to perform an

interactive interference analyses

• Automatically attribute cell identities to new cells

Trang 28

Methodology

28

Finally, a description of the designing of the toolkit is done

Trang 29

• State of the Art

• Generation of Interference Matrices

Study of Formats of ATOLL Interference Matrices Study of Log-files from Drive Tests (TEMS Investigations) Cross-compatibility between ATOLL Interference Matrices and TEMS Log-files

Transformation of Exported Log-file fields to Interference Matrices Evaluation of Generated matrices (Correlation Analyses)

• Automatic Allocation of Cell Identities

Obtaining list of existing Cell Ids Constraints for site being created New mechanism for Cell Id allocation Notification

• Toolkit Designing

Toolkit Architecture Database Modelling Programming languages and tools used

Trang 30

Methodology

30

2.1 State of the art

A similar project was undertaken in the same company Orange Cameroon This was

undertaken by NGAN TAFAM Aaron Achille on the occasion of his end of course

dissertation to obtain a Masters degree in Telecommunications Engineering (year 2009) at the

National Advanced School of Engineering Yaoundé Cameroon The theme of the project

was: “l’interopérabilité des outils RNO et Atoll”

Among other project specifications, the student had to:

- Permit the production of frequency plans on Atoll by using interference matrices

issued from Radio Measurement Statistics (RMS) The RMS were gotten from a tool

called Radio Network Optimisation (RNO) used then for network optimisation (N.B

as of the time of writing of this document, a more evolved version of RNO called

NPO was in use)

The table below compares the 2 projects in terms of their inputs and outputs:

Table 6: Comparison with NGAN TAFAM

Taken from the tool TEMS Investigation as log files

These statistics already contain C/I threshold values grouped in different ranges

Statistics not yet done Log files contain signal levels which are used to get C/I

Output Data

Only one Atoll compatible interference matrix format was generated, the CLC format

Four output formats required:

IM0, IM1, IM2, CLC

We notice from table 6 above that the two projects have different input data As such,

although the required output is Atoll compatible interference matrices for both cases, the

methods of approach differ

Trang 31

Another similar project was done by William NDEFO TAKA on the occasion of his end of

course dissertation to obtain a Masters degree in Telecommunications Engineering (year

2005) at the National Advanced School of Engineering Yaoundé Cameroon The theme of the

project was: “Résolution du problème d’interférence par la mise en oeuvre d’un outil de

gestion de fréquence CAS D’ORANGE CAMEROUN”

Among other project specifications, the student was expected to:

- Generate interference matrices with data issued from drive tests

- Use these interference matrices to detect co-channel and adjacent-channel interference

problems

- Propose frequency modifications to be made in other to reduce these interference

problems

The table below compares the 2 projects in terms of their inputs and outputs

Table 7: Comparison with NDEFO TAKA

Table 7 above shows us that the 2 projects have different expected results As such, the

methodology employed for each case is different

Description of major tools

Here, a description of the each software that intervenes in the scope of this project is given

They are: ATOLL and TEMS Investigation

ATOLL

It is a powerful radio frequency design platform It is also a multi-technology design platform

(i.e used for several technologies like GSM, CDMA, WIMAX and UMTS) It supports a

wireless operator throughout the network lifecycle, from initial design to densification

Trang 32

ATOLL has several modules The figure below shows some of the key modules in the

software:

ATOLL calculates Prediction based

How does ATOLL calculate its predictio

In ATOLL, predictions calculations are performed in 3 phases:

• ATOLL calculates path loss (L

The propagation model used by Orange Cameroon is the

model proposed by the company SIRADEL The model is calibrated for Cameroon

The formula used for path loss calculation is:

: Path loss

: Transmitter receiver losses calculated through propagation model

: Transmitter antenna attenuation

: Receiver antenna attenuation

• If shadowing is taken into consideration, ATOLL evaluates a

• Cell couverage analyses

• Interference analyses

• Handover analyses

Figure 9: Some ATOLL modules

Prediction based interference matrices

How does ATOLL calculate its prediction based interference matrices?

In ATOLL, predictions calculations are performed in 3 phases:

ATOLL calculates path loss (Lpath) using selected propagation model

The propagation model used by Orange Cameroon is the Volcano Macrocell

proposed by the company SIRADEL The model is calibrated for Cameroon

The formula used for path loss calculation is:

= " "

: Transmitter receiver losses calculated through propagation model

: Transmitter antenna attenuation

: Receiver antenna attenuation

If shadowing is taken into consideration, ATOLL evaluates a shadowing margin

!#$%&'() *$+)'( ∶ *-./01

Service planning and analyses

Cell couverage

Interference

Handover analyses

Interference matrix generation

• Prediction and drivetest based interference matrices supported

• Open format allowing the editing and modifications

of interference matrices using third party applications

Automatic Frequency Planning

• Allocation of FH parameters

• Intergrate with 3rd party AFP tools

n based interference matrices?

) using selected propagation model

Volcano Macrocell This is a

proposed by the company SIRADEL The model is calibrated for Cameroon

2G/3G planning

co-• Site sharing

• Simultaneous display and analyses of 2G and 3G networks

Trang 33

234 − *-./01− /6+ 7 − 38 (5)

Where

234 : Signal level at receiver

/6 : Indoor losses (for indoor coverage considerations)

7 : Receiver antenna gain

38 : Receiver Losses

Thus, the Signal level at the receiver for both the serving and interfering cells can be

calculated These 2 signal levels are used to calculate the C/I

The prediction based interference matrices calculated by ATOLL rely on

- Propagation models

- Digital Terrain Models (DTM)

- Clutter height and class considerations

These prediction calculations have some limitations since the geographical databases (DTM)

are not very precise in some regions, hence the need to complete them with real terrain data

TEMS Investigation

It is an air interface test tool for cellular networks It supports several wireless networks like

GSM and CDMA networks

It has 2 major modules:

- Data Collection

This module interfaces with the mobile phone and other measurement devices like the GPS

It collects data and records them in log files It also does analyses of single log files at a time

- Route Analyses

This module simply permits rapid analyses of multiple log files

Trang 34

The first aspect treated was the generation of ATOLL compatible interference matrices The

steps followed for this are illustrated by

Generation of Interference Matrices

llowed for this are illustrated by the figure below

10: Procedure for interference matrix generation

Study of the log files from drive tests (log files from TEMS Investigation)

Definition of fields from log files needed for generation of interference histograms

fields from log files needed for generation of

Transformation of the fields from the log file to the ATOLL interference histogram proper

Trang 35

There are four different file formats for ATOLL interference matrices These are:

• One histogram per line (.im0) format

• One value per line with dictionary file (.clc) format

• One value per line (transmitter name repeated) (.im1) format

• Only co-channel and adjacent-channel values (.im2) format

One histogram per line (.im0) format

This file contains one histogram per line for each interfered/interfering sub-cell pair The

histogram is a list of C/I values with associated probabilities

The im0 file consists of 2 main parts:

• The first part is a header used for format identification It must start with and contain

the following lines:

• The second part details interference histograms for each interfered sub-cell/interferer

sub-cell pair The lines after the header are considered as comments if they start with

“#” If not, they must have the following format:

The 4 tab-separated columns are defined in the table below:

Table 8: columns definition for im0 files format interference matrices

Column1 Interfered transmitter Name of the interfered transmitter

Column2 Interfering transmitter Name of the interferer transmitter

Column3

Interfered TRX type

Interfered sub-cell In order to save storage, all sub-cells with no power offset are not duplicated (e.g BCCH, TCH)

Trang 36

Methodology

36

Sample

Figure 11: Sample im0 file format interference matrix

One value per line with dictionary file (.clc) format

CLC file

The clc file consists of two main parts:

Trang 37

Table 9: Columns definition for clc files format interference matrices

Column1

Interfered transmitter

Identification number of interfered transmitter If the column is empty, its value

is identical to that of the above line

Column2 Interfering transmitter Identification number of interferer

transmitter If the column is null, its value is identical to that of the above line

Column5

Probability C/I > Threshold

Probability to have C/I the value specified in column 4 (C/I threshold) This field must not

be empty

Sample

Figure 12: Sample clc file format interference matrix

Trang 38

Methodology

38

Note: The columns 1, 2 and 3 must be defined only in the first line of each histogram

DCT File (Dictionary file)

Description

The DCT file is divided into 2 parts:

• The second part provides information about transmitters taken into account in the

AFP The lines after the header are considered as comments if they start with the “#”

symbol If not, they must start with and contain the following lines:

Table 10: Columns definition for dct files

Column1 Transmitter name Text Name of transmitter

Column2 Transmitter identifier Integer Identification number of the transmitter

Column3 BCCH during calculation Integer BCCH used in calculations

Column4 BSIC during calculation Integer BSIC used in calculations

Column5 % of vic’ coverage Float Percentage of overlap of the victim

service area

Column6 % of int’ coverage Float Percentage of overlap of the interferer

service area

The last 4 columns describe the interference matrix scope

Note:

A dictionary file simply associates transmitter names to transmitter identifier for a

given CLC interference matrix

When importing interference matrices with CLC format, you must specify the clc file

to be imported ATOLL looks for the associated dct file in the same directory and

uses it to decode transmitter identifiers If the dct file is unavailable, ATOLL assumes

Trang 39

of the clc file must contain the names of the interfered and interfering transmitter

instead of their identifiers It’s like this that we generate our CLC interference

matrices No associated DCT file is needed

Sample

Figure 13: Sample dct file format interference matrix

One value per line (Transmitter name repeated) (.im1) format

This file contains one C/I threshold and probability pair value per line for each

interfered/interfering sub-cell pair The histogram is a list of C/I values with associated

probabilities

The im1 file consists of 2 main parts:

The 5 tab-separated columns are defined in the table below:

<Column1><tab><Column2><tab><Column3><tab><Column4><tab><Column5><newline

Trang 40

Methodology

40

Table 11: Columns definition for im1 files format interference matrices

Column1 Interfered transmitter Name of the interfered transmitter

Column2 Interfering transmitter Name of the interferer transmitter

Column3

Interfered TRX type

Interfered sub-cell In order to save storage, all sub-cells with no power offset are not duplicated (e.g BCCH, TCH)

Column4 C/I probability C/I value This column cannot be null

Column5

Probability C/I > Threshold

Probability to have C/I the value specified in column 4 (C/I threshold) This field must not

End of. ..

End of. .. Implementation of a Toolkit for the Processing of files issued from Radio Measurements and the Automatic Allocation of Cell Ids to New Cells

End of course Dissertation

Định dạng
Số trang	80
Dung lượng	13,46 MB