Atoll 2.8 Chapters1-6

7/30/2019 Atoll 2.8 Chapters1-6

http://slidepdf.com/reader/full/atoll-28-chapters1-6 1/304

Technical Reference Guide

v e r s i o n 2.8.2

AT282_TRG_E2






© Forsk 2010 AT282_TRG_E2 3

Contact Information

Atoll 2.8.2 Technical Reference Guide Release AT282_TRG_E2

© Copyright 1997 - 2010 by Forsk

The software described in this document is provided under a licence agreement. The software may only be used/copied

under the terms and conditions of the licence agreement. No part of this document may be copied, reproduced or

distributed in any form without prior authorisation from Forsk.

The product or brand names mentioned in this document are trademarks or registered trademarks of their respective

registering parties.

About the Technical Reference Guide

This document is targeted at readers with a prior knowledge of Atoll, its operation and basic functioning. It is not the User

Manual for Atoll, and does not teach how to operate and use Atoll. It is a supplementary document containing detailed

descriptions of models, algorithms and concepts adopted in Atoll. Therefore, it concerns only the appropriate personnel.

The Atoll Technical Reference Guide is divided into three parts with each part comprising similar topics. The first part

contains descriptions of general terms, entities, ideas and concepts in Atoll that are encountered throughout its use. It is

followed by the second part that consists of descriptions of entities common to all types of networks and the algorithms

that are technology independent and are available in any network type. Lastly, the guide provides detailed descriptions of

each basic type of network that can be modelled and studied in Atoll.

Forsk (Head Office) 7 rue des Briquetiers

31700 Blagnac

France

L"«

¬

www.forsk.com

[email protected]

[email protected]

+33 (0) 562 74 72 10

+33 (0) 562 74 72 25

+33 (0) 562 74 72 11

Web

Sales and pricing information

Technical support

General

Technical support

Fax

Forsk (USA Office) 200 South Wacker Drive

Suite 3100

Chicago, IL 60606

USA

L"«

¬

[email protected]

[email protected]

+1 312 674 4846

+1 888 GoAtoll (+1 888 462 8655)

+1 312 674 4847

Sales and pricing information

Technical support

General

Technical support

Fax

Forsk (China Office) Suite 302, 3/F, West Tower,

Jiadu Commercial Building,

No.66 Jianzhong Road,

Tianhe Hi-Tech Industrial Zone,

Guangzhou, 510665,

People’s Republic of China

L«

¬

www.forsk.com.cn

[email protected]

+86 20 8553 8938

+86 20 8553 8285

+86 10 6513 4559

Web

Information and enquiries

Telephone

Fax (Guangzhou)

Fax (Beijing)




4 AT282_TRG_E2 © Forsk 2010



Table of Contents


Table of Contents

1 Coordinate Systems and Units ....................................................... 271.1 Coordinate Systems............................................................................................................................... 27

1.1.1 Description of Coordinate Systems .................................................................................................. 271.1.1.1 Geographic Coordinate System.................................................................................................. 27

1.1.1.2 Datum ......................................................................................................................................... 27

1.1.1.3 Meridian ...................................................................................................................................... 27

1.1.1.4 Ellipsoid ...................................................................................................................................... 27

1.1.1.5 Projection .................................................................................................................................... 28

1.1.1.6 Projection Coordinate System .................................................................................................... 28

1.1.2 Coordinate Systems in Atoll ............................................................................................................. 28

1.1.2.1 Projection Coordinate System .................................................................................................... 28

1.1.2.2 Display Coordinate System ........................................................................................................ 28

1.1.2.3 Internal Coordinate Systems ...................................................................................................... 28

1.1.3 File Formats ..................................................................................................................................... 29

1.1.3.1 Unit Codes .................................................................................................................................. 29

1.1.3.2 Datum Codes.............................................................................................................................. 30

1.1.3.3 Projection Method Codes ........................................................................................................... 31

1.1.3.4 Ellipsoid Codes ........................................................................................................................... 311.1.3.5 Projection Parameter Indices...................................................................................................... 32

1.1.4 Creating a Coordinate System ......................................................................................................... 32

1.2 Units ....................................................................................................................................................... 32

1.2.1 Power Units ...................................................................................................................................... 32

1.2.2 Length Units ..................................................................................................................................... 33

1.3 BSIC Format .......................................................................................................................................... 33

2 Geographic and Radio Data ........................................................... 372.1 Geographic Data .................................................................................................................................... 37

2.1.1 Data Type ......................................................................................................................................... 37

2.1.1.1 Digital Terrain Model (DTM) ....................................................................................................... 37

2.1.1.2 Clutter (Land Use) ...................................................................................................................... 38

2.1.1.2.1 Clutter Classes...................................................................................................................... 38

2.1.1.2.2 Clutter Heights ...................................................................................................................... 382.1.1.3 Traffic Data ................................................................................................................................. 38

2.1.1.3.1 User Profile Environment Based Traffic Maps ...................................................................... 38

2.1.1.3.2 User Profile Traffic Maps....................................................................................................... 38

2.1.1.3.3 Sector Traffic Maps ............................................................................................................... 38

2.1.1.3.4 User Density Traffic Maps..................................................................................................... 39

2.1.1.4 Vector Data ................................................................................................................................. 39

2.1.1.5 Scanned Images......................................................................................................................... 39

2.1.1.6 Population ................................................................................................................................... 39

2.1.1.7 Other Geographic Data............................................................................................................... 39

2.1.2 Supported Geographic Data Formats .............................................................................................. 39

2.2 Radio Data ............................................................................................................................................. 40

2.2.1 Site ................................................................................................................................................... 40

2.2.2 Antenna ............................................................................................................................................ 40

2.2.3 Transmitter ....................................................................................................................................... 40

2.2.4 Repeater ........................................................................................................................................... 402.2.5 Remote Antenna .............................................................................................................................. 41

2.2.6 Station .............................................................................................................................................. 41

2.2.7 Hexagonal Design ............................................................................................................................ 41

2.2.8 GSM GPRS EGPRS Documents ..................................................................................................... 41

2.2.8.1 TRX............................................................................................................................................. 41

2.2.8.2 Subcell ........................................................................................................................................ 41

2.2.8.3 Cell Type..................................................................................................................................... 41

2.2.9 All CDMA, WiMAX, and LTE Documents ......................................................................................... 41

2.2.9.1 Cell.............................................................................................................................................. 41

3 File Formats .................................................................................... 453.1 BIL Format ............................................................................................................................................. 45





3.1.1 HDR Header File...............................................................................................................................45

3.1.1.1 Description ..................................................................................................................................45

3.1.1.2 Samples ......................................................................................................................................46

3.1.1.2.1 Digital Terrain Model..............................................................................................................46

3.1.1.2.2 Clutter Classes File................................................................................................................46

3.1.1.2.3 BIL File...................................................................................................................................46

3.2 TIF Format start here ..............................................................................................................................47

3.2.1 TFW Header File...............................................................................................................................47

3.2.2 Sample ..............................................................................................................................................48

3.2.2.1 Clutter Classes File .....................................................................................................................48

3.3 BMP Format............................................................................................................................................48

3.3.1 BMP File Description.........................................................................................................................48

3.3.1.1 BMP File Structure ......................................................................................................................48

3.3.1.2 BMP Raster Data Encoding ........................................................................................................50

3.3.1.2.1 Raster Data Compression Descriptions.................................................................................50

3.3.2 BPW/BMW Header File Description..................................................................................................51

3.3.3 Sample ..............................................................................................................................................51


3.4 PNG Format............................................................................................................................................51

3.4.1 PGW Header File Description ...........................................................................................................51

3.5 Generic Raster Header File (.wld) ..........................................................................................................52

3.5.1 WLD File Description ........................................................................................................................52

3.5.2 Sample ..............................................................................................................................................52


3.6 DXF Format ............................................................................................................................................52

3.7 SHP Format............................................................................................................................................52

3.8 MIF Format .............................................................................................................................................53

3.9 TAB Format ............................................................................................................................................53

3.10 ECW Format ...........................................................................................................................................54

3.11 Erdas Imagine Format ............................................................................................................................54

3.12 Planet EV/Vertical Mapper Geographic Data Format .............................................................................54

3.13 ArcView Grid Format ..............................................................................................................................55

3.13.1 ArcView Grid File Description ...........................................................................................................55

3.13.2 Sample ..............................................................................................................................................55

3.14 Other Supported Geographic Data File Formats ....................................................................................55

3.15 Planet Format .........................................................................................................................................55

3.15.1 DTM File............................................................................................................................................56

3.15.1.1 Description ..................................................................................................................................56

3.15.1.2 Sample ........................................................................................................................................56

3.15.2 Clutter Class Files .............................................................................................................................56

3.15.2.1 Description ..................................................................................................................................563.15.2.2 Sample ........................................................................................................................................57

3.15.3 Vector Files .......................................................................................................................................57

3.15.3.1 Description ..................................................................................................................................57

3.15.3.2 Sample ........................................................................................................................................58

3.15.4 Image Files........................................................................................................................................58

3.15.5 Text Data Files ..................................................................................................................................58

3.16 MNU Format ...........................................................................................................................................59

3.16.1 Description ........................................................................................................................................59

3.16.2 Sample ..............................................................................................................................................59

3.17 XML Table Export/Import Format ...........................................................................................................59

3.17.1 Index.xml File ....................................................................................................................................59

3.17.2 XML File ............................................................................................................................................60

3.18 Externalised Propagation Results Format ..............................................................................................61

3.18.1 DBF File ............................................................................................................................................61

3.18.1.1 DBF File Format ..........................................................................................................................613.18.1.1.1 DBF Structure ........................................................................................................................61

3.18.1.1.2 DBF Header (Variable Size - Depends on Field Count) ........................................................61

3.18.1.1.3 Each DBF Record (Fixed Length)..........................................................................................63

3.18.1.2 DBF File Content .........................................................................................................................63

3.18.2 LOS File ............................................................................................................................................64

3.19 Externalised Tuning Files .......................................................................................................................64

3.19.1 DBF File ............................................................................................................................................64

3.19.1.1 DBF File Format ..........................................................................................................................64

3.19.1.1.1 DBF Structure ........................................................................................................................65

3.19.1.1.2 DBF Header (Variable Size - Depends on Field Count) ........................................................65

3.19.1.1.3 Each DBF Record (Fixed Length)..........................................................................................66

3.19.1.2 DBF File Content .........................................................................................................................66



Table of Contents


3.19.2 PTS File............................................................................................................................................ 67

3.20 Interference Histograms File Formats.................................................................................................... 67

3.20.1 One Histogram per Line (.im0) Format............................................................................................. 67

3.20.1.1 Sample........................................................................................................................................ 68

3.20.2 One Value per Line with Dictionary File (.clc) Format ...................................................................... 68

3.20.2.1 CLC File ...................................................................................................................................... 69

3.20.2.1.1 Description ............................................................................................................................ 69

3.20.2.1.2 Sample .................................................................................................................................. 69

3.20.2.2 DCT File...................................................................................................................................... 70

3.20.2.2.1 Description ............................................................................................................................ 70

3.20.2.2.2 Sample .................................................................................................................................. 70

3.20.3 One Value per Line (Transmitter Name Repeated) (.im1) Format ................................................... 71

3.20.3.1 Sample........................................................................................................................................ 71

3.20.4 Only Co-Channel and Adjacent Values (.im2) Format ..................................................................... 72

3.20.4.1 Sample........................................................................................................................................ 72

3.21 Antenna Formats.................................................................................................................................... 73

3.21.1 2D Antenna Pattern Format ............................................................................................................. 73

3.21.2 3D Antenna Pattern Format ............................................................................................................. 74

4 Calculations .................................................................................... 794.1 Overview ................................................................................................................................................ 79

4.2 Path Loss Matrices................................................................................................................................. 80

4.2.1 Calculation Area Determination........................................................................................................ 81

4.2.1.1 Computation Zone ...................................................................................................................... 81

4.2.1.2 Use of Polygonal Zones in Coverage Prediction Reports........................................................... 81

4.2.2 Calculate / Force Calculation Comparison ....................................................................................... 824.2.2.1 Calculate..................................................................................................................................... 82

4.2.2.2 Force Calculation ........................................................................................................................ 82

4.2.3 Matrix Validity ................................................................................................................................... 82

4.3 Path Loss Calculations........................................................................................................................... 83

4.3.1 Ground Altitude Determination ......................................................................................................... 83

4.3.2 Clutter Determination ....................................................................................................................... 84

4.3.2.1 Clutter Class ............................................................................................................................... 84

4.3.2.2 Clutter Height.............................................................................................................................. 84

4.3.3 Geographic Profile Extraction........................................................................................................... 84

4.3.3.1 Extraction Methods ..................................................................................................................... 84

4.3.3.1.1 Radial Extraction ................................................................................................................... 84

4.3.3.1.2 Systematic Extraction ........................................................................................................... 85

4.3.3.2 Profile Resolution: Multi-Resolution Management ...................................................................... 86

4.4 Propagation Models ............................................................................................................................... 88

4.4.1 Okumura-Hata and Cost-Hata Propagation Models......................................................................... 89

4.4.1.1 Hata Path Loss Formula ............................................................................................................. 89

4.4.1.2 Corrections to the Hata Path Loss Formula................................................................................ 89

4.4.1.3 Calculations in Atoll .................................................................................................................... 89

4.4.2 ITU 529-3 Propagation Model .......................................................................................................... 90

4.4.2.1 ITU 529-3 Path Loss Formula..................................................................................................... 90

4.4.2.2 Corrections to the ITU 529-3 Path Loss Formula ....................................................................... 90

4.4.2.2.1 Environment Correction ........................................................................................................ 90

4.4.2.2.2 Area Size Correction ............................................................................................................. 90

4.4.2.2.3 Distance Correction .............................................................................................................. 91


4.4.3 Standard Propagation Model (SPM) ................................................................................................ 91

4.4.3.1 SPM Path Loss Formula............................................................................................................. 91


4.4.3.2.1 Visibility and Distance Between Transmitter and Receiver ................................................... 92

4.4.3.2.2 Effective Transmitter Antenna Height ................................................................................... 92

4.4.3.2.3 Effective Receiver Antenna Height ....................................................................................... 95

4.4.3.2.4 Correction for Hilly Regions in Case of LOS ......................................................................... 95

4.4.3.2.5 Diffraction .............................................................................................................................. 96

4.4.3.2.6 Losses due to Clutter ............................................................................................................ 96

4.4.3.2.7 Recommendations ................................................................................................................ 97

4.4.3.3 Automatic SPM Calibration......................................................................................................... 97

4.4.3.3.1 General Algorithm ................................................................................................................. 98

4.4.3.3.2 Sample Values for SPM Path Loss Formula Parameters ..................................................... 98

4.4.3.4 Unmasked Path Loss Calculation ............................................................................................... 99

4.4.4 WLL Propagation Model ................................................................................................................. 100

4.4.4.1 WLL Path Loss Formula ........................................................................................................... 100

4.4.4.2 Calculations in Atoll .................................................................................................................. 100





4.4.4.2.1 Free Space Loss..................................................................................................................100

4.4.4.2.2 Diffraction.............................................................................................................................100

4.4.5 ITU-R P.526-5 Propagation Model..................................................................................................100

4.4.5.1 ITU 526-5 Path Loss Formula ...................................................................................................100

4.4.5.2 Calculations in Atoll ...................................................................................................................101

4.4.5.2.1 Free Space Loss..................................................................................................................101

4.4.5.2.2 Diffraction.............................................................................................................................101

4.4.6 ITU-R P.370-7 Propagation Model..................................................................................................101

4.4.6.1 ITU 370-7 Path Loss Formula ...................................................................................................101


4.4.6.2.1 Free Space Loss..................................................................................................................101

4.4.6.2.2 Corrected Standard Loss .....................................................................................................101

4.4.7 Erceg-Greenstein (SUI) Propagation Model ...................................................................................102

4.4.7.1 SUI Terrain Types .....................................................................................................................103

4.4.7.2 Erceg-Greenstein (SUI) Path Loss Formula..............................................................................103


4.4.8 ITU-R P.1546-2 Propagation Model................................................................................................104


4.4.8.1.1 Step 1: Determination of Graphs to be Used .......................................................................105

4.4.8.1.2 Step 2: Calculation of Maximum Field Strength...................................................................105

4.4.8.1.3 Step 3: Determination of Transmitter Antenna Height .........................................................105

4.4.8.1.4 Step 4: Interpolation/Extrapolation of Field Strength ...........................................................105

4.4.8.1.5 Step 5: Calculation of Correction Factors ............................................................................107

4.4.8.1.6 Step 6: Calculation of Path Loss..........................................................................................108

4.4.9 Sakagami Extended Propagation Model.........................................................................................108

4.4.10 Appendices.....................................................................................................................................110

4.4.10.1 Free Space Loss .......................................................................................................................110

4.4.10.2 Diffraction Loss..........................................................................................................................110

4.4.10.2.1 Knife-Edge Diffraction ..........................................................................................................110

4.4.10.2.2 3 Knife-Edge Deygout Method.............................................................................................111

4.4.10.2.3 Epstein-Peterson Method ....................................................................................................112

4.4.10.2.4 Deygout Method with Correction .........................................................................................112

4.4.10.2.5 Millington Method.................................................................................................................113

4.5 Path Loss Tuning ..................................................................................................................................113

4.5.1 Transmitter Path Loss Tuning .........................................................................................................113

4.5.2 Repeater Path Loss Tuning ............................................................................................................114

4.6 Antenna Attenuation Calculation ..........................................................................................................115

4.6.1 Calculation of Azimuth and Tilt Angles............................................................................................115

4.6.2 Antenna Pattern 3-D Interpolation...................................................................................................116

4.6.3 Additional Electrical Downtilt Modelling...........................................................................................117

4.6.4 Antenna Pattern Smoothing ............................................................................................................1174.6.4.1 Smoothing Algorithm .................................................................................................................119

4.7 Shadowing Model .................................................................................................................................119

4.7.1 Shadowing Margin Calculation........................................................................................................123

4.7.1.1 Shadowing Margin Calculation in Predictions ...........................................................................124

4.7.1.2 Shadowing Margin Calculation in Monte-Carlo Simulations......................................................125

4.7.2 Macro-Diversity Gains Calculation ..................................................................................................126

4.7.2.1 Uplink Macro-Diversity Gain Evaluation ....................................................................................126

4.7.2.1.1 Shadowing Error PDF (n Signals)........................................................................................126

4.7.2.1.2 Uplink Macro-Diversity Gain ................................................................................................128

4.7.2.2 Downlink Macro-Diversity Gain Evaluation ...............................................................................128

4.7.2.2.1 Shadowing Error PDF (n Signals)........................................................................................128

4.7.2.2.2 Downlink Macro-Diversity Gain............................................................................................131

4.8 Appendices ...........................................................................................................................................131

4.8.1 Transmitter Radio Equipment .........................................................................................................131

4.8.1.1 UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, and TD-SCDMA Documents..............................1324.8.1.2 GSM GPRS EGPRS Documents ..............................................................................................133

4.8.1.3 WiMAX 802.16d and WiMAX 802.16e Documents ...................................................................133

4.8.1.4 LTE Documents.........................................................................................................................134

4.8.2 Secondary Antennas.......................................................................................................................135

5 GSM GPRS EDGE Networks ........................................................1395.1 Signal Level Calculations......................................................................................................................139

5.1.1 Point Analysis..................................................................................................................................139

5.1.1.1 Profile Tab .................................................................................................................................139

5.1.1.2 Reception Tab ...........................................................................................................................139

5.1.2 Signal Level-based Coverage Predictions ......................................................................................140

5.1.2.1 Service Area Determination ......................................................................................................140



Table of Contents


5.1.2.1.1 All Servers........................................................................................................................... 140

5.1.2.1.2 Best Signal Level and a Margin .......................................................................................... 140

5.1.2.1.3 Second Best Signal Level and a Margin ............................................................................. 140

5.1.2.1.4 Best Signal Level per HCS Layer and a Margin.................................................................. 141

5.1.2.1.5 Second Best Signal Level per HCS Layer and a Margin .................................................... 141

5.1.2.1.6 HCS Servers and a Margin ................................................................................................. 141

5.1.2.1.7 Highest Priority HCS Server and a Margin.......................................................................... 142

5.1.2.1.8 Best Idle Mode Reselection Criterion (C2).......................................................................... 142

5.1.2.2 Coverage Display ..................................................................................................................... 143

5.1.2.2.1 Coverage Resolution .......................................................................................................... 143

5.1.2.2.2 Display Types ..................................................................................................................... 143

5.2 Interference-based Calculations .......................................................................................................... 144

5.2.1 Carrier-to-Interference Ratio Calculation........................................................................................ 144

5.2.2 Point Analysis ................................................................................................................................. 147

5.2.3 Interference-based Coverage Predictions ...................................................................................... 147

5.2.3.1 Service Area Determination ...................................................................................................... 148

5.2.3.2 Coverage Area Determination .................................................................................................. 148

5.2.3.2.1 Interference Condition Satisfied by At Least One TRX ....................................................... 148

5.2.3.2.2 Interference Condition Satisfied by The Worst TRX ........................................................... 148



5.2.3.3.2 Display Types ..................................................................................................................... 148

5.3 GPRS/EDGE Calculations ................................................................................................................... 149

5.3.1 Coding Scheme Selection and Throughput Calculation Without Ideal Link Adaptation ................. 149

5.3.1.1 Calculations Based on C .......................................................................................................... 149

5.3.1.2 Calculations Based on C/I........................................................................................................ 150

5.3.1.3 Calculations Based on C/(I+N) ................................................................................................. 150

5.3.2 Coding Scheme Selection and Throughput Calculation With Ideal Link Adaptation ...................... 151

5.3.2.1 Calculations Based on C .......................................................................................................... 151

5.3.2.2 Calculations Based on C/I ........................................................................................................ 151


5.3.3 Application Throughput Calculation................................................................................................ 152

5.3.4 BLER Calculation ........................................................................................................................... 152

5.3.5 GPRS/EDGE Coverage Predictions............................................................................................... 152


5.3.5.1.1 All Servers........................................................................................................................... 152





5.3.5.1.6 HCS Servers and a Margin ................................................................................................. 1535.3.5.1.7 Highest Priority HCS Server and a Margin.......................................................................... 154

5.3.5.1.8 Best Idle Mode Reselection Criterion (C2).......................................................................... 154



5.3.5.2.2 Display Types ..................................................................................................................... 154

5.4 Codec Mode Selection and CQI Calculations ...................................................................................... 156

5.4.1 Circuit Quality Indicator Calculations.............................................................................................. 158

5.4.2 CQI Calculation Without Ideal Link Adaptation .............................................................................. 158

5.4.2.1 Calculations Based on C/N....................................................................................................... 158


5.4.3 CQI Calculation With Ideal Link Adaptation ................................................................................... 159

5.4.3.1 Calculations Based on C/N....................................................................................................... 159


5.4.4 Circuit Quality Indicators Coverage Predictions ............................................................................. 160

5.4.4.1 Service Area Determination ...................................................................................................... 1605.4.4.1.1 All Servers........................................................................................................................... 160





5.4.4.1.6 HCS Servers and a Margin ................................................................................................. 161

5.4.4.1.7 Highest Priority HCS Server and a Margin.......................................................................... 161



5.4.4.2.2 Display Types ..................................................................................................................... 162

5.5 Traffic Analysis..................................................................................................................................... 162

5.5.1 Traffic Distribution .......................................................................................................................... 162




10 AT282_TRG_E2 © Forsk 2010

5.5.1.1 Normal Cells (Nonconcentric, No HCS Layer) ..........................................................................162

5.5.1.1.1 Circuit Switched Services ....................................................................................................162

5.5.1.1.2 Packet Switched Services ...................................................................................................163

5.5.1.2 Concentric Cells ........................................................................................................................163



5.5.1.3 HCS Layers ...............................................................................................................................163



5.5.2 Calculation of the Traffic Demand per Subcell................................................................................163

5.5.2.1 User Profile Traffic Maps ...........................................................................................................163

5.5.2.1.1 Normal Cells (Nonconcentric, No HCS Layer).....................................................................163

5.5.2.1.2 Concentric Cells...................................................................................................................164

5.5.2.1.3 HCS Layers .........................................................................................................................165

5.5.2.2 Sector Traffic Maps ...................................................................................................................168

5.5.2.2.1 Normal Cells (Nonconcentric, No HCS Layer).....................................................................168

5.5.2.2.2 Concentric Cells...................................................................................................................169

5.5.2.2.3 HCS Layers .........................................................................................................................169

5.6 Network Dimensioning ..........................................................................................................................173

5.6.1 Dimensioning Models and Quality Graphs......................................................................................173

5.6.1.1 Circuit Switched Traffic..............................................................................................................174

5.6.1.2 Packet Switched Traffic .............................................................................................................174

5.6.1.2.1 Throughput ..........................................................................................................................174

5.6.1.2.2 Delay....................................................................................................................................176

5.6.1.2.3 Blocking Probability .............................................................................................................176

5.6.2 Network Dimensioning Process......................................................................................................178

5.6.2.1 Network Dimensioning Engine ..................................................................................................178

5.6.2.1.1 Inputs ...................................................................................................................................178

5.6.2.1.2 Outputs ................................................................................................................................178

5.6.2.2 Network Dimensioning Steps ....................................................................................................178

5.6.2.2.1 Step 1: Timeslots Required for CS Traffic ...........................................................................178

5.6.2.2.2 Step 2: TRXs Required for CS Traffic and Dedicated PS Timeslots ...................................179

5.6.2.2.3 Step 3: Effective CS Blocking, Effective CS Traffic Overflow and Served CS Traffic..........179

5.6.2.2.4 Step 4: TRXs to Add for PS Traffic ......................................................................................179

5.6.2.2.5 Step 5: Served PS Traffic ....................................................................................................181

5.6.2.2.6 Step 6: Total Traffic Load ....................................................................................................182

5.7 Key Performance Indicators Calculation...............................................................................................182

5.7.1 Circuit Switched Traffic ...................................................................................................................182

5.7.1.1 Erlang B.....................................................................................................................................182

5.7.1.2 Erlang C ....................................................................................................................................182

5.7.1.3 Served Circuit Switched Traffic .................................................................................................1835.7.2 Packet Switched Traffic...................................................................................................................183

5.7.2.1 Case 1: Total Traffic Demand > Dedicated + Shared Timeslots ...............................................183

5.7.2.1.1 Traffic Load ..........................................................................................................................183

5.7.2.1.2 Packet Switched Traffic Overflow ........................................................................................183

5.7.2.1.3 Throughput Reduction Factor ..............................................................................................183

5.7.2.1.4 Delay....................................................................................................................................183


5.7.2.1.6 Served Packet Switched Traffic ...........................................................................................183

5.7.2.2 Case 2: Total Traffic Demand < Dedicated + Shared Timeslots ...............................................184

5.7.2.2.1 Traffic Load ..........................................................................................................................184

5.7.2.2.2 Packet Switched Traffic Overflow ........................................................................................184

5.7.2.2.3 Throughput Reduction Factor ..............................................................................................184

5.7.2.2.4 Delay....................................................................................................................................184


5.7.2.2.6 Served Packet Switched Traffic ...........................................................................................1845.8 Neighbour Allocation.............................................................................................................................184

5.8.1 Global Allocation for All Transmitters ..............................................................................................185

5.8.2 Allocation for a Group of Transmitters or One Transmitter .............................................................188

5.9 AFP Appendices ...................................................................................................................................188

5.9.1 The AFP Cost Function...................................................................................................................188

5.9.1.1 Cost Function ............................................................................................................................189

5.9.1.2 Cost Components......................................................................................................................190

5.9.1.2.1 Separation Violation Cost Component.................................................................................190

5.9.1.2.2 Interference Cost Component..............................................................................................191

5.9.1.2.3 I_DIV, F_DIV and Other Advanced Cost Parameters..........................................................193

5.9.2 The AFP Blocked Traffic Cost.........................................................................................................193

5.9.2.1 Calculation of New Traffic Loads Including Blocked Traffic Loads............................................194



Table of Contents

© Forsk 2010 AT282_TRG_E2 11

5.9.2.2 Recalculation of CS and PS From Traffic Loads ...................................................................... 195

5.9.2.3 Testing the Blocked Cost Using Traffic Analysis ...................................................................... 196

5.9.3 Interferences .................................................................................................................................. 196

5.9.3.1 Using Interferences................................................................................................................... 196

5.9.3.2 Cumulative Density Function of C/I Levels ............................................................................... 196

5.9.3.3 Precise Definition ...................................................................................................................... 197

5.9.3.4 Precise Interference Distribution Strategy ................................................................................ 197

5.9.3.4.1 Direct Availability of Precise Interference Distribution to the AFP....................................... 197

5.9.3.4.2 Efficient Calculation and Storage of Interference Distribution............................................. 197

5.9.3.4.3 Robustness of the IM .......................................................................................................... 197

5.9.3.5 Traffic Load and Interference Information Discrimination ......................................................... 197

6 UMTS HSPA Networks ................................................................. 2016.1 General Prediction Studies .................................................................................................................. 201

6.1.1 Calculation Criteria ......................................................................................................................... 201

6.1.2 Point Analysis ................................................................................................................................. 201

6.1.2.1 Profile Tab ................................................................................................................................ 201

6.1.2.2 Reception Tab .......................................................................................................................... 201

6.1.3 Coverage Studies ........................................................................................................................... 202


6.1.3.1.1 All Servers........................................................................................................................... 202




6.1.3.2.1 Plot Resolution .................................................................................................................... 203

6.1.3.2.2 Display Types ..................................................................................................................... 2036.2 Definitions and Formulas ..................................................................................................................... 204

6.2.1 Inputs.............................................................................................................................................. 204

6.2.2 Ec/I0 Calculation ............................................................................................................................ 209

6.2.3 DL Eb/Nt Calculation ...................................................................................................................... 210

6.2.4 UL Eb/Nt Calculation ...................................................................................................................... 211

6.3 Active Set Management ....................................................................................................................... 212

6.4 Simulations........................................................................................................................................... 212

6.4.1 Generating a Realistic User Distribution ........................................................................................ 212

6.4.1.1 Simulations Based on User Profile Traffic Maps ...................................................................... 213

6.4.1.1.1 Circuit Switched Service (i) ................................................................................................. 213

6.4.1.1.2 Packet Switched Service (j) ................................................................................................ 213

6.4.1.2 Simulations Based on Sector Traffic Maps............................................................................... 216

6.4.1.2.1 Throughputs in Uplink and Downlink................................................................................... 216

6.4.1.2.2 Total Number of Users (All Activity Statuses) ..................................................................... 217

6.4.1.2.3 Number of Users per Activity Status ................................................................................... 217

6.4.2 Power Control Simulation ............................................................................................................... 218

6.4.2.1 Algorithm Initialization ............................................................................................................... 219

6.4.2.2 R99 Part of the Algorithm ......................................................................................................... 219

6.4.2.3 HSDPA Part of the Algorithm.................................................................................................... 223

6.4.2.3.1 HSDPA Power Allocation .................................................................................................... 223

6.4.2.3.2 Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users............. 224

6.4.2.3.3 HSDPA Bearer Allocation Process ..................................................................................... 224

6.4.2.3.4 Fast Link Adaptation Modelling ........................................................................................... 226

6.4.2.3.5 MIMO Modelling .................................................................................................................. 236

6.4.2.3.6 Scheduling Algorithms ........................................................................................................ 236

6.4.2.4 HSUPA Part of the Algorithm.................................................................................................... 238

6.4.2.4.1 Admission Control ............................................................................................................... 238

6.4.2.4.2 HSUPA Bearer Allocation Process ..................................................................................... 240

6.4.2.4.3 Noise Rise Scheduling ........................................................................................................ 242

6.4.2.4.4 Radio Resource Control...................................................................................................... 245

6.4.2.5 Convergence Criteria ................................................................................................................ 245

6.4.3 Results ........................................................................................................................................... 246

6.4.3.1 R99 Related Results................................................................................................................. 246

6.4.3.2 HSPA Related Results.............................................................................................................. 247

6.4.3.2.1 Statistics Tab ...................................................................................................................... 247

6.4.3.2.2 Mobiles Tab ........................................................................................................................ 248

6.4.3.2.3 Cells Tab ............................................................................................................................. 250

6.4.3.2.4 Sites Tab ............................................................................................................................. 252

6.4.4 Appendices..................................................................................................................................... 252

6.4.4.1 Admission Control in the R99 Part............................................................................................ 252

6.4.4.2 Resources Management........................................................................................................... 252

6.4.4.2.1 OVSF Codes Management ................................................................................................. 252




12 AT282_TRG_E2 © Forsk 2010

6.4.4.2.2 Channel Elements Management .........................................................................................254

6.4.4.2.3 Iub Backhaul Throughput.....................................................................................................254

6.4.4.3 Downlink Load Factor Calculation .............................................................................................255

6.4.4.3.1 Downlink Load Factor per Cell.............................................................................................255

6.4.4.3.2 Downlink Load Factor per Mobile ........................................................................................256

6.4.4.4 Uplink Load Factor Due to One User ........................................................................................256

6.4.4.5 Inter-carrier Power Sharing Modelling .......................................................................................258

6.4.4.6 Best Server Determination in Monte Carlo Simulations - Old Method ......................................259

6.5 UMTS HSPA Prediction Studies ...........................................................................................................261

6.5.1 Point Analysis..................................................................................................................................261

6.5.1.1 AS Analysis Tab ........................................................................................................................261

6.5.1.1.1 Bar Graph and Pilot Sub-Menu............................................................................................261

6.5.1.1.2 Downlink Sub-Menu.............................................................................................................263

6.5.1.1.3 Uplink Sub-Menu .................................................................................................................268

6.5.2 Coverage Studies............................................................................................................................271

6.5.2.1 Pilot Reception Analysis ............................................................................................................271

6.5.2.1.1 Prediction Study Inputs ........................................................................................................272

6.5.2.1.2 Study Display Options .........................................................................................................272

6.5.2.2 Downlink Service Area Analysis ................................................................................................273



6.5.2.3 Uplink Service Area Analysis ....................................................................................................274



6.5.2.4 Downlink Total Noise Analysis ..................................................................................................276

6.5.2.4.1 Study Inputs.........................................................................................................................276

6.5.2.4.2 Analysis on All Carriers........................................................................................................276

6.5.2.4.3 Analysis on a Specific Carrier ..............................................................................................277

6.5.2.5 HSDPA Prediction Study ...........................................................................................................277



6.5.2.6 HSUPA Prediction Study ...........................................................................................................281


6.5.2.6.2 Calculation Options..............................................................................................................282

6.5.2.6.3 Display Options....................................................................................................................282

6.6 Automatic Neighbour Allocation............................................................................................................284

6.6.1 Neighbour Allocation for All Transmitters........................................................................................284

6.6.2 Neighbour Allocation for a Group of Transmitters or One Transmitter ............................................288

6.6.3 Importance Calculation ...................................................................................................................288

6.6.3.1 Importance of Intra-carrier Neighbours .....................................................................................288

6.6.3.2 Importance of Inter-carrier Neighbours .....................................................................................2896.7 Primary Scrambling Code Allocation ....................................................................................................290

6.7.1 Automatic Allocation Description.....................................................................................................290

6.7.1.1 Options and Constraints ............................................................................................................290

6.7.1.2 Allocation Process .....................................................................................................................291

6.7.1.2.1 Single Carrier Network.........................................................................................................292

6.7.1.2.2 Multi-Carrier Network ...........................................................................................................292

6.7.1.3 Priority Determination ................................................................................................................293

6.7.1.3.1 Cell Priority ..........................................................................................................................293

6.7.1.3.2 Transmitter Priority ..............................................................................................................295

6.7.1.3.3 Site Priority ..........................................................................................................................295

6.7.2 Allocation Examples........................................................................................................................295

6.7.2.1 Allocation Strategies and Use a Maximum of Codes ................................................................295

6.7.2.1.1 Strategy: Clustered ..............................................................................................................296

6.7.2.1.2 Strategy: Distributed ............................................................................................................297

6.7.2.1.3 Strategy: ‘One Cluster per Site ............................................................................................2976.7.2.1.4 Strategy: ‘Distributed per Site ..............................................................................................298

6.7.2.2 Allocate Carriers Identically .......................................................................................................298

6.8 Automatic GSM-UMTS Neighbour Allocation .......................................................................................299

6.8.1 Overview .........................................................................................................................................299


6.8.2.1 Algorithm Based on Distance ....................................................................................................299

6.8.2.2 Algorithm Based on Coverage Overlapping ..............................................................................300

6.8.2.3 Appendices................................................................................................................................302

6.8.2.3.1 Delete Existing Neighbours Option ......................................................................................302

6.8.2.3.2 Calculation of Inter-Transmitter Distance ............................................................................302



Table of Contents

© Forsk 2010 AT282_TRG_E2 13

7 CDMA2000 Networks ................................................................... 3057.1 General Prediction Studies .................................................................................................................. 305

7.1.1 Calculation Criteria ......................................................................................................................... 305

7.1.2 Point Analysis ................................................................................................................................. 305

7.1.2.1 Profile Tab ................................................................................................................................ 305

7.1.2.2 Reception Tab .......................................................................................................................... 306



7.1.3.1.1 All Servers........................................................................................................................... 306


7.1.3.1.3 Second Best Signal Level and a Margin ............................................................................. 3067.1.3.2 Coverage Display ..................................................................................................................... 307

7.1.3.2.1 Plot Resolution .................................................................................................................... 307

7.1.3.2.2 Display Types ..................................................................................................................... 307

7.2 Definitions and Formulas ..................................................................................................................... 308

7.2.1 Parameters Used for CDMA2000 1xRTT Modelling ...................................................................... 308

7.2.1.1 Inputs ........................................................................................................................................ 308

7.2.1.2 Ec/I0 Calculation....................................................................................................................... 312

7.2.1.3 DL Eb/Nt Calculation ................................................................................................................ 312

7.2.1.4 UL Eb/Nt Calculation ................................................................................................................ 313

7.2.1.5 Simulation Results .................................................................................................................... 315

7.2.2 Parameters Used for CDMA2000 1xEV-DO Modelling .................................................................. 316

7.2.2.1 Inputs ........................................................................................................................................ 316

7.2.2.2 Ec/I0 and Ec/Nt Calculations .................................................................................................... 319

7.2.2.3 UL Eb/Nt Calculation ................................................................................................................ 320

7.2.2.4 Simulation Results .................................................................................................................... 321

7.3 Active Set Management ....................................................................................................................... 322

7.4 Simulations........................................................................................................................................... 323


7.4.1.1 Number of Users, User Activity Status and User Data Rate..................................................... 323

7.4.1.1.1 Simulations Based on User Profile Traffic Maps ................................................................. 323

7.4.1.1.2 Simulations Based on Sector Traffic Maps ......................................................................... 326

7.4.1.2 Transition Flags for 1xEV-DO Rev.0 User Data Rates ............................................................. 331

7.4.1.3 User Geographical Position ...................................................................................................... 331

7.4.2 Network Regulation Mechanism..................................................................................................... 331

7.4.2.1 AtollCDMA2000 1xRTT Power Control Simulation Algorithm................................................... 331

7.4.2.1.1 Algorithm Initialization ......................................................................................................... 332

7.4.2.1.2 Presentation of the Algorithm.............................................................................................. 332

7.4.2.1.3 Convergence Criterion ........................................................................................................ 338

7.4.2.2 CDMA2000 1xEV-DO Power/Data Rate Control Simulation Algorithm .................................... 339

7.4.2.2.1 Algorithm Initialization ......................................................................................................... 340

7.4.2.2.2 Presentation of the Algorithm.............................................................................................. 340

7.4.2.2.3 Convergence Criterion ........................................................................................................ 345

7.4.3 Appendices..................................................................................................................................... 346

7.4.3.1 Admission Control..................................................................................................................... 346

7.4.3.2 Resources Management........................................................................................................... 346

7.4.3.2.1 Walsh Code Management .................................................................................................. 346

7.4.3.2.2 Channel Element Management .......................................................................................... 347

7.4.3.3 Downlink Load Factor Calculation ............................................................................................ 347

7.4.3.3.1 Downlink Load Factor per Cell ............................................................................................ 347

7.4.3.3.2 Downlink Load Factor per Mobile........................................................................................ 349

7.4.3.4 Best Server Determination in Monte Carlo Simulations - Old Method ...................................... 349

7.5 CDMA2000 Prediction Studies............................................................................................................. 351

7.5.1 Point Analysis: The AS Analysis Tab ............................................................................................. 351

7.5.1.1 Bar Graph and Pilot Sub-Menu................................................................................................. 351

7.5.1.2 Downlink Sub-Menu.................................................................................................................. 353

7.5.1.2.1 CDMA2000 1xRTT.............................................................................................................. 353

7.5.1.2.2 CDMA2000 1xEV-DO ......................................................................................................... 357

7.5.1.3 Uplink Sub-Menu ...................................................................................................................... 358

7.5.1.3.1 CDMA2000 1xRTT.............................................................................................................. 358

7.5.1.3.2 CDMA2000 1xEV-DO ......................................................................................................... 362


7.5.2.1 Pilot Reception Analysis ........................................................................................................... 365

7.5.2.2 Downlink Service Area Analysis ............................................................................................... 366

7.5.2.2.1 CDMA2000 1xRTT.............................................................................................................. 366

7.5.2.2.2 CDMA2000 1xEV-DO ......................................................................................................... 368

7.5.2.3 Uplink Service Area Analysis.................................................................................................... 369




14 AT282_TRG_E2 © Forsk 2010

7.5.2.3.1 CDMA2000 1xRTT ..............................................................................................................369

7.5.2.3.2 CDMA2000 1xEV-DO ..........................................................................................................370

7.5.2.4 Downlink Total Noise Analysis ..................................................................................................373

7.5.2.4.1 Analysis on all Carriers ........................................................................................................373

7.5.2.4.2 Analysis on a Specific Carrier ..............................................................................................374

7.6 Automatic Neighbour Allocation............................................................................................................374

7.6.1 Neighbour Allocation for all Transmitters ........................................................................................374

7.6.2 Neighbour Allocation for a Group of Transmitters or One Transmitter ............................................377

7.6.3 Importance Calculation ...................................................................................................................377

7.6.3.1 Importance of Intra-carrier Neighbours .....................................................................................377

7.6.3.2 Importance of Inter-carrier Neighbours .....................................................................................378

7.7 PN Offset Allocation..............................................................................................................................379


7.7.1.1 Options and Constraints ............................................................................................................379

7.7.1.2 Allocation Process .....................................................................................................................380



7.7.1.2.3 Difference between Adjacent and Distributed PN-Clusters .................................................381


7.7.1.3.1 Cell Priority ..........................................................................................................................382


7.7.1.3.3 Site Priority ..........................................................................................................................384

7.7.2 Allocation Examples........................................................................................................................384

7.7.2.1 Strategy: PN Offset per Cell ......................................................................................................385

7.7.2.2 Strategy: Adjacent PN-Clusters Per Site ...................................................................................385

7.7.2.3 Strategy: ‘Distributed PN-Clusters Per Site...............................................................................386

7.8 Automatic GSM-CDMA Neighbour Allocation.......................................................................................386

7.8.1 Overview .........................................................................................................................................386




7.8.2.3 Delete Existing Neighbours Option ...........................................................................................389

8 TD-SCDMA Networks....................................................................3938.1 Definitions and Formulas ......................................................................................................................393

8.1.1 Inputs ..............................................................................................................................................393

8.1.2 P-CCPCH Eb/Nt and C/I Calculation ..............................................................................................397

8.1.3 DwPCH C/I Calculation ...................................................................................................................398

8.1.4 DL TCH Eb/Nt and C/I Calculation..................................................................................................398

8.1.5 UL TCH Eb/Nt and C/I Calculation..................................................................................................398

8.1.6 Interference Calculation ..................................................................................................................399

8.1.7 HSDPA Dynamic Power Calculations .............................................................................................399

8.1.8 Smart Antenna Models....................................................................................................................399

8.1.8.1 Downlink Beamforming .............................................................................................................399

8.1.8.2 Uplink Beamforming ..................................................................................................................400

8.1.8.3 Uplink Beamforming and Interference Cancellation (MMSE) ....................................................400

8.2 Signal Level Based Calculations ..........................................................................................................401

8.2.1 Point Analysis..................................................................................................................................401

8.2.1.1 Profile Tab .................................................................................................................................401

8.2.1.2 Reception Tab ...........................................................................................................................401

8.2.2 RSCP Based Coverage Predictions................................................................................................402

8.2.2.1 Calculation Criteria ....................................................................................................................402

8.2.2.2 P-CCPCH RSCP Coverage Prediction .....................................................................................402

8.2.2.2.1 Coverage Condition .............................................................................................................402

8.2.2.2.2 Coverage Display ................................................................................................................403

8.2.2.3 Best Server P-CCPCH Coverage Prediction.............................................................................403

8.2.2.4 P-CCPCH Pollution Coverage Prediction..................................................................................403

8.2.2.5 DwPCH RSCP Coverage Prediction .........................................................................................404



8.2.2.6 UpPCH RSCP Coverage Prediction..........................................................................................404



8.2.2.7 Baton Handover Coverage Prediction .......................................................................................405



8.2.2.8 Scrambling Code Interference Analysis ....................................................................................405

8.3 Monte Carlo Simulations.......................................................................................................................406



Table of Contents

© Forsk 2010 AT282_TRG_E2 15


8.3.1.1 Simulations Based on User Profile Traffic Maps ...................................................................... 406

8.3.1.1.1 Circuit Switched Service (i) ................................................................................................. 407

8.3.1.1.2 Packet Switched Service (j) ................................................................................................ 407

8.3.1.2 Simulations Based on Sector Traffic Maps............................................................................... 410

8.3.1.2.1 Throughputs in Uplink and Downlink................................................................................... 410

8.3.1.2.2 Total Number of Users (All Activity Statuses) ..................................................................... 410

8.3.1.2.3 Number of Users per Activity Status ................................................................................... 411

8.3.2 Power Control Simulation ............................................................................................................... 411

8.3.2.1 Algorithm Initialisation ............................................................................................................... 412

8.3.2.2 R99 Part of the Algorithm ......................................................................................................... 412

8.3.2.2.1 Determination of Mi’s Best Server (SBS(Mi))...................................................................... 412

8.3.2.2.2 Dynamic Channel Allocation ............................................................................................... 413

8.3.2.2.3 Uplink Power Control .......................................................................................................... 414

8.3.2.2.4 Downlink Power Control...................................................................................................... 416

8.3.2.2.5 Uplink Signals Update......................................................................................................... 418

8.3.2.2.6 Downlink Signals Update .................................................................................................... 418

8.3.2.2.7 Control of Radio Resource Limits (Downlink Traffic Power and Uplink Load) .................... 418

8.3.2.3 HSDPA Part of the Algorithm.................................................................................................... 419

8.3.2.3.1 HSDPA Power Allocation .................................................................................................... 419

8.3.2.3.2 Connection Status and Number of HSDPA Users .............................................................. 421

8.3.2.3.3 HSDPA Admission Control.................................................................................................. 421

8.3.2.3.4 HSDPA Dynamic Channel Allocation.................................................................................. 422

8.3.2.3.5 Ressource Unit Saturation .................................................................................................. 422

8.3.2.4 Convergence Criteria ................................................................................................................ 422

8.4 TD-SCDMA Prediction Studies............................................................................................................ 423

8.4.1 P-CCPCH Reception Analysis (Eb/Nt) or (C/I) ............................................................................... 423

8.4.2 DwPCH Reception Analysis (C/I) ................................................................................................... 424

8.4.3 Downlink TCH RSCP Coverage ..................................................................................................... 426

8.4.4 Uplink TCH RSCP Coverage ......................................................................................................... 426

8.4.5 Downlink Total Noise...................................................................................................................... 427

8.4.6 Downlink Service Area (Eb/Nt) or (C/I)........................................................................................... 428

8.4.7 Uplink Service Area (Eb/Nt) or (C/I) ............................................................................................... 430

8.4.8 Effective Service Area (Eb/Nt) or (C/I) ........................................................................................... 431

8.4.9 Cell to Cell Interference .................................................................................................................. 432

8.4.10 UpPCH Interference ....................................................................................................................... 433

8.4.11 HSDPA Coverage .......................................................................................................................... 433

8.5 Smart Antenna Modelling..................................................................................................................... 434

8.5.1 Modelling in Simulations................................................................................................................. 435

8.5.1.1 Grid of Beams Modelling .......................................................................................................... 435

8.5.1.2 Adaptive Beam Modelling ......................................................................................................... 4368.5.1.3 Statistical Modelling .................................................................................................................. 437

8.5.1.4 Beamforming Smart Antenna Models ....................................................................................... 437

8.5.1.4.1 Downlink Beamforming ....................................................................................................... 438

8.5.1.4.2 Uplink Beamforming............................................................................................................ 439

8.5.1.4.3 Uplink Beamforming and Interference Cancellation (MMSE).............................................. 440

8.5.1.5 3rd Party Smart Antenna Modelling .......................................................................................... 441

8.5.2 Construction of the Geographic Distributions ................................................................................. 441

8.5.3 Modelling in Coverage Predictions ................................................................................................. 443

8.5.4 HSDPA Coverage Prediction ......................................................................................................... 443

8.5.4.1 Fast Link Adaptation Modelling................................................................................................. 443

8.5.4.1.1 CQI Based on P-CCPCH Quality ........................................................................................ 444

8.5.4.1.2 CQI Based on HS-PDSCH Quality...................................................................................... 447

8.5.4.2 Coverage Prediction Display Options ....................................................................................... 449

8.5.4.2.1 Colour per CQI .................................................................................................................... 449

8.5.4.2.2 Colour per Peak Throughput............................................................................................... 4498.5.4.2.3 Colour per HS-PDSCH Ec/Nt.............................................................................................. 449

8.6 N-Frequency Mode and Carrier Allocation........................................................................................... 449

8.6.1 Automatic Carrier Allocation ........................................................................................................... 449

8.7 Neighbour Allocation ............................................................................................................................ 450

8.7.1 Neighbour Allocation for All Transmitters ....................................................................................... 451

8.7.2 Neighbour Allocation for a Group of Transmitters or One Transmitter ........................................... 453

8.7.3 Importance Calculation................................................................................................................... 453

8.8 Scrambling Code Allocation ................................................................................................................. 454

8.8.1 Automatic Allocation Description .................................................................................................... 455

8.8.1.1 Allocation Constraints and Options........................................................................................... 455

8.8.1.2 Allocation Strategies ................................................................................................................. 455

8.8.1.3 Allocation Process .................................................................................................................... 456




16 AT282_TRG_E2 © Forsk 2010




8.8.1.4.1 Cell Priority ..........................................................................................................................457


8.8.1.4.3 Site Priority ..........................................................................................................................460

8.8.2 IScrambling Code Allocation Example ............................................................................................460

8.8.2.1 Single Carrier Network ..............................................................................................................460

8.8.2.1.1 Strategy: Clustered ..............................................................................................................461

8.8.2.1.2 Strategy: Distributed per Cell...............................................................................................461

8.8.2.1.3 Strategy: One SYNC_DL Code per Site ..............................................................................462

8.8.2.1.4 Strategy: Distributed per Site...............................................................................................462

8.8.2.2 Multi Carrier Network.................................................................................................................462

8.9 Automatic GSM/TD-SCDMA Neighbour Allocation ..............................................................................463




8.9.1.3 Appendices................................................................................................................................466

8.9.1.3.1 Delete Existing Neighbours Option ......................................................................................466

8.9.1.3.2 Calculation of Inter-Transmitter Distance ............................................................................466

9 WiMAX BWA Networks..................................................................4699.1 Definitions and Formulas ......................................................................................................................469

9.1.1 Input ................................................................................................................................................469

9.1.2 Co- and Adjacent Channel Overlaps Calculation............................................................................473

9.1.3 Preamble Signal Quality Calculations .............................................................................................4739.1.3.1 Preamble Signal Level Calculation............................................................................................473

9.1.3.2 Preamble Noise Calculation ......................................................................................................474

9.1.3.3 Preamble Interference Calculation ............................................................................................474

9.1.3.4 Preamble C/N Calculation .........................................................................................................474

9.1.3.5 Preamble C/(I+N) Calculation....................................................................................................474

9.1.4 Traffic and Pilot Signal Quality Calculations ...................................................................................475

9.1.4.1 Traffic and Pilot Signal Level Calculation (DL) ..........................................................................475

9.1.4.2 Traffic and Pilot Noise Calculation (DL) ....................................................................................475

9.1.4.3 Traffic and Pilot Interference Calculation (DL) ..........................................................................475

9.1.4.4 Traffic and Pilot C/N Calculation (DL) .......................................................................................476

9.1.4.5 Traffic and Pilot C/(I+N) Calculation (DL) ..................................................................................477

9.1.4.6 Traffic Signal Level Calculation (UL) .........................................................................................477

9.1.4.7 Traffic Noise Calculation (UL) ...................................................................................................477

9.1.4.8 Traffic Interference Calculation (UL) .........................................................................................478

9.1.4.9 Traffic C/N Calculation (UL) ......................................................................................................478

9.1.4.10 Traffic C/(I+N) Calculation (UL) .................................................................................................478

9.1.5 Throughput Calculation ...................................................................................................................478

9.1.5.1 Calculation of Total Cell Resources ..........................................................................................478

9.1.5.2 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation...........480

9.1.6 Scheduling and Radio Resource Management...............................................................................481

9.1.6.1 User Throughput Calculation.....................................................................................................482

9.1.7 Smart Antenna Models....................................................................................................................483

9.1.7.1 Downlink Beamforming .............................................................................................................483

9.1.7.2 Uplink Beamforming ..................................................................................................................483

9.1.7.3 Uplink Beamforming and Interference Cancellation (MMSE) ....................................................484

9.2 Calculation Processes ..........................................................................................................................484

9.2.1 Point Analysis: Profile Tab ..............................................................................................................484

9.2.2 Point Analysis: Reception Tab ........................................................................................................484

9.2.3 Point Analysis: Interference Tab .....................................................................................................485

9.2.4 Preamble Signal Level Coverage Predictions.................................................................................485

9.2.4.1 Coverage Area Determination ...................................................................................................486

9.2.4.1.1 All Servers ...........................................................................................................................486

9.2.4.1.2 Best Signal Level and a Margin ...........................................................................................486

9.2.4.1.3 Second Best Signal Level and a Margin ..............................................................................486

9.2.4.2 Coverage Display ......................................................................................................................486

9.2.4.2.1 Coverage Resolution ...........................................................................................................486

9.2.4.2.2 Display Types ......................................................................................................................486

9.2.5 Effective Signal Analysis Coverage Predictions..............................................................................487


9.2.5.2 Coverage Parameter Calculation ..............................................................................................488





Table of Contents

© Forsk 2010 AT282_TRG_E2 17

9.2.5.3.2 Effective Signal Analysis (DL) Display Types ..................................................................... 488

9.2.5.3.3 Effective Signal Analysis (UL) Display Types ..................................................................... 489

9.2.6 Calculations on Subscriber Lists .................................................................................................... 490

9.2.7 Monte Carlo Simulations ................................................................................................................ 490

9.2.7.1 Generating a Realistic User Distribution ................................................................................... 490

9.2.7.1.1 Simulations Based on User Profile Traffic Maps and Subscriber Lists ............................... 491

9.2.7.1.2 Simulations Based on Sector Traffic Maps ......................................................................... 492

9.2.7.2 Simulation Process ................................................................................................................... 494

9.2.8 C/(I+N)-Based Coverage Predictions ............................................................................................. 498

9.2.8.1 Coverage Area Determination .................................................................................................. 498

9.2.8.2 Coverage Parameter Calculation.............................................................................................. 498



9.2.8.3.2 Coverage by C/(I+N) Level (DL) Display Types .................................................................. 500

9.2.8.3.3 Coverage by Best Bearer (DL) Display Types .................................................................... 500

9.2.8.3.4 Coverage by Throughput (DL) Display Types..................................................................... 501

9.2.8.3.5 Coverage by Quality Indicator (DL) Display Types ............................................................. 501

9.2.8.3.6 Coverage by C/(I+N) Level (UL) Display Types .................................................................. 501

9.2.8.3.7 Coverage by Best Bearer (UL) Display Types .................................................................... 502

9.2.8.3.8 Coverage by Throughput (UL) Display Types..................................................................... 502

9.2.8.3.9 Coverage by Quality Indicator (UL) Display Types ............................................................. 503

9.3 Calculation Algorithms ......................................................................................................................... 504

9.3.1 Co- and Adjacent Channel Overlaps Calculation ........................................................................... 504

9.3.1.1 Conversion From Channel Numbers to Start and End Frequencies ........................................ 504

9.3.1.2 Co-Channel Overlap Calculation .............................................................................................. 505

9.3.1.3 Adjacent Channel Overlap Calculation..................................................................................... 506

9.3.1.4 FDD – TDD Overlap Ratio Calculation ..................................................................................... 506

9.3.1.5 Total Overlap Ratio Calculation ................................................................................................ 507

9.3.2 Preamble Signal Level and Quality Calculations............................................................................ 508

9.3.2.1 Preamble Signal Level Calculation ........................................................................................... 508

9.3.2.2 Preamble Noise Calculation ..................................................................................................... 509

9.3.2.3 Preamble Interference Calculation ........................................................................................... 510

9.3.2.4 Preamble C/N Calculation ........................................................................................................ 512

9.3.2.5 Preamble C/(I+N) Calculation ................................................................................................... 512

9.3.3 Best Server Determination ............................................................................................................. 513

9.3.4 Service Area Calculation ................................................................................................................ 513

9.3.5 Permutation Zone Selection (WiMAX 802.16e).............................................................................. 514

9.3.6 Traffic and Pilot Signal Level and Quality Calculations .................................................................. 515

9.3.6.1 Traffic and Pilot Signal Level Calculation (DL) ......................................................................... 515

9.3.6.2 Traffic and Pilot Noise Calculation (DL) .................................................................................... 516

9.3.6.3 Traffic and Pilot Interference Calculation (DL) .......................................................................... 5189.3.6.3.1 Traffic and Pilot Interference Signal Levels Calculation (DL).............................................. 518

9.3.6.3.2 Effective Traffic and Pilot Interference Calculation (DL) ..................................................... 522

9.3.6.4 Traffic and Pilot C/N Calculation (DL) ....................................................................................... 525

9.3.6.5 Traffic and Pilot C/(I+N) and Bearer Calculation (DL) .............................................................. 527

9.3.6.6 Traffic Signal Level Calculation (UL) ........................................................................................ 529

9.3.6.7 Traffic Noise Calculation (UL) ................................................................................................... 530

9.3.6.8 Traffic Interference Calculation (UL) ......................................................................................... 531

9.3.6.8.1 Traffic Interference Signal Levels Calculation (UL)............................................................. 531

9.3.6.8.2 Noise Rise Calculation (UL) ................................................................................................ 532

9.3.6.9 Traffic C/N Calculation (UL)...................................................................................................... 532

9.3.6.10 Traffic C/(I+N) and Bearer Calculation (UL) ............................................................................. 536

9.3.7 Throughput Calculation .................................................................................................................. 539

9.3.7.1 Calculation of Total Cell Resources.......................................................................................... 539

9.3.7.1.1 Calculation of Sampling Frequency .................................................................................... 539

9.3.7.1.2 Calculation of Symbol Duration........................................................................................... 5409.3.7.1.3 Calculation of Total Cell Resources - TDD Networks ......................................................... 540

9.3.7.1.4 Calculation of Total Cell Resources - FDD Networks ......................................................... 542

9.3.7.2 Channel Throughput, Cell Capacity, and Allocated Bandwidth Throughput Calculation .......... 542

9.3.8 Scheduling and Radio Resource Management .............................................................................. 546

9.3.8.1 Scheduling and Radio Resource Allocation.............................................................................. 546

9.3.8.2 User Throughput Calculation .................................................................................................... 552

9.3.9 Smart Antenna Models ................................................................................................................... 553

9.3.9.1 Downlink Beamforming ............................................................................................................. 554

9.3.9.2 Uplink Beamforming ................................................................................................................. 555

9.3.9.3 Uplink Beamforming and Interference Cancellation (MMSE) ................................................... 557

9.4 Automatic Allocation Algorithms........................................................................................................... 558

9.4.1 Automatic Neighbour Allocation ..................................................................................................... 558




18 AT282_TRG_E2 © Forsk 2010

9.4.2 Automatic Inter-Technology Neighbour Allocation ..........................................................................561

9.4.3 Automatic Frequency Planning .......................................................................................................563

9.4.3.1 Separation Constraint and Relationship Weights ......................................................................564

9.4.3.2 Calculation of Cost Between TBA and Related Cells ................................................................564

9.4.3.3 AFP Algorithm ...........................................................................................................................566

9.4.4 Automatic Preamble Index Allocation .............................................................................................566

9.4.4.1 Constraint and Relationship Weights ........................................................................................567

9.4.4.2 Calculation of Cost Between TBA and Related Cells ................................................................568

9.4.4.3 Automatic Allocation Algorithm..................................................................................................570

10 LTE Networks ................................................................................57310.1 Definitions and Formulas ......................................................................................................................573

10.1.1 Input ................................................................................................................................................573

10.1.2 Downlink Transmission Powers Calculation ...................................................................................577

10.1.3 Co- and Adjacent Channel Overlaps Calculation............................................................................578

10.1.4 Signal Level and Signal Quality Calculations..................................................................................579

10.1.4.1 Signal Level Calculation (DL) ....................................................................................................579

10.1.4.2 Noise Calculation (DL) ..............................................................................................................580

10.1.4.3 Interference Calculation (DL) ....................................................................................................580

10.1.4.4 C/N Calculation (DL) .................................................................................................................581

10.1.4.5 C/(I+N) Calculation (DL) ............................................................................................................581

10.1.4.6 Signal Level Calculation (UL) ....................................................................................................582

10.1.4.7 Noise Calculation (UL) ..............................................................................................................583

10.1.4.8 Interference Calculation (UL) ....................................................................................................583

10.1.4.9 Noise Rise Calculation (UL) ......................................................................................................583

10.1.4.10 C/N Calculation (UL) .................................................................................................................58410.1.4.11 C/(I+N) Calculation (UL) ............................................................................................................584

10.1.5 Throughput Calculation ...................................................................................................................584

10.1.5.1 Calculation of Downlink Cell Resources....................................................................................584

10.1.5.2 Calculation of Uplink Cell Resources ........................................................................................585

10.1.5.3 Channel Throughput, Cell Capacity, Allocated Bandwidth Throughput, and Average User

Throughput Calculation586

10.1.6 Scheduling and Radio Resource Management...............................................................................587

10.1.6.1 User Throughput Calculation.....................................................................................................588

10.2 Calculation Processes ..........................................................................................................................589

10.2.1 Point Analysis: Profile Tab ..............................................................................................................589

10.2.2 Point Analysis: Reception Tab ........................................................................................................589

10.2.3 Point Analysis: Interference Tab .....................................................................................................589

10.2.4 Downlink Reference Signal Level Coverage Predictions ................................................................590


10.2.4.1.1 All Servers ...........................................................................................................................590

10.2.4.1.2 Best Signal Level and a Margin ...........................................................................................590

10.2.4.1.3 Second Best Signal Level and a Margin ..............................................................................590



10.2.4.2.2 Display Types ......................................................................................................................591

10.2.5 Effective Signal Analysis Coverage Predictions..............................................................................592





10.2.5.3.2 Effective Signal Analysis (DL) Display Types ......................................................................593

10.2.5.3.3 Effective Signal Analysis (UL) Display Types ......................................................................594

10.2.6 Calculations on Subscriber Lists .....................................................................................................594

10.2.7 Monte Carlo Simulations .................................................................................................................595

10.2.7.1 Generating a Realistic User Distribution ...................................................................................595

10.2.7.1.1 Simulations Based on User Profile Traffic Maps and Subscriber Lists ................................595

10.2.7.1.2 Simulations Based on Sector Traffic Maps..........................................................................597

10.2.7.2 Simulation Process....................................................................................................................598

10.2.8 C/(I+N)-Based Coverage Predictions..............................................................................................601





10.2.8.3.2 Coverage by C/(I+N) Level (DL) Display Types ..................................................................603

10.2.8.3.3 Coverage by Best Bearer (DL) Display Types .....................................................................604

10.2.8.3.4 Coverage by Throughput (DL) Display Types .....................................................................605

10.2.8.3.5 Coverage by Quality Indicator (DL) Display Types ..............................................................606



Table of Contents

© Forsk 2010 AT282_TRG_E2 19

10.2.8.3.6 Coverage by C/(I+N) Level (UL) Display Types .................................................................. 606

10.2.8.3.7 Coverage by Best Bearer (UL) Display Types .................................................................... 606

10.2.8.3.8 Coverage by Throughput (UL) Display Types..................................................................... 607

10.2.8.3.9 Coverage by Quality Indicator (UL) Display Types ............................................................. 608

10.3 Calculation Algorithms ......................................................................................................................... 608

10.3.1 Downlink Transmission Powers Calculation................................................................................... 608

10.3.2 Co- and Adjacent Channel Overlaps Calculation ........................................................................... 612

10.3.2.1 Conversion From Channel Numbers to Start and End Frequencies ........................................ 613

10.3.2.2 Co-Channel Overlap Calculation .............................................................................................. 613

10.3.2.3 Adjacent Channel Overlap Calculation ..................................................................................... 614

10.3.2.4 Total Overlap Ratio Calculation ................................................................................................ 615

10.3.3 Signal Level and Signal Quality Calculations ................................................................................. 615

10.3.3.1 Signal Level Calculation (DL) ................................................................................................... 616

10.3.3.2 Noise Calculation (DL).............................................................................................................. 618

10.3.3.3 Interference Calculation (DL).................................................................................................... 619

10.3.3.4 C/N Calculation (DL)................................................................................................................. 623

10.3.3.5 C/(I+N) and Bearer Calculation (DL) ........................................................................................ 625

10.3.3.6 Signal Level Calculation (UL) ................................................................................................... 629

10.3.3.7 Noise Calculation (UL).............................................................................................................. 631

10.3.3.8 Interference Calculation (UL).................................................................................................... 631

10.3.3.8.1 Interfering Signal Level Calculation (UL)............................................................................. 632

10.3.3.8.2 Noise Rise Calculation (UL) ................................................................................................ 633

10.3.3.9 C/N Calculation (UL)................................................................................................................. 634

10.3.3.10 C/(I+N) and Bearer Calculation (UL) ........................................................................................ 636

10.3.4 Best Server Determination ............................................................................................................. 639

10.3.5 Service Area Calculation................................................................................................................ 640

10.3.6 Throughput Calculation .................................................................................................................. 641

10.3.6.1 Calculation of Total Cell Resources.......................................................................................... 641

10.3.6.1.1 Calculation of Downlink Cell Resources ............................................................................. 641

10.3.6.1.2 Calculation of Uplink Cell Resources .................................................................................. 643

10.3.6.2 Channel Throughput, Cell Capacity, Allocated Bandwidth Throughput, and Average User

Throughput Calculation644

10.3.7 Scheduling and Radio Resource Management .............................................................................. 647

10.3.7.1 Scheduling and Radio Resource Allocation.............................................................................. 648

10.3.7.2 User Throughput Calculation .................................................................................................... 653

10.4 Automatic Allocation Algorithms........................................................................................................... 654

10.4.1 Automatic Neighbour Allocation ..................................................................................................... 654

10.4.2 Automatic Inter-Technology Neighbour Allocation ......................................................................... 657

10.4.3 Automatic Frequency Planning ...................................................................................................... 659

10.4.3.1 Separation Constraint and Relationship Weights ..................................................................... 660

10.4.3.2 Calculation of Cost Between TBA and Related Cells ............................................................... 66010.4.3.3 AFP Algorithm........................................................................................................................... 662

10.4.4 Automatic Physical Cell ID Allocation............................................................................................. 662

10.4.4.1 Constraint and Relationship Weights........................................................................................ 663

10.4.4.2 Calculation of Cost Between TBA and Related Cells ............................................................... 664

10.4.4.3 Automatic Allocation Algorithm ................................................................................................. 666

11 Repeaters and Remote Antennas................................................. 66911.1 Modelling Repeaters ............................................................................................................................ 669

11.1.1 CDMA Documents.......................................................................................................................... 669

11.1.1.1 Over the Air ............................................................................................................................... 669

11.1.1.1.1 Signal Level Received From Repeaters.............................................................................. 669

11.1.1.1.2 Gain Automatic Calculation................................................................................................. 670

11.1.1.1.3 Donor Side Parameter Automatic Calculation..................................................................... 671

11.1.1.2 Microwave Link ......................................................................................................................... 672



11.1.1.3 Fibre Link .................................................................................................................................. 674



11.1.1.4 Appendices ............................................................................................................................... 675

11.1.1.4.1 Automatic Controls.............................................................................................................. 675

11.1.1.4.2 Carrier Power and Interference Calculation ........................................................................ 676

11.1.1.4.3 Consideration of Repeater Noise Figure............................................................................. 678

11.1.2 GSM Documents ............................................................................................................................ 678

11.1.2.1 Over the Air ............................................................................................................................... 678


11.1.2.1.2 EIRP Automatic Calculation ................................................................................................ 679




20 AT282_TRG_E2 © Forsk 2010

11.1.2.1.3 Donor Side Parameter Automatic Calculation .....................................................................680

11.1.2.2 Microwave Link..........................................................................................................................681

11.1.2.2.1 Signal Level Received From Repeaters ..............................................................................681

11.1.2.2.2 EIRP Automatic Calculation.................................................................................................681

11.1.2.3 Fibre Link...................................................................................................................................682

11.1.2.3.1 Signal Level Received From Repeaters ..............................................................................682

11.1.2.3.2 EIRP Automatic Calculation.................................................................................................683

11.1.2.4 Appendices................................................................................................................................683

11.1.2.4.1 Automatic Controls ..............................................................................................................683

11.2 Modelling Remote Antennas.................................................................................................................684

11.2.1 CDMA Documents ..........................................................................................................................684

11.2.1.1 Signal Level Received From Remote Antennas........................................................................684

11.2.1.2 Gain Automatic Calculation .......................................................................................................684

11.2.2 GSM Documents.............................................................................................................................685

11.2.2.1 Signal Level Received From Remote Antennas........................................................................685

11.2.2.2 EIRP Automatic Calculation ......................................................................................................686



© Forsk 2010 AT282_TRG_E2 21

List of Figures

List of Figures

Figure 2.1: Digital Terrain Model.................................................................................................................................. 37

Figure 2.2: Schematic view of a DTM file .................................................................................................................... 37

Figure 2.3: Clutter Classes .......................................................................................................................................... 38

Figure 3.1: 2D Antenna Pattern Format Containing Co-polar Diagrams Only............................................................. 73

Figure 3.2: 2D Antenna Pattern Format Containing Co-polar and Cross-polar Diagrams........................................... 74

Figure 4.1: Example 1: Single Calculation Area .......................................................................................................... 81

Figure 4.2: Example 2: Multiple Calculation Areas ...................................................................................................... 81

Figure 4.3: Ground Altitude Determination - 1............................................................................................................. 83




Figure 4.7: Clutter Height............................................................................................................................................. 84

Figure 4.8: Radial calculation method.......................................................................................................................... 85

Figure 4.9: Site-bin centre profile................................................................................................................................. 85

Figure 4.10: Radial calculation method.......................................................................................................................... 86

Figure 4.11: Enhanced Slope at Receiver ..................................................................................................................... 93

Figure 4.12: Losses due to Clutter................................................................................................................................. 96Figure 4.13: Tx-Rx profile .............................................................................................................................................. 97

Figure 4.14: Knife-Edge Diffraction.............................................................................................................................. 110

Figure 4.15: Deygout Construction – 1 Obstacle ......................................................................................................... 111

Figure 4.16: Deygout Construction – 3 Obstacles ....................................................................................................... 112

Figure 4.17: Epstein-Peterson Construction................................................................................................................ 112

Figure 4.18: Millington Construction ............................................................................................................................ 113

Figure 4.19: Azimuth and Tilt Computation.................................................................................................................. 115

Figure 4.20: Vertical Pattern Transformation due to Electrical Downtilt....................................................................... 117

Figure 4.21: Vertical Antenna Pattern.......................................................................................................................... 118

Figure 4.22: Peaks and Nulls in the Antenna Pattern.................................................................................................. 118

Figure 4.23: Log-normal Probability Density Function ................................................................................................. 119

Figure 4.24: Normalised Margin .................................................................................................................................. 125

Figure 4.25: Margin - Probability (Case of 2 Signals) .................................................................................................. 130

Figure 4.26: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 1dB) ............................................... 130

Figure 4.27: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 2dB) ............................................... 131

Figure 4.28: Reference Point - Location of the Transmission/Reception parameters ................................................. 131

Figure 5.1: FER vs. C/I Graphs.................................................................................................................................. 157

Figure 5.2: BER vs. C/I Graphs ................................................................................................................................. 157

Figure 5.3: MOS vs. C/I Graphs................................................................................................................................. 157

Figure 5.4: Representation of a Concentric Cell TXi.................................................................................................. 164

Figure 5.5: Representation of Micro and Macro Layers............................................................................................. 165

Figure 5.6: Concentric Cells....................................................................................................................................... 166

Figure 5.7: Concentric Cells....................................................................................................................................... 171

Figure 5.8: Reduction of Throughput per Timeslot .................................................................................................... 175

Figure 5.9: Reduction Factor for Different Packet Switched Traffic Loads (Lp, X-axis)............................................. 176

Figure 5.10: Blocking Probability for Different Packet Switched Traffic Loads (Lp, X-axis)......................................... 177

Figure 5.11: Network Dimensioning Process............................................................................................................... 178Figure 5.12: Minimum Throughput Reduction Factor .................................................................................................. 181

Figure 5.13: Overlapping Zones .................................................................................................................................. 186

Figure 5.14: The Advanced tab of the AFP module Properties dialogue..................................................................... 193

Figure 5.15: The cumulative density of C/I levels between [TX1, BCCH] and [TX2, BCCH]....................................... 196

Figure 6.1: Description of a Packet Session.............................................................................................................. 214

Figure 6.2: UMTS HSPA Power Control Algorithm.................................................................................................... 218

Figure 6.3: Connection status of HSDPA bearer users ............................................................................................. 224

Figure 6.4: HSDPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users..................... 225

Figure 6.5: HSDPA Bearer Allocation Process for Packet (HSDPA) and Packet (HSPA) Service Users ................. 226

Figure 6.6: HSDPA UE Categories Table.................................................................................................................. 231

Figure 6.7: HSDPA Radio Bearers Table .................................................................................................................. 232




22 AT282_TRG_E2 © Forsk 2010

Figure 6.8: HSUPA UE Categories Table................................................................................................................... 239

Figure 6.9: HSUPA Radio Bearers Table ................................................................................................................... 239

Figure 6.10: HSUPA Bearer SelectionTable ................................................................................................................ 240

Figure 6.11: HSUPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users...................... 241

Figure 6.12: HSUPA Bearer Allocation Process for Packet (HSPA) Service Users..................................................... 241

Figure 6.13: OVSF Code Tree Indices (Not OVSF Code Numbers) ............................................................................ 253

Figure 6.14: Overlapping Zone for Intra-carrier Neighbours......................................................................................... 285

Figure 6.15: Overlapping Zone for Inter-carrier Neighbours - 1st Case ....................................................................... 286

Figure 6.16: Overlapping Zone for Inter-carrier Neighbours - 2nd Case...................................................................... 287

Figure 6.17: Neighbourhood Constraints...................................................................................................................... 294

Figure 6.18: Primary Scrambling Codes Allocation...................................................................................................... 296

Figure 6.19: Inter-Transmitter Distance Computation .................................................................................................. 302

Figure 7.1: CDMA2000 1xRTT Power Control Algorithm ........................................................................................... 332

Figure 7.2: CDMA2000 1xEVDO Power Control Algorithm........................................................................................ 339

Figure 7.3: Walsh Code Tree Indices (Not Walsh Code Numbers)............................................................................ 346

Figure 7.4: Overlapping Zones - 1st Case.................................................................................................................. 375

Figure 7.5: Overlapping Zones - 2nd Case ................................................................................................................ 376


Figure 7.7: PN Offset Allocation ................................................................................................................................. 384

Figure 8.1: Description of a Packet Session............................................................................................................... 408

Figure 8.2: TD-SCDMA Power Control Algorithm ...................................................................................................... 412

Figure 8.3: Grid Of Beams Modelling ......................................................................................................................... 435

Figure 8.4: GOB Modelling - Determination of the Best Beam................................................................................... 436

Figure 8.5: Adaptive Beam Modelling - Determination of the Best Beam .................................................................. 436Figure 8.6: Linear Adaptive Antenna Array ................................................................................................................ 437

Figure 8.7: Downlink Beamforming ............................................................................................................................ 438

Figure 8.8: Uplink Beamforming ................................................................................................................................. 439

Figure 8.9: Uplink Adaptive Algorithm ........................................................................................................................ 440

Figure 8.10: Construction of the Geographic Distribution of Downlink Traffic Power................................................... 442

Figure 8.11: Geographic Distribution of Downlink Traffic Power .................................................................................. 442

Figure 8.12: Geographic Distribution of downlink traffic power and uplink load ........................................................... 443

Figure 8.13: Radio Bearers Table ................................................................................................................................ 446

Figure 8.14: UE Categories Table................................................................................................................................ 447

Figure 8.15: Weighted Distance Between Transmitters ............................................................................................... 450

Figure 8.16: N-frequency Neighbour Allocation............................................................................................................ 452

Figure 8.17: Overlapping Coverages............................................................................................................................ 452


Figure 8.19: Scrambling Code Allocation Example ...................................................................................................... 460Figure 8.20: Scrambling Code Allocation to All Carriers .............................................................................................. 463

Figure 8.21: Inter-Transmitter Distance Computation .................................................................................................. 466

Figure 9.1: WiMAX Simulation Algorithm ................................................................................................................... 494

Figure 9.2: Victim and Interfering Mobiles.................................................................................................................. 495

Figure 9.3: Simulation Convergence Stability Factor ................................................................................................. 496

Figure 9.4: Co-Channel and Adjacent Channel Overlaps .......................................................................................... 504

Figure 9.5: Downlink C/(I+N) calculation in Simulations............................................................................................. 520

Figure 9.6: Downlink C/(I+N) calculation in Coverage Predictions............................................................................. 521

Figure 9.7: Segmentation ........................................................................................................................................... 523

Figure 9.8: Segmentation Interference Scenarios ...................................................................................................... 524

Figure 9.9: Linear Adaptive Antenna Array ................................................................................................................ 554

Figure 9.10: Downlink Beamforming ............................................................................................................................ 554

Figure 9.11: Uplink Beamforming ................................................................................................................................. 555Figure 9.12: Uplink Adaptive Algorithm ........................................................................................................................ 557

Figure 9.13: Determination of Adjacent Cells............................................................................................................... 559

Figure 9.14: Overlapping Zones ................................................................................................................................... 560

Figure 9.15: Inter-Transmitter Distance Calculation ..................................................................................................... 562

Figure 9.16: Weighted Distance Between Cells ........................................................................................................... 566

Figure 9.17: Importance Based on Distance Relation.................................................................................................. 566

Figure 9.18: Weighted Distance Between Cells ........................................................................................................... 570

Figure 9.19: Importance Based on Distance Relation.................................................................................................. 570

Figure 10.1: LTE Simulation Algorithm......................................................................................................................... 599

Figure 10.2: Co-Channel and Adjacent Channel Overlaps .......................................................................................... 612

Figure 10.3: Determination of Adjacent Cells............................................................................................................... 655



© Forsk 2010 AT282_TRG_E2 23

List of Figures

Figure 10.4: Overlapping Zones .................................................................................................................................. 656

Figure 10.5: Inter-Transmitter Distance Calculation .................................................................................................... 658

Figure 10.6: Weighted Distance Between Cells........................................................................................................... 662

Figure 10.7: Importance Based on Distance Relation ................................................................................................. 662

Figure 10.8: Weighted Distance Between Cells........................................................................................................... 666

Figure 10.9: Importance Based on Distance Relation ................................................................................................. 666

Figure 11.1: CDMA Documents - Over the Air Repeater............................................................................................. 669

Figure 11.2: Over the Air Repeater - Downlink Total Gain .......................................................................................... 670

Figure 11.3: Over the Air Repeater - Uplink Total Gain............................................................................................... 671

Figure 11.4: Angle from North (Azimuth) ..................................................................................................................... 671

Figure 11.5: Positive/Negative Mechanical Downtilt .................................................................................................... 672

Figure 11.6: Tilt Angle Computation ............................................................................................................................ 672

Figure 11.7: CDMA Documents - Microwave Link Repeater ....................................................................................... 672

Figure 11.8: Microwave Link Repeater - Downlink Total Gain..................................................................................... 673

Figure 11.9: Microwave Link Repeater - Uplink Total Gain ......................................................................................... 673

Figure 11.10: CDMA Documents - Fibre Link Repeater ................................................................................................ 674

Figure 11.11: Fibre Link Repeater - Downlink Total Gain.............................................................................................. 675

Figure 11.12: Fibre Link Repeater - Uplink Total Gain .................................................................................................. 675

Figure 11.13: GSM Documents - Over the Air Repeater ............................................................................................... 679

Figure 11.14: Over the Air Repeater - EIRP .................................................................................................................. 679

Figure 11.15: Angle from North (Azimuth) ..................................................................................................................... 680

Figure 11.16: Positive/Negative Mechanical Downtilt.................................................................................................... 680

Figure 11.17: Tilt Angle Computation ............................................................................................................................ 680

Figure 11.18: GSM Documents - Microwave Link Repeater.......................................................................................... 681Figure 11.19: Microwave Link Repeater - EIRP............................................................................................................. 682

Figure 11.20: GSM Documents - Fibre Link Repeater................................................................................................... 682

Figure 11.21: Fibre Link Repeater - EIRP...................................................................................................................... 683

Figure 11.22: CDMA Documents - Remote Antenna Signal Level ................................................................................ 684

Figure 11.23: Remote Antennas - Downlink Total Gain................................................................................................. 685

Figure 11.24: Remote Antennas - Uplink Total Gain ..................................................................................................... 685

Figure 11.25: GSM Documents - Remote Antenna Signal Level................................................................................... 685

Figure 11.26: Remote Antennas - EIRP ........................................................................................................................ 686




24 AT282_TRG_E2 © Forsk 2010



Chapter 1Coordinate Systems and Units

This chapter presents the different coordinate systems available in Atoll by default. It describes the projection, display, and internal coordinate

systems, and describes the format of the coordinate systems files. This chapter also provides details of the different power and length units

available in Atoll .


http://slidepdf.com/reader/full/atoll-28-chapters1-6 26/30426 AT282_TRG_E2 © Forsk 2010




© Forsk 2010 AT282_TRG_E2 27

Chapter 1: Coordinate Systems and Units

1 Coordinate Systems and Units

1.1 Coordinate Systems A map or a geo-spatial database is a flat representation of data collected from a curved surface. A projection is a means

for producing all or part of a spheroid on a flat sheet. This projection cannot be done without distortion. Therefore, the

cartographer must choose the characteristic (distance, direction, scale, area, or shape) that he wants to be shown accu-

rately at the expense of the other characteristics, or compromise on several characteristics [1-3]. The projected zones are

referenced using cartographic coordinates (meter, yard, etc.). Two projection methods are widely used:• The Lambert Conformal-Conic Method: A portion of the earth is mathematically projected on a cone conceptu-

ally secant at one or two standard parallels. This projection method is useful for representing countries or regions

that have a predominant east-west expanse.

• The Universal Transverse Mercator (UTM) Method: A portion of the earth is mathematically projected on a cyl-

inder tangent to a meridian (which is transverse or crosswise to the equator). This projection method is useful for

mapping large areas that are oriented north-south.

The geographic system is not a projection. It is only a representation of a location on the surface of the earth in geographic

coordinates (degree-minute-second, grade) giving the latitude and longitude in relation to the meridian origin (e.g., Paris

for NTF system and Greenwich for ED50 system). The locations in the geographic system can be converted into other

projections.

1.1.1 Description of Coordinate Systems

A Geographic coordinate system is a latitude and longitude coordinate system. The latitude and longitude are related to

an ellipsoid, a geodetic datum, and a prime meridian. The geodetic datum provides the position and orientation of the ellip-

soid relative to the earth.

Cartographic coordinate systems are obtained by transforming each (latitude, longitude) value into an (easting, northing)

value. A projection coordinate system is obtained by transforming each (latitude, longitude) value into an (easting, north-

ing) value. Projection coordinate systems are geographic coordinate systems that provide longitude and latitude, and the

transformation method characterised by a set of parameters. Different methods may require different sets of parameters.

For example, the parameters required for Transverse Mercator coordinate systems are:• The longitude of the natural origin (central meridian)

• The latitude of the natural origin

• The False Easting value

• The False Northing value

• A scaling factor at the natural origin (central meridian)

Basic definitions are presented below.

1.1.1.1 Geographic Coordinate System

The geographic coordinate system is a datum and a meridian. Atoll enables you to choose the most suitable geographic

coordinate system for your geographic data.

1.1.1.2 Datum

The datum consists of the ellipsoid and its position relative to the WGS84 ellipsoid. In addition to the ellipsoid, translation,rotation, and distortion parameters define the datum.

1.1.1.3 Meridian

The standard meridian is Greenwich, but some geographic coordinate systems are based on other meridians. These

meridians are defined by the longitude with respect to Greenwich.

1.1.1.4 Ellipsoid

The ellipsoid is the pattern used to model the earth. It is defined by i ts geometric parameters.

References:[1] Snyder, John. P., Map Projections Used by the US Geological Survey, 2nd Edition, United States Government

Printing Office, Washington, D.C., 313 pages, 1982.

[2] http://www.colorado.edu/geography/gcraft/notes/gps/gps_f.html

[3] http://www.posc.org/Epicentre.2_2/DataModel/ExamplesofUsage/eu_cs34.html

[4] http://www.ign.fr/telechargement/Pi/SERVICES/transfo.pdf (Document in French)




1.1.1.5 Projection

The projection is the transformation applied to project the ellipsoid of the earth on to a plane. There are different projection

methods that use specific sets of parameters.

1.1.1.6 Projection Coordinate System

The projection coordinate system is the result of the application of a projection to a geographic coordinate system. It asso-

ciates a geographic coordinate system and a projection. Atoll enables you to choose the projection coordinate system

matching your geographic data.

1.1.2 Coordinate Systems in AtollDepending on the working environment, there can be either two or four coordinate systems used in Atoll. If you are work-

ing with stand-alone documents, i.e., documents not connected to databases, there are two coordinate systems used in

Atoll:

• Projection coordinate system

• Display coordinate system

If you are working in a multi-user environment, Atoll uses four coordinate systems:

• Projection coordinate system for the Atoll document

• Display coordinate system for the Atoll document

• Internal projection coordinate system for the database

• Internal display coordinate system for the database

1.1.2.1 Projection Coordinate System

The projection coordinate system is the coordinate system of the available raster geographic data files. You should set the

projection coordinate system of your Atoll document so that it corresponds to the coordinate system of the available raster

geographic data. You can set the projection coordinate system of your document in the Options dialog.

All the raster geographic data files that you want to import and use in an Atoll document must have the same coordinate

system. You cannot work with raster geographic data files with different coordinate systems in the same document.

The projection coordinate system is used to keep the coordinates of sites (radio network data) consistent with the

geographic data.

When you import a raster geographic data file, Atoll reads the geo-referencing information from the file (or from its header

file, depending on the geographic data file format), i.e., its Northwest pixel, to determine the coordinates of each pixel.Atoll does not use any coordinate system during the import process. However, the geo-referencing information of

geographic data files are considered to be provided in the projection coordinate system of the document.

1.1.2.2 Display Coordinate System

The display coordinate system is the coordinate system used for the display, e.g., in dialogs, in the Map window rulers, in

the status bar, etc. The coordinates of each pixel of geographic data are converted to the display coordinate system from

the projection coordinate system for display. The display coordinate system is also used for sites (radio network data). You

can set the display coordinate system of your document in the Options dialog.

If you import sites data, the coordinate system of the sites must correspond to the display coordinate system of your Atoll

document.

If you change the display coordinate system in a document which is not connected to a database, the coordinates of all

the sites are converted to the new display system.

1.1.2.3 Internal Coordinate Systems

The internal coordinate systems are the projection and the display coordinate systems stored in a database. The projection

and display coordinate systems set by the administrator in the central Atoll project are stored in the database when the

database is created, and cannot be modified by users. Only the administrator can modify the internal coordinate systems

manually by editing the entries in the CoordSys and the Units tables. All Atoll documents opened from a database will

have the internal coordinate systems of the database as their default projection and display coordinate systems.

Note:

• If you import vector geographic data (e.g., traffic, measurements, etc.) with different

coordinate systems, it is possible to convert the coordinate systems of these data into the

projection coordinate system of your Atoll document.

Note:

• If the coordinate systems of all your geographic data files and sites (radio network data) are

the same, you do not have to define the projection and display coordinate systems

separately. By default, the two coordinate systems are the same.



© Forsk 2010 AT282_TRG_E2 29


When exporting an Atoll project to a database, the currently chosen display coordinate system becomes the internal

display coordinate system for the database, and the currently chosen projection coordinate system becomes the internal

projection coordinate system for the database.

Although Atoll stores both the coordinate systems in the database, i.e., the projection and the display coordinate systems,

the only relevant coordinate system for the database is the internal display coordinate system because this coordinate

system is the one used for the coordinates of sites (radio network data).

Users working on documents connected to a database can modify the coordinate systems in their documents locally, and

save these changes in their documents, but they cannot modify the coordinate systems stored in the database.

If you change the display coordinate system in a document which is not connected to a database, the coordinates of all

the sites are converted to the new display system.

If you change the display coordinate system in a document which is connected to a database, the coordinates of all thesites are converted to the new coordinate system in the Atoll document locally but not in the database because the internal

coordinate systems cannot be changed.

Atoll uses the internal coordinates systems in order to keep the site coordinates consistent in the database which is

usually accessed by a large number of users in a multi-user environment.

1.1.3 File Formats

The Coordsystems folder located in the Atoll installation directory contains all the coordinate systems, both geographic

and cartographic, offered in the tool. Coordinate systems are grouped by regions. A catalogue per region and a "Favour-

ites" catalogue are available in Atoll. The Favourites catalogue is initially empty and can be filled by the user by adding

coordinate systems to it. Each catalogue is described by an ASCII text file with .cs extension. In a .cs file, each coordinate

system is described in one line. The line syntax for describing a coordinate system is:

Examples:

You should keep the following points in mind when editing or creating .cs files:

• The identification code enables Atoll to differentiate coordinates systems. In case you create a new coordinate

system, its code must be an integer value higher than 32767.

• When describing a new datum, you must enter the ellipsoid code and parameters instead of the datum code in

brackets. There can be 3 to 7 parameters defined in the following order: Dx, Dy, Dz, Rx, Ry, Rz, S. The syntax of

the line in the .cs file will be:

• There can be up to seven projection parameters. These parameters must be ordered according to the parameter

index (see "Projection Parameter Indices" on page 32). Parameter with index 0 is the first one. Projection param-

eters are delimited by commas.

• For UTM projections, you must provide positive UTM zone numbers for north UTM zones and negative numbers

for south UTM zones.

• You can add all other information as comments (such as usage or region).

Codes of units, data, projection methods, and ellipsoids, and projection parameter indices are listed in the tables below.

1.1.3.1 Unit Codes

Code = "Name of the system"; Unit Code; Datum Code; Projection Method Code,

Projection Parameters; "Comments"

4230 = "ED50"; 101; 230; 1; "Europe - west"

32045 = "NAD27 / Vermont"; 2; 267; 6, -72.5, 42.5, 500000, 0, 0.9999643; "United

States - Vermont"

Code = "Name of the system"; Unit Code; {Ellipsoid Code, Dx, Dy, Dz, Rx, Ry,

Rz, S}; Projection Method Code, Projection Parameters; "Comments"

Code Cartographic Units Code Geographic Units

0 Metre 100 Radian

1 Kilometre 101 Degree

2 Foot 102 Grad

3 Link 103 ArcMinute

4 Chain 104 ArcSecond

5 Yard

6 Nautical mile

7 Mile

-1 Unspecified -1 Unspecified




1.1.3.2 Datum Codes

Code Datum Code Datum

121 Greek Geodetic Reference System 1987 260 Manoca

125 Samboja 261 Merchich

126 Lithuania 1994 262 Massawa

130 Moznet (ITRF94) 263 Minna

131 Indian 1960 265 Monte Mario

201 Adindan 266 M'poraloko

202 Australian Geodetic Datum 1966 267 North American Datum 1927

203 Australian Geodetic Datum 1984 268 NAD Michigan

204 Ain el Abd 1970 269 North American Datum 1983

205 Afgooye 270 Nahrwan 1967

206 Agadez 271 Naparima 1972

207 Lisbon 272 New Zealand Geodetic Datum 1949

208 Aratu 273 NGO 1948

209 Arc 1950 274 Datum 73

210 Arc 1960 275 Nouvelle Triangulation Française

211 Batavia 276 NSWC 9Z-2

212 Barbados 277 OSGB 1936

213 Beduaram 278 OSGB 1970 (SN)214 Beijing 1954 279 OS (SN) 1980

215 Reseau National Belge 1950 280 Padang 1884

216 Bermuda 1957 281 Palestine 1923

217 Bern 1898 282 Pointe Noire

218 Bogota 283 Geocentric Datum of Australia 1994

219 Bukit Rimpah 284 Pulkovo 1942

221 Campo Inchauspe 285 Qatar

222 Cape 286 Qatar 1948

223 Carthage 287 Qornoq

224 Chua 288 Loma Quintana

225 Corrego Alegre 289 Amersfoort

226 Cote d'Ivoire 290 RT38

227 Deir ez Zor 291 South American Datum 1969

228 Douala 292 Sapper Hill 1943

229 Egypt 1907 293 Schwarzeck

230 European Datum 1950 294 Segora

231 European Datum 1987 295 Serindung

232 Fahud 296 Sudan

233 Gandajika 1970 297 Tananarive 1925

234 Garoua 298 Timbalai 1948

235 Guyane Francaise 299 TM65

236 Hu Tzu Shan 300 TM75

237 Hungarian Datum 1972 301 Tokyo

238 Indonesian Datum 1974 302 Trinidad 1903

239 Indian 1954 303 Trucial Coast 1948

240 Indian 1975 304 Voirol 1875

241 Jamaica 1875 305 Voirol Unifie 1960

242 Jamaica 1969 306 Bern 1938

243 Kalianpur 307 Nord Sahara 1959

244 Kandawala 308 Stockholm 1938

245 Kertau 309 Yacare

247 La Canoa 310 Yoff

248 Provisional South American Datum 1956 311 Zanderij



© Forsk 2010 AT282_TRG_E2 31


1.1.3.3 Projection Method Codes

1.1.3.4 Ellipsoid Codes

249 Lake 312 Militar-Geographische Institut

250 Leigon 313 Reseau National Belge 1972

251 Liberia 1964 314 Deutsche Hauptdreiecksnetz

252 Lome 315 Conakry 1905

253 Luzon 1911 322 WGS 72

254 Hito XVIII 1963 326 WGS 84

255 Herat North 901 Ancienne Triangulation Française

256 Mahe 1971 902 Nord de Guerre

257 Makassar 903 NAD 1927 Guatemala/Honduras/Salvador (Panama Zone)

258 European Reference System 1989

Code Projection Method Code Projection Method

0 Undefined 8 Oblique Stereographic

1 No projection > Longitude / Latitude 9 New Zealand Map Grid

2 Lambert Conformal Conical 1SP 10 Hotine Oblique Mercator

3 Lambert Conformal Conical 2SP 11 Laborde Oblique Mercator

4 Mercator 12 Swiss Oblique Cylindrical

5 Cassini-Soldner 13 Oblique Mercator

6 Transverse Mercator 14 UTM Projection

7 Transverse Mercator South Oriented

Code Name Major Axis Minor Axis

1 Airy 1830 6377563.396 6356256.90890985

2 Airy Modified 1849 6377340.189 6356034.44761111

3 Australian National Spheroid 6378160 6356774.71919531

4 Bessel 1841 6377397.155 6356078.96261866

5 Bessel Modified 6377492.018 6356173.50851316

6 Bessel Namibia 6377483.865 6356165.38276679

7 Clarke 1858 6378293.63924683 6356617.98173817

8 Clarke 1866 6378206.4 6356583.8

9 Clarke 1866 Michigan 6378693.7040359 6357069.45104614

10 Clarke 1880 (Benoit) 6378300.79 6356566.43

11 Clarke 1880 (IGN) 6378249.2 6356515

12 Clarke 1880 (RGS) 6378249.145 6356514.86954978

13 Clarke 1880 (Arc) 6378249.145 6356514.96656909

14 Clarke 1880 (SGA 1922) 6378249.2 6356514.99694178

15 Everest 1830 (1937 Adjustment) 6377276.345 6356075.41314024

16 Everest 1830 (1967 Definition) 6377298.556 6356097.5503009

17 Everest 1830 (1975 Definition) 6377301.243 6356100.231

18 Everest 1830 Modified 6377304.063 6356103.03899315

19 GRS 1980 6378137 6356752.31398972

20 Helmert 1906 6378200 6356818.16962789

21 Indonesian National Spheroid 6378160 6356774.50408554

22 International 1924 6378388 6356911.94612795

23 International 1967 6378160 6356774.71919530

24 Krassowsky 1940 6378245 6356863.01877305

25 NWL 9D 6378145 6356759.76948868

26 NWL 10D 6378135 6356750.52001609

27 Plessis 1817 6376523 6355862.93325557

28 Struve 1860 6378297 6356655.84708038




1.1.3.5 Projection Parameter Indices

1.1.4 Creating a Coordinate SystemAtoll provides a large catalogue of default coordinate systems. Nevertheless, it is possible to add the description of

geographic and cartographic coordinate systems. New coordinate systems can be created from scratch or initialised on

the basis of an existing one.

To create a new coordinate system from scratch:

1. Select Tools > Options. The Options dialog opens.

2. Select the Coordinates tab.

3. Click the browse button (...) on the right of the Projection field.

4. Click the New button. The Coordinate System dialog opens.

5. In the Coordinate System dialog,

a. Select the coordinate systems catalogue to which you want to add the new coordinate system.

b. In the General properties section: Enter a name for the new coordinate system, select a unit. You can also

enter any comments about its usage. Atoll assigns the code automatically.

c. In the Category section: Select the type of coordinate system. Enter the longitude and latitude for a geograph-

ic coordinate system, or the type of projection and its set of associated parameters for a cartographic coordi-

nate system (false easting and northing, and the first and second parallels).d. In the Geo section: Specify the meridian and choose a datum for the coordinate system. The associated el-

lipsoid is automatically selected. You can also describe a geodetic datum by selecting "..." in the Datum list.

In this case, you must provide parameters (Dx, Dy, Dz, Rx, Ry, Rz, and S) needed for the transformation of

the datum into WGS84, and an ellipsoid.

6. Click OK. The new coordinate system is added to the selected coordinate system catalogue.

To create a new coordinate system based on an existing system, select a coordinate system in the Coordinate Systems

dialog before clicking New in step 4. The new coordinate system is initialised with the values of the selected coordinate

system.

1.2 Units

1.2.1 Power Units

Depending on the working environment, there can be either one or two types of units for transmission and reception

powers. If you are working with stand-alone documents, i.e., documents not connected to databases, there is only one unit

used in Atoll:

• Display power units

If you are working in a multi-user environment, Atoll uses two type of units:

• Display power units for the Atoll document

• Internal power units for the database

The display units are used for the display in dialogs and tables, e.g., reception thresholds (coverage prediction properties,

etc.), and received signal levels (measurements, point analysis, coverage predictions etc.). You can set the display units

for your document in the Options dialog.

29 War Office 6378300.583 6356752.27021959

30 WGS 84 6378137 6356752.31398972

31 GEM 10C 6378137 6356752.31398972

32 OSU86F 6378136.2 6356751.51667196

33 OSU91A 6378136.3 6356751.61633668

34 Clarke 1880 6378249.13884613 6356514.96026256

35 Sphere 6371000 6371000

Index Projection Parameter Index Projection Parameter

0 UTM zone number 4 Scale factor at origin

0 Longitude of origin 4 Latitude of 1st parallel

1 Latitude of origin 5 Azimuth of central line

2 False Easting 5 Latitude of 2nd parallel

3 False Northing 6 Angle from rectified to skewed grid



© Forsk 2010 AT282_TRG_E2 33


The internal units are the power units stored in a database. The power units set by the administrator in the central Atoll

project are stored in the database when the database is created, and cannot be modified by users. Only the administrator

can modify the internal units manually by editing the entries in the Units tables. All Atoll documents opened from a data-

base will have the internal units of the database as their default power units.

Users working on documents connected to a database can modify the units in their documents locally, and save these

changes in their documents, but they cannot modify the units stored in the database.

1.2.2 Length Units

There are two types of units for distances, heights, and offsets:

• Display length units• Internal length units

The display length units are used to display distances, heights, and offsets in dialogs, tables, and the status bar. You can

set the display units for your document in the Options dialog.

The internal unit for lengths is metre for all Atoll documents whether they are connected to databases or not. The internal

unit is not stored in the databases. The internal unit cannot be changed.

1.3 BSIC FormatDepending on the working environment, there can be either one or two types of BSIC formats. If you are working with

stand-alone documents, i.e., documents not connected to databases, there is only one BSIC format:

• Display BSIC format

If you are working in a multi-user environment, Atoll uses two type of formats:

• Display BSIC format for the Atoll document

• Internal BSIC format for the database

The display format is used for the display in dialogs and tables. You can set the display format for your document from the

Transmitters folder’s context menu.

The internal format is the BSIC format stored in a database. The BSIC format set by the administrator in the central Atoll

project is stored in the database when the database is created, and cannot be modified by users. Only the administrator

can modify the internal format manually by editing the corresponding entry in the Units tables. All Atoll documents opened

from a database will have the internal format of the database as their default BSIC format.

Users working on documents connected to a database can modify the format in their documents locally, and save this

change in their documents, but they cannot modify the format stored in the database.






Chapter 2Geographic and Radio Data

This chapter defines the different types of data with which you can work in Atoll. These data can be geographic data, such as maps, and radio

network data, such as sites, antennas, other equipment and parameters.






© Forsk 2010 AT282_TRG_E2 37

Chapter 2: Geographic and Radio Data

2 Geographic and Radio Data

2.1 Geographic Data

2.1.1 Data Type

Atoll manages several geographic data types; DTM (Digital Terrain Model), clutter (Land-Use), scanned images, vector

data, traffic data, population, and any other generic data.

2.1.1.1 Digital Terrain Model (DTM)

The DTM (Digital Terrain Model or height) files describe the ground elevation above the sea level. DTM files supported by

Atoll are 16 bits/pixel relief maps in .tif, .bil, Planet© and Erdas Imagine formats and 8 bits/pixel relief maps in .tif, .bil,

Erdas Imagine and .bmp formats. DTM maps are taken into account in path loss calculations by Atoll propagation models.

DTM file provides altitude value (z stated in metre) on evenly spaced points. Abscissa and ordinate axes are respectively

oriented in right and downwards directions. Space between points is defined by pixel size (P stated in metre). Pixel size

must be the same in both directions. First point given in the file corresponds to the centre of the upper-left pixel of the map.

This point refers to the northwest point geo-referenced by Atoll. Four points (hence, four altitude values) are necessary

to describe a “bin”; these points are bin vertices.

Therefore, a n*n bin DTM file requires (n)2 points (altitude values).

Figure 2.1: Digital Terrain Model

Figure 2.2: Schematic view of a DTM file

Notes:

• Altitude values differ within a bin. Method used to calculate altitudes is described in the

Path loss calculations: Altitude determination part. Concerning DTM map display, Atoll

takes altitude of the southwest point of each bin to determine its colour.

• In most documents, Digital Elevation Model (DEM) and Digital Terrain Model (DTM) are

differentiated and do not have the same meaning. By definition, DEM refers to altitude

above sea level including, both, ground and clutter while DTM just corresponds to the

ground height above sea level. In Atoll, the DEM term may be used instead of DTM term.




2.1.1.2 Clutter (Land Use)

You may import two types of clutter files in ATL documents. These files indicate either the clutter class or the clutter height

on each bin of the map.

2.1.1.2.1 Clutter Classes

Atoll supports 8 bits/pixel (255 classes) raster maps in .tif, .bil, .bmp, Erdas Imagine formats or 16 bits/pixel raster maps

in Planet© format. This kind of clutter file describes the land cover (dense urban, buildings, residential, forest, open,

villages, …). A grid map represents ground and each bin of the map is characterised by a code corresponding to a main

type of cover (a clutter class). Atoll automatically lists all the clutter classes of the map. It is possible to specify an average

clutter height for each clutter class manually during the map description step. Clutter maps are taken into account in path

loss calculations by Atoll propagation models.

Clutter file provides a clutter code per bin. Bin size is defined by pixel size (P stated in metre). Pixel size must be the same

in both directions. Abscissa and ordinate axes are respectively oriented in right and downwards directions. First point given

in the file corresponds to the centre of the upper-left pixel of the image. This point refers to the northwest point geo-refer-

enced by Atoll.

Therefore, a n*n bin Clutter file requires (n)2 code values.

2.1.1.2.2 Clutter Heights

Files supported by Atoll for clutter heights are 8 or 16 bits/pixel raster maps in .tif, .bil and Erdas Imagine formats. The file

provides clutter height value on evenly spaced points. Abscissa and ordinate axes are respectively oriented in right anddownwards directions. Space between points is defined by pixel size (P in metre). Pixel size must be the same in both

directions. First point given in the file corresponds to the centre of the upper-left pixel of the map. This point refers to the

northwest point geo-referenced by Atoll.

These maps are taken into account in path loss calculations by Atoll propagation models.

2.1.1.3 Traffic Data

Atoll offers different kinds of traffic data:

2.1.1.3.1 User Profile Environment Based Traffic MapsAtoll supports 8 bits/pixel (256 class) traffic raster maps in .tif, .bil, .bmp, Erdas Imagine formats. These maps provide

macroscopic traffic estimation. Each pixel is assigned an environment class, which is a list of user profiles with a defined

mobility type and a density.

2.1.1.3.2 User Profile Traffic Maps

Atoll supports vector traffic maps with .dxf®, Planet©, .shp, .mif and .agd formats. These maps are detailed traffic estima-

tions (lines, polygons or points carrying a specific traffic). Each polygon, line or point is assigned a specific user profile with

associated mobility type and density. They can be built from population density vector maps.

2.1.1.3.3 Sector Traffic Maps

Atoll supports maps with .agd format. This kind of map is based on the network feedback. It provides actual information

on connections (and not just subscriber estimation) from the network. It is built from a coverage by transmitter prediction

Figure 2.3: Clutter Classes

Note:

• The clutter code is the same inside a bin.

Note:

• Atoll considers the clutter height of the nearest point in calculations (see Path loss

calculations: Clutter determination part). For map display, Atoll takes clutter height of the

southwest point of each bin to determine its colour.



© Forsk 2010 AT282_TRG_E2 39


study that defines sector boundaries for the traffic distribution in each sector. In UMTS and CDMA, either data rates or the

number of users per service are indicated for each transmitter service area. In GSM/TDMA, Atoll expects a number of

Erlangs in case of voice service and data rate values for packet-switched services for each transmitter service area.

2.1.1.3.4 User Density Traffic Maps

This kind of map is only available in GSM/TDMA documents. Atoll supports 16 and 32 bits/pixel traffic raster maps in .tif,

.bil, .bmp, Planet© and Erdas Imagine formats. This map is also based on the network feedback as it deals with network

users information as well. Each pixel is assigned a number of users with a given service, terminal and mobility type.

In GSM documents, traffic maps are taken into account for traffic analysis and network dimensioning. In UMTS and CDMA

documents, they are used by the Monte-Carlo simulator to model user distributions and evaluate related network param-

eters (cell power, mobile terminal power, …).

2.1.1.4 Vector Data

These data represent either polygons (regions, etc.), lines (roads, coastlines, etc.) or points (towns, etc.). Atoll supports

vector data files in .dxf®, Planet©, .shp, .mif and .agd formats. These maps are only used for display and provide informa-

tion about the geographic environment.

2.1.1.5 Scanned Images

These geographic data include the road maps and the satellite images. They are only used for display and provide infor-

mation about the geographic environment. Atoll supports scanned image files in .tif (1, 4, 8, 24-bits/pixel), .bil (1, 4, 8, 24-

bits/pixel), Planet© (1, 4, 8, 24-bits/pixel), .bmp (1-24-bits/pixel), Erdas Imagine (1, 4, 8, 24-bits/pixel) and .ecw (24-bits/

pixel) formats.

2.1.1.6 Population

Atoll deals with vector population files (polygons, lines or points) in .mif, .shp and .agd formats or 8, 16, 32 bits/pixel raster

population files in .tif, .bil, .bmp and Erdas Imagine formats. Population map describes the population distribution. They

are considered in clutter statistics and in coverage prediction reports.

2.1.1.7 Other Geographic Data

It is possible to import generic geographic data types, other than those listed above, (Customer density, revenue density,

etc.) in Atoll. These data can be either vector files in .mif, .shp and .agd formats or 8, 16, 32 bits/pixel raster files in .tif,

.bil, .bmp and Erdas Imagine formats. These maps are taken into account in clutter statistics and in coverage prediction

reports.

The ArcView Grid format (.txt) is an ASCII format dedicated to define raster maps. It may be used to export any raster map

such as DTM, images, Clutter Classes and/or Heights, Population, Generic data maps and even coverage predictions.

The contents of an ArcView Grid file are in ASCII and consist of a header, describing the content, followed by the content

in the form of cell values.

2.1.2 Supported Geographic Data FormatsAtoll offers Import/Export filters for the most commonly used geographic data formats. The different filters are:

Notes:

• The minimum resolut ion supported by Atoll is 1m for any raster maps, excepted for

scanned images, for which it is unlimited.

• DTM and clutter map resolution must be an integer.

• All the raster maps you want to import in an ATL document must be represented in the

same projection system.

File formatImport/

ExportCan contain Georeferenced

.bil Both

DTM, Clutter classes and heights, Traffic,

Image, Population, Other data Yes via .hdr files

.tif BothDTM, Clutter classes and heights, Traffic,

Image, Population, Other data

Yes via associated .tfw files if

they exist

Planet© Both DTM, Clutter classes, Image, Vector data Yes via index files

.bmp BothDTM, Clutter heights, Clutter classes, Traffic,

Image, Population, Other dataYes via .bpw (or .bmw) files

.dxf® Import Only Vector data, Vector traffic Yes

.shp BothVector data, Vector traffic, Population, Other

dataYes

.mif/.mid BothVector data, Vector traffic, Population, Other

dataYes




Thus, to sum up, you can import:

• DTM files in .tif (16-bits, 8-bits), .bil (16-bits, 8-bits), Planet© (16-bits), Erdas Imagine (16-bits, 8-bits), Vertical

Mapper (.grd, .grc) and .bmp (8-bits) formats.

• Clutter heights files in .tif (16-bits, 8-bits), .bil (16-bits, 8-bits), Planet© (16-bits), Erdas Imagine (16-bits, 8-bits),

Vertical Mapper (.grd, .grc) and .bmp (8-bits) formats.

• Clutter classes and traffic raster files in .tif (8-bits), .bil (8-bits), .bmp (8-bit), Erdas Imagine (8-bits) and Vertical

Mapper (.grd, .grc) and Planet© format (16-bits) are also supported.

• Vector data files in .dxf®, Planet©, .shp, .mif and .agd formats.

• Vector traffic files in .dxf®, Planet©, .shp, .mif and .agd formats.

• Scanned image files in .tif (1, 4, 8, 24-bits), .bil (1, 4, 8, 24-bits), Planet© (1, 4, 8, 24-bits), .bmp (1-24-bits), Erdas

Imagine (1, 4, 8, 24-bits), Vertical Mapper (.grd, .grc) and .ecw (Enhanced Compressed Wavelet) (24 bits) formats.• Population files in .mif, .shp, .agd, .tif (8, 16, 32-bits), .bil (8, 16, 32-bits), .bmp (8, 32-bits), Vertical Mapper (.grd,

.grc) and Erdas Imagine (8, 16, 32-bits) formats.

• Other generic data types in .mif, .shp, .agd, .tif (8, 16, 32-bits), .bil (8, 16, 32-bits), .bmp (8, 32-bits), Vertical

Mapper (.grd, .grc) and Erdas Imagine (8, 16, 32-bits) formats.

2.2 Radio DataAtoll manages several radio data types; sites, transmitters, antennas, stations and hexagonal designs. Data definition in

Atoll is detailed hereafter.

2.2.1 Site

A site is a geographical point where one or several transmitters (multi-sectored site or station) equipped with antennas are

located.

2.2.2 Antenna

An antenna is a device used for transmitting or receiving electromagnetic waves.

2.2.3 Transmitter

A transmitter is a group of radio devices located at a site. Transmitters are equipped with antenna(s) and other equipment

such as feeder, tower mounted amplifiers (TMA) and BTS.

2.2.4 Repeater

A repeater is a device that receives, amplifies and transmits the radiated or conducted RF carrier both in downlink and

uplink. It comprises a donor side and a server side. The donor side receives the signal from a donor transmitter. This signal

may be carried by different types of links such as radio link, microwave link, or optic fibre. The server side transmits the

repeated signal.

Erdas Imagine Import OnlyDTM, Clutter classes and heights, Traffic,

Image, Population, Other dataYes

ArcView Grid Export OnlyDTM, Clutter classes and heights, Traffic,


Yes automatically embedded in

the data file

.agd BothVector data, Vector traffic, Population, Other

data


the data file

Vertical Mapper

(.grd, .grc)Both

DTM, Clutter classes and heights, Traffic,



the data file

.ecw Import Only Images Yes via ers file (not mandatory)

Note:

• The .wld files may be used as georeferencement file for any type of binary raster file.

• Tiled .tif format is not supported.

Note:

• It is possible to import Packbit, FAX-CCITT3 and LZW compressed .tif files. However, in

case of DTM and clutter, we recommend not to use compressed files in order to avoid poor

performances. If uncompressed files are too big, it is better to split them.



© Forsk 2010 AT282_TRG_E2 41


2.2.5 Remote Antenna

The use of remote antennas allows antenna positioning at locations that would normally require prohibitively long runs of

feeder cable. A remote antenna is connected to the base station via an optic fibre. The main difference from a repeater is

that a remote antenna generates its own cell whereas a repeater extends the coverage of an existing cell.

2.2.6 Station

A station can represent one transmitter on a site or a group of transmitters on a same site sharing the same properties.

You can define station templates and build your network from stations instead of single transmitters.

2.2.7 Hexagonal Design

A hexagonal design is a group of stations created from the same station template.

2.2.8 GSM GPRS EGPRS Documents

2.2.8.1 TRX

A base station (transmitter) consists of several transceivers or TRXs. One TRX supports as many timeslots as the multi-

plexing factor defined in properties of your frequency band (8 timeslots in GSM networks). Three types of TRXs are

modelled in Atoll:

• The BCCH TRX type: carries the BCCH,

• The TCH TRX type: which is the default traffic carrier,

• The TCH_INNER TRX type: this TRX type is an inner traffic carrier.

2.2.8.2 Subcell

A subcell corresponds to a group of TRXs having the same radio characteristics, the same quality (C/I) requirements, and

common settings. A subcell is characterised by the ‘transmitter-TRX type’ pair. Each transmitter may have one or more

subcells. The most common configurations are the {BCCH, TCH} configuration or the {BCCH, TCH, TCH_INNER} one.

2.2.8.3 Cell Type

A cell type describes the subcells (types of TRXs) that a cell can use and their parameters, which can be different. In the

current Atoll version, the cell type definition must include a TRX type as the BCCH carrier (BCCH TRX type) and another

TRX type as the default traffic carrier (TCH TRX type). Only one TRX type carrying the broadcast and only one TRX type

carrying the default TCH are supported.

2.2.9 All CDMA, WiMAX, and LTE Documents

2.2.9.1 Cell

Cell comprises the carrier characteristics of a transmitter. Cell is characterised by the ‘ transmitter-carrier’ pair. The trans-

mitter-carrier pair must be unique.






Chapter 3File Formats

Atoll supports a set of file formats for each type of data, may it be geographic data or calculation results. This chapter contains details of these file

formats, their usage, availability, and limitations.






© Forsk 2010 AT282_TRG_E2 45

Chapter 3: File Formats

3 File Formats

3.1 BIL FormatBand Interleaved by Line is a method of organizing image data for multi-band images. It is a schema for storing the actual

pixel values of an image in a file. The pixel data is typically preceded by a file header that contains auxiliary data about the

image, such as the number of rows and columns in the image, a colour map, etc. .bil data stores pixel information band

by band for each line, or row, of the image. Although .bil is a data organization schema, it is treated as an image format.

An image description (number of rows and columns, number of bands, number of bits per pixel, byte order, etc.) has to beprovided to be able to display the .bil file. This information is included in the header .hdr file associated with the .bil file. A

.hdr file has the same name as the .bil file it refers to, and should be located in the same directory as the source file. The

.hdr structure is simple; it is an ASCII text file containing eleven lines. You can open a .hdr file using any ASCII text editor.

Atoll supports the following objects in .bil format:

• Digital Terrain Model (8 or 16 bits)

• Clutter heights (8 or 16 bits)

• Clutter classes and User profile environment based traffic maps (8 bits)

• User density traffic maps (16 or 32 bits)

• Raster images (1, 4, 8, 24 bits)

• Population maps (8, 16, 32 bits)

• Other generic geographic data (8, 16, 32 bits)

• Path loss or received signal level value matrices (16 bits)

3.1.1 HDR Header File

3.1.1.1 Description

The header file is a text file that describes how data are organised in the .bil file. The header file is made of rows, each

row having the following format:

where ‘keyword’ corresponds to an attribute type, and ‘value’ defines the attribute value.

Keywords required by Atoll are described below. Other keywords are ignored.

Four additional keywords may be optionally managed.

which can be :

keyword value

nrows Number of rows in the image.

ncols Number of columns in the image.

nbands Number of spectral bands in the image, (1 for DTM data and

8 bit pictures).nbits Number of bits per pixel per band; 8 or 16 for DTMs or

Clutter heights (altitude in metres), 8 for clutter class-

es file (clutter code), 16 for path loss matrices (path

loss in dB, field value in dBm, dBµV and DBµV/m).

byteorder Byte order in which image pixel values are stored. Accept-

ed values are M (Motorola byte order) or I (Intel byte or-

der).

layout Must be ‘bil’.

skipbytes Byte to be skipped in the image file in order to reach the

beginning of the image data. Default value is 0.

ulxmap x coordinate of the centre of the upper-left pixel.

ulymap y coordinate of the centre of the upper-left pixel.

xdim x size in metre of a pixel.

ydim y size in metre of a pixel.

pixeltype Type of data read (in addition to the length)

UNSIGNDINT Undefined 8, 16, 24 or 32 bits

SIGNEDINT Integer 16 or 32 bits

FLOAT Real 32 or 64 bits




in some cases, this keyword can be replace by datatype defined as follows:

It can be:

The other optional keywords are :valueoffset, valuescale and nodatavalue.

By default, integer data types are chosen with respect to the pixel length (nbits).

So, we have

3.1.1.2 Samples

Here, the data is 20m.

3.1.1.2.1 Digital Terrain Model

3.1.1.2.2 Clutter Classes File

3.1.1.2.3 BIL File

.bil files are usually binary files without header. Data are stored starting from the Northwest corner of the area. The skip-

bytes value defined in the header file allows to skip records if the data do not start at the beginning of the file.

datatype Type of data read (in addition to the length)

Un Undefined n bits (8, 16, 24 or 32 bits)

In Integer n bits (16 or 32 bits)

Rn Real n bits (32 or 64 bits)

RGB24 Integer 3 colour components on 24 bits

valueoffset Real value to be added to the read value (V read )

valuescale Scaling factor to be applied to the read value

nodatavalue Value corresponding to “NO DATA”

V V read valuescale valueoffset +u=

nrows 1500

ncols 1500

nbands 1

nbits 8 or 16

byteorder M

layout bil

skipbytes 0

ulxmap 975000

ulymap 1891000

xdim 20.00

ydim 20.00

nrows 1500

ncols 1500

nbands 1

nbits 8

byteorder M

layout bil

skipbytes 0

ulxmap 975000

ulymap 1891000

xdim 20.00

ydim 20.00



© Forsk 2010 AT282_TRG_E2 47


3.2 TIF Format start hereTagged Image File Format graphics filter supports all image types (monochrome, greyscale, palette colour, and RGB full

colour images) and Packbit, LZW or fax group 3-4 compressions. .tif files are not systematically geo-referenced. You have

to enter spatial references of the image manually during the import procedure (x and y-axis map coordinates of the centre

of the upper-left pixel, pixel size); an associated file with .tfw extension will be simultaneously created with the same name

and in the same directory as the .tif file it refers to. Atoll will then use the .tfw file during the import procedure for an auto-

matic geo-referencing.

You can modify the colour palette convention used by Atoll when exporting .tif files. This can be helpful when working on

.tif files exported by Atoll in other tools. In the default palette, the first colour indexes represent the useful information and

the remaining colour indexes represent the background. It is possible to export .tif files with a palette which defines the

background colour at the colour index 0, and then the colour indexes necessary to represent useful information. Add the

following lines in the Atoll.ini file to set up the new palette convention:

Please refer to the Administrator Manual for more details about the Atoll.ini file.

Atoll supports the following objects in .tif format:

• Digital Terrain Model (8 or 16 bits)

• Clutter heights (8 or 16 bits)

• Clutter classes and User profile environment based traffic maps (8 bits)

• User density traffic maps (16 or 32 bits)


• Population maps (8, 16, 32 bits)

• Other generic geographic data (8, 16, 32 bits)

.tfw file contains the spatial reference data of an associated .tif file. The .tfw file structure is simple; it is an ASCII text file

that contains six lines. You can open a .tfw file using any ASCII text editor.

3.2.1 TFW Header File

The .tfw files contain spatial reference data for the associated .tif file. The header file is a text file that describes how data

are organised in the .tif file. You can open a .tfw file using any ASCII text editor. The header file consists of six lines, wi th

each line having the following description:

Note:

• Atoll also supports .tif files using the Packbit, FAX-CCITT3 and LZW compression modes.

[TiffExport]

PaletteConvention=Gis

Notes:

• Using compressed geo data formats (compressed .tif, Erdas Imagine, or .ecw) can cause

performance loss due to real-time decompression. However, you can recover this loss in

performance by:

- Either, hiding the status bar, which provides geographic data information in real time, by

unchecking the Status Bar item in the View menu.

- Or, not displaying some of the information, such as altitude, clutter class and clutter

height, in the status bar. This can be done through the Atoll.ini file, by adding the following

lines:

[StatusBar]

DisplayZ=0

DisplayClutterClass=0

DisplayClutterHeight=0

• You can also save the produced map in an uncompressed format.

• Please refer to the Administrator Manual for more details about the Atoll.ini file.

Line Description

1 x dimension of a pixel in map units

2 amount of translation

3 amount of rotation

4 negative of the y dimension of a pixel in map units

5 x-axis map coordinate of the centre of the upper-left pixel

6 y-axis map coordinate of the centre of the upper-left pixel




3.2.2 Sample

3.2.2.1 Clutter Classes File

3.3 BMP FormatThis is the MS-Windows standard format. It holds black & white, 16-, 256- and True-colour images. The palletized 16-

colour and 256-colour images may be compressed via run length encoding (though compressed .bmp files are quite rare).

The image data itself can either contain pointers to entries in a colour table or literal RGB values. .bmp files are not system-

atically geo-referenced. You have to enter spatial references of the image manually during the import procedure (x and y-

axis map coordinates of the centre of the upper-left pixel, pixel size). When exporting (saving) a .bmp file, an associated

file with .bpw extension is created with the same name and in the same directory as the .bmp file it refers to. Atoll storesthe georeferencing information in this file for future imports of the .bmp so that the .bpw file can be used during the import

procedure for automatic geo-referencing. Atoll also supports .bmw extension for the .bmp related world files.

Atoll supports the following objects in .bmp format:

• Digital Terrain Model (8 bits)

• Clutter Heights (8 bits)

• Clutter classes and User density traffic maps (8 bits)


• Population maps (8, 32 bits)

• Other generic geographic data (8, 32 bits)

3.3.1 BMP File Description

A .bmp file contains of the following data structures:

• BITMAPFILEHEADER bmfh Contains some information about the bitmap file (about the fi le, notabout the bitmap itself).

• BITMAPINFOHEADER bmih Contains information about the bi tmap (such as size, colours, etc.).

• RGBQUAD aColors[] Contains a colour table.

• BYTE aBitmapBits[] Image data (whose format is specified by the bmih structure).

3.3.1.1 BMP File Structure

The following tables give exact information about the data structures. The Start-value is the position of the byte in the file

at which the explained data element of the structure starts, the Size-value contains the number of bytes used by this data

element, the Name column contains both generic name and the name assigned to this data element by the Microsoft API

documentation, and the Description column gives a short explanation of the purpose of this data element.

• BITMAPFILEHEADER (Header - 14 bytes):

• BITMAPINFOHEADER (InfoHeader - 40 bytes):

Note:

• Atoll does not use the lines 2 and 3 when importing a .tif format geographic file.

100.00

0.00

0.00

-100.00

60000.00

2679900.00

Start SizeName

DescriptionGeneric MS API

1 2 Signature bfType Must always be set to 'BM' to declare that this is a .bmp-file.

3 4 FileSize bfSize Specifies the size of the file in bytes.

7 2 Reserved1 bfReserved1 Unused. Must be set to zero.

9 2 Reserved2 bfReserved2 Unused. Must be set to zero.

11 4 DataOffset bfOffBitsSpecifies the offset from the beginning of the file to the bitmap (raster)

data.



© Forsk 2010 AT282_TRG_E2 49


• RGBQUAD array (ColorTable):

• Pixel data:

The interpretation of the pixel data depends on the BITMAPINFOHEADER structure. It is important to know that the rows

of a .bmp are stored upside down meaning that the uppermost row which appears on the screen is actually the lowermost

row stored in the bitmap. $QRWKHU LPSRUWDQW WKLQJ LV WKDW WKHQXPEHURIE\WHV LQRQH URZPXVWDOZD\VEH DGMXVWHG by

appending zero bytesWRILWLQWRWKHERUGHURIDPXOWLSOHRIIRXU (16-bit or 32-bit rows).

Start SizeName


15 4 Size biSizeSpecifies the size of the BITMAPINFOHEADER structure, in

bytes (= 40 bytes).

19 4 Width biWidth Specifies the width of the image, in pixels.

23 4 Height biHeight Specifies the height of the image, in pixels.

27 2 Planes biPlanesSpecifies the number of planes of the target device, must be

set to zero or 1.

29 2 BitCount biBitCount

Specifies the number of bits per pixel.

1 = monochrome pallete. # of colours = 14 = 4-bit palletized. # of colours = 16

8 = 8-bit palletized. # of colours = 256

16 = 16-bit palletized. # of colours = 65536

24 = 24-bit palletized. # of colours = 16M

31 4 Compression biCompression

Specifies the type of compression, usually set to zero.

0 = BI_RGB no compression

1 = BI_RLE8 8-bit RLE encoding

2 = BI_RLE4 4-bit RLE encoding

35 4 ImageSize biSizeImageSpecifies the size of the image data, in bytes. If there is no

compression, it is valid to set this element to zero.

39 4 XpixelsPerM biXPelsPerMeter Specifies the the horizontal pixels per meter.

43 4 YpixelsPerM biYPelsPerMeter Specifies the the vertical pixels per meter.

47 4 ColoursUsed biClrUsed

Specifies the number of colours actually used in the bitmap. If

set to zero the number of colours is calculated using thebiBitCount element.

51 4 ColoursImportant biClrImportantSpecifies the number of colour that are 'important' for the

bitmap. If set to zero, all colours are considered important.

Note:

• biBitCount actually specifies the colour resolution of the bitmap. It also decides if there is a

colour table in the file and how it looks like.

- In 1-bit mode the colour table has to contain 2 entries (usually white and black). If a bit in

the image data is clear, it points to the first palette entry. If the bit is set, it points to the

second.

- In 4-bit mode the colour table must contain 16 colours. Every byte in the image data

represents two pixels. The byte is split into the higher 4 bits and the lower 4 bits and each

value of them points to a palette entry.

- In 8-bit mode every byte represents a pixel. The value points to an entry in the colour

table which contains 256 entries.- In 24-bit mode three bytes represent one pixel. The first byte represents the red part, the

second the green and the third the blue part. There is no need for a palette because every

pixel contains a literal RGB-value, so the palette is omitted.

Start SizeName


1 1 Blue rgbBlue Specifies the blue part of the colour.

2 1 Green rgbGreen Specifies the green part of the colour.

3 1 Red rgbRed Specifies the red part of the colour.

4 1 Reserved rgbReserved Must always be set to zero.

Note:

• In a colour table (RGBQUAD), the specification for a colour starts with the blue byte, while

in a palette a colour always starts with the red byte.




3.3.1.2 BMP Raster Data Encoding

Depending on the image BitCount and on the Compression flag there are 6 different encoding schemes. In all of them,

• Pixels are stored bottom-up, left-to-right.

• Pixel lines are padded with zeros to end on a 32-bit boundary.

• For uncompressed formats every line will have the same number of bytes.

• Colour indices are zero based, meaning a pixel colour of 0 represents the first colour table entry, a pixel colour of

255 (if there are that many) represents the 256th entry. For images with more than 256 colours there is no colour

table.

3.3.1.2.1 Raster Data Compression Descriptions

• 4-bit / 16 colour images

• 8-bit / 256 colour images

Encoding type

BitCoun

t

Compressio

n Remarks

1-bit

B&W images1 0

Every byte holds 8 pixels, its highest order bit representing the

leftmost pixel of these 8. There are 2 colour table entries. Some

readers assume that 0 is black and 1 is white. If you are storing

black and white pictures you should stick to this, with any other 2

colours this is not an issue. Remember padding with zeros up to a

32-bit boundary.

4-bit

16 colour images4 0

Every byte holds 2 pixels, its high order 4 bits representing the left of

those. There are 16 colour table entries. These colours do not have

to be the 16 MS-Windows standard colours. Padding each l ine with

zeros up to a 32-bit boundary will result in up to 28 zeros = 7 'wasted

pixels'.

8-bit

256 colour images8 0

Every byte holds 1 pixel. There are 256 colour table entries.

Padding each line with zeros up to a 32-bit boundary will result in up

to 3 bytes of zeros = 3 'wasted pixels'.

16-bit

High colour images16 0

Every 2 bytes hold 1 pixel. There are no colour table entries.

Padding each line with zeros up to a 16-bit boundary will result in up

to 2 zero bytes.

24-bit

True colour images24 0

Every 4 bytes hold 1 pixel. The first holds i ts red, the second its

green, and the third its blue intensity. The fourth byte is reserved

and should be zero. There are no colour table entries. No zero

padding necessary.

4-bit

16 colour images4 2

Pixel data is stored in 2-byte chunks. The first byte specifies the

number of consecutive pixels with the same pair of colour. The

second byte defines two colour indices. The resulting pixel pattern

will have interleaved high-order 4-bits and low order 4 bits

(ABABA...). If the first byte is zero, the second defines an escape

code. The End-of-Bitmap is zero padded to end on a 32-bit

boundary. Due to the 16bit-ness of this structure this will always be

either two zero bytes or none.

8-bit

256 colour images8 1

The pixel data is stored in 2-byte chunks. The first byte specifies the

number of consecutive pixels with the same colour. The second byte

defines their colour indices. If the first byte is zero, the second

defines an escape code. The End-of-Bitmap is zero padded to end

on a 32-bit boundary. Due to the 16bit-ness of this structure this will

always be either two zero bytes or none.

n (Byte 1) c (Byte 2) Description

>0 any

n pixels to be drawn. The 1st, 3rd, 5th, ... pixels' colour is in c's high-order 4 bits, the

even pixels' colour is in c's low-order 4 bits. If both colour indices are the same, it

results in just n pixels of colour c.

0 0 End-of-line

0 1 End-of-Bitmap

0 2Delta. The following 2 bytes define an unsigned offset in x and y direction (y being up).

The skipped pixels should get a colour zero.

0 >=3The following c bytes will be read as single pixel colours just as in uncompressed files.

Up to 12 bits of zeros follow, to put the file/memory pointer on a 16-bit boundary again.

n (Byte 1) c (Byte 2) Description

>0 any n pixels of colour number c



© Forsk 2010 AT282_TRG_E2 51


3.3.2 BPW/BMW Header File Description

The header file is a text file that describes how data are organised in the .bmp file. The header file is made of rows, eachrow having the following description:

Atoll supports .bpw and .bmw header file extensions for Import, but exports headers with .bpw file extensions.

3.3.3 Sample3.3.3.1 Clutter Classes File

3.4 PNG FormatPortable Network Graphics (PNG) is a bitmapped image format that employs lossless data compression. PNG supports

palette-based (palettes of 24-bit RGB or 32-bit RGBA colors), greyscale, RGB, or RGBA images. PNG was designed for

transferring images on the Internet, not professional graphics, and so does not support other color spaces (such as

CMYK). PNG files nearly always use file extension .PNG or .png.

When exporting (saving) a .png file, an associated file with .pgw extension is created with the same name and in the same

directory as the .png file it refers to. Atoll stores the georeferencing information in this file for future imports of the .png so

that the .pgw file can be used during the import procedure for automatic geo-referencing.

For more information on the PNG file format, see www.w3.org/TR/PNG/.

3.4.1 PGW Header File Description

A PNG World file (.pgw file extension) is a plain text file used by geographic information systems (GIS) to provide georef-

erencing information for raster map images in .png format. The world file parameters are:

0 0 End-of-line

0 1 End-of-Bitmap

0 2Delta. The following 2 bytes define an unsigned offset in x and y direction (y being up).

The skipped pixels should get a colour zero.

0 >=3The following c bytes will be read as single pixel colours just as in uncompressed files.

A zero follows, if c is odd, putting the file/memory pointer on a 16-bit boundary again.

Line Description







100.00

0.00

0.00

-100.00

60000.00

2679900.00

Line Description











© Forsk 2010 AT282_TRG_E2 53


3.8 MIF FormatMapInfo Interchange Format (.mif) allows various types of data to be attached to a variety of graphical items. These ASCII

files are editable, easy to generate, and work on all platforms supported by MapInfo. Vector objects with a .mif extension

may be imported in Atoll.

Two files, a .mif and a .mid, contain MapInfo data. Graphics reside in the .mif file while the text contents are stored in the

.mid file. The text data is delimited with one row per record, and Carriage Return, Carriage Return plus Line Feed, or Line

Feed between lines. The .mif file has two sections, the file header and the data section. The .mid file is optional. When

there is no .mid file, all fields are blank.

You can find more information at http://www.mapinfo.com.

You can define mappings between the coordinate system used for the MapInfo vector files, defined in the corresponding

.mif files, and Atoll. In this way, when you import a vector file, Atoll can detect the correct coordinate system automatically.

For more information about defining the mapping between coordinate systems, please refer to the Administrator Manual.

3.9 TAB FormatTAB files (MapInfo Tables) are the native format of MapInfo. They actually consist of a number of files with extensions

such as .TAB, .DAT and .MAP. All of these files need to be present and kept together for the table to work. These are

defined as follows:

• .TAB: table structure in ASCII format

• .DAT: table data storage in binary format

• .MAP: storage of map objects in binary format

• .ID: index to the MapInfo graphical objects (.MAP) file

• .IND: index to the MapInfo tabular (DAT) fileYou can find more information at http://www.mapinfo.com.

You can define mappings between the coordinate system used for the MapInfo vector files, defined in the corresponding

.mif files, and Atoll. In this way, when you import a vector file, Atoll can detect the correct coordinate system automatically.

For more information about defining the mapping between coordinate systems, please refer to the Administrator Manual.

TAB files are also supported as georeference information files for raster files (.bmp and .tif). The .TAB file must have the

following format:

The fields in bold are described below:

!table

!version 300

!charset WindowsLatin1

Definition Table

File "raster.bmp"

Type "RASTER"

(ulxmap,ulymap) (0,0) Label "Pt 1",

(llxmap,llymap) (0,nrows) Label "Pt 2",

(lrxmap,lrymap) (ncols,nrows) Label "Pt 3",

(urxmap,urymap) (ncols,0) Label "Pt 4"

Field Description

File "raster.bmp" Name of the raster file (e.g., raster.bmp)

ulxmap x coordinate of the centre of the upper-left pixel in metres

ulymap y coordinate of the centre of the upper-left pixel in metres

llxmap x coordinate of the centre of the lower-left pixel in metres

llymap y coordinate of the centre of the lower-left pixel in metres

lrxmap x coordinate of the centre of the lower-right pixel in metres

lrymap y coordinate of the centre of the lower-right pixel in metres

urxmap x coordinate of the centre of the upper-right pixel in metres

urymap y coordinate of the centre of the upper-right pixel in metres

nrows Number of rows in the image

ncols Number of columns in the image




3.10 ECW FormatThe Enhanced Compressed Wavelet file format is supported in Atoll. .ecw files are geo-referenced image files, which can

be imported in Atoll. This is an Open Standard wavelet compression technology, developed by Earth Resource Mapping,

which can compress images with up to a 100-to-1 compression ratio. Each compressed image file contains a header carry-

ing the following information about the image:

• The image size expressed as the number of cells across and down

• The number of bands (RGB images have three bands)

• The image compression rate

• The cell measurement units (meters, degrees or feet)

• The size of each cell in measurement units• Coordinate space information (Projection, Datum etc.)

3.11 Erdas Imagine FormatAtoll supports Erdas Imagine data files in order to import DTM (8 or 16 bit/pixel), clutter (8 bit/pixel), traffic (8 bit/pixel),

and image (1-24 bit/pixel) files with the .img format. These files use the Erdas Imagine Hierarchical File Format (HFA)

structure. For any type of file, if there are pyramids (storage of different resolution layers), they are used to enhance

performance when decreasing the resolution of the display. Some aspects of working with Erdas Imagine format in Atoll

are:

• Atoll supports uncompressed as well as compressed (or partially compressed) DTM .img files.

• You can create a .mnu file to improve the clutter class map loading.

• The colour-to-code association (raster maps) may be automatically imported from the .img file.

• These files are automatically geo-referenced, i.e., they do not require any additional file for geo-reference.

For image files, the number of supported bands is either 1 (colour palette is defined separately) or 3 (no colour palette but

direct RGB information for each pixel). In case of 3 bands, only 8 bit per pixel format is supported. Therefore, 8-bit images,

containing RGB information (three bands are provided: the first band is for Blue, the second one is for Green and the third

for Red), can be considered as 24 bit per pixel files. 32 bit per pixel files are not supported.

3.12 Planet EV/Vertical Mapper Geographic Data FormatVertical Mapper offers two types of grids:

• Numerical continuous grids, which contain numerical information (such as DTM), and are stored in files with the

.grd extension.

• Classified grids, which contain alphanumeric (characters) information, and are stored in files with the .grc exten-

sion.

Atoll is capable of supporting the Vertical Mapper Classified Grid (GRC) and Vertical Mapper Continuous Grid (GRD) file

formats in order to import and export:

• GRD: DTM, image, population, traffic density, and other data types.

• GRC: DTM, clutter classes, clutter heights, environment traffic, image, population, and other data types.

It is also possible to export coverage prediction studies in GRD and GRC formats.

This is the geographic data format used by Planet EV. So, it is possible to directly import geographic data from Planet EV

to Atoll using this format.

Notes:

• Using compressed geo data formats (compressed .tif, Erdas Imagine, or .ecw) can cause

performance loss due to real-time decompression. However, you can recover this loss in

performance by:

- Either, hiding the status bar, which provides geographic data information in real time, by

unchecking the Status Bar item in the View menu.

- Or, not displaying some of the information, such as altitude, clutter class and clutter

height, in the status bar. This can be done through the Atoll.ini file, by adding the following

lines:

[StatusBar]DisplayZ=0

DisplayClutterClass=0

DisplayClutterHeight=0

• You can also save the produced map in an uncompressed format.

• Please refer to the Administrator Manual for more details about the Atoll.ini file.



© Forsk 2010 AT282_TRG_E2 55


3.13 ArcView Grid FormatThe ArcView Grid format (.txt) is an ASCII format dedicated to defining raster maps. It may be used to export any raster

map such as DTM, images, clutter classes and/or heights, population, other data maps, and even coverage predictions.

The contents of an ArcView Grid file are in ASCII and consist of a header, describing the content, followed by the content

in the form of cell values.

3.13.1 ArcView Grid File Description

The format of this file is as follows:

3.13.2 Sample

3.14 Other Supported Geographic Data File FormatsOther than the .bil, .tif, Planet, .dxf, .shp, .mif, .img, and .ecw formats, Atoll supports 3 other formats.

The .ist and .dis formats are ASCII files used for Digital Terrain Model only. .ist images come from Istar, whereas .dis

images come from IGN (Institut Géographique National). The .ist format works in exactly the same way as the .bil format,

except for DTM images. For DTM images, the .ist format uses a decimetric coding for altitudes, whereas .bil images use

only a metric coding.

3.15 Planet Format

The Planet geographic data are described by a set of files grouped in a Planet directory. The directory structure dependson the geographic data type.

Atoll supports the following objects in Planet format:

• Digital Terrain Model (8 and 16 bits)

• Clutter class maps (16 bits)

• Raster images (1, 4, 8 and 24 bits)

• Vector data

• Text data

ncols XXX Number of columns of the grid (XXX columns).

nrows XXX Number of rows of the grid (XXX rows).

xllcenter XXX OR

xllcorner XXX Significant value relative to the bin centre or corner.

yllcenter OR

yllcorner XXX Significant value relative to the bin centre or corner.

cellsize XXX Grid resolution.

nodata_value XXX Optional value corresponding to no data (no information).

//Row 1 Top of the raster. Description of the first row. Syntax:

ncols number of values separated by spaces.

:

:

//Row N Bottom of the raster.

ncols 303

nrows 321

xllcorner 585300.000000

yllcorner 5615700.000000

cellsize 100.000000

nodata_value 0

...




3.15.1 DTM File

3.15.1.1 Description

The DTM directory consists of three files; the height file and two other files detailed below:

• The index file structure is simple; it is an ASCII text file that holds position information about the file. It contains

five columns. You can open an index file using any ASCII text editor. The format of the index file is as follows:

• The projection file provides information about the projection system used. This file is optional. It is an ASCII text

file with four lines maximum.

3.15.1.2 Sample

Index file associated with height file (DTM data):

Projection file associated with height file (DTM data):

3.15.2 Clutter Class Files


The Clutter directory consists of three files; the clutter file and two other files detailed below:

• The menu file, an ASCII text file, defines the feature codes for each type of clutter. It consists of as many lines

(with the following format) as there are clutter codes in the clutter data files. This file is optional.

• The index file gives clutter spatial references. The structure of clutter index file is the same as the structure of DTM

index file.

Field Acceptable values Description

File name Text Name of file referenced by the index fileEast min Float x-axis map coordinate of the centre of the upper-left pixel in meters

East max Float x-axis map coordinate of the centre of the upper-right pixel in meters

North min Float y-axis map coordinate of the centre of the lower-left pixel in meters

North max Float y-axis map coordinate of the centre of the upper-left pixel in meters

Square size Float Dimension of a pixel in meters

Line Description

Spheroid

Zone

Projection

Central meridianLatitude and longitude of projection central meridian and equivalent x and y coordinates in meters

(optional)

Note:

• In the associated binary file, the value -9999 corresponds to ‘No data’ which is supported

by Atoll.

sydney1 303900 343900 6227900 6267900 50

Australian-1965

56

UTM

0 153 500000 10000000

Field Type Description

Clutter-code Integer (>1) Identification code for clutter class

Feature-name Text (up to 32 characters in length)Name associated with the clutter-code. (It may contain

spaces)

Note:

• In the associated binary file, the value -9999 corresponds to ‘No data’ which is supported

by Atoll.



© Forsk 2010 AT282_TRG_E2 57


3.15.2.2 Sample

Menu file associated with the clutter file:

3.15.3 Vector Files


Vector data comprises terrain features such as coastlines, roads, etc. Each of these features is stored in a separate vector

file. Four types of files are used, the vector file, where x and y coordinates of vector paths are stored, and three other files

detailed below:

• The menu file, an ASCII text file, lists the vector types stored in the database. The menu file is composed of one

or more records with the following structure:

The fields are separated by space character.

• The index file, an ASCII text file, lists the vector files and associates each vector file with one vector type, and

optionally with one attribute file. The index file consists of one or more records with the following structure:

The fields are separated by spaces.

• The attribute file stores the height and description properties of vector paths. This file is optional.

1 open

2 sea

3 inlandwater

4 residential

5 meanurban

6 denseurban7 buildings

8 village

9 industrial

10 openinurban

11 forest

12 parks

13 denseurbanhigh

14 blockbuildings

15 denseblockbuild

16 rural

17 mixedsuburban


Vector type code Integer > 0 Identification code for the vector typeVector type name Text (up to 32 characters in length) Name of the vector type


Vector file name Text (up to 32 characters in length) Name of the vector file

Attribute file name Text (up to 32 characters in length)Name of attribute file associated with the vector file

(optional)

Dimensions Real

vector file eastmin: minimum x-axis coordinate of all

vector path points in the vector file

vector file eastmax: maximum x-axis coordinate of all

vector path points in the vector filevector file northmin: minimum y-axis coordinate of all

vector path points in the vector file

vector file northmax: maximum y-axis coordinate of all

vector path points

Vector type name Text (up to 32 characters in length)

Name of the vector type with which the vector file is

associated. This one must match exactly a vector type

name field in the menu file.




3.15.3.2 Sample

Index file associated with the vector files

3.15.4 Image Files

The image directory consists of two files, the image file with .tif extension and an index file with the same structure as the

DTM index file structure.

3.15.5 Text Data Files

The text data directory consists of:

• The text data files are ASCII text files with the following format:

Each file contains a line of text followed by easting and northing of that text, etc.

• The index file, an ASCII text file, stores the position of each text file. It consists of one or more records with the

following structure:

The fields are separated by spaces.

• The menu file, an ASCII text file, contains the text features. This file is optional.

sydney1.airport 313440 333021 6239426 6244784 airport

sydney1.riverlake 303900 342704 6227900 6267900 riverlake

sydney1.coastline 322837 343900 6227900 6267900 coastline

sydney1.railways 303900 336113 6227900 6267900 railways

sydney1.highways 303900 325155 6240936 6267900 highways

sydney1.majstreets 303900 342770 6227900 6267900 majstreetssydney1.majorroads 303900 342615 6227900 6267900 majorroads

Airport637111.188 3094774.00

Airport

628642.688 3081806.25


File name Text (up to 32 characters in length) File name of the text data file

East Min RealMinimum x-axis coordinate of all points listed in the text

data file

East Max RealMaximum x-axis coordinate of all points listed in the text

data file

North Min RealMinimum y-axis coordinate of all points listed in the text

data file

North Max RealMaximum y-axis coordinate of all points listed in the text

data file

Text feature Text (up to 32 characters in length) This field is omitted in case no menu file is available.

railwayp.txt -260079 693937 2709348 3528665 Railway_Station

airport.txt -307727 771663 2547275 3554675 Airport

ferryport.txt 303922 493521 2667405 3241297 Ferryport

1 Airport

2 Ferryport

3 Railway_Station



© Forsk 2010 AT282_TRG_E2 59


3.16 MNU Format

3.16.1 Description

A .mnu file is useful when importing clutter classes or raster traffic files in .tif, .bil and .img formats. It gives the correspond-

ence between the clutter (or traffic) code and the class name. It is a text file with the same name as the clutter (or traffic)

file with .mnu extension. It must be stored at the same location as the clutter (or traffic) file. It has the same structure as

the menu file used in the Planet format.

Separator used can either be a space character or a tab.

3.16.2 Sample

A .mnu file associated to a clutter classes file:

3.17 XML Table Export/Import Format All the data tables in an Atoll document can be exported to XML files.

Atoll creates the following files when exporting data tables to XML files:

• One index.xml file which contains the mapping between the data tables in Atoll and the corresponding XML file

created by the export.

• One XML file per data table which contains the data table format (schema) and the data.

The XML import does not modify the active document table and field definitions. Therefore, the Networks and Custom-

Fields tables, although exported, are not imported.

The following sections describe the structures of these two types of XML files created at export.

3.17.1 Index.xml File

The index.xml file stores the system (GSM, UMTS, etc.) and the technology (TDMA, CDMA, etc.) of the document, and

the version of Atoll used for exporting the data tables to XML files. It also contains the mapping between the data tables

in the Atoll document and the XML file corresponding to each data table.

The root tag <Atoll_XML_Config...> of the index.xml file contains the following attributes:

The index file also contains a list of mapping between the tables exported from Atoll and the XML files corresponding to

each table. This list is sorted in the order the Atoll tables are to be imported.

The list is composed of <XML_Table.../> tags with the following attributes:

A sample extract of the index.xml is given below:


Class code Integer > 0 Identification code for the clutter (or traffic) class

Class name Text (up to 50 characters in length) Name of the clutter (or traffic) class. It may contain spaces.

0 none

1 open

2 sea

3 inland_water

4 residential

5 meanurban

Attribute Description

Atoll_File_System Corresponds to the SYSTEM_ field of the Networks table of the exported document

Atoll_File_TechnologyCorresponds to the TECHNOLOGY field of the Networks table of the exported

document

Atoll_File_Version Corresponds to the Atoll version

Attribute Description

XML_File Corresponds to the exported XML file name (e.g., "Sites.xml")

Atoll_Table Corresponds to the exported Atoll table name (e.g., "Sites")




Note that no closing tag </XML_Table> is required.

3.17.2 XML FileAtoll creates an XML file per exported data table. This XML file has two sections, one for storing the description of the

table structure, and the second for the data itself. The XML file uses the standard XML rowset schema (schema included

in the XML file between <s:Schema id=’RowsetSchema’> and </s:Schema> tags).

Rowset Schema

The XML root tag for XML files using the rowset schema is the following:

The schema definition follows the root tag and is enclosed between the following tags:

In the rowset schema, after the schema description, the data are enclosed between <rs:data> and </rs:data>.

Between these tags, each record is handled by a <z:row … /> tag having its attributes set to the record field values since

in the rowset schema, values are handled by attributes. Note that no closing tag </z:row> is required.

A sample extract of a Sites.xml file containing the Sites table with only one site is given below:

<Atoll_XML_Config Atoll_File_System="UMTS" Atoll_File_Technology="CDMA"

Atoll_File_Version="2.x.x build xxxx">

<XML_Table XML_File="CustomFields.xml" Atoll_Table="CustomFields" />

<XML_Table XML_File="CoordSys.xml" Atoll_Table="CoordSys" />

...

</Atoll_XML_Config>

<xml xmlns:s='uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882'

xmlns:dt='uuid:C2F41010-65B3-11d1-A29F-00AA00C14882'

xmlns:rs='urn:schemas-microsoft-com:rowset'

xmlns:z='#RowsetSchema'>

<s:Schema id=’RowsetSchema’>

<!-Schema is defined here, using <s:ElementType> and <s:AttributeType> tags ->

</s:Schema>

<xml xmlns:s='uuid:BDC6E3F0-6DA3-11d1-A2A3-00AA00C14882'

xmlns:dt='uuid:C2F41010-65B3-11d1-A29F-00AA00C14882'

xmlns:rs='urn:schemas-microsoft-com:rowset'

xmlns:z='#RowsetSchema'>

<s:Schema id='RowsetSchema'>

<s:ElementType name='row' content='eltOnly' rs:updatable='true'>

<s:AttributeType name='NAME' rs:number='1' rs:maydefer='true' rs:writeun-

known='true' rs:basetable='Sites' rs:basecolumn='NAME' rs:keycolumn='true'>

<s:datatype dt:type='string' dt:maxLength='50'/>

</s:AttributeType>

<s:AttributeType name='LONGITUDE' rs:number='2' rs:maydefer='true' rs:wri-

teunknown='true' rs:basetable='Sites' rs:basecolumn='LONGITUDE'>

<s:datatype dt:type='float' dt:maxLength='8' rs:precision='15' rs:fix-

edlength='true'/>

</s:AttributeType>

<s:AttributeType name='LATITUDE' rs:number='3' rs:maydefer='true' rs:write-

unknown='true' rs:basetable='Sites' rs:basecolumn='LATITUDE'>

<s:datatype dt:type='float' dt:maxLength='8' rs:precision='15' rs:fix-

edlength='true'/>

</s:AttributeType>

<s:AttributeType name='ALTITUDE' rs:number='4' rs:nullable='true' rs:mayde-

fer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecolumn='ALTI-

TUDE'>



© Forsk 2010 AT282_TRG_E2 61


3.18 Externalised Propagation Results FormatPropagation results, i.e. the path loss matrices, may be stored in an external folder. This folder consists of a dBASE III

based file named ‘pathloss.dbf’ that contains calculation parameters of all the transmitters considered and one file (or two

when calculating main and extended path loss matrices) per transmitter taken into account. This is a binary file with .los

extension and contains the path loss values for a transmitter.

3.18.1 DBF File

dBASE III file (pathloss.dbf) has a standard .dbf format described below. Its content can be checked by opening it in MS-

Access. The format is detailed hereafter.

3.18.1.1 DBF File FormatFor general information, the format of .dbf files in any Xbase language is described.

Following notations are used in tables:

3.18.1.1.1 DBF Structure

3.18.1.1.2 DBF Header (Variable Size - Depends on Field Count)

<s:datatype dt:type='r4' dt:maxLength='4' rs:precision='7' rs:fix-

edlength='true'/>

</s:AttributeType>

<s:AttributeType name='COMMENT_' rs:number='5' rs:nullable='true' rs:mayde-

fer='true' rs:writeunknown='true' rs:basetable='Sites' rs:basecol-

umn='COMMENT_'>

<s:datatype dt:type='string' dt:maxLength='255'/>

</s:AttributeType>

<s:extends type='rs:rowbase'/>

</s:ElementType>

</s:Schema>

<rs:data>

<rs:insert>

<z:row NAME='Site0' LONGITUDE='8301' LATITUDE='-9756'/>

</rs:insert>

</rs:data>

</xml>

Note:

• Each transmitter path loss matrix is calculated on the area where calculation radius

intersects the computation zone (see: Computation zone).

FS = FlagShip D3 = dBaseIII+

Fb = FoxBase D4 = dBaseIV

Fp = FoxPro D5 = dBaseV

CL = Clipper

Byte Description

0...n .dbf header (see next part for size, byte 8)

n+1

1st record of fixed length (see next parts)

2nd record (see next part for size, byte10) …

last record

If .dbf is not empty

last optional: 0x1a (eof byte)

Byte Size Contents Description Applies for (supported by)

00 1 0x03 plain .dbf FS, D3, D4, D5, Fb, Fp, CL

0x04 plain .dbf D4, D5 (FS)

0x05 plain .dbf D5, Fp (FS)




• Field descriptor array in the .dbf header (32 bytes for each field)

• Field type and size in the .dbf header, field descriptor (1 byte)

0x43 with .dbv memo var size FS

0xB3 with .dbv and .dbt memo FS

0x83 with .dbt memo FS, D3, D4, D5, Fb, Fp, CL

0x8B with .dbt memo in D4 format D4, D5

0x8E with SQL table D4, D5

0xF5 with .fmp memo Fp

01 3 YYMMDD Last update digits All

04 4 ulong Number of records in file All

08 2 ushort Header size in bytes All10 2 ushort Record size in bytes All

12 2 0,0 Reserved All

14 1 0x01 Begin transaction D4, D5

0x00 End Transaction D4, D5

0x00 ignored FS, D3, Fb, Fp, CL

15 1 0x01 Encrypted D4, D5

0x00 normal visible All

16 12 0 (1) multi-user environment use D4,D5

28 1 0x01 production index exists Fp, D4, D5

0x00 index upon demand All

29 1 n language driver ID D4, D5

0x01 codepage437 DOS USA Fp

0x02 codepage850 DOS Multi ling Fp

0x03 codepage1251 Windows ANSI Fp

0xC8 codepage1250 Windows EE Fp


30 2 0,0 reserved All

32 n*32 Field Descriptor, (see next paragraph) all

+1 1 0x0D Header Record Terminator all


0 11 ASCI field name, 0x00 termin all11 1 ASCI field type (see next paragraph) all

12 4 n,n,n,n Fld address in memory D3

n,n,0,0 offset from record begin Fp

0,0,0,0 ignored FS, D4, D5, Fb, CL

16 1 byte Field length, bin (see next paragraph) all \ FS,CL: for C field type

17 1 byte decimal count, bin all / both used for fld lng

18 2 0,0 reserved all

20 1 byte Work area ID D4, D5

0x00 unused FS, D3, Fb, Fp, CL

21 2 n,n multi-user dBase D3, D4, D5

0,0 ignored FS, Fb, Fp, CL

23 1 0x01 Set Fields D3, D4, D5

0x00 ignored FS, Fb, Fp, CL

24 7 0...0 reserved all

31 1 0x01 Field is in .mdx index D4, D5


Size Type Description/Storage Applies for (supported by)

C 1...n Char ASCII (OEM code page chars)

rest= space, not \0 term.all



© Forsk 2010 AT282_TRG_E2 63


3.18.1.1.3 Each DBF Record (Fixed Length)

3.18.1.2 DBF File Content

The .dbf file provides information that is needed to check validity of each path loss matrix.

n = 1...64kb (using deci count) FS

n = 1...32kb (using deci count) Fp, CL

n = 1...254 all

D 8 Date 8 ASCII digits (0...9) in the YYYYMMDD format all

F 1...n Numeric

ASCII digits (-.0123456789)

variable pos. of float.point

n = 1...20

FS, D4, D5, Fp

N 1...n Numeric ASCII digits (-.0123456789)

fix posit/no float.pointall

n = 1...20 FS, Fp, CL

n = 1...18 D3, D4, D5, Fb

L 1 Logical ASCII chars (YyNnTtFf space) FS, D3, Fb, Fp, CL

ASCII chars (YyNnTtFf?) D4, D5 (FS)

M 10 Memo10 digits repres. the start block posit. in .dbt file, or 10 spaces if

no entry in memoall

V 10 Variable

Variable, bin/asc data in .dbv

4bytes bin= start pos in memo

4bytes bin= block size

1byte = subtype

1byte = reserved (0x1a)

10 spaces if no entry in .dbv

FS

P 10 Picturebinary data in .ftp

structure like MFp

B 10 Binarybinary data in .dbt

structure like MD5

G 10 GeneralOLE objects

structure like MD5, Fp

2 2 short int binary int max +/- 32767 FS

4 4 long int binary int max +/- 2147483647 FS

8 8 double binary signed double IEEE FS

Byte Size Description Applies for (supported by)

0 1 deleted flag "*" or not deleted " " all

1…n 1…x-times contents of fields, fixed length, unterminated.

For n, see (2) byte 10…11

All


TX_NAME Text Name of the transmitter

FILE_NAME Text Name (and optionally, path) of .los file

MODEL_NAME Text Name of propagation model used to calculate path loss

MODEL_SIG Text

Signature (identity number) of model used in calculations. You may check it in the

propagation model properties (General tab).

The Model_SIG is used for the purpose of validity. A unique Model_SIG is

assigned to each propagation model. When model parameters are modified, the

associated model ID changes. This enables Atoll to detect path loss matrix

invalidity. In the same way, two identical propagation models in different projects

do not have the same model IDa.

ULXMAP Float X-coordinate of the top-left corner of the path loss matrix upper-left pixel

ULYMAP Float Y-coordinate of the top-left corner of the path loss matrix upper-left pixel

RESOLUTION Float Resolution of path loss matrix in metre

NROWS Float Number of rows in path loss matrix

NCOLS Float Number of columns in path loss matrix

FREQUENCY Float Frequency band

TILT Float Transmitter antenna mechanical tilt

AZIMUTH Float Transmitter antenna azimuth




3.18.2 LOS File

The data file is a 16 bits binary row file organized in a standard row-column structure. It contains an integer path loss value,

with a 1/16 dB unit. Data are stored starting from the southwest to the northeast corner of the area.

3.19 Externalised Tuning FilesAtoll can tune path loss matrices obtained from propagation results by the use of real measurements (CW Measurements

or Test Mobile Data). For each measured transmitter, Atoll tries to merge measurements and predictions on the same

points and to smooth the surrounding points of the path loss matrices for homogeneity reasons. A transmitter path loss

matrix can be tuned several times by the use of several measurement paths. All these tuning paths are stored in a cata-

logue. This catalogue is stored under a .tuning folder containing a .dbf file and one .pts file per corrected transmitter. Since

a tuning file can contain several measurement paths, all these measurements are added to the tuning file.

For more information on the path loss tuning algorithm, See "Path Loss Tuning" on page 113.

3.19.1 DBF File

dBASE III file (pathloss.dbf) has a standard .dbf format described below. Its content can be checked by opening it in MS- Access. The format is detailed hereafter.

3.19.1.1 DBF File Format

For general information, the format of .dbf files in any Xbase language is described.

Following notations are used in tables:

TX_HEIGHT Float Transmitter height in metre

TX_POSX Float X-coordinate of the transmitter

TX_POSY Float Y-coordinate of the transmitter

ALTITUDE Float Ground height above sea level at the transmitter in metre

RX_HEIGHT Float Receiver height in metre

ANTENNA_SI FloatLogical number referring to antenna pattern. Antennas with the same pattern will

have the same number.

MAX_LOS FloatMaximum path loss stated in 1/16 dB. This information is used, when no

calculation radius is set, to check the matrix validity.

CAREA_XMIN Float Lowest x-coordinate of centre pixel located on the calculation radiusb

CAREA_XMAX Float Highest x-coordinate of centre pixel located on the calculation radius

CAREA_YMIN Float Lowest y-coordinate of centre pixel located on the calculation radius

CAREA_YMAX Float Highest y-coordinate of centre pixel located on the calculation radius

WAREA_XMIN Float Lowest x-coordinate of centre pixel located in the computation zonec

WAREA_XMAX Float Highest x-coordinate of centre pixel located in the computation zone

WAREA_YMIN Float Lowest y-coordinate of centre pixel located in the computation zone

WAREA_YMAX Float Highest y-coordinate of centre pixel located in the computation zone

LOCKED Boolean

Locking status

0: path loss matrix is not locked

1: path loss matrix is locked.

INC_ANT Boolean

Atoll indicates if losses due to the antenna pattern are taken into account in the

path loss matrix.0: antenna losses not taken into account

1: antenna losses included

a. In order to benefit from the calculation sharing feature, users must retrieve the propagation models from the same

central database. This can be done using the Open from database command for a new document or the Refresh

command for an existing one. Otherwise, Atoll generates different model_ID (even if same parameters are applied

on the same kind of model) and calculation sharing become unavailable due to inconsistency.

b. These coordinates enable Atoll to determine the area of calculation for each transmitter.

c. These coordinates enable Atoll to determine the rectangle including the computation zone.

FS = FlagShip D3 = dBaseIII+

Fb = FoxBase D4 = dBaseIV

Fp = FoxPro D5 = dBaseV

CL = Clipper



© Forsk 2010 AT282_TRG_E2 65


3.19.1.1.1 DBF Structure

3.19.1.1.2 DBF Header (Variable Size - Depends on Field Count)

• Field descriptor array in the .dbf header (32 bytes for each field)

Byte Description

0...n .dbf header (see next part for size, byte 8)

n+1

1st record of fixed length (see next parts)

2nd record (see next part for size, byte10) …

last record

If .dbf is not empty

last optional: 0x1a (eof byte)


00 1 0x03 plain .dbf FS, D3, D4, D5, Fb, Fp, CL

0x04 plain .dbf D4, D5 (FS)

0x05 plain .dbf D5, Fp (FS)

0x43 with .dbv memo var size FS

0xB3 with .dbv and .dbt memo FS

0x83 with .dbt memo FS, D3, D4, D5, Fb, Fp, CL

0x8B with .dbt memo in D4 format D4, D5

0x8E with SQL table D4, D5

0xF5 with .fmp memo Fp

01 3 YYMMDD Last update digits All

04 4 ulong Number of records in file All

08 2 ushort Header size in bytes All

10 2 ushort Record size in bytes All

12 2 0,0 Reserved All

14 1 0x01 Begin transaction D4, D5

0x00 End Transaction D4, D5


15 1 0x01 Encrypted D4, D5

0x00 normal visible All

16 12 0 (1) multi-user environment use D4,D5

28 1 0x01 production index exists Fp, D4, D5

0x00 index upon demand All

29 1 n language driver ID D4, D5

0x01 codepage437 DOS USA Fp

0x02 codepage850 DOS Multi ling Fp

0x03 codepage1251 Windows ANSI Fp

0xC8 codepage1250 Windows EE Fp


30 2 0,0 reserved All

32 n*32 Field Descriptor, (see next paragraph) all

+1 1 0x0D Header Record Terminator all


0 11 ASCI field name, 0x00 termin all

11 1 ASCI field type (see next paragraph) all

12 4 n,n,n,n Fld address in memory D3

n,n,0,0 offset from record begin Fp

0,0,0,0 ignored FS, D4, D5, Fb, CL

16 1 byte Field length, bin (see next paragraph) all \ FS,CL: for C field type

17 1 byte decimal count, bin all / both used for fld lng

18 2 0,0 reserved all

20 1 byte Work area ID D4, D5




• Field type and size in the .dbf header, field descriptor (1 byte)

3.19.1.1.3 Each DBF Record (Fixed Length)

3.19.1.2 DBF File Content

The .dbf file provides information about the measured transmitters participating in the tuning.

0x00 unused FS, D3, Fb, Fp, CL

21 2 n,n multi-user dBase D3, D4, D5

0,0 ignored FS, Fb, Fp, CL

23 1 0x01 Set Fields D3, D4, D5

0x00 ignored FS, Fb, Fp, CL

24 7 0...0 reserved all

31 1 0x01 Field is in .mdx index D4, D5


Size Type Description/Storage Applies for (supported by)

C 1...n Char ASCII (OEM code page chars)

rest= space, not \0 term.all

n = 1...64kb (using deci count) FS

n = 1...32kb (using deci count) Fp, CL

n = 1...254 all

D 8 Date 8 ASCII digits (0...9) in the YYYYMMDD format all

F 1...n Numeric


variable pos. of float.point

n = 1...20

FS, D4, D5, Fp

N 1...n Numeric


fix posit/no float.point all

n = 1...20 FS, Fp, CL

n = 1...18 D3, D4, D5, Fb

L 1 Logical ASCII chars (YyNnTtFf space) FS, D3, Fb, Fp, CL

ASCII chars (YyNnTtFf?) D4, D5 (FS)

M 10 Memo10 digits repres. the start block posit. in .dbt file, or 10 spaces if

no entry in memoall

V 10 Variable

Variable, bin/asc data in .dbv

4bytes bin= start pos in memo

4bytes bin= block size

1byte = subtype

1byte = reserved (0x1a)

10 spaces if no entry in .dbv

FS

P 10 Picture binary data in .ftpstructure like M

Fp

B 10 Binarybinary data in .dbt

structure like MD5

G 10 GeneralOLE objects

structure like MD5, Fp

2 2 short int binary int max +/- 32767 FS

4 4 long int binary int max +/- 2147483647 FS

8 8 double binary signed double IEEE FS

Byte Size Description Applies for (supported by)

0 1 deleted flag "*" or not deleted " " all

1…n 1…x-times contents of fields, fixed length, unterminated.

For n, see (2) byte 10…11 All


TX_NAME Text Name of the transmitter

FILE_NAME Text Name (and optionally, path) of .pts file

AREA_XMIN Float Not used

AREA_XMAX Float Not used



© Forsk 2010 AT282_TRG_E2 67


3.19.2 PTS File

The tuning file contains a header and the l ist of points.

The contents of the header is:

• 4 bytes : version

• 4 bytes : flag (can be used to manage flags like active flag)

• 50 bytes : GUID• 4 bytes : Number of points

• 255 bytes : original measurements name (with prefix Num : for test mobile data and CW: for CW measurements)

• 256 bytes : comment

• 4 bytes : X_RADIUS

• 4 bytes : Y_RADIUS

• 4 bytes : Gain : measurement gain - losses

• 4 bytes : Global error

• 4 bytes : Rx height

• 4 bytes : Frequency

• 8 bytes : Tx Position

The list of points contains following 4-uplet for all points

• 4 bytes : X

• 4 bytes : Y

• 4 bytes : Measurement value

• 4 bytes : Incidence angle.

3.20 Interference Histograms File FormatsInterference histograms required by automatic frequency planning tools can be imported and exported.

3.20.1 One Histogram per Line (.im0) Format

This file contains one histogram per line for each interfered/interfering subcell pair. The histogram is a list of C/I values

with associated probabilities.

The .im0 file consists of two 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 histogram of each interfered subcell-interferer subcell pair.

The lines after the header are considered as comments if they start with the symbol "#". If not, they must have the following

format:

AREA_YMIN Float Not used

AREA_YMAX Float Not used

Notes:

• No validity check is carried out when importing an interference histogram file.

• Atoll only imports interference histograms related to loaded transmitters.

• The lines starting with the symbol "#" are considered as comments.

• The interferer TRX type is not specified. In fact, the subcells of the interferer transmitter

differ by their power offsets. If the power offset of a subcell is X with respect to the BCCH,

then its interference C/I histogram will be shifted by X with respect to the BCCH

interference histogram. It contains no further information; therefore, the interferer TRX type

is always BCCH.

• For each interfered subcell-interferer subcell pair, Atoll saves probabilities for several C/I

values (between 6 to 24 values). Five of these values are fixed; probabilities are calculated

for C/I values equal to –9, 1, 8, 14, and 22 dB. Then, between each fixed C/I value, there

can be up to three additional values (this number depends on the probability variation

between the fixed values). The C/I values have 0.5 dB accuracy and probability values are

calculated and stored with an accuracy of 0.002 for probabilities between 1 and 0.05, and

with an accuracy of 0.0001 for probabilities lower than 0.05.

• If no power offset is defined on the Interfered TRX type, it is possible to use the "All" value.

• The values of probability should be absolute (between 0 and 1), and not in precentage

(between 0 and 100%).

# Calculation Results Data File.

# Version 1.1, Tab separated format. Commented lines start with #.




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

3.20.1.1 Sample

3.20.2 One Value per Line with Dictionary File (.clc) FormatAtoll creates two ASCII text files in a specified directory: xxx.dct and xxx.clc (xxx is the user-specified name).

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

Column name Description

Column1 Interfered transmitter Name of the interfered transmitter.

Column2 Interfering transmitter Name of the interferer transmitter.

Column3 Interfered TRX typeInterfered subcell. In order to save storage, all subcells with no power

offset are not duplicated (e.g. BCCH, TCH).

Column4 C/I Probability

C/I value and the probability associated to this value separated by a space

character. This entry cannot be null.



# Remark: C/I results do not incorporate power offset values.

# Fields are:

#------------------------------------------------------------------------

#Transmitter Interferer TRX type {C/I Probability} values

#------------------------------------------------------------------------

#

# Warning, The parameter settings of this header can be wrong if

# the "export" is performed following an "import". They

# are correct when the "export" follows a "calculate".

#

# Service Zone Type is "Best signal level of the highest priority HCS layer".

# Margin is 5.

# Cell edge coverage probability 75%.

# Traffic spreading was Uniform

##---------------------------------------------------------------------#

#

Site0_2 Site0_1 BCCH,TCH -10 1 -9 0.996 -6 0.976 -4 0.964 -1 0.936

0 0.932 1 0.924 4 0.896 7 0.864 8 0.848

9 0.832 10 0.824 11 0.804 14 0.712 17 0.66

Site0_2 Site0_3 BCCH,TCH -10 1 -9 0.996 -6 0.976 -4 0.972 -1 0.948

0 0.94 1 0.928 4 0.896 7 0.856 8 0.84

11 0.772 13 0.688 14 0.636 15 0.608 18 0.556

Site0_3 Site0_1 BCCH,TCH -10 1 -9 0.996 -6 0.98 -3 0.948 0 0.932

1 0.924 4 0.892 7 0.852 8 0.832 9 0.816

10 0.784 11 0.764 14 0.644 15 0.616 18 0.564

Site0_3 Site0_2 BCCH,TCH -9 1 -6 0.972 -3 0.964 -2 0.96 0 0.94

1 0.932 4 0.904 7 0.876 8 0.86 9 0.844

11 0.804 13 0.744 14 0.716 15 0.692 18 0.644

Note:

• When importing interference histograms with standard 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 this file is unavailable, Atoll assumes that the transmitter

identifiers are the transmitter names. In this case, the columns 1 and 2 of the .clc file must

contain the names of the interfered and interferer transmitters instead of their identification

numbers.



© Forsk 2010 AT282_TRG_E2 69


3.20.2.1 CLC File

3.20.2.1.1 Description

The .clc file consists of two parts:




format:


3.20.2.1.2 Sample



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


Column1 Interfered transmitter Identification number of the interfered transmitter. If the column is empty,

its value is identical to the one of the line above.

Column2 Interfering transmitter Identification number of the interferer transmitter. If the column is null, its

value is identical to the one of the line above.

Column3 Interfered TRX type

Interfered subcell. If the column is null, its value is identical to the one of

the line above. In order to save storage, all subcells with no power offset

are not duplicated (e.g. BCCH, TCH).

Column4 C/I threshold C/I value. This column cannot be null.

Column5 Probability C/I > ThresholdProbability to have C/I the value specified in column 4 (C/I threshold). This

field must not be empty.

Note:

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




# Fields are:

##------------#------------#------------#-----------#------------------#

#| Interfered | Interfering| Interfered | C/I | Probability |

#| Transmitter| Transmitter| Trx type | Threshold | C/I >= Threshold |

##------------#------------#------------#-----------#------------------#

#




#


# Margin is 5.



##---------------------------------------------------------------------#

1 2 TCH_INNER 8 1

9 0.944

10 0.904

11 0.892

14 0.844




3.20.2.2 DCT File

3.20.2.2.1 Description

The .dct file is divided into two parts:


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


format:

The last four columns describe the interference matrix scope. One transmitter per line is described separated with a tab

character.

3.20.2.2.2 Sample

15 0.832

16 0.812

17 0.752

22 0.316

25 0.292

1 2 BCCH,TCH 8 1

9 0.944

10 .904

13 0.872

14 0.84

17 0.772

Note:

• If the TCH and BCCH histograms are the same, they are not duplicated. A single record

indicates that the histograms belong to TCH and BCCH both. For example, instead of:

1 2 TCH -9.5 1 - 9 1 - 6 1

1 2 BCCH -9.5 1 - 9 1 - 6 1

We have:

1 2 TCH,BCCH -9.5 1 - 9 1 - 6 1

# Calculation Results Dictionary File.


<Column1><tab><Column2><newline>

Column name Type Description

Column1 Transmitter name Text Name of the 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

# Calculation Results Dictionary File.


# Fields are:

##-----------#-----------#-----------#-----------#---------#---------#

#|Transmitter|Transmitter|BCCH during|BSIC during|% of vic'|% of int'|

#|Name |Identifier |calculation|calculation|coverage |coverage |

##-----------#-----------#-----------#-----------#---------#---------#

#




© Forsk 2010 AT282_TRG_E2 71


3.20.3 One Value per Line (Transmitter Name Repeated) (.im1)

FormatThis file contains one C/I threshold and probability pair value per line for each interfered/interfering subcell pair. The histo-

gram is a list of C/I values with associated probabilities.

The .im1 file consists of two parts:




format:


3.20.3.1 Sample



#

# Service Zone Type is "Best signal level per HCS layer".

# Margin is 5.

# Cell edge coverage probability is 75%.

# Traffic spreading was Uniform (percentage of interfered area)

##---------------------------#

Site0_0 1 -1 -1 100 100

Site0_1 2 -1 -1 100 100

Site0_2 3 -1 -1 100 100

Site1_0 4 -1 -1 100 100

Site1_1 5 -1 -1 100 100

Site1_2 6 -1 -1 100 100

Site2_0 7 -1 -1 100 100

Site2_1 8 -1 -1 100 100



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




Column3 Interfered TRX typeInterfered subcell. In order to save storage, all subcells with no power

offset are not duplicated (e.g. BCCH, TCH).

Column4 C/I threshold C/I value. This column cannot be null.

Column5 Probability C/I > ThresholdProbability to have C/I the value specified in column 4 (C/I threshold). This

field must not be empty.




# Fields are:

#------------------------------------------------------------------------

#Transmitter Interferer TRX type C/I Probability

#------------------------------------------------------------------------

#




3.20.4 Only Co-Channel and Adjacent Values (.im2) Format

In this case, there is only one .im2 file containing co-channel and adjacent channel interference probabilities specified for

each interfered transmitter – interferer transmitter pair. There is only one set of values for all the subcells of the interfered

transmitter.

Each line must have the following format:

Where the separator (<SEP>) can either be a tab or a semicolon.

The four columns are defined in the table below:

corresponds to the required C/I threshold. This parameter is defined for each subcell.

is the adjacent channel protection level.

3.20.4.1 Sample




#


# Margin is 5.



##---------------------------------------------------------------------#

Site0_2 Site0_1 BCCH,TCH -10 1

Site0_2 Site0_1 BCCH,TCH -9 0.996




Site0_2 Site0_1 BCCH,TCH 0 0.932







...

<Column1><SEP><Column2><SEP><Column3><SEP><Column4><newline>




Column3Co-channel interference

probabilityProbability of having

Column4 Adjacent channel

interference probabilityProbability of having

C I e Ma x BCCH TCH ,

C I re q e d

C I e Ma x BCCH TCH ,

C I re q e F – d

C I req e

F




# Fields are:

#------------------------------------------------------------------------

#Transmitter Interferer Co-channel Adjacent channel

#------------------------------------------------------------------------



© Forsk 2010 AT282_TRG_E2 73


The columns in the sample above are separated with a tab. These columns can also be separated with a semilcolon:

3.21 Antenna FormatsThe following sections describe the 2D and 3D antenna pattern formats in Atoll. Antenna patterns can be imported from

txt, csv, or xls files.

3.21.1 2D Antenna Pattern Format

The 2D antenna pattern is stored in the DIAGRAM field of the Antennas table.

The format of 2D antenna patterns containing co-polar diagrams only can be understood from Figure 3.1: on page 73.

The contents of the DIAGRAM field are formatted as follows:

• Pattern Descriptor 1: Space-separated list of parameters.

- First entry: The number of co-polar diagrams. For example, 2.

- Second entry: First co-polar diagram type = 0 for azimuth (horizontal) diagram.

- Third entry: The elevation angle of the azimuth diagram.

- Fourth entry: The number of angle-attenuation pairs in the first co-polar diagram. For example, 360.

• Co-polar Horizontal Diagram: Horizontal co-polar diagram (the second entry in the preceding descriptor is 0).

The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....

• Pattern Descriptor 2: Space-separated list of parameters.- First entry: Second co-polar diagram type = 1 for elevation (vertical) diagram.

- Second entry: The azimuth angle of the elevation diagram.

- Third entry: The number of angle-attenuation pairs in the second co-polar diagram. For example, 360.

• Co-polar Vertical Diagram: Vertical co-polar diagram (the first entry in the preceding descriptor is 1). The format

is space-separated angle attenuation pairs. For example, 0 0 1 0.1....

• End: The number cross-polar diagrams = 0.

The format of 2D antenna patterns containing co-polar and cross-polar diagrams can be understood from Figure 3.2: on

page 74.

#




#


# Margin is 5.



##---------------------------------------------------------------------#

Site0_2 Site0_1 0.226667 0.024

Site0_2 Site0_3 0.27 0.024

Site0_3 Site0_1 0.276 0.02

Site0_3 Site0_2 0.226 0.028

Site0_2;Site0_1;0.226667;0.024

Site0_2;Site0_3;0.27;0.024

Site0_3;Site0_1;0.276;0.02

Site0_3;Site0_2;0.226;0.028

Figure 3.1: 2D Antenna Pattern Format Containing Co-polar Diagrams Only

PatternEndCo-polar Horizontal Diagram Co-polar Vertical Diagram

2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 … 0

Discriptor 1Pattern

Discriptor 2




The contents of the DIAGRAM field are formatted as follows:


- First entry: The number of co-polar diagrams. For example, 2.

- Second entry: First co-polar diagram type = 0 for azimuth (horizontal) diagram.


- Fourth entry: The number of angle-attenuation pairs in the first co-polar diagram. For example, 360.

• Co-polar Horizontal Diagram: Horizontal co-polar diagram (the second entry in the preceding descriptor is 0).

The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....


- First entry: Second co-polar diagram type = 1 for elevation (vertical) diagram.


-Third entry:

The number of angle-attenuation pairs in the second co-polar diagram. For example, 360.• Co-polar Vertical Diagram: Vertical co-polar diagram (the first entry in the preceding descriptor is 1). The format

is space-separated angle attenuation pairs. For example, 0 0 1 0.1....


- First entry: The number of cross-polar diagrams. For example, 2.

- Second entry: First cross-polar diagram type = 0 for azimuth (horizontal) diagram.


- Fourth entry: The number of angle-attenuation pairs in the first cross-polar diagram. For example, 360.

• Cross-polar Horizontal Diagram: Horizontal cross-polar diagram (the second entry in the preceding descriptor

is 0). The format is space-separated angle attenuation pairs. For example, 0 0 1 0 2 0.1....


- First entry: Second cross-polar diagram type = 1 for elevation (vertical) diagram.


- Third entry: The number of angle-attenuation pairs in the second cross-polar diagram. For example, 360.

• Cross-polar Vertical Diagram: Vertical cross-polar diagram (the first entry in the preceding descriptor is 1). The

format is space-separated angle attenuation pairs. For example, 0 0 1 0.1....

An antenna pattern described in a text format may be converted to a binary format using a converter. In binary, each

antenna is described by a header and a list of value pairs.

The header is defined as follows:

• flag : (Integer, 32 bits) 0 for omni diagrams, 1 for directional

• num: (Short integer, 16 bits) Number of diagrams (0, 1, 2, 3, 4)

• siz0 : (Short integer, 16 bits) Size of the first diagram (horizontal co-polar section, elevation = 0°)

• siz1: (Short integer, 16 bits) Size of the second diagram (vertical co-polar section, azimuth = 0°)

• siz2 : (Short integer, 16 bits) Size of the third diagram (horizontal cross-polar)

• siz3: (Short integer, 16 bits) Size of the fourth diagram (vertical cross-polar)

• prec : (Short integer, 16 bits) Precision of the following angle values (100)

Then follows the content of each of the defined diagrams, i.e., the diagrams whose sizes (siz0 , siz1, siz2 , siz3) are not

zero. Each diagram consists of a list of value pairs. The number of value pairs in a list depends on the value of the siz0 ,

siz1, siz2 , and siz3 parameters. For example, siz2 = 5 means there are five value pairs in the third diagram.

The value pairs in each list are:

• ang : (Short integer, 16 bits) The first component of the value pair is the angle in degrees multiplied by 100. For

example, 577 means 5.77 degrees.

• loss: (Short integer, 16 bits) The second component of the value pair is the loss in dB for the given angle ang .

All the lists of value pairs are concatenated without a separator.

3.21.2 3D Antenna Pattern Format

The 3D antenna pattern format is as follows:

• Header: The text file may contain a header with additional information. When you import the antenna pattern you

can indicate the row number in the file where the header ends and the antenna pattern begins.

Figure 3.2: 2D Antenna Pattern Format Containing Co-polar and Cross-polar Diagrams

Pattern Co-polar Horizontal Diagram Co-polar Vertical Diagram

2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 …

Discriptor 1Pattern

Discriptor 2

Pattern Cross-polar Horizontal Diagram Cross-polar Vertical DiagramDiscriptor 3

PatternDiscriptor 4

2 0 0 360 0 0 1 0 2 0.1 … 1 0 360 0 0 1 0.1 …



© Forsk 2010 AT282_TRG_E2 75


• Antenna Pattern: Each row contains three values to describe the 3D antenna pattern. The columns containing

the values can be in any order:

- Azimuth: Allowed range of values is from 0° to 360°. The smallest increment allowed is 1°.

- Tilt: Allowed range of values is from -90° to 90° or from 0° to 180°. The smallest increment allowed is 1°.

- Attenuation: The attenuation in dB.






Chapter 4Calculations

This chapter describes in detail the calculation of path losses, the propagation models implemented in Atoll by default, the calculation of antenna

attenuation according to antenna patterns, and other calculation algorithms in Atoll .






© Forsk 2010 AT282_TRG_E2 79

Chapter 4: Calculations

4 Calculations

4.1 OverviewThree kinds of predictions are available in Atoll:

• Point analysis enables you to visualise transmitter-receiver profile and to get predictions for a user-defined

receiver in real time anywhere on a geographic map (Point analysis window: Profile tab).

• Coverage studies consider each bin of calculation areas as a potential receiver you can define. Therefore, covered

bins correspond to areas where a criterion on the predicted received signal is fulfilled.• Point analysis based on path loss matrices enables you to get parameters derived from predicted values in cov-

erage studies (field received, path loss, C/I, UMTS parameters) for a receiver anywhere inside a calculation area

(Point analysis window: Reception, Interference, AS analysis tabs ).

An overview of different analysis methods is presented in the table below:

In any case, prediction is performed in three steps:

1st step: First of all, Atoll calculates the path loss ( ), using the selected propagation model.

is the loss on the transmitter-receiver path calculated through the propagation model. value depends on

the selected propagation model.

is the transmitter antenna attenuation (from antenna patterns).

is the receiver antenna attenuation ( ) (from antenna patterns).

2nd step: When the option “Shadowing taken into account” is selected, Atoll evaluates a shadowing margin,

, from the user-defined model standard deviation at the receiver and the cell edge coverage probability.

Coverage studies Point analysisPoint analysis based on path loss

matrices

Any study ProfileReception, Results,

Interference, AS analysis

Receiver

position

At the centre of each

calculation bin within

calculation areas

Anywhere. Even beyond

computation zone Anywhere inside the calculation areas

CalculationPath loss matrix

calculationReal time

No calculation: result coming from path

loss matrices

Profile

extractionaRadial except when

using SPMSystematic

Method used for coverage studies: radial

except when using SPM

ResultOne value inside a

calculation bin

Different values inside a

calculation binOne value inside a calculation bin

a. When using SPM, you can choose either radial or systematic calculation option.

Notes:

• In coverage studies, Atoll calculates path loss for every bin within calculation areas.

However, only results on calculation bins inside the computation zone are displayed.

• Profile point analysis is calculated in real time. Therefore, prediction is always consistent

with the network. On the other hand, if you modify any parameter (radio or geo), which may

make matrices invalid, consider updating the matrices before using point analysis based onpath loss matrices.

• Due to different calculation methods, you can get different results at a same point when

performing a point analysis in profile or reception mode.

Notes:

• In any project, Atoll considers that the receiver antenna is in the transmitter antenna axis.

Therefore, the receiver antenna attenuation is supposed to be zero.

• Transmitter antenna attenuation may not be considered in this step. It depends on

propagation model provider, who may choose to include this parameter in

calculation. However, all the propagation models available in Atoll calculate by

considering transmitter antenna attenuation.

L pat h

L pat h Lmodel Lant Tx Lant Rx

+ +=

Lmodel Lmodel

Lant Tx

Lant Rx Lant Rx 0 =

L pat h

L pat h

M Shadowing model –




3rd step: Then, Atoll determines the prediction criterion and displays coverage.

For a signal level study,

The signal level at the receiver ( ) is calculated. We have (in dBm):

Where

is the effective isotropic radiated power of the transmitter.

is the transmitter power.

is the transmitter antenna gain.

are transmitter losses.

is the shadowing margin.

are the indoor losses, taken into account when the option “Indoor coverage” is selected,

are receiver losses.

is the receiver antenna gain.

The prediction is performed for a user-defined cell edge coverage probability (x%). This means that the measured criterion

exceeds the predicted criterion for x% of time. The prediction is reliable during x% of time.

4.2 Path Loss MatricesAtoll is able to calculate two path loss matrices per transmitter, a first matrix over a smaller radius computed with a high

resolution and a propagation model (main matrix), and a second matrix over a larger radius computed with a low resolution

and another propagation model (extended matrix).

To be considered for calculations, a transmitter must fulfil the following conditions:

• It must be act ive,

• It must satisfy filter criteria defined in the Transmitters folder, and• It must have a calculation area.

In the rest of the document, a transmitter fulfilling the conditions detailed above will be called TBC transmitter.

The path loss matrix size of a TBC transmitter depends on its calculation area. Atoll determines a path loss value ( )

on each calculation bin (calculation bin is defined by the resolution) of the calculation area of the TBC transmitter. You may

have one or two path loss matrices per TBC transmitter.

Note:

• For a cell edge coverage probability of 50%, the shadowing margin is always zero. In this

case, Atoll still works as above.

Notes:

• In UMTS and CDMA documents, and .

• In UMTS and CDMA documents, Atoll considers that and equal zero when

calculating the received signal level (in point analysis, Profile and Reception tabs, and in

common coverage studies such as Coverage per transmitter, Coverage by field level,

Overlapping).

• In GSM_EGPRS documents, .

• In GSM_EGPRS documents, receiver is equipped with an antenna with zero gain.

Note:

• In case of interference studies, only signal from interfered transmitter (C) is downgraded by

the shadowing margin. We consider that interference value (I) is not altered by the

shadowing margin.

P Rec

P Rec E IR P L pat h – M Shadowing model – – LIndoor – Gant Rx LRx – +=

EIRP P Tx Gant Tx LTx – +=

EIRP

P Tx

Gant Tx

LTx


LIndoor

LRx

Gant Rx

P Tx P Pilot = LTx Lto ta l DL – =

Gant Rx LRx

LTx Lto ta l DL – =

L pat h



© Forsk 2010 AT282_TRG_E2 81


4.2.1 Calculation Area Determination

4.2.1.1 Computation Zone

Transmitter calculation area is made of a rectangle or a square depending on transmitter calculation radius and the compu-

tation zone.

Calculation radius enables Atoll to define a square around the transmitter. One side of the square equals twice the entered

calculation radius.

Since the computation zone can be made of one or several polygons, transmitter calculation area corresponds to the inter-

section area between its calculation square and the rectangle containing the computation zone area(s).

4.2.1.2 Use of Polygonal Zones in Coverage Prediction Reports

Prediction statistics are evaluated over the focus zone, if existing, then over the computation zone, if existing, or over the

whole covered area. The area of the focus and computation zones are calculated by decomposition in triangles.

The area of each prediction is calculated by counting its pixels inside the focus (resp. computation) zone. This number of

pixels multiplied by the area of one of its pixels gives the total area.

This area depends on the study resolution. At the border of the focus (resp. computation) zone, pixels are considered

either IN or OUT of the zone. A pixel is IN if its centre is inside the focus zone.

If a prediction covers the entire focus (resp. computation) zone, its area should be equal to the focus (resp. computation)

zone area, but as these 2 different methods differ, the results may be slightly different. If it happens that the value of the

prediction area is higher than the focus zone area, then the calculated percentage value is higher than 100%. In that case,

Atoll automatically replaces it by 100%.

Figure 4.1: Example 1: Single Calculation Area

Figure 4.2: Example 2: Multiple Calculation Areas

Computation zone(s)

Rectangle containing the computation zone(s)

Calculation area defined (square)

Transmitter

Calculation area: real area for which Atoll calculates path losses





© Forsk 2010 AT282_TRG_E2 83


4.3 Path Loss Calculations

4.3.1 Ground Altitude Determination

Atoll determines reception and transmission site altitude from Digital Terrain Model map. The method used to evaluate

site altitude is based on a bil inear interpolation. It is described below.

Let us suppose a site S located inside a bin. Atoll knows the altitudes of four bin vertices, S’1, S’’1, S’2 and S’’2, from the

DTM file (Centre of each DTM pixel).

1st step: Atoll draws a vertical line through S. This line respectively intersects (S’1,S’’1) and (S’2, S’’2) lines at S1 and S2.

2nd step: Atoll determines the S1 and S2 altitudes using a linear interpolation method.

New clutter class edition Invalid Path loss matrices Insufficientb Necessary

Coverage study resolution Valid Prediction study Sufficient Not necessary

Cell edge coverage probability Valid Prediction study Sufficient Not necessary

Coverage study conditions Valid Prediction study Sufficient Not necessary

Coverage study display options Valid Prediction study Sufficient Not necessary

a.Modification of any parameter related to main or other antennas makes matrix invalid.

b.Except if this action has an impact on the site positions/altitudes.

Tip 1Calculate or Force Calculation?

If you modify radio data or calculation areas, use the Calculate button. On the other hand, if you change geographic

data, it is necessary to use Force calculation.

Tip 2Calculation area management

When performing prediction studies, it is recommended to follow this methodology to minimise recalculations:

1st step: Calculate without computation zone.

2 nd step: Draw a computation zone and calculate.

3rd step: Decrease the calculation radius and calculate.

Figure 4.3: Ground Altitude Determination - 1





3rd step: Atoll performs a second linear interpolation to evaluate the S altitude.

4.3.2 Clutter Determination

Some propagation models need clutter class and clutter height as information at receiver or along a transmitter-receiver

profile.

4.3.2.1 Clutter Class

Atoll uses clutter classes file to determine the clutter class.

4.3.2.2 Clutter Height

To evaluate the clutter height, Atoll uses clutter heights file if available in the .atl document; clutter height of a site is the

height of the nearest point in the file.

Example: Let us suppose a site S. In the clutter heights file, Atoll reads clutter heights of four points around the site, S’1,

S’’1, S’2 and S’’2. Here, the nearest point to S is S”2; therefore Atoll takes the S”2 clutter height as clutter height of S.

If you do not have any clutter height file, Atoll takes clutter height information in clutter classes file. In this case, clutter

height is an average height related to a clutter class.

4.3.3 Geographic Profile Extraction

Geographic profile extraction is needed in order to calculate diffraction losses. Profiles can be based on DTM only or on

DTM and clutter both. In fact, it depends on the selected propagation model.

4.3.3.1 Extraction Methods

4.3.3.1.1 Radial Extraction

Atoll draws radials from the site (where transmitter is located) to each calculation bin located along the transmitter calcu-

lation area border. In other words, Atoll determines a geographic profile between site and each bin centre.



Figure 4.7: Clutter Height



© Forsk 2010 AT282_TRG_E2 85


The receiver may be located either anywhere within a calculation bin (Point prediction) or at the centre of a calculation bin

(Coverage study). Therefore, according to the receiver position, Atoll chooses the nearest profile and uses it (receiver is

considered as located on the profile) to perform prediction study at the receiver.

4.3.3.1.2 Systematic Extraction

In this case, Atoll systematically extracts a geographic profile between the site (where transmitter resides) and the

receiver.

Figure 4.8: Radial calculation method

Transmitter

Radial: Atoll will extract a geographic profile for each radial

Centre of a bin located on the calculation border

Receiver: it may be anywhere in point analysis or at the centre of each calculation bin in coverage studies

Figure 4.9: Site-bin centre profile




4.3.3.2 Profile Resolution: Multi-Resolution Management

Geographic profile resolution depends on resolution of geographic data used by the propagation model (DTM and/or clut-

ter).

1. 1st case: If the chosen propagation model considers both DTM and clutter heights along the profile, the profile

resolution will be the highest of the two.Example 1: Standard Propagation Model is used to perform predictions. A DTM map with a 40 m resolution and

a clutter heights map with a 20 m resolution are available.

Both DTM and clutter maps are considered when using the Standard propagation model. Therefore, here, the

profile resolution will be 20 m. It means that Atoll will extract geographic information, ground altitude and clutter

height, every 20 m. To get ground altitude every 20m, Atoll uses the bilinear interpolation method described in

"Ground Altitude Determination" on page 83. Clutter heights are read from the clutter heights map.Atoll takes the

clutter height of the nearest point every 20m (see Path loss calculations: Clutter determination).

Example 2: Standard Propagation Model is used to perform predictions. A DTM map with a 40 m resolution and

a clutter classes map with a 20 m resolution are available. No clutter height file has been imported in .atl document.

Both DTM and clutter maps are considered when using the Standard propagation model. Therefore, here, the

profile resolution will be 20 m. It means that Atoll will extract geographic information, ground altitude and clutter

height, every 20 m. To get ground altitude every 20 m, Atoll uses the bilinear interpolation method described in

"Ground Altitude Determination" on page 83. Atoll uses the clutter classes map to determine clutter height. Every

20 m, it determines clutter class and takes associated average height.

2. 2nd case: If the chosen propagation model takes into account only DTM map along the profile, profile resolution

will be the highest resolution among the DTM files.

Example: Cost-Hata is used to perform predictions. Both DTM maps with 40 m and 25 m resolutions and a clutter

map with a 20 m resolution are available.

Figure 4.10: Radial calculation method

Transmitter

Geographic profiles

Receiver: it may be anywhere in point analysis or at the centre of each calculation bin in coverage studies



© Forsk 2010 AT282_TRG_E2 87


Only DTM maps are considered along the whole profile when using Cost-Hata model. Therefore, here, the profile

resolution will be 25 m. It means that Atoll will extract geographic information, only the ground altitude, every

25 m. DTM 1 is on the top of DTM 2. Thus, Atoll will consider ground elevation read from DTM 1 in the definition

area of DTM 1 and DTM 2 elsewhere. To get ground altitude every 25 m, Atoll uses the bilinear interpolation

method described in "Ground Altitude Determination" on page 83.

Explorer window Work space

DTM

• DTM 1 (25m)

• DTM 2 (40m)

Clutter

• Clutter (20m)

Notes:

• The selected profile resolution does not depend on the geographic layer order. In the last

example, whatever the DTM file order you choose, profile resolution will always be 25m.

On the other hand, the geographic layer order will influence the usage of data to establish

the profile.

• The calculation bin of path loss matrices defined by the grid resolution is independent of

geographic file resolution.




4.4 Propagation ModelsPropagation models available in Atoll are listed in the table below along with their main characteristics.

C O S T - H a t a

O

k u m u r a - H a t a

1 5 0 - 2 0 0 0 M H z

L ( d , f , H R x )

( p e r e n v i r o n m e n t )

D i f f r a c t i o n l o s s

D e y g o u t

( 1 o b s t a c l e )

D T M

R a d i a l

M a c r o c e l l

M i n i c e l l

S t r e e t

M o b i l e

G S

M 9 0 0 , G S M 1 8 0 0 ,

U M T S , C D M A 2 0 0 0 ,

L T E

E r c e g - G r e e n s t e i n

( S U I )

1 9 0 0 - 6 0 0 0 M H z

L ( d , f , H T x , H R x )



D e y g o u t

( 1 o b s t a c l e )

D T M

R a d i a l

M a c r o c e l l

M i n i c e l l

S t r e e t

F i x e d

U r b a n a n d s u b u r b a n

a r e a s

1 0 0 m < d < 8 k m

F i x e d W i M A X

I T U 5 2 9 - 3

3 0 0 - 1 5 0 0 M H z

L ( d , f , H R x )



D e y g o u t

( 1 o b s t a c l e )

D T M

R a d i a l

M a c r o c e l l

M i n i c e l l

S t r e e t

M o b i l e

1 < d < 1 0 0 k m

G

S M , C D M A 2 0 0 0 ,

L T E

S t a n d a r d P r o p a g a t i o n M o d e l

1 5 0 - 3 5 0 0 M H z

L ( d , H T x e f f , H R x e f f , D i f f l o s s , c l u t t e r )

D e y g o u t ( 3 o b s t a c l e s )

E p s t e i n - P e t e r s o n ( 3 o b s t a c l e s )

D e y g o u t c o r r e c t e d ( 3 o b s t a c l e s )

M i l l i n g t o n ( 1 o b s t a c l e )

D T M

C l u t t e r

R a d i a l

S y s t e m a t i c

M a c r o c e l l

M i n i c e l l

S t r e e t

R o o f t o p

M o b i l e a n d F i x e d

1 < d < 2 0 k m

G S M , U M T S ,

C D M A 2 0 0 0 , W i M A X , L T E

W L L

3 0 - 1 0 0 0 0

M H z

F r e e s p a c

e l o s s


D e y g o u t

( 3 o b s t a

c l e s )

D T M C l u t t e r

R a d i a l

- S t r e e t

R o o f t o p

F i x e d

F i x e d r e c

e i v e r s

W L L , M i c r o w a v e

l i n k s , W i M A X

I T U 5 2 6 - 5

3 0 - 1 0 0 0 0 M H z

F r e e s p a c e l o s s


D e y g o u t

( 3 o b s t a c l e s )

D e y g o u t c o r r e c t e d

( 3 o b s t a c l e s )

D T M

R a d i a l

M a c r o c e l l

S t r e e t

F i x e d

F i x e d r e c e i v e r s

W L L

I T U 1 5 4 6

3 0

- 3 0 0 0 M H z

F r e e

s p a c e l o s s +

C

o r r e c t i o n s

- - -

M

a c r o c e l l

R o o f t o p

M o b i l e

1 <

d < 1 0 0 0 k m

L a n d

a n d m a r i t i m e

m o b i l e , b r o a d c a s t

I T U 3 7 0 - 7

( V i e n n a 9 3 )

1 0 0 - 4 0 0 M H z

F r e e s p a c e l o s s

C o r r e c t e d

s t a n d a r d

l o s s

- - -

M a c r o c e l l

R o o f t o p

F i x e d

d > 1 0 k m

L o w f r e q u e n c i e s

B r o a d c a s t

P r o p a g a t i o n

m o d e l

F r e q u e n c y

b a n d

P h y s i c a l

p h e n o m e n a

D i f f r a c t i o n

c a l c u l a t i o n

m e t h o d

P r o f i l e

b a s e d o n

P r o f i l e

e x t r a c t i o n

m o d e

C e l l s i z e

R e c e i v e r

l o c a t i o n

R e c e i v e r

U s e



© Forsk 2010 AT282_TRG_E2 89


4.4.1 Okumura-Hata and Cost-Hata Propagation Models

4.4.1.1 Hata Path Loss Formula

Hata formula empirically describes the path loss as a function of frequency, receiver-transmitter distance and antenna

heights for an urban environment. This formula is valid for flat, urban environments and 1.5 metre mobile antenna height.

Path loss (Lu ) is calculated (in dB) as follows:

f is the frequency (MHz).

hTx is the transmitter antenna height above ground (m) (Hb notation is also used in Atoll).

d is the distance between the transmitter and the receiver (km).

The parameters A1, A2 , A3, B1, B2 , and B3 can be user-defined. Default values are proposed in the table below:

4.4.1.2 Corrections to the Hata Path Loss Formula

As described above, the Hata formula is valid for urban environment and a receiver antenna height of 1.5m. For other envi-

ronments and mobile antenna heights, corrective formulas must be applied.

• For urban areas:

• For suburban areas:

• For quasi-open rural areas:

• For open rural areas:

a(hRx ) is a correction for a receiver antenna height different from 1.5m.

4.4.1.3 Calculations in Atoll

Hata models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the receiver.

1st step: For each calculation bin, Atoll determines the clutter bin on which the receiver is located. This clutter bin corre-

sponds to a clutter class. Then, it uses the Hata formula assigned to this clutter class to evaluate .

2nd step: This step depends on whether the ‘Add diffraction loss’ option is checked.

• If the ‘Add diffraction loss’ option is unchecked, Atoll stops calculations.

Notes:

• In formulas described above, is stated in dB.

• Under Physical phenomena, L(...) expressions refer to formulas customisable in Atoll.

• SUI stands for Stanford University Interim models.

Lmodel

Lu A1 A2 f log A3 hTx log B1 B2 hTx log B3hTx + + d log + + +=

ParametersOkumura-Hata

f d1500 MHz

Cost-Hata

f > 1500 MHz

A1 69.55 49.30

A2 26.16 33.90

A3 -13.82 -13.82

B1 44.90 44.90

B2 -6.55 -6.55

B3 0 0

Lmodel1 Lu a hRx – =

Lmodel1 Lu a hRx – 2 f 28 ------

© ¹§ ·log © ¹§ · 2 – 5.4 – =

Lmodel1 Lu a hRx – 4.78 f log 2 – 18.33 f log 35.94 – +=

Lmodel1 Lu a hRx – 4.78 f log 2 – 18.33 f log 40.94 – +=

Environment a(hRx )

Rural/Small city

Large city

1.1 f log 0.7 – hRx 1.56 f log 0.8 – –

3.2 11.75 hRx log 2 4.97 –

Note:

• When receiver antenna height equals 1.5m, a(hRx ) is close to 0 dB regardless of frequency.

Lmodel1

Lmodel Lmodel1=




• If the ‘Add diffraction loss’ option is selected, Atoll proceeds as follows:

a. It extracts a geographic profile between the transmitter and the receiver based on the radial calculation mode.

b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates loss-

es due to diffraction .

4.4.2 ITU 529-3 Propagation Model

4.4.2.1 ITU 529-3 Path Loss Formula

The ITU 529.3 model is a Hata-based model. For this reason, its formula empirically describes the path loss as a function

of frequency, receiver-transmitter distance and antenna heights for a urban environment. This formula is valid for flat,

urban environments and 1.5 metre mobile antenna height.

The standard ITU 529-3 formula, for a receiver located on a urban environment, is given by:

where:

E is the field strength for 1 kW ERP

f is the frequency (MHz).

is the transmitter antenna height above ground (m) (H b notation is also used in Atoll)

is the receiver antenna height above ground (m)

d is the distance between the transmitter and the receiver (km)

b is the distance correction

The domain of validity of such is formula is:

• Frequency range: 300-1500 MHz

• Base Station height: 30-200 m

• Mobile height: 1-10 m

• Distance range: 1-100 km

Since Atoll needs the path loss (Lu ) formula, a conversion has to be made. One can find the following conversion formula:

which gives the following path loss formula for the ITU 529-3 model:

4.4.2.2 Corrections to the ITU 529-3 Path Loss Formula

4.4.2.2.1 Environment Correction

As described above, the Hata formula is valid for urban environment. For other environments and mobile antenna heights,

corrective formulas must be applied.

for large city and urban environments

for suburban area

for rural area

4.4.2.2.2 Area Size Correction

In the formulas above, is the environment correction and is defined according to the area size

Note:

• Like for any Hata-based model, is, by default, limited to the computed free space

loss value. It is also possible to avoid this option (option in the related scrolling menu of

Configuration tab).

Lmodel2

Lmodel Lmodel1 Lmodel2 +=

Lmodel

E 69.82 6.16 f log – 13.82 hTx log 44.9 6.55 hTx log – d log b – +=

hTx

hRx

L u 139.37 20 f log E – +=

L u 69.55 26.16 f log 13.82 hTx log – 44.9 6.55 hTx log – d log b

+ +=

Lmodel1 L u a hRx – =

Lmodel1 L u a hRx 2 f 28 ------

© ¹§ ·log © ¹§ · 2 – 5.4 – – =

Lmodel1 L u a hRx – 4.78 f log 2

– 18.33 f log 40.94 – +=

Environment a(Hr)

Rural/Small city

Large city

a hRx

1.1 f log 0.7 – hRx 1.56 f log 0.8 – –

3.2 11.75 hRx log 2 4.97 –



© Forsk 2010 AT282_TRG_E2 91


4.4.2.2.3 Distance Correction

The distance correction refers to the term b above.


Hata-based models take into account topo map (DTM) between transmitter and receiver and morpho map (clutter) at the

receiver.

1st step: For each calculation bin, Atoll determines the clutter bin on which the receiver is located. This clutter bin corre-

sponds to a clutter class. Then, it uses the ITU 529-3 formula assigned to this clutter class to evaluate .

2nd step: This step depends on whether the ‘Add diffraction loss’ option is checked.

• If the ‘Add diffraction loss’ option is unchecked, Atoll stops calculations.


a. It extracts a geographic profile between the transmitter and the receiver based on the radial calculation mode.b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates loss-

es due to diffraction .

4.4.3 Standard Propagation Model (SPM)

4.4.3.1 SPM Path Loss FormulaSPM is based on the following formula:

with,

K1: constant offset (dB).

K2: multiplying factor for log(d).

d: distance between the receiver and the transmitter (m).

K3: multiplying factor for log(HTxeff ).

HTxeff: effective height of the transmitter antenna (m).

K4: multiplying factor for diffraction calculation. K4 has to be a positive number.

Diffraction loss: loss due to diffraction over an obstructed path (dB).

K5: multiplying factor for

K6: multiplying factor for .

K7: multiplying factor for .

: effective mobile antenna height (m).

Kclutter : multiplying factor for f(clutter).

f(clutter): average of weighted losses due to clutter.

Distance b

d<20 km 1

d>20 km

where

b 1 0.14 1.87 1 0 4 – f u 1.07 10

3 – h'Tx u+ + d

20 ------log © ¹

§ · 0.8 u+=

h'Tx

hTx

1 7 10 6 – hTx

2 u+

--------------------------------------------=

Note:



Configuration tab)

Lmodel1

Lmodel Lmodel1=

Lmodel2

Lmodel Lmodel1 Lmodel2 +=

Lmodel

Lmodel K 1 K 2 d log K 3 H Txeff log K 4 DiffractionLossu K 5 d log H Txeff log u + + + + +

K 6 H Rxeff K 7 H Rxeff log K clutter f c lutter + +

=

d log H Txeff log u

H Rxeff

H Rxeff log

H Rxeff





4.4.3.2.1 Visibility and Distance Between Transmitter and Receiver

For each calculation bin, Atoll determines:

• The distance between the transmitter and the receiver.

If the distance Tx-Rx is less than the maximum user-defined distance (break distance), the receiver is considered to be

near the transmitter. Atoll will use the set of values marked “Near transmitter”.

If the distance Tx-Rx is greater than the maximum distance, receiver is considered far from transmitter. Atoll will use the

set of values “Far from transmitter”.

• Whether the receiver is in the transmitter line of sight or not.

If the receiver is in the transmitter line of sight, Atoll will take into account the set of values (K1,K2)LOS. The LOS is defined

by no obstruction along the direct ray between the transmitter and the receiver.

If the receiver is not in the transmitter line of sight, Atoll will use the set of values (K1,K2)NLOS.

4.4.3.2.2 Effective Transmitter Antenna Height

Effective transmitter antenna height (H Txeff ) may be calculated with six different methods.

Height Above Ground

The transmitter antenna height is above the ground (H Tx in m).

H Txeff = H Tx

Height Above Average Profile

The transmitter antenna height is determined relative to an average ground height calculated along the profile between a

transmitter and a receiver. The profile length depends on distance min and distance max values and is limited by the trans-

mitter and receiver locations. Distance min and Distance max are minimum and maximum distances from the transmitter

respectively.

where,

is the ground height (ground elevation) above sea level at transmitter (m).

is the average ground height above sea level along the profile (m).

Slope at Receiver Between 0 and Minimum Distance

The transmitter antenna height is calculated using the ground slope at receiver.

where,

is the ground height (ground elevation) above sea level at receiver (m).

is the ground slope calculated over a user-defined distance (Distance min). In this case, Distance min is a distance from

receiver.

Spot H t

If then,

If then,

Note:

• If the profile is not located between the transmitter and the receiver, H Txeff equals H Tx only.

Notes:

• If then, Atoll uses 20m in calculations.

• If then, Atoll takes 200m.

H Txeff H Tx H 0Tx H 0 – +=

H 0Tx

H 0

H Txeff H Tx H 0Tx + H 0Rx – K d u+=

H 0Rx

K

H Txeff 20m

H Txeff 200m!

H 0Tx H 0Rx ! H Txeff H Tx H 0Tx H 0Rx – +=

H 0Tx H 0Rx d H Txeff H Tx =



© Forsk 2010 AT282_TRG_E2 93


Absolute Spot H t

These values are only used in the two last methods and have different meanings according to the method.

Enhanced Slope at Receiver

Atoll offers a new method called “Enhanced slope at receiver” to evaluate the effective transmitter antenna height.

Let x-axis and y-axis respectively represent positions and heights. We assume that x-axis is oriented from the transmitter

(origin) towards the receiver.

This calculation is achieved in several steps:

1st step: Atoll determines line of sight between transmitter and receiver.

The LOS line equation is:

where,

is the receiver antenna height above the ground (m).

i is the point index.

Res is the profile resolution (distance between two points).

2nd step: Atoll extracts the transmitter-receiver terrain profile.

3rd step: Hills and mountains are already taken into account in diffraction calculations. Therefore, in order for them not to

unfavourably influence the regression line calculation, Atoll filters the terrain profile.

Atoll calculates two filtered terrain profiles; one established from the transmitter and another from the receiver. It deter-

mines filtered height of every profile point. Profile points are evenly spaced on the basis of profile resolution. To determine

filtered terrain height at a point, Atoll evaluates ground slope between two points and compares it with a threshold set to

0.05; where three cases are possible.

Some notations defined hereafter are used in next part.

is the filtered height.

is the original height. Original terrain height is determined from extracted ground profile.

- Filter starting from transmitter

Let us assume that

For each point, we have three different cases:

1st case: If and ,

Then,

Note:

• Distance min and distance max are set to 3000 and 15000 m according to ITU

recommendations (low frequency broadcast f < 500 Mhz) and to 0 and 15000 m according

Okumura recommendations (high frequency mobile telephony).

Figure 4.11: Enhanced Slope at Receiver

H Txeff H Tx H 0Tx H 0Rx – +=

Los i H 0Tx H Tx + H 0Tx H Tx + H 0Rx H Rx + –

d ------------------------------------------------------------------------------- Res i – =

H Rx

H filt

H orig

H f i l t Tx – Tx H orig Tx =

H orig i H orig i 1 – !H orig i H orig i 1 – –

Re s------------------------------------------------------ 0.05 d

H f i lt Tx – i H f i lt Tx – i 1 – H orig i H orig i 1 – – +=




2nd case: If and

Then,

3rd case: If

Then,

If additionally

Then,

- Filter starting from receiver

Let us assume that

For each point, we have three different cases:

1st case: If and ,

Then,

2nd case: If and

Then,

3rd case: If

Then,

If additionally

Then,

Then, for every point of profile, Atoll compares the two filtered heights and chooses the higher one.

4th step: Atoll determines the influence area, R. It corresponds to the distance from receiver at which the original terrain

profile plus 30 metres intersects the LOS line for the first time (when beginning from transmitter).

The influence area must satisfy additional conditions:

• ,

• ,• R must contain at least three bins.

5th step: Atoll performs a linear regression on the filtered profile within R in order to determine a regression line.

The regression line equation is:

and

where,

i is the point index. Only points within R are taken into account.

d(i) is the distance between i and the transmitter (m).

Then, Atoll extends the regression line to the transmitter location. Therefore, its equation is:

Notes:

• When several inf luence areas are possible, Atoll chooses the highest one.

• If d < 3000m, R = d .

H orig i H orig i 1 – !H orig i H orig i 1 – –

Re s------------------------------------------------------ 0.05 !

H f i l t Tx – i H f i lt Tx – i 1 – =

H orig i H orig i 1 – d

H f i l t Tx – i H f i lt Tx – i 1 – =

H filt i H orig i !

H f i l t Tx – i H orig i =

H filt Rx H orig Rx =

H orig i H orig i 1+ !H orig i H orig i 1+ –

Re s------------------------------------------------------- 0.05 d

H f i l t Rx – i H f i l t Rx – i 1+ H orig i H orig i 1+ – +=

H orig i H orig i 1+ !H orig i H orig i 1+ –

Re s------------------------------------------------------- 0.05 !

H f i l t Rx – i H f i l t Rx – i 1+ =

H orig

i H orig

i 1+ d

H f i l t Rx – i H f i l t Rx – i 1+ =

H filt i H orig i !

H f i l t Rx – i H orig i =

H filt i max H f i lt Tx – i H f i l t Rx – i =

R 3000mt

R 0.01 d t

y ax b+=

a

d i d m – H filt i H m –

i

¦

d i d m –

2

¦

------------------------------------------------------------------------= b H m ad m – =

H m1

n--- H filt i

i

¦=

d m d R

2 ---- – =

regr i a i Res b+=



© Forsk 2010 AT282_TRG_E2 95


6th step: Then, Atoll calculates effective transmitter antenna height, (m).

If H Txeff is less than 20m, Atoll recalculates it with a new influence area, which begins at transmitter.

7th step: If is still less than 20m (even negative), Atoll evaluates path loss using and applies a

correction factor.

Therefore, if ,

where,

4.4.3.2.3 Effective Receiver Antenna Height

where,

is the receiver antenna height above the ground (m).

is the ground height (ground elevation) above sea level at the receiver (m).

is the ground height (ground elevation) above sea level at the transmitter (m).

4.4.3.2.4 Correction for Hilly Regions in Case of LOS

An optional corrective term enables Atoll to correct path loss for hilly regions when the transmitter and the receiver are in

Line-of-sight.

Therefore, if the receiver is in the transmitter line of sight and the Hilly terrain correction option is active, we have:

When the transmitter and the receiver are not in line of sight, the path loss formula is:

is determined in three steps. Influence area, R, and regression line are supposed available.

1st step: For every profile point within influence area, Atoll calculates height deviation between the original terrain profile

and regression line. Then, it sorts points according to the deviation and draws two lines (parallel to the regression line),

one which is exceeded by 10% of the profile points and the other one by 90%.

2nd step: Atoll evaluates the terrain roughness, 'h; it is the distance between the two lines.

3rd

step: Atoll calculates .

We have

If ,

Else

If ,

Else

i Rx is the point index at receiver.

Notes:

• In case , 1000m wil l be used in calculations.

• If is still less than 20m, an additional correction is taken into account (7th step).

H Txeff

H Txeff

H 0Tx H Tx b – +

1 a2

+

--------------------------------------=

H Txeff 1000m!

H Txeff

H Txeff H Txeff 20m=

H Txeff 20m

Lmodel Lmodel H Txeff 20m= d f K lowant +=

K lowant d

10 5

--------- 0.3 H Txeff 20 – – 20 1 H Txeff 20 – –

9.63d

1000 -------------+© ¹

§ · 6.93d

1000 -------------+© ¹

§ ·------------------------------------------------------------------------------ – =

Note:

• The calculation of effective antenna heights ( and ) is based on extracted

DTM profiles. They are not properly performed if you have not imported heights (DTM file)

beforehand.

H Rxeff H Rx H 0Rx + H 0Tx – =

H Rx

H 0Rx

H 0Tx

H Rxeff H Txeff

Lmodel K 1 LOS K 2 LOS d log K 3 H Txeff log K 5 H Txeff d log log K 6 H Rx K clutter f c lutter K hill LOS+ + + + + +=

Lmodel K 1 NLOS K 2 NLOS d log K 3 H Txeff log K 4 Diffraction K 5 H Txeff d log log K 6 H Rx K clutter f c lutter + + + + + +=

K hill LOS

K hill LOS

K hill LOS K h K hf +=

0 h' 20md K h 0 =

K h 7.73 h' log 2

15.29 h' log – 6.746 +=

0 h' 10md K hf 2 – 0.1924 H 0Rx H Rx regr i Rx – + =

K hf 2 – 1.616 h' log 2 – 14.75 h' log 11.21 – +

H 0Rx H Rx regr i Rx – +

h'------------------------------------------------------------ =




4.4.3.2.5 Diffraction

Four methods are available to calculate diffraction loss over the transmitter-receiver profile. They are detailed in the

Appendices.

Along the transmitter-receiver profile, you may consider:

• Either ground altitude and clutter height (Consider heights in diffraction option),

In this case, Atoll uses clutter height information from clutter heights file if available in the .atl document. Other-

wise, it considers average clutter height specified for each clutter class in the clutter classes file description.

• Or only ground altitude.

4.4.3.2.6 Losses due to Clutter

Atoll calculates f(clutter) over a maximum distance from receiver:

where,

L: loss due to clutter defined in the Clutter tab by the user (in dB).

w : weight determined through the weighting function.

n: number of points taken into account over the profile. Points are evenly spaced depending on the profile resolution.

Four weighting functions are available:

• Uniform weighting function:

• Triangular weighting function:

• , where d’ i is the distance between the receiver and the ith point and D is the maximum distance

defined.

• Logarithmic weighting function:

• Exponential weighting function:

The chart below shows the weight variation with the distance for each weighting function.

Figure 4.12: Losses due to Clutter

f c lutter Li w i

i 1=

¦=

w i 1

n---=

w i d i

d j

n

¦

--------------=

d i D d 'i – =

w i

d i D---- 1+© ¹

§ ·log

d j D---- 1+© ¹

§ ·log

n

¦

--------------------------------------=

w i e

d i D----

1 –

e

d j D----

1 –

j 1=

n

¦

--------------------------=



© Forsk 2010 AT282_TRG_E2 97


4.4.3.2.7 Recommendations

Beware that the clutter influence may be taken into account in two terms, Diffraction loss and f(clutter) at the same time.

To avoid this, we advise:

1. Not to consider clutter heights to evaluate diffraction loss over the transmitter-receiver profile if you specify losses

per clutter class.

This approach is recommended if the clutter height information is statistical (clutter roughly defined, no alti-

tude).

Or

2. Not to define any loss per clutter class if you take clutter heights into account in the diffraction loss.

In this case, f(clutter)=0 . Losses due to clutter are only taken into account in the computed Diffraction loss term.

This approach is recommended if the clutter height information is either semi-deterministic (clutter roughly

defined, altitude defined with an average height per clutter class) or deterministic (clutter sharply defined, altitude

defined with an average height per clutter class or - even better - via a clutter height file).

In case of semi-deterministic clutter information, specify receiver clearance (m) per clutter class. Both ground altitude and

clutter height are considered along the whole transmitter-receiver profile except over a specific distance around the

receiver (clearance), where Atoll proceeds as if there was only the DTM map. The clearance information is used to model

streets.

In the above figure, the ground altitude and clutter height (in this case, average height specified for each clutter class in

the clutter classes map description) are taken into account along the profile.

Clearance definition is not necessary in case of deterministic clutter height information. Clutter height information is accu-

rate enough to be used directly without additional information such as clearance. Two cases can be considered:

1. If the receiver is in the street (clutter height lower than receiver height), Atoll calculates the path loss by consid-

ering potentially some diffraction loss at reception.

2. If the receiver is supposed to be inside a building (clutter height higher than receiver height), Atoll does not con-

sider any difraction (and clearance) from the building but takes into account the indoor loss as an additional pen-

etration loss.

4.4.3.3 Automatic SPM Calibration

The goal of this tool is to calibrate parameters and methods of the SPM formula in a simple and reproducible way. Cali-

bration is based on imported CW measurement data. It is the process of limiting the difference between predicted and

measured values. For a complete description of the calibration procedure (including the very important prerequisite filtering

work on the CW measurement points), please refer to the User Manual and the SPM Calibration Guide.

The following SPM formula parameters can be estimated:

• K 1, K 2 , K 3, K 4, K 5 , K 6 and K 7

Figure 4.13: Tx-Rx profile

Notes:

• To consider indoor losses in building only when using a deterministic clutter map (clutter

height map), the 'Indoor Coverage' box must not be checked in predictions unless this loss

will be counted twice inside buildings (on the entire reception clutter class and not only

inside the building).



Configuration tab)

• Even with no clearance, the clutter height (extracted either from clutter class or clutter

height folders) is never considered at the last profile point.

Lmodel




• Losses per clutter class (K clutter must be user-defined)

• Effective antenna height method

• Diffraction method

Automatic model calibration provides a mathematical solution. The relevance of this mathematical solution with a physical

and realistic solution must be determined before committing these results.

You must keep in mind that the model calibration and its result (standard deviation and root mean square) strongly depend

on the CW measurement samples you use. A calibrated model must restore the behaviour of CW measurements depend-

ing on their configuration on a large scale, and not just totally coincide with a few number of CW measurements. The cali-

brated model has to give correct results for every new CW measurement point in the same geographical zone, without

having been calibrated on these new CW measurements.

4.4.3.3.1 General AlgorithmPropagation model calibration is a special case of the more general Least-Square problems, i.e. given a real m x n matrix

A, and a real m-vector b, find a real n-vector x0 that minimises the Euclidean length of Ax - b.

Here,

m is the number of measurement points,

n is the number of parameters to calibrate,

A is the values of parameter associated variables (log(d), log(heff), etc.) at each measurement point, and

b is the vector of measurement values.

The vector x0 is the set of parameters found at the end of the calibration.

The theoretical mathematical solution of this problem was found by Gauss (around 1830). Further enhancements to the

original method were proposed in the 60's in order to solve the numerical instability problem.

In 1974, Lawson & Hanson [2] proposed a theoretical solution of the least-square problem with general linear inequality

constraints on the vector x0. Atoll implementation is based on this method, which is explained in detail in [1].

4.4.3.3.2 Sample Values for SPM Path Loss Formula Parameters

The following tables list some sample orders of magnitudes for the different parameters composing the Standard Propa-

gation Model formula.

K1 depends on the frequency and the technology. Here are some sample values:

The above K1 values for WiMAX are extrapolated estimates for different frequency ranges. It is highly recommended to

calibrate the SPM using measurement data collected on the field for WiMAX networks before using the SPM for predic-

References:[1] Björck A. “Numerical Methods for Least Square Problems”, SIAM, 1996.

[2] Lawson C.L., Hanson R.J. “Solving Least Squares Problems”, SIAM, 1974.

Minimum Typical Maximum

K 1 Variable Variable Variable

K 2 20 44.9 70

K 3 -20 5.83 20

K 4 0 0.5 0.8

K 5 -10 -6.55 0

K 6 -1 0 0

K 7 -10 0 0

Project type Frequency (MHz) K 1

GSM 900 935 12.5

GSM 1800 1805 22

GSM 1900 1930 23UMTS 2110 23.8

1xRTT 1900 23

WiMAX

2300 24.7

2500 25.4

2700 26.1

3300 27.8

3500 28.3



© Forsk 2010 AT282_TRG_E2 99


tions.

All K paramaters can be defined by the automatic calibration wizard. Since K clutter is a constant, its value is strongly

dependant on the values given to the losses per clutter classes. From experienced users, the typical losses (in dB) per

clutter class are:

These values have to be entered only when considering statistical clutter class maps only.

4.4.3.4 Unmasked Path Loss Calculation

You can use the SPM to calculate unmasked path losses. Unmasked path losses are calculated by not taking into account

the transmitter antenna patterns, i.e., the attenuation due to the transmitter antenna pattern is not included. Such path

losses are useful when using path loss matrices calculated by Atoll with automatic optimisation tools.

The instance of the SPM available by default, under the Propagation Models folder in the Modules tab, has the following

characteristics:

• Signature: {D5701837-B081-11D4-931D-00C04FA05664}

• Type: Atoll.StdPropagModel.1

You can access these parameters in the Propagation Models table by double-clicking the Propagation Models folder in

the Modules tab.

To make the SPM calculate path losses excluding the antenna pattern attenuation, you have to change the type of the

SPM to:

• Type: Atoll.StdPropagModelUnmasked.1

However, changing the type only does not invalidate the already calculated path loss matrices, because the signature of

the propagation model is still the same. If you want Atoll to recognize that the SPM has changed, and to invalidate thepath loss matrices calculated with this model, you have to change the signature of the model as well. The default signature

for the SPM that calculates unmasked path loss matrices is:

• Signature: {EEE060E5-255C-4C1F-B36C-A80D3D972583}

The above signature is a default signature. Atoll automatically creates different signatures for different instances of the

same propagation model. Therefore, it is possible to create different instances of the SPM, with different parameter

settings, and create unmasked versions of these instances.

You can change the signature and type of the original instance of the SPM, but it is recommended to make a copy of the

SPM in order not to lose the original SPM parameters. So, you will be able to keep different versions of the SPM, those

that calculate path losses with antenna pattern attenuation, and others that calculate path losses without it.

The usual process flow of an ACP working on an Atoll document through the API would be to:

1. Backup the storage directory of path loss matrices.

2. Set a different storage directory for calculating and storing unmasked path loss matrices.

3. Select the SPM used, backup it’s signature, and change its signature and type as shown above.

4. Perform optimisation using the path loss matrices calculated by the unmasked version of the SPM.

5. Restore the type and the signature of the SPM.

6. Reset the path loss storage directory to the original one.

Dense urban From 4 to 5

Woodland From 2 to 3

Urban 0

Suburban From -5 to -3

Industrial From -5 to -3

Open in urban From -6 to -4

Open From -12 to -10

Water From -14 to -12

Note:

• The Standard Propagation Model is deduced from the Hata formulae, valid in the case of

an urban environment. The above values are consistent since they are normalized with

respect to the urban clutter class (0 dB for urban clutter class). Positive values correspond

to denser clutter classes and negative values to less dense clutter classes.

Notes:

• It is not possible to calibrate the unmasked version of the SPM using measurement data.

• You can also use Atoll.ini options, AngleCalculation = 2000 and AngleCalculation = 3000,

for calculating unmasked path losses and angles of incidence, respectively. These options

are only available for the propagation models available with Atoll by default. Please refer to

the Administrator Manual for details.




4.4.4 WLL Propagation Model

4.4.4.1 WLL Path Loss Formula

Where is the free space loss calculated using the formula entered in the model properties, is the diffraction loss

calculated using the 3-obstacle Deygout method, and is the diffraction multiplying factor defined in the model prop-

erties.


4.4.4.2.1 Free Space Loss

Please refer to the Appendices for further details about free space loss calculation.


Atoll calculates diffraction loss along the transmitter-receiver profile built from DTM and clutter maps. Therefore, losses

due to clutter are taken into account in diffraction losses. Atoll takes clutter height information from the clutter heights file

if available in the .atl document. Otherwise, it considers average clutter height specified for each clutter class in the clutter

classes file description.

The Deygout construction (considering 3 obstacles) is used. This method is detailed in the Appendices. The final diffrac-

tion losses are determined by multiplying the diffraction losses calculated using the Deygout method by the Diffraction

multiplying factor defined in the model properties.

Receiver Clearance

Define receiver clearance (m) per clutter class when clutter height information is either statistical or semi-deter-ministic. Both ground altitude and clutter height are considered along the whole profile except over a specific distance

around the receiver (clearance), where Atoll proceeds as if there was only the DTM map (see SPM part). Atoll uses the

clearance information to model streets.

If the clutter is deterministic, do not define any receiver clearance (m) per clutter class. In this case, clutter height

information is accurate enough to be used directly without additional information such as clearance ( Atoll can locate

streets).

Receiver Height

Entering receiver height per clutter class enables Atoll to consider the fact that receivers are fixed and located on the roofs.

Visibility

If the option ‘Line of sight only’ is not selected, Atoll computes Lmodel on each calculation bin using the formula defined

above. When selecting the option ‘Line of sight only’, Atoll checks for each calculation bin if the Diffraction loss (as defined

in the Diffraction loss: Deygout part) calculated along profile equals 0.

• In this case, receiver is considered in ‘line of sight’ and Atoll computes Lmodel on each calculation bin using the

formula defined above.

• Otherwise, Atoll considers that Lmodel tends to infinity.

4.4.5 ITU-R P.526-5 Propagation Model


• Using the SPM, you can also calculate the angles of incidence by creating a new instance

of the SPM with the following characteristics:

Type: Atoll.StdPropagModelIncidence.1

Signature: {659F0B9E-2810-4e59-9F0D-DA9E78E1E64B}

Important:

• The "masked" version of the algorithm has not been changed. It still takes into account

Atoll.ini options. However, the "unmasked" version does not take Atoll.ini options into

account.

• It’s highly recommended to use one method (Atoll.ini options) or the other one (new

identifier & signature) but not to combine both.

Lmodel LFS F Diff LDiff u+=

LFS LDiff

F Diff

Lmodel LFS LDiff +=



© Forsk 2010 AT282_TRG_E2 101


Where is the free space loss calculated using the formula entered in the model properties and is the diffraction

loss calculated using the 3-obstacle Deygout method.





Atollcalculates diffraction loss along the transmitter-receiver profile is built from the DTM map. The Deygout construction(considering 3 obstacles), with or without correction, is used. These methods are detailed in the Appendices.



If d<1 km,

If d>1000 km,

If 1<d<1000 km,

d is the distance between the transmitter and the receiver (km).




4.4.6.2.2 Corrected Standard Loss

This formula is given for a 60 dBm (1kW) transmitter power.

where,

Cn is the field strength received in dBPV/m,

is a correction factor for effective receiver antenna height (dB),

Acl is the correction for terrain clearance angle (dB),

f is the frequency in MHz.

Cn Calculation

The C n value is determined from charts C n=f(d, H Txeff ).

In the following part, let us assume that C n=E n(d,H Txeff ) (where E n(d,H Txeff ) is the field received in dBPV/m) is read from

charts for a distance, d (in km), and an effective transmitter antenna height, H Txeff (in m).

First of all, Atoll evaluates the effective transmitter antenna height, , as follows:

If ,

If ,

If ,

where,

is the transmitter antenna height above the ground (m).

is the ground height (ground elevation) above sea level at the transmitter (m).

is the average ground height (m) above sea level for the profile between a point 3 km from transmitter and the

receiver (located at d km from transmitter).

is the average ground height (m) above sea level for the profile between a point 3 km and another 15 km from

transmitter.

Then, depending on d and H Txeff , Atoll determines C n using bilinear interpolation as follows.

LFS LDiff

Lmodel LFS=

Lmodel 1000 =

Lmodel max LFS CorrectedS dardLosstan =

CorrectedS dardLosstan 60 C n – AH Rxeff – Acl – 108.75 – 31.54 20 f log – +=

AH Rxeff

H Txeff

0 d 3kmd H Txeff H 0Tx H Tx H 0Rx – +=

3 d 15 kmd H Txeff H 0Tx H Tx H 0 3 d ; – +=

15 d H Txeff H 0Tx H Tx H 0 3 15 ; – +=

H Tx

H 0Tx

H 0 3 d ;

H 0 3 15 ;




If 37.5 H Txeff 1200 , C n= E n(d,H Txeff )

Otherwise, Atoll considers (d is stated in km)

Therefore,

If H Txeff < 37.5

If , we have

Else C n=E n(d, 37.5) – E n(d horizon, 37.5) + E n(25, 37.5)

If H Txeff > 1200

If , we have

Else C n=E n(d, 1200) – E n(d horizon, 1200) + E n(142, 1200)

AHRxeff Calculation

where,

HRx is the user-defined receiver height,

c is the height gain factor.

Acl Calculation

If f 300 MHz,

Otherwise,

With

where,

is the clearance angle (in radians) determined according to the recommendation 370-7 (figure 19),

f is the frequency stated in MHz.

4.4.7 Erceg-Greenstein (SUI) Propagation Model

Erceg-Greenstein propagation model is a statistical path loss model derived from experimental data collected at 1.9 GHz

in 95 macrocells. The model is for suburban areas, and it distinguishes between different terrain categories called the Stan-

ford University Interim Terrain Models. This propagation model is well suited for distances and base station antenna

heights that are not well-covered by other models. The path loss model applies to base antenna heights from 10 to 80 m,

base-to-terminal distances from 0.1 to 8 km, and three distinct terrain categories.

The basic path loss equation of the Erceg-Greenstein propagation model is:

Where . This is a fixed quantity which depends upon the frequency of operation. d is the distance

between the base station antenna and the receiver terminal and d 0 is a fixed reference distance (100 m). a(H BS ) is thecorrection factor for base station antenna heights, H BS:

Where , and a, b, and c are correction coefficients which depend on the SUI terrain type.

The Erceg-Greenstein propagation model is further developed through the correction factors introduced by the Stanford

University Interim model. The standards proposed by the IEEE working group 802.16 include channel models developed

by Stanford University. The basic path loss equation with correction factors is presented below:

Note:

• c values are provided in the recommendation 370-7; for example, c=4 in a rural case.

d horizon 4.1 H Txeff =

d d horizont C n E n d 25 d horizon – + 37.5 =

d d horizont C n E n d 142 d horizon – + 1200 =

AH Rxeff

c

6 --- 20

H Rx

10 ----------

© ¹§ ·log =

d Acl 8.1 6.9 20 Q 0.1 – 2 1+ Q 0.1 – + log +> @ – =

Acl 14.9 6.9 20 Q 0.1 – 2

1+ Q 0.1 – + log +> @ – =

Q T – 4000 f

300 ----------=

T

PL A 10 a H BS Lo g 10 d

d 0

------© ¹§ ·+=

A 20 Lo g 10

4Sd 0

O-------------

© ¹§ ·=

a H BS a b H BSc

H BS

----------+ – =

10 m H BS 80 md d

PL A 10 a H BS Lo g 10 d

d 0

------© ¹§ · a f a H R – ++=



© Forsk 2010 AT282_TRG_E2 103


Where a(f) is the correction factor for the operating frequency, , with f being the operating

frequency in MHz. a(H R ) is the correction factor for the receiver antenna height, , where d

depends on the terrain type.

4.4.7.1 SUI Terrain Types

The SUI models are divided into three types of terrains1, namely A, B and C.

• Type A is associated with maximum path loss and is appropriate for hilly terrain with moderate to heavy tree

densities.

• Type B is characterised with either mostly flat terrains with moderate to heavy tree densities or hilly terrains

with light tree densities.

• Type C is associated with minimum path loss and applies to flat terrain with light tree densities.

The constants used for a, b, and c are given in the table below.

4.4.7.2 Erceg-Greenstein (SUI) Path Loss Formula

The Erceg-Greenstein (SUI) propagation model formula can be simplified from the following equation:

(1)

to the equation below:

(2)

Where,

• f is the operating frequency in MHz

• d is the distance from the transmitter to the received in m in equation (1) and in km in equation (2)

• H BS is the transmitter height in m

• H R is the receiver height in m

The above equation is divided into two parts in Atoll:

Where,

The above path loss formulas are valid for d > d 0 , i.e. d > 100 m. For d < 100 m, the path loss has been restricted to the

free space path loss with correction factors for operating frequency and receiver height:

instead of

Where a(f) and a(Hr) have the same definition as given above. Simplifying the above equation, we get,

, or

Note:

• a(H R ) = 0 for H R = 2 m.

References:[1] V. Erceg et. al, “An empirically based path loss model for wireless channels in suburban environments,” IEEE J.

Select Areas Commun., vol. 17, no. 7, July 1999, pp. 1205-1211.

[2] Abhayawardhana, V.S.; Wassell, I.J.; Crosby, D.; Sellars, M.P.; Brown, M.G.; "Comparison of empirical propagation

path loss models for fixed wireless access systems," Vehicular Technology Conference, 2005. IEEE 61st Volume 1, 30

May-1 June 2005 Page(s):73 - 77 Vol. 1

a f 6 Log 10 f

2000 -------------

© ¹§ ·=

a H R X Log 10

H R

2 -------

© ¹§ ·=

1. The word ‘terrain’ is used in the original definition of the model rather than ‘environment’. Hence it is used

interchangeably with ‘environment’ in this description.

Model Parameter Terrain A Terrain B Terrain C

a 4.6 4.0 3.6

b (m-1) 0.0075 0.0065 0.005

c (m) 12.6 17.1 20

X 10.8 10.8 20

P L 20 L og 10

4Sd 0

O-------------

© ¹§ · 10 a H BS Lo g 10

d

d 0

------© ¹§ · a f a H R – ++=

P L 7.366 – 26 Log 10 f 10 a H BS 1 Log 10 d + a H R – ++=

PL Lu a H R – =

L u 7.366 – 26 Log 10

f 10 a H BS

1 L og 10

d + ++=

P L 20 L og 10 4 S d

O------------------

© ¹§ · a f a H R – += PL 20 L og 10

4 S d O

------------------© ¹§ ·=

PL 12.634 26 Log 10 f 20 Log 10 d a H R – ++= Lu 12.634 26 Log 10 f 20 Log 10 d ++=




The above equation is not user-modifiable in Atoll except for the coefficient of , i.e. 26. Atoll uses the same coef-

ficient as the one you enter for in Atoll for the case d > d 0 .


The Erceg-Greenstein (SUI) propagation model takes DTM into account between the transmitter and the receiver, and it

can also take clutter into account at the receiver location.

1st step: For each pixel in the calculation radius, Atoll determines the clutter bin on which the receiver is located. This

clutter bin corresponds to a clutter class. Atoll uses the Erceg-Greenstein (SUI) path loss formula assigned to this clutter

class to evaluate path loss.

2nd step: This step depends on whether the ‘Add diffraction loss’ option is selected or not.

• If the ‘Add diffraction loss’ option is not selected, 1st step gives the final path loss result.


a. It extracts a geographic profile between the transmitter and the receiver using the radial calculation method.

b. It determines the largest obstacle along the profile in accordance with the Deygout method and evaluates loss-

es due to diffraction . For more information on the Deygout method, see "3 Knife-Edge Deygout

Method" on page 111.

The final path loss is the sum of the path loss determined in 1st step and .

Shadow fading is computed in Atoll independent of the propagation model. For more information on the shadow fading

calculation, see "Shadowing Model" on page 119.


This propagation model is based on the P.1546-2 recommendations of the ITU-R. These recommendations extend the

P.370-7 recommendations, and are suited for operating frequencies from 30 to 3000 MHz. The path loss is calculated by

this propagation model with the help of graphs available in the recommendations. The graphs provided in the recommen-

dations represent field (or signal) strength, given in , as a function of distance for:

• Nominal frequencies, : 100, 600, and 1000 MHz

The graphs provided for 100 MHz are applicable to frequencies from 30 to 300 MHz, those for 600 MHz are appli-

cable to frequencies from 300 to 1000 MHz, and the graphs for 1000 MHz are applicable to frequencies from 1000

to 3000 MHz. The method for interpolation is described in the recommendations (Annex 5, § 6).

• Transmitter antenna heights, : 10, 20, 37.5, 75, 150, 300, 600, and 1200 m

For any values of from 10 to 3000 m, an interpolation or extrapolation from the appropriate two curves is used,

as described in the recommendations (Annex 5, § 4.1). For below 10 m, the extrapolation to be applied is given

in Annex 5, § 4.2. It is possible for the value of to be negative, in which case the method is given in Annex 5,

§ 4.3.

• Time variability, : 1, 10, and 50 %

The propagation curves represent the field strength values exceeded for 1, 10 and 50 % of time.

• Receiver antenna height, : 10 m

For land paths, the graphs represent field strength values for a receiver antenna height above ground, equal to

the representative height of the clutter around the receiver. The minimum value of the representative height of clut-

ter is 10 m. For sea paths, the graphs represent field strength values for a receiver antenna height of 10 m.

For other values of receiver antenna height, a correction is applied according to the environment of the receiver.The method for calculating this correction is given in Annex 5, § 9.

These recommendations are not valid for transmitter-receiver distances less than 1 km or greater than 1000 km. Therefore

in Atoll, the path loss between a transmitter and a receiver over less than 1 km is the same as the path loss over 1 km.

Similarly, the path loss between a transmitter and a receiver over more than 1000 km is the same as the path loss over

1000 km.

Moreover, these recommendations are not valid for transmitter antenna heights less than the average clutter height

surrounding the transmitter.

Note:

• You can get the same resulting equation by setting a(hBS) = 2.

Lo g 10 f

Lo g 10 f

LDiffraction

LDiffraction

Notes:

• The cold sea graphs are used for calculations over warm and cold sea both.

• The mixture of land and sea paths is not supported by Atoll.

db PV m e

f n

h1

h1

h1

h1

t

h2



© Forsk 2010 AT282_TRG_E2 105



The input to the propagation model are the transmission frequency, transmitter and receiver heights, the distance between

the transmitter and the receiver, the precentage of time the field strength values are exceeded, the type of environment

(i.e., land or sea), and the clutter at the receiver location.

In the following calculations, is the transmission frequency, is the transmitter-receiver distance, and is the percent-

age of time for which the path loss has to be calculated.

The following calculations are performed in Atoll to calculate the path loss using this propagation model.

4.4.8.1.1 Step 1: Determination of Graphs to be Used

First of all, the upper and lower nominal frequencies are determined for any given transmission frequency. The upper andlower nominal frequencies are the nominal frequencies (100, 600, and 2000 MHz) between which the transmission

frequency is located, i.e., .

Once and are known, along with the information about the percentage of time and the type of path (land or sea),

the sets of graphs which will be used for the calculation are also known.

4.4.8.1.2 Step 2: Calculation of Maximum Field Strength

A field strength must not exceed a maximum value, , which is given by:

for land paths, and

for sea paths.

Where is the free space field strength for 1 kW ERP, is an enhancement for sea graphs.

4.4.8.1.3 Step 3: Determination of Transmitter Antenna Height

The transmitter antenna height to be used in the calculation depends on the type and length of the path.

• Land paths

• Sea paths

Here, all antenna heights (i.e., , , and ) are in expressed in m. is the antenna height above ground and

is the effective height of the transmitter antenna, which is its height over the average level of the ground between distances

of and d km from the transmitter in the direction of the receiver.

4.4.8.1.4 Step 4: Interpolation/Extrapolation of Field StrengthThe interpolations are performed in series in the same order as described below. The first interpolation/extrapolation is

performed over the field strength values, , from the graphs for transmitter antenna height to determine . The second

interpolation/extrapolation is performed over the interpolated/extrapolated values of to determine . And, the thrid

and final interpolation/extrapolation is performed over the interpolated/extrapolated values of to determine .

Step 4.1: Interpolation/Extrapolation of Field Strength for Transmitter Antenna Height

If the value of coincides with one of the eight heights for which the field strength graphs are provided, namely 10, 20,

37.5, 75, 150, 300, 600, and 1200 m, the required field strength is obtained directly from the corresponding graph. Other-

wise:

• If

The field strength is interpolated or extrapolated from field strengths obtained from two curves using the following

equation:

Where if , otherwise is the nearest nominal effective height below ,

if , otherwise is the nearest nominal effective height above , is the field

strength value for at the required distance, and is the field strength value for at the required

distance.

• If

f d t

f n1 f f n2

f n1 f n1 t

E Max

E Ma x E FS 106.9 20 Log d u – = =

E Ma x E FS E SE + 106.9 20 Log d u – 2.38 1 d 8.94 e – exp – ^ ` Log 50 t e u+= =

E FS E SE

h1 heff =

h1 Max 1 ha =

h1 heff ha ha heff

0.2 d u

E E h1

E h1 E d

E d E f

h1

10 m h1 3000 m

E h1 E Low E Up E Low – Log h1 hLow e

Log hUp hLow e ------------------------------------------u+=

hLow 600 m= h1 1200 m! hLow h1

hUp 1200 m= h1 1200 m! hUp h1 E Low

hLow E Up hUp

0 m h1 10 m




- For land path if the transmitter-receiver distance is less than the smooth-Earth horizon distance

, i.e., if ,

, or

because

- For land path if the transmitter-receiver distance is greater than or equal to the smooth-Earth horizon distance

, i.e., if ,

, or because

Where is the field strength value read for the transmitter-receiver distance of y from the graph available

for the transmitter antenna height of x .

If in the above equation, even though , the field strength is de-

termined from linear extrapolation for Log (distance) of the graph given by:

Where is penultimate tabulation distance (km), is the final tabulation distance (km), is the

field strength value for , and is the field strength value for .

- For sea path, should not be less than 1 m. This calculation requires the distance at which the path has 0.6

of the first Fresnel zone just unobstructed by the sea surface. This distance is given by:

(km)

Where (km) with (frequency-dependent term),

and (asymptotic term defined by the horizon distance).

If the 0.6 Fresnel clearance distance for the sea path where the transmitter antenna height is 20 m is

also calculated as:

(km)

Once and are known, the field strength for the required distance is given by:

Where is the maximum field strength at the required distance as calculated in "Step 2: Calculation of

Maximum Field Strength" on page 105, is for ,

, ,

and is the field strength calculated as described for land paths. and are field strengths

interpolated for distance y and , respectively, and .

• If

A correction is applied to the field strength, , calculated in the above description in order to take into accountthe diffraction and tropospheric scattering. This correction is the maximum of the diffraction correction,, and trop-

ospheric scattering correction, .

Where with and ,

, and is 1.35 for 100 MHz, 3.31 for 600 MHz, 6.00 for 2000 MHz.

with , (radius of the Earth), and is the effec-

tive Earth radius factor for mean refractivity conditions.

d H h1 4.1 h1u= d 4.1 h1u

E h1 E 10 d H 10 E 10 d E 10 d H h1 – +=

E h1 E 10 12.9 km E 10 d E 10 d H h1 – += d H 10 12.9 km=

d H h1 4.1 h1u= d 4.1 h1ut

E h1 E 10 d H 10 d d H h1 – + = E h1 E 10 12.9 km d d H h1 – + = d H 10 12.9 km=

E x y

d H 10 d d H h1 – + 1000 km! d 1000 kmd

E h1 E Low E Up E Low – Log d DLow e

Log DUp DLow e --------------------------------------------u+=

DLow DUp E Low

DLow E Up DUp

h1

Dh1 D0.6 f h1 h2 10 m= =

D0.6 Max 0.001Df Dhu

Df Dh+-------------------© ¹

§ ·= Df 0.0000389 f h1 h2 uuu=

Dh 4.1 h1 h2 + u=

d Dh1!

D20 D0.6 f h1 20 m= h2 10 m= =

Dh1 D20

E h1

E Max for d Dh1d

E Dh1E D20

E Dh1 –

Log d Dh1 e Log D20 Dh1 e ---------------------------------------u+ for Dh1 d D20

E ' 1 F S – u E '' F Su+ for d D20 t¯°°®°°

-

=

E Ma x

E Dh1E Max d Dh1=

E D20 E 10 D20 E 20 D20 E 10 D20 –

Log h1 10 e Log 20 10 e ----------------------------------u+= E ' E 10 d E 20 d E 10 d –

Log h1 10 e Log 20 10 e ----------------------------------u+=

E '' E 10 y E 20 y

h1 10 m and 20 m= F S d D20 – d e =

h1 0 m

E h1

C h1 Max C h1d C h1t =

C h1d 6.03 J Q – = J Q 6.9 20 Log Q 0.1 – 2 1+ Q 0.1 – + u+> @= Q K Q Teff2 u=

Teff2 ar c h1 –

9000 -------------

© ¹§ ·tan= K Q

C h1t 30 L og Te

Te Teff2 +------------------------

© ¹§ ·u= Te

180 d ua S k uu----------------------= a 6370 km= k 4 3 e =



© Forsk 2010 AT282_TRG_E2 107


Step 4.2: Interpolation/Extrapolation of Field Strength for Transmitter-Receiver Distance

In the field strength graphs in the recommendations, the field strength is plotted against distance from 1 km to 1000 km.

The distance values for which field strengths are tabulated are 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,

19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200,

225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, 500, 525, 550, 575, 600, 625, 650, 675, 700, 725, 750, 775, 800,

825, 850, 875, 900, 925, 950, 975, 1000. If the transmitter-receiver distance is a value from this list, then interpolation of

field strength is not required and the field strength can be directly read from the graphs.

If the transmitter-receiver distance does not coincide with the list of distances for which the field strengths are accurately

available from the graphs, the field strength are linearly interpolated or extrapolated for the logarithm of the distance using

the following equation:

Where is the lower value of the nearest tabulated distance to , is the higher value of the nearest tabulated

distance to , is the field strength value for , and is the field strength value for .

Step 4.3: Interpolation/Extrapolation of Field Strength for Transmission Frequency

The field strength at the transmission frequency is interpolated from the graphs available for the upper and lower nominal

frequencies as follows:

Where is the lower nominal frequency (100 MHz if f < 600 MHz, 600 MHz otherwise), is the higher nominal

frequency (600 MHz if f < 600 MHz, 2000 MHz otherwise), is the field strength value for , and is the field

strength value for .

In the case of transmission frequencies below 100 MHz or above 2000 MHz, the field strength values are extrapolated

from the two nearer nominal frequency values. The above equation is used for all land paths and sea paths.

4.4.8.1.5 Step 5: Calculation of Correction Factors

Step 5.1: Correction for Receiver Antenna Height

The receiver antenna height correction depends on the type of path and clutter in which the receiver is located. The field

strength values given by the graphs for land paths are for a reference receiver antenna at a height, (m), representative

of the height of the clutter surrounding the receiver, subject to a minimum height value of 10 m. Examples of reference

heights are 20 m for an urban area, 30 m for a dense urban area, and 10 m for a suburban area. For sea paths the notional

value of is 10 m.

For land paths, the elevation angle of the arriving ray is taken into account by calculating a modified representative clutter

height , given by .

Note that for , .

The different correction factors are calculated as follows:

• For land path in urban and suburban zones

With and .

If , is reduced by .

• For land path other zones

• For sea path

and are determined as distances at which at which the path has 0.6 of the first Fresnel zone just unob-

structed by the sea surface with and variable , respectively. These distances are given by

E d E Low E Up E Low – Log d d Low e

Log d Up d Low e ------------------------------------------u+=

d Low d d Up

d E Low d Low E Up d Up

E f E Low E Up E Low – Log f f Low e

Log f Up f Low e ---------------------------------------u+=

f Low f Up

E Low f Low E Up

f Up

R

R

R ' R ' Max 11000 d R uu 15 h1u –

1000 d u 15 – ---------------------------------------------------------------© ¹

§ ·=

h1 6.5 d u R + R ' R |

C Receiver

6.03 J Q – for h2 R '

3.2 6.2 Log f u+ Lo g h2

R '------

© ¹§ ·u for h2 R 't

¯°®°-

=

J Q 6.9 20 Log Q 0.1 – 2 1+ Q 0.1 – + u+> @= Q 0.0108 f u R ' h2 – ar c

R ' h2 –

27

-----------------© ¹§ ·tanuu=

R ' 10 m C Receiver 3.2 6.2 Log f u+ Lo g 10

R '------© ¹

§ ·u

C Receiver 3.2 6.2 Log f u+ Lo g h2

10 ------

© ¹§ ·u=

d 10 d h2

h2 10 m= h2




and (km), respectively. Here

as explained earlier.

- If ,

- If and ,

- If and and ,

- If and and ,

Step 5.2: Correction for Short Urban/Suburban Paths

This correction is only applied when the path loss is to be calculated over land paths, over a transmitter-receiver distance

less than 15 km, in urban and suburban zones. This correction takes into account the presence of buildings in these zones.

The buildings are assumed to be of uniform height.

The correction represents a reduction in the field strength due to building clutter. It is added to the field strength and is

given by:

Where is the antenna height above the ground, and R is the clutter height of the clutter class where the receiver is

located. This correction is only applied when and .

Step 5.3: Correction for Receiver Clearance Angle

This correction is only applied when the path loss is to be calculated over land paths, and over a transmitter-receiver

distance less than 16 km. This correction gives more precise field strength prediction over small reception areas. The

correction is added to the field strength and is given by:

Where , , and

is the clearance angle in degrees determined from:

• : The elevation angle of the line from the receiver which just clears all terrain obstructions in the direction of the

transmitter over a distance of up to 16 km but not going beyond the transmitter.

• : The reference angle, .

Where and are the heights of the transmitter and the receiver above sea level, respectively.

4.4.8.1.6 Step 6: Calculation of Path Loss

First, the final field strength is calculated from the interpolated/extrapolated field strength, , by applying the corrections

calculated earlier. The calculated field strength is given by:

The resulting field strength is given by , from which the path loss (basic transmission loss, ) is

calculated as follows:

4.4.9 Sakagami Extended Propagation ModelThe Sakagami extended propagation model is based on the simplification of the extended Sakagami-Kuboi propagation

model. The Sakagami extended propagation model is valid for frequencies above 3 GHz. Therefore, it is only available in

WiMAX 802.16d and WiMAX 802.16e documents by default.

The Sakagami-Kuboi propagation model requires detailed information about the environment, such as widths of the streets

where the receiver is located, the angles formed by the street axes and the directions of the incident waves, heights of the

buildings close to the receiver, etc. The path loss formula for the Sakagami-Kuboi propagation model is [1]:

d 10 D0.6 f h1 h2 10 m= = d h2 D0.6 f h1 h2 = D0.6 Max 0.001Df Dhu

Df Dh+-------------------© ¹

§ ·=

h2 10 m! C Receiver 3.2 6.2 Log f u+ Lo g h2

10 ------

© ¹§ ·u=

h2 10 m d d 10 ! C Receiver 3.2 6.2 Log f u+ Lo g h2

10 ------

© ¹§ ·u=

h2 10 m d d 10 d d h2 C Receiver 0 =

h2 10 m d d 10 d d h2 ! C Receiver 3.2 6.2 Log f u+ Lo g

h2

10 ------© ¹§ ·u

Log d d h2 e

Log d 10 d h2 e -------------------------------------© ¹§ ·u=

C Building 3.3 – Log f 1 0.85 Log d u – 1 0.46 L og 1 ha R – + u – =

ha

d 15 km h1 R – 150 m

C Clearance J Q' J Q – =

J Q 6.9 20 Log Q 0.1 – 2 1+ Q 0.1 – + u+> @= Q' 0.036 f u= Q 0.065 TClearance f uu=

TClearance

T

TRef TRef ar c h1S h2S –

1000 d u

------------------------© ¹§ ·tan=

h1S h2S

E f

E Calc E f C Receiver C Building C Clearance+ + +=

E Min E Calc E Max = LB

LB 139 E – 20 Log f u+=

LModel 100 7.1 Log W u – 0.023 Mu 1.4 Log hs u 6.1 Log H 1 u 24.37 3.7 H

hb0

--------© ¹§ · 2

u – Log hb u – + + + +

43.2 3.1 Log hb u – > @ Log d u 20 Log f u e13 L og f 3.23 – u

+ +

=



© Forsk 2010 AT282_TRG_E2 109


Where,

• W is the width (in meters) of the streets where the receiver is located

• is the angle (in degrees) formed by the street axes and the direction of the incident wave

• hs is the height (in meters) of the buildings close to the receiver

• H 1 is the average height (in meters) of the buildings close to the receiver

• hb is the height (in meters) of the transmitter antenna with respect to the observer

• hb0 is the height (in meters) of the transmitter antenna with respect to the ground level

• H is the average height (in meters) of the buildings close to the base station

• d is the separation (in kilometres) between the transmitter and the receiver

• f is the frequency (in MHz)

The Sakagami-Kuboi propagation model is valid for:

Studies [2] have shown that the Sakagami-Kuboi propagation model can be extended to frequencies higher than 3 GHz,

which also allows a simplification in terms of the input required by the model.The path loss formula for the extended Sakagami-Kuboi propagation model is:

Where a is a corrective factor with three components:

• W is the width (in meters) of the streets where the receiver is located

• H 0 (= hs = H 1) is the height (in meters) of the buildings close to the receiver

• hb (= hb0 ) is the height (in meters) of the transmitter antenna with respect to the ground

• hm is the height (in meters) of the receiver antenna

• H is the average height (in meters) of the buildings close to the base station

• d is the separation (in metres) between the transmitter and the receiver

• f is the frequency (in GHz)

The extended Sakagami-Kuboi propagation model is valid for:

Studies also show that above 3 GHz, the path loss predicted by the extended model is almost independant of the input

parameters such as street widths and angles. Therefore, the extended Sakagami-Kuboi propagation model can be simpli-

fied to the extended Sakagami propagation model:

The extended Sakagami propagation model is valid for:

The path loss calculation formula of the Sakagami extended propagation model resembles the formula of the Standard

Propagation Model. In Atoll, this model is in fact a copy of the Standard Propagation Model with the following values

assigned to the K coefficients:

5 m < W < 50 m

0° < < 90°

5 m < hs < 80 m

5 m < H 1 < 50 m

20 m < hb < 100 m

0.5 km < d < 10 km

450 MHz < f < 2200 MHz

5 m < W < 50 m

10 m < H 0 < 30 m

10 m < hb < 100 m

0.1 km < d < 3 km

0.8 GHz < f < 8 GHz

1.5 m < hm < 5 m

10 m < hb < 100 m

0.1 km < d < 3 km

3 GHz < f < 8 GHz

1.5 m < hm < 5 m

K1 65.4 (calculated for 3.5 GHz)

M

M

hb0 H t

LModel 54 40 L og d u 30 Log hb u – 21 Log f u a+ + +=

a a H 0 a W a hm + + 11 Log H 0

20 -------© ¹

§ ·u 7.1 Log W

20 ------© ¹

§ ·u – 5 Log hm

1.5 --------© ¹

§ ·u – = =

LModel

54 40 L og d u 30 Log hb

u – 21 Log f u 5 L og hm

u – + +=




For more information on the Standard Propagation Model, see "Standard Propagation Model (SPM)" on page 91.

4.4.10 Appendices

4.4.10.1 Free Space Loss

The calculation of free space loss is based on ITU 525 recommendations.

where,

f is the frequency in MHz,

d is the Tx-Rx distance in km,

Free space loss is stated in dB.

4.4.10.2 Diffraction Loss

The calculation of diffraction is based on ITU 526-5 recommendations. General method for one or more obstacles (knife-

edge diffraction) is used to evaluate diffraction losses (Diffraction loss in dB). Four construction modes are implemented

in Atoll. All of them are based on this same physical principle presented hereafter, but differ in the way they consider one

or several obstacles. Calculations take the earth curvature into account through the effective Earth radius concept (K

factor=1.333).

4.4.10.2.1 Knife-Edge Diffraction

The procedure checks whether a knife-edge obstructs the first Fresnel zone constructed between the transmitter and thereceiver. The diffraction loss, J( Q ), depends on the obstruction parameter (Q), which corresponds to the ratio of the obstruc-

tion height (h) and the radius of the Fresnel zone (R ).

where,

n is the Fresnel zone index,

K2 40

K3 -30

K4 0

K5 0

K6 0

K7 -5

References:[1] Manuel F. Catedra, Jesus Perez-Arriaga, "Cell Planning for Wireless Communications," Artech House Publishers,

1999.

[2] Koshiro Kitao, Shinichi Ichitsubo, "Path Loss Prediction Formula for Urban and Suburban Areas for 4G Systems,"

IEEE, 2006.

LFS 32.4 20 f log 20 d log + +=

Figure 4.14: Knife-Edge Diffraction

R c 0 n d 1 d 2

f d 1 d 2 + ----------------------------------=



© Forsk 2010 AT282_TRG_E2 111


c 0 is the speed of light (2.99792 x108 ms-1),

f is the frequency in Hz

d 1 is the distance from the transmitter to obstacle in m,

d 2 is the distance from obstacle to receiver in m.

We have:

where,

h is the obstruction height (height from the obstacle top to the Tx-Rx axis).

Hence,

For 1 knife-edge method, if ,

Else,

4.4.10.2.2 3 Knife-Edge Deygout Method

The Deygout construction, limited to a maximum of three edges, is applied to the entire profile from transmitter to receiver.

This method is used to evaluate path loss incurred by multiple knife-edges. Deygout method is based on a hierarchical

knife-edge sorting used to distinguish the main edges, which induce the largest losses, and secondary edges, which have

a lesser effect. The edge hierarchy depends on the obstruction parameter (Q) value.

1 Obstacle

A straight line between transmitter and receiver is drawn and the height of the obstacle above the Tx-Rx axis, hi, is calcu-

lated. The obstruction position, di, is also recorded. Qi are evaluated from these data. The point with the highest Q value is

termed the principal edge, p, and the corresponding loss is J(Qp).

Therefore, we have

3 Obstacles

Then, the main edge (point p) is considered as a secondary transmitter or receiver. Therefore, the profile is divided in two

parts: one half profile, between the transmitter and the knife-edge section, another half, constituted by the knife-edge-

receiver section.

Note:

• In case of multiple-knife edge method, the minimum required to estimate diffraction loss

is -0.78.

Q h

r ---=

r R

2 -------=

Q 0.7 – t J Q 6.9 20 Q 0.1 – 2 1+ Q 0.1 – + log +=

J Q 0 =

Q

Figure 4.15: Deygout Construction – 1 Obstacle

Di f f rac tionLoss J QP =




The same procedure is repeated on each half profile to determine the edge with the higher Q. The two obstacles found,

(points t and r), are called ‘secondary edges’. Losses induced by the secondary edges, J (Qt ) and J (Qr ), are then calculated.

Once the edge hierarchy is determined, the total loss is evaluated by adding all the intermediary losses obtained.

Therefore, if

we have

Otherwise, If ,

4.4.10.2.3 Epstein-Peterson Method

The Epstein-Peterson construction is limited to a maximum of three edges. First, Deygout construction is applied to deter-

mine the three main edges over the whole profile as described above. Then, the main edge height, h p, is recalculated

according to the Epstein-Peterson construction. h p is the height above a straight line connecting t and r points. The main

edge position d p is recorded and Q p and J (Q p) are evaluated from these data.

Therefore, we have

4.4.10.2.4 Deygout Method with Correction

The Deygout method with correction (ITU 526-5) is based on the Deygout construction (3 obstacles) plus an empirical

correction, C .

Figure 4.16: Deygout Construction – 3 Obstacles

Note:

• In case of ITU 526-5 and WLL propagation models, Diffraction loss term is determined as

follows:

- If , we have

Where,

- Otherwise

QP 0 !

Dif frac tionLoss J QP J Qt J Qr + +=

QP

0.7 – ! Dif frac tionLoss J QP

=

QP 0.78 – ! Dif frac tionLoss J QP J Qt J Qr + t +=

t minJ QP

6 -------------- 1© ¹

§ ·=

Dif frac tionLoss 0 =

Figure 4.17: Epstein-Peterson Construction

Dif f rac t ionLoss J QP J Qt J Qr + +=



© Forsk 2010 AT282_TRG_E2 113


Therefore, If ,

we have

Otherwise

4.4.10.2.5 Millington Method

The Millington construction, limited to a single edge, is applied over the entire profile. Two horizon lines are drawn at the

transmitter and at the receiver. A straight line between the transmitter and the receiver is defined and the height of the

intersection point between the two horizon lines above the Tx-Rx axis, hh, is calculated. The position d h is recorded and

then, from these values, Qh and J (Qh) are evaluated using the same previous formulas.

Therefore, we have

4.5 Path Loss TuningAtoll can tune path loss matrices obtained from propagation results by the use of real measurements (CW Measurements

or Test Mobile Data). For each measured transmitter, Atoll tries to merge measurements and predictions on the same

points and to smooth the surrounding points of the path loss matrices for homogeneity reasons. A transmitter path loss

matrix can be tuned several times by the use of several measurement paths. All these tuning paths are stored in a cata-

logue. This catalogue is stored under a .tuning folder containing a .dbf file and one .pts file per tuned transmitter. Since a

tuning file can contain several measurement paths, all these measurements are added to the tuning file.

For more information on the tuning files, See "Externalised Tuning Files" on page 64.

4.5.1 Transmitter Path Loss TuningThe same algorithm is used for CW Measurement and Test Mobile Data. It is also the same for main and extended matri-

ces.

Path Losses tuning will be done using two steps.

1. Total matrix correction

A mean error is calculated between each measured value and the corresponding bin in the pathloss matrix. Mean error is

calculated for each pathloss matrix (main and extended) of each transmitter. This mean error is then applied to all the

matrix bins. This tuning is done to smooth the local corrections (step 2) of measured values and not the tuned bins.

2. Local correction for each measured value

For each measured value, an ellipse is used to define the pathloss area which has to be tuned. The main axis of the ellipse

is oriented to the transmitter.The ellipse is user-defined by two parameters :

Note:

• In case of ITU 526-5 propagation model, Diffraction loss term is determined as follows:

- If , we have

Where,

(d : distance stated in km between the transmitter and the receiver).

- Otherwise

QP 0 !

Dif frac tionLoss J QP J Qt J Qr C + + +=

Dif f rac t ionLoss J QP C +=

QP 0.78 – ! Dif f rac tionLoss J QP t J Qt J Qr C + + +=

t min

J QP

6 -------------- 1© ¹§ ·

=

C 8.0 0.04 d +=

Dif f rac tionLoss 0 =

Figure 4.18: Millington Construction

Dif f rac tionLoss J Qh =




• The radius of the axis parallel to the Profile ( A)

• The radius of the axis perpendicular to the Profile (B)

Let’s take M a measurement value and the path loss value at point i , before any tuning.

The squared elliptic distance between i and M is given by:

Where:

and are the X-coordinates of i and M respectively

and are the Y-coordinates of i and M respectively

The mean error for the first tuning is given by:

Where is the error between measurement and prediction at point i

Then, the path loss value is tuned using E :

Finally, a second tuning ( ) is applied where:

so

Where is (measurement gain - losses).

So, the final tuned path loss is:

so

When several ellipses overlap a pathloss bin, the final tuned path loss is given by:

Where is the number of overlapping ellipses

4.5.2 Repeater Path Loss Tuning

In the case of repeaters, Atoll provides only a composite measured value per pixel which is a combination of the contri-

bution of both a transmitter and one or several repeaters. In order to tune the path loss matrices of donor transmitters and

repaters, its is mandatory to split the contribution of each element in the measured value as starting point.

Let’s take M the measured value.

where :

represents the contribution of the donor transmitter in the measured value.

represents the contribution of the repeater in the measured value.

Note:

• M is limited by the minimum measurement threshold defined in the interface.

Note:

• E is limited by the maximum total correction defined in the interface.

Note:

• is limited by the maximum local correction defined in the interface.

P i

Di

X i X M – 2

A2

--------------------------Y i Y M – 2

B2

--------------------------+=

X i X M

Y i Y M

E 1

n---

© ¹§ · ei

i

¦u=

ei

P i new

P i ol d

E +=

R i

R i 1 Di – M g P i – new

– u= R i 1 Di – M g P i ol d

E + – – u=

g

R i

P i tuned

P i new

R i += P i tuned

P i ol d

E R i + +=

P i tuned

1 d j – P j tuned

j

¦© ¹¨ ¸

n d j ¦ – © ¹¨ ¸§ ·

----------------------------------------------------=

n

Note:

• All the values are used in Watts

M M d M r +=

M d

M r



© Forsk 2010 AT282_TRG_E2 115


If and represent respectively the filtered signal level from the donor transmitter and the repeater on a pixel, one

can define the contribution of each element as follows:

and .

Following the path loss tuning process described in "Transmitter Path Loss Tuning" on page 113, the donor transmitter

(resp. the repeater) is then tuned using (resp. ) values.

4.6 Antenna Attenuation Calculation

The modelling method used to evaluate transmitter antenna attenuation, , is described below. Atoll calculates the

accurate azimuth and tilt angles and then, performs a 3-D interpolation of horizontal and vertical patterns to determine the

attenuation of antenna.

Furthermore, you will find explanations about the remote electrical downtilt modelling.

4.6.1 Calculation of Azimuth and Tilt Angles

From the direction of the transmitter antenna and the receiver position relative to the transmitter, Atoll determines the

receiver position relative to the direction of the transmitter antenna (i.e. the direction of the transmitter-receiver path in the

transmitter antenna coordinate system).

aTx and eTx are respectively the transmitter (Tx) antenna azimuth and tilt in the coordinate system .

aRx and eRx are respectively the azimuth and tilt of the receiver (Rx) in the coordinate system .

d is the distance between the transmitter (Tx) and the receiver (Rx).

In the coordinate system , the receiver coordinates are:

(1)

Let az and el respectively be the azimuth and tilt of the receiver in the transmitter antenna coordinate system

. These angles describe the direction of the transmitter-receiver path in the transmitter antenna coordinate

system. Therefore, the receiver coordinates in are:

(2)

C d C r

M d M C d

C d C r +-------------------u= M r M

C r C d C r +-------------------u=

M d M r

LantTx

Figure 4.19: Azimuth and Tilt Computation

S0 x y z

S0 x y z

S0 x y z

x Rx

y Rx

z Rx

eRx cos aRx sin d eRx cos aRx cos d

eRx sin – d

=

STx x '' y '' z ''

STx x '' y '' z ''

x ''Rx

y ''Rx

z ''Rx

el cos az sin d el cos az cos d

el sin – d

=




According to the figure above, we have the following relations:

(3)

and

(4)

Therefore, the relation between the system and the transmitter antenna system is:

(5)

We get,

(6)

Then, substituting the receiver coordinates in the system S0 from Eq. (1) and the receiver coordinates in the system STx

from Eq. (2) in Eq. (6) leads to a system where two solutions are possible:

1st solution: If , then and

2nd solution: If , then

and

If , then

4.6.2 Antenna Pattern 3-D Interpolation

The direction of the transmitter-receiver path in the transmitter antenna coordinate system is given by angle values, az and

el. Atoll considers these values in order to determine transmitter antenna attenuations in the horizontal and vertical

patterns. It reads the attenuation H(az) in the horizontal pattern for the calculated azimuth angle az and the attenuation

V(el) in the vertical pattern for the calculated tilt angle el . Then, it calculates the antenna total attenuation, .

x '

y '

z '

aTx cos aTx sin – 0

aTx sin aTx cos 0

0 0 1

x

y

z

x=

x ''

y ''

z ''

1 0 0

0 eTx cos eTx sin –

0 eTx sin eTx cos

x '

y '

z '

x=

S0 x y z STx x '' y '' z ''

x ''

y ''

z ''

1 0 0

0 eTx cos eTx sin –

0 eTx sin eTx cos


aTx sin aTx cos 0

0 0 1

x x

y

z

x=

x ''

y ''

z ''


eTx cos aTx sin eTx cos aTx cos eTx sin –

eTx sin aTx sin eTx sin aTx cos eTx cos

x

y

z

x=

aRx aTx = az 0 = el eRx eTx – =

aRx aTx z

az 1

eTx cos

aRx aTx – tan--------------------------------------

eTx sin eRx tan

aRx aTx – sin--------------------------------------------------+

------------------------------------------------------------------------------------------------atan=

el az sineTx sin –

aRx aTx – tan--------------------------------------

eTx cos eRx tan

aRx aTx – sin----------------------------------------------------+

¯ ¿® ¾- ½

atan=

az sin aRx aTx – sin 0 az az 180 +=

Notes:

• Atoll assumes that the horizontal and vertical patterns are two cross-sections of the 3-D

pattern. In other words, the description of the antenna pattern must satisfy the following:

H(0)=V(0) and H( )=V( )

In case of an electrical tilt, D, the horizontal pattern is a conical section with a D degrees

elevation off the horizontal plane. Here, horizontal and vertical patterns must satisfy the

following:

H(0)=V( D ) and H( )=V( -D )

If the constraints listed above are satisfied, this implies that:

1. Interpolated horizontal and vertical patterns respectively fit in with the entered horizontal

and vertical patterns, even in case of electrical tilt,

2. The contribution of both the vertical pattern back and front parts are taken into account.

Otherwise, only the second point is guaranteed.

• The above interpolation is performed in dBs.

LantTx az el

LantTx az el H az 180 a z –

180 ------------------------- H 0 V el –

az

180 ---------- H 180 V 180 e l – – + – =



© Forsk 2010 AT282_TRG_E2 117


4.6.3 Additional Electrical Downtilt Modelling

The additional electrical downtilt, AEDT , also referred to as remote electrical downtilt or REDT , introduces a conical trans-

formation of the 3-D antenna pattern in the vertical axis. In order to take it into account, the vertical pattern is transformed

as follows:

when

when

Where, the angle values are in degrees.

The vertical pattern transformation is represented below. The left picture shows the initial vertical pattern when there is no

electrical downtilt and the right one shows the vertical pattern transformation due to an electrical downtilt of 10°.

Then, Atoll proceeds as explained in the previous section. It determines the antenna attenuation in the transformed verti-

cal pattern for the calculated tilt angle (V(el)) and applies the 3-D interpolation formula in order to calculate the antenna

total attenuation, .

4.6.4 Antenna Pattern Smoothing

Empirical propagation models, like the Standard Propagation Model (SPM), require antenna pattern smoothing in the verti-

cal plane to simulate the effects of reflections and diffractions. Signal level predictions can be improved by smoothing the

high-attenuation points of the vertical pattern. You can smooth vertical as well as horizontal antenna patterns in Atoll.

The antenna pattern smoothing algorithm in Atoll first determines the peaks and nulls in the pattern within the smoothing

angle ( ASmoothing ) defined by the user. Peaks (P ) are the lowest attenuation angles and nulls (N ) are the highest attenua-

tion angles in the pattern. Then, it determines the nulls to be smoothed (N Smoothing ) and their corresponding angles accord-

ing to the defined Peak-to-Null Deviation (DPeak-to-Null ). DPeak-to-Null is the minimum difference of attenuation in dBs

between two peaks and a null between them. Finally, Atoll smooths the pattern between 0 and the smoothing angle

( ASmoothing ) by applying the smoothing to a certain smoothing factor (F Smoothing ) defined by the user.

Let’s take an example of an antenna pattern to be smoothed, as shown in Figure 4.21: on page 118. Let DPeak-to-Null be

10 dB, ASmoothing = 90 degrees, and F Smoothing = 0.5.

• Angle values in formulas are stated in degrees.

• The above interpolation is not used in case the transmitter antenna has a 3-D antenna

pattern.

Figure 4.20: Vertical Pattern Transformation due to Electrical Downtilt

V x V x AEDT – = x 90 – 90 [ , ]

V x V x A ED T + = x 90 270 [ , ]

LantTx az el




Atoll first determines the peaks and nulls in the part of the pattern to be smoothed by verifying the slopes of the pattern

curve at each angle.

Peaks (P ) and Nulls (N ):

Then, Atoll verifies whether the difference of attenuation at a given angle is DPeak-to-Null less than the before and after it.

This comparison determines the nulls to be smoothed (N Smoothing ).

Nulls to be smoothed (N Smoothing ):

Once the nulls are known, Atoll applies the smoothing algorithm to all the attenuation values at all the angles between the

first peak, the null, and the last peak.

Figure 4.21: Vertical Antenna Pattern

Figure 4.22: Peaks and Nulls in the Antenna Pattern

Angle (°) Attenuation (dB)

1 0.1

15 33.5

21 13.2

30 37.6

38 16.9

49 32.2

67 15.6

Angle (°) Attenuation (dB)

15 33.5

30 37.6

49 32.2



© Forsk 2010 AT282_TRG_E2 119


4.6.4.1 Smoothing Algorithm

For all nulls surrounded by two peaks P 1 and P 2 at angles and ,

Where,

i is the angle in degrees from to incremented by 1 degree,

A Angle is the attenuation at any given angle which can be i, or , and

F Smoothing is the smoothing factor defined by the user.

4.7 Shadowing ModelPropagation models predict the mean path loss as a function of transmission and reception parameters such as frequency,

antenna heights, and distance, etc. Therefore, the predicted path loss between a transmitter and a receiver is constant, in

a given environment and for a given distance. However, in reality different types of clutter may exist in the transmitter-

receiver path. Therefore, the path losses for the same distance could be different along paths that pass throught different

types of environments. The location of the receiver in different types of clutter causes variations with respect to the mean

path loss values given by the path loss models. Some paths undergo more loss while others are less obstructed and may

have higher received signal strength. The variation of path loss with respect to the mean path loss values predicted by the

propagation models, depending on the type of environment is called shadow fading (shadowing) or slow fading. "Slow"

fading implies that the variations in the path loss due to shadow fading occur comparatively slower than the fast fading

effect (Rayleigh fading), which is due to the mobile receiving multipath copies of a signal.

Different types of clutter (buildings, hills, etc.) make large shadows that cause variations in the path loss over long

distances. As a mobile passes under a shadow, the path loss to the mobile keeps varying from point to point. Shadow

fading varies as the mobile moves, while fast fading can vary even if the mobile remains at the same location or moves

over very small distances. It is crucial to account for the shadow fading in order to predict the reliability of coverage

provided by any mobile cellular system.

The shadowing effect is modelled by a log-normal (Gaussian) distribution, as shown in Figure 4.23: on page 119, whose

standard deviation depends on the type of clutter.

Different clutter types have different shadowing effects. Therefore, each clutter type in Atoll can have a different standard

deviation representing its shadowing characteristics. For different standard deviations, the shape of the Gaussian distri-

bution curve remains similar, as shown in Figure 4.23: on page 119.

The accuracy of this model depends upon:

• The suitability of the range of standard deviation used for each clutter class,

• The definition (bin size) of the digital map,

• How up-to-date the digital map is,

• The number of clutter classes,

• The accuracy of assignment of clutter classes.

Shadowing is applied to the predicted path loss differently depending on the technology, and whether it is applied to predic-

tions or simulations. The following sections explain how shadowing margins are calculated and applied to different tech-

nology documents.

n N Smoothing D1 D2

Ai Smoothed A i F Smoothing Ai AD1

AD2 AD1

–

D2 D1 – ------------------------

© ¹¨ ¸§ ·

i D1 – +¯ ¿® ¾- ½

– – =

D1 D2

D1 D2

Figure 4.23: Log-normal Probability Density Function

V




Shadowing margins are calculated for a given cell edge coverage probability. The cell edge coverage probability is the

probability of coverage at a pixel located at the cell edge, and corresponds to the reliability of coverage that you are plan-

ning to achieve at the cell edge. For example, a cell edge coverage probability of 75 % means that the users located at

the cell edge will receive adequate signal level during 75 % of the time. Therefore, a coverage prediction with a cell edge

coverage probability of x % means that the signal level predicted on each pixel is reliable x % of the time, and the overall

predicted coverage area is reliable at least x % of the time.

GSM GPRS EGPRS Documents

The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 124, and

applied to signal level or C/I as explained below.

• Signal Level-Based Predictions

Signal level-based predictions include coverage predictions (Coverage by Transmitter, Coverage by Signal Level,

and Overlapping Zones) and calculations in point analysis tabs (Profile and Reception) that require calculation of

the received signal level only, and do not depend on interference.

In these calculations (signal level calculations), a shadowing margin ( ) is appl ied to the

received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverageprobability, and depends on the model standard deviation ( in dB) associated to the clutter class where the

receiver is located.

• Interference-Based Predictions

Interference-based predictions include coverage predictions (Coverage by C/I Level, Interfered Zones, GPRS/

EGPRS Coding Schemes, RLC/MAC Throughout/Timeslot, Application Throughput/Timeslot, Circuit Quality Indi-

cators) and calculations in point analysis window’s Interference tab that require calculation of the received signal

level and interference received from other base stations.

In these calculations, ( calculations), the shadowing margin ( ) is appl ied to the ratio of the

carrier power (C ) and the interfering signal levels (I ) received from the interfering base stations. This shadowing

margin is calculated for a given cell edge coverage probability and depends on the C/I standard deviation (

in dB) associated to the clutter class where the receiver is located.

UMTS HSPA and CDMA2000 1xRTT 1xEV-DO Documents

The shadowing margins are calculated as explained in "Shadowing Margin Calculation in Predictions" on page 124 and

"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 125, and applied to signal level, Ec/I0, or Eb/Nt as

explained below.





In these calculations (signal level calculations), a shadowing margin ( ) is appl ied to the

received signal level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage

probability, and depends on the model standard deviation ( in dB) associated to the clutter class where the


• Interference+noise-Based Predictions

Interference+noise-based predictions include coverage predictions (Pilot Reception Analysis, Downlink Total

Noise, Service Area Analyses, Handoff Status, etc.) and point analysis (AS Analysis tab) that require calculation

of the received signal level and interference and noise received from other base stations.

In these calculations, the shadowing margins ( , , or

) are applied to Ec/I0 or Eb/Nt. These shadowing margins are calculated for a given cell

edge coverage probability and depend on the Ec/I0 or Eb/Nt standard deviations ( , , or

, in dB) associated to the clutter class where the receiver is located.

• Macro-Diversity Gains

References:[1] Saunders S. “Antennas and propagation for Wireless Communication Systems” pp. 180-198

[2] Holma H., Toskala A. “WCDMA for UMTS”

[3] Jhong S., Leonard M. “CDMA systems engineering handbook” pp. 309-315, 1051-1053”

[4] Remy J.G., Cueugnet J., Siben C. “Systèmes de radiocommunications avec les mobiles” pp. 309-310

[5] Laiho J., Wacker A., Novosad T. “Radio network planning and optimisation for UMTS” pp. 80-81


Vmodel

C I e M Shadowing C I e –

VC I e


Vmodel

M Shadowing Ec Io e – M Shadowing Eb Nt e DL –

M Shadowing Eb Nt e UL –

VEc Io e V Eb Nt e DL

V Eb Nt e UL



© Forsk 2010 AT282_TRG_E2 121


Atoll calculates the uplink and downlink macro-diversity gains ( and ) depending

on the receiver handover status. These gains are respectively taken into account to evaluate the uplink Eb/Nt in

case of soft handover and the downlink Ec/Io from best server. For detailed description of the calculation of macro-

diversity gains, please refer to "Macro-Diversity Gains Calculation" on page 126.

• Monte-Carlo Simulations

Random values for shadowing margins are calculated for each transmitter-receiver link and applied to the

predicted signal level. A shadowing margin for each transmitter-receiver link in each simulation is obtained by

taking a random value from the probability density distribution for the appropriate clutter class. The probability

distribution is a log-normal distribution as explained above.

TD-SCDMA Documents


"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 125, and applied to signal level or interference+noise

predictions as explained below.


Signal level-based predictions include coverage predictions (Best Server and RSCP P-CCPCH Coverages, P-

CCPCG Pollution, Baton Handover Coverage, DwPCH and UpPCH Coverages, Cell to Cell Interference, and

Scrambling Code Interference) and calculations in point analysis tabs (Profile and Reception) that require calcu-

lation of the received signal level only, and do not depend on interference.

In these calculations (signal level calculations), a shadowing margin ( ) is applied to the





Interference+noise-based predictions include coverage predictions (P-CCPCH Eb/Nt and C/I Coverages, Service

Area Analsyses for downlink and uplink Eb/Nt and C/I, etc.) that require calculation of the received signal level and

interference received from other base stations.

In these calculations, the shadowing margins ( , , or

) are applied to Eb/Nt. These shadowing margins are calculated for a given cell edge cover-

age probabi lity and depend on the Eb/Nt standard deviations ( , , or , in dB)

associated to the clutter class where the receiver is located.




taking a random value from the probability density distribution for the appropriate clutter class. The probabilitydistribution is a log-normal distribution as explained above.

WiMAX 802.16d and WiMAX 802.16e Documents


"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 125 , and applied to signal level or C/(I+N) as

explained below.





In these calculations (signal level calculations), a shadowing margin ( ) is applied to the





Interference-based predictions include coverage predictions (Coverage by C/(I+N) Level, Coverage by Best

Bearer, Coverage by Throughput, etc.) that require calculation of the received signal level and interference.

In these calculations, (C/(I+N) calculat ions), in addition to the shadowing margin ( ) applied to

the received signal level calculated for each pixel, the ratio is applied to the

interfering signal levels (I ). is calculated for a given cell edge coverage probability and depends

on the C/I standard deviation ( in dB) associated to the clutter class where the receiver is located.

Gmacro diversity –

ULGmacro diversity –

DL


Vmodel

M Shadowing Eb Nt e P CCPCH – – M Shadowing Eb Nt e DL –


V Eb Nt e P CCPCH – V Eb Nt e DL

V Eb Nt e UL


Vmodel


M Shadowing model – M Shadowing C I e – –

M Shadowing C I e –

VC I e




The reason why the ratio is used can be understood from the following deri-

vation (linear, not it dB):

Inputs

- : The predicted received carrier power without any shadowing margin.

- : The predicted received interference power without any shadowing margin.

- : Shadowing margin based on the model standard deviation ( )

- : Shadowing margin based on the C/I standard deviation ( )

- : Thermal noise

Calculations

The effective received carrier power is given by:

The effective C/I is given by:

The above equations lead to:

Where corresponds to in dB.

Therefore, the effective C/(I+N) is given by:






LTE Documents


"Shadowing Margin Calculation in Monte-Carlo Simulations" on page 125 , and applied to signal level or C/(I+N) as

explained below.





In these calculations (signal level calculations), a shadowing margin ( ) is applied to the signal

level calculated for each pixel. The shadowing margin is calculated for a given cell edge coverage probability, and

depends on the model standard deviation ( in dB) associated to the clutter class where the receiver is

located.


Interference-based predictions include coverage predictions (Coverage by C/(I+N) Level, Coverage by Best

Bearer, Coverage by Throughput, etc.) that require calculation of the received signal level and received interfer-

ence.

In these calculations, (C/(I+N) calculations), in addi tion to the shadowing margin ( ) appl ied to

the signal level calculated for each pixel, the ratio is applied to the interfering

signal levels (I ). is calculated for a given cell edge coverage probability and depends on the C/I

standard deviation ( in dB) associated to the clutter class where the receiver is located.


C P

I P

mC 10


10 -----------------------------------------------------

mC I e 10


10

----------------------------------------------

N

C mC C P u=

C

I ---- mC I e

C P

I P

-------u=

I C

mC I e C P

I P

-------u

--------------------------mC C P u

mC I e C P

I P

-------u

--------------------------mC

mC I e

------------ I P u= = =

mC

mC I e ------------ M Shadowing model – M Shadowing C I e – –

C

I N + -----------------

mC C P u

mC

mC I e ------------ I P u N +© ¹

§ ·----------------------------------------=


Vmodel




VC I e



© Forsk 2010 AT282_TRG_E2 123


The reason why the ratio is used can be understood from the following deri-

vation (linear, not it dB):

Inputs

- : The predicted received carrier power without any shadowing margin.

- : The predicted received interference power without any shadowing margin.

- : Shadowing margin based on the model standard deviation ( )

- : Shadowing margin based on the C/I standard deviation ( )

- : Thermal noise

Calculations

The effective received carrier power is given by:

The effective C/I is given by:

The above equations lead to:

Where corresponds to in dB.

Therefore, the effective C/(I+N) is given by:






4.7.1 Shadowing Margin Calculation

The following sections describe the calculation method used for determining different shadowin margins.

The following shadowing margins are calculated using the method described below:


C P

I P

mC 10


10 -----------------------------------------------------

mC I e 10


10

----------------------------------------------

N

C mC C P u=

C

I ---- mC I e

C P

I P

-------u=

I C

mC I e C P

I P

-------u

--------------------------mC C P u

mC I e C P

I P

-------u

--------------------------mC

mC I e

------------ I P u= = =

mC

mC I e ------------ M Shadowing model – M Shadowing C I e – –

C

I N + -----------------

mC C P u

mC

mC I e ------------ I P u N +© ¹

§ ·----------------------------------------=

Network TypeStandard

DeviationM Shadowing Applied to

GSM GPRS EGPRSC

C/I

UMTS HSPA

C

Ec/I0

Eb/Nt (DL)

Eb/Nt (UL)

CDMA2000 1xRTT 1xEV-DO

C

Ec/I0

Eb/Nt (DL)

Eb/Nt (UL)

Vmodel M Shadowing model –

VC I e M Shadowing C I e –


VEc Io e M Shadowing Ec Io e –

V Eb Nt e DLM Shadowing Eb Nt e DL –

V Eb Nt e ULM Shadowing Eb Nt e UL –


VEc Io e M Shadowing Ec Io e –






4.7.1.1 Shadowing Margin Calculation in Predictions

Shadowing margins, M Shadowing , are calculated from standard deviation values defined for the clutter class where the pixel

(probe mobile) is located, and required cell edge coverage probability, and applied to the path loss, L path.

Shadowing Error PDF (1 Signal)

The measured path loss in dB can be expressed as a Gaussian random variable:

where,

• L path is the predicted path loss,

• VdB is the user-defined standard deviation of the error,

• G(0,1) is a zero-mean unit-variance Gaussian random variable.

Therefore, the probability density function (pdf) for the random (shadowing) part of path loss is:

The probability that the shadowing error exceeds z dB is

Normalising x by dividing it byVdB:

where Q is the complementary cumulative function.

To ensure a given cell edge coverage probability, , for the predicted value, a shadowing margin, , is added

to the link budget.

Confidence in the prediction can be expressed as:

where,

• is the signal level predicted at the receiver.

•

• EIRP is the effective isotropic radiated power of the transmitter.

• are receiver losses.

• is the receiver antenna gain.

The shadowing margin is calculated such that:

A lookup table is used for mapping the values of Q vs. a set of cell edge coverage probabilities.

TD-SCDMA

C

Eb/Nt P-CCPCH

Eb/Nt (DL)

Eb/Nt (UL)

WiMAX 802.16d

WiMAX 802.16e

C and C/(I+N)

C/(I+N)

LTEC and C/(I+N)

C/(I+N)


V Eb Nt e P CCPCH – M Shadowing Eb Nt e P CCPCH – –







L L pat h VdB G 0 1 u+=

pL x 1

VdB 2 S--------------------- e

x 2

2 VdB2

-------------- –

u=

P L x z ! pL x x d

z

f

³ 1

VdB 2 S--------------------- e

x 2

2 VdB2

-------------- –

x d

z

f

³ u= =

P L x z ! 1

2 S----------- e

x 2

2 ------ –

x d

z

VdB

----------

f

³ u Qz

VdB

---------© ¹§ ·= =

R L M Shadowing

C d P 'Tx L – P re c t L P 'Tx d P rec – G 0 1 VdBu M Shadowing d=

P rec P re c P 'Tx L pat h – M Shadowing – =

P 'Tx E IR P GantRx LRx – +=

LRx

GantRx

P C d P rec t R L M Shadowing 1 P L x M Shadowing – 0 ! – 1 QM Shadowing

VdB

------------------------------© ¹§ · – = = =



© Forsk 2010 AT282_TRG_E2 125


In interference-based predictions, where signal to noise ratio is calculated, the shadowing margin is only applied to the

signal from the interfered transmitter (C). We consider that the interference value is not altered by the shadowing margin.

Random variations also exist in the interfering signals, but taking only the average interference gives accurate results. [3]

explains how a certain level of interference is maintained by congestion control in CDMA-based networks.

4.7.1.2 Shadowing Margin Calculation in Monte-Carlo Simulations

Shadowing margins, M Shadowing , are calculated from standard deviation values defined for the clutter class where the pixel

(probe mobile) is located, and required cell edge coverage probability, and added to the path loss, L path.

Random values are generated during Monte-Carlo simulation. Each user is assigned a service, a mobility type, an activitystatus, a geographic position and a random shadowing value.

For each link, path loss (L) can be broken down to .

Here, is a zero mean gaussian random variable representing variation due to shadowing. It can be

expressed as the sum of two uncorrelated zero mean gaussian random variables, and . models the error related

to the receiver’s location (surrounding environment), and remains the same for all links between the receiver and the base

stations from which it is receiving signals. models the error related to the path between the transmitter and the receiver.

Therefore, in case of two links, we have:

for link 1

for link 2

Standard deviations of and can be calculated from , the model standard deviation , and the

correlation coefficient between and .

Assuming all have the same standard deviations, we have:

Therefore,

is set to 0.5 in Atoll, which gives:

and

Therefore, to model shadowing error common to all the signals received at a receiver ( ) , values are

randomly generated for each receiver. These values have a zero-mean gaussian distribution with a standard deviation of

, where is the model standard deviation associated with the receiver’s clutter class.

Figure 4.24: Normalised Margin M inarg M Shadowing

VdB

------------------------------=

L L pat h [+=

[ G 0 VdB

[L [P [L

[P

[1 [L [P 1

+=

[2 [L [P 2

+=

[L VL [P i VP [i Vmodel

U [1 [2

[P

Vmodel 2 VL

2 VP 2

+=

UVL

2

Vmodel 2

-----------------=

VP 2 Vmodel

2 1 U – u=

VL

2

Vmodel

2

Uu=

U

VL

Vmodel

2 -----------------= VP

Vmodel

2 -----------------=

E Shadowing model – Receiver

Vmodel

2 -----------------

© ¹§ · Vmodel




Next, Atoll generates another random value for each transmitter-receiver pair. This values represents the shadowing error

not related to the location of the receiver ( ) . These values also have a zero-mean gaussian distribution

with a standard deviation .

So, we have:

Random shadowing error has its mean value at zero. Hence, this shadowing modelling method has no impact on the simu-

lated network load. On the other hand, as shadowing errors on the transmitter-receiver links are uncorrelated, the method

influences the calculated macro-diversity gain in case the mobile is in soft handover.

4.7.2 Macro-Diversity Gains Calculation

The following sections explain how uplink and downlink macro-diversity gains are calculated in UMTS HSPA and

CDMA2000 1xRTT 1xEV-DO documents for predictions and AS Analysis tab of the point analysis tool.

4.7.2.1 Uplink Macro-Diversity Gain Evaluation

In UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, mobiles may be in soft handoff (mobile connected to cells located on

different sites). In this case, we can consider the shadowing error pdf described below.

4.7.2.1.1 Shadowing Error PDF (n Signals)

For each link, path loss (L) can be broken down as:

is a zero mean gaussian random variable representing variation due to shadowing. It can be expressed as

the sum of two uncorrelated zero mean gaussian random variables, and . models error related to the receiver

local environment; it is the same whichever the link. models error related to the path between transmitter and receiver.


for the link 1

for the link 2

Knowing , the uplink Eb/Nt standard deviation and the correlation coefficient between and , we

can calculate standard deviations of and (assuming all have the same standard deviations).

We have:

Therefore,

2 Signals Without Recombination

In technologies supporting soft handoff (UMTS and CDMA2000), cell is interference limited. As for one link, to ensure a

required cell edge coverage probability for the prediction, we add to each link budget a shadowing margin,

.

Prediction reliability in order to have Eb/Nt higher or equal to Eb/Nt from the best server can be expressed as:

E Shadowing model – Path

Vmodel

2 -----------------

© ¹§ ·

E Shadowing model – E Shadowing model – Receiver

E Shadowing model – Path

+=

Note:

• The calculation and use of macro-diversity gains can be disabled through the Atoll.ini file.

For more information, see the Administrator Manual .

L L pat h [+=

[ G 0 VdB

[L [P [L

[P

[1 [L [P 1

+=

[2 [L [P 2

+=

[i V Eb Nt e UL U [1 [2

[L VL [P VP [P

V Eb Nt e UL

2 VL

2 VP

2 +=

UVL

2

V Eb Nt e UL

2 --------------------------=

VP 2 V Eb Nt e UL

2 1 U – u=

VL2 V Eb Nt e UL2 Uu=

R L


2signals

C d 1

N 1--------- P 'Tx1 L1 – N 1 – CI pre d

1t [1 P 'Tx1 L pat h1 – N 1 – CI pre d

1 – d=



© Forsk 2010 AT282_TRG_E2 127


or

where

is the quality level (signal to noise ratio) predicted at the receiver for link i.

N i is the noise level for link i.

We note:

and

is the minimum needed margin on each link.

Therefore, the probability of having a quality at least equal to the best predicted one is:

We can express it using , and

Then, we have:

If we introduce user defined standard deviation and correlation coefficient , and consider that is a

Gaussian pdf:

C d 2

N 2

--------- P 'Tx2 L2 – N 2 – CI pre d 1t [2 P 'Tx2 L pat h2

– N 2 – CI pred 1

– d=

CI pred i

M Shadowing Eb Nt e UL – 2signals P 'Tx i L pat hi

– N i – CI pre d i – =

'12

CI pre d 1

CI pre d 2

– =

'12

R LnoMRC


2signals 1 P L1 L2

C d 1

N 1--------- CI pre d

1C d 2

N 2

--------- CI pre d 1

© ¹¨ ¸§ ·

– =

R LnoMRC


2signals 1 P [1 [2 [1 M Shadowing Eb Nt e UL –

2signals! [2 M Shadowing Eb Nt e UL –

2signals'1

2 – ! – =

[L [P 1

[P 2

P [1 [2 [1 M Shadowing Eb Nt e UL – 2signals! [2 M Shadowing Eb Nt e UL –

2signals '12

– ! [L 'L=

P [L'L P

[P 1

[P 2

u [P

1M Shadowing Eb Nt e UL –

2signals 'L – ! [P 2

M Shadowing Eb Nt e UL – 2signals '1

2 – 'L – ! =

P [1 [2 [1 M Shadowing Eb Nt e UL – 2signals! [2 M Shadowing Eb Nt e UL –

2signals '12

– ! [L 'L=

P [L'L P [P

u 'P 1

M Shadowing Eb Nt e UL – 2signals 'L – ! P [P

'P 2


2 – 'L – ! u=

R LnoMRC

M Shadowing Eb Nt e UL – 2signals

1 P [L'L P [P

u 'P 1

M Shadowing Eb Nt e UL – 2signals 'L – ! P [P

'P 2


2 – 'L – ! u 'Ld

f –

f

³ –

© ¹¨ ¸¨ ¸§ ·

=

P [P 'P

i M Shadowing Eb Nt e UL –

2signals'L – !

1

VP 2 S------------------ e

x 2

–

2 VP 2

----------

x d

M Shadowing Eb Nt e UL

– 2signals

'L – © ¹

§ ·

f

³ QM Shadowing Eb Nt e UL –

2signals 'L –

VP

---------------------------------------------------------------------© ¹¨ ¸§ ·

=

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

=

R LnoMRC


1 P [L'L Q

M Shadowing Eb Nt e UL – 2signals 'L –

VP

---------------------------------------------------------------------

© ¹

¨ ¸§ ·

u QM Shadowing Eb Nt e UL –

2signals '12

– 'L –

VP

----------------------------------------------------------------------------------

© ¹

¨ ¸§ ·

u 'Ld

f –

f

³ –

© ¹¨ ¸¨ ¸§ ·

=

V Eb Nt e UL U P [L

R LnoMRC


11

2 S----------- e

x L2

–

2 ---------

QM Shadowing Eb Nt e UL –

2signals x LV Eb Nt e UL

U –

V Eb Nt e UL1 U –

------------------------------------------------------------------------------------------------------- -

© ¹¨ ¸¨ ¸§ ·


2signals x LV Eb Nt e UL

U – '12

–

V Eb Nt e UL1 U –

--------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸§ ·

u x Ld

f –

f

³ –

© ¹¨ ¸¨ ¸§ ·

=




n Signals Without Recombination

We can generalize the previous expression to n signals (n is the number of available signals - Atoll may consider up to 3

signals):

The case where softer handoff occurs (two signals from co-site cells) is equivalent to the one signal case. The Softer/softcase is equivalent to the two signals case. For the path associated with the softer recombination, we will use combined

SNR to calculate the availability of the link.

Correlation Coefficient Determination

There is currently no agreed model for predicting correlation coefficient between and . Two key variables influ-

ence correlation:

• The angle between the two signals. If this angle is small, correlation is high.

• The relative values of the two signal lengths. If angle is 0 and lengths are the same, correlation is zero. Correlation

is different from zero when path lengths differ.

A simple model has been found [1]:

when

is a function of the mean size of obstacles near the receiver and J is also linked to the receiver environment.

In a normal handover status, assuming a hexagonal design for sites, is close to S (+/- S/3) and D1/D2 is close to 1.

In [1,5], when and .

In Atoll, is set to 0.5.

4.7.2.1.2 Uplink Macro-Diversity Gain

Atoll determines the uplink macro-diversity gain ( ) from the shadowing margins calculated in case of one

signal and n signals.

Therefore, we have:

Where n is the number of cell-mobile signals.

4.7.2.2 Downlink Macro-Diversity Gain Evaluation

In UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, in case of soft handoff, mobiles are able to switch from one cell to

another if the best pilot drastically fades. To model this function, we have to consider the probability of fading over the

shadowing margin, both for the best signal and for all the other available signals, in the shadowing margin calculation.

Let us consider the shadowing error pdf described below.

4.7.2.2.1 Shadowing Error PDF (n Signals)

For each link, path loss (L) can be broken down as:

is a zero mean gaussian random variable representing variation due to shadowing. It can be expressed as

the sum of two uncorrelated zero mean gaussian random variables, and . models the error related to the receiver

local environment, which is the same for all links. models the error related to the path between the transmitter and the

receiver.


for the link 1

for the link 2

R LnoMRC

M Shadowing Eb Nt e UL – nsignals

11

2 S----------- e

x L2

–

2 ---------

QM Shadowing Eb Nt e UL –

nsignals x LV Eb Nt e UL

U –

V Eb Nt e UL1 U –

------------------------------------------------------------------------------------------------------- -

© ¹¨ ¸¨ ¸§ ·


nsignals x LV Eb Nt e UL

U – '12

–

V Eb Nt e UL1 U –

--------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸§ ·

u x Ld

f –

f

³ –

© ¹¨ ¸¨ ¸§ ·

=

U [1 [2

UMT

M------

© ¹§ ·

JD1

D2 --------= IT I Sd d

MT

M

U 0.5 = J 0.3= MT S

10 ------=

U

Gmacro diversity – UL


M Shadowing Eb Nt e UL – nsignals

M Shadowing Eb Nt e UL – – =

L L pat h [+=

[ G 0 VdB

[L [P [L

[P

[1 [L [P 1

+=

[2 [L [P 2

+=



© Forsk 2010 AT282_TRG_E2 129


Knowing , the Ec/Io standard deviation and the correlation coefficient between and , we can calculate

standard deviations of and (assuming all have the same standard deviations).

We have:

Therefore,

2 Available Signals

In technologies supporting soft handoff (UMTS and CDMA2000) cells are interference limited. As for one link, to ensure a

required cell edge coverage probability for the prediction, we add a shadowing margin, , to each

link budget.

Prediction reliability to have for the best server can be expressed as:

Or

We note:

is the minimum needed margin on each link.

Therefore, probability of having a quality at least equal to the best predicted one is:

We can express it by using , and

Then, we have:

[i VEc I e o U [1 [2

[L VL [P VP [P

VEc I e o2

VL2

VP 2

+=

UVL

2

VEc I e o2

----------------=

VP 2 VEc I e o

2 1 U – u=

VL2 VEc I e o

2 Uu=

R L M Shadowing Ec Io e – 2signals

Ec

Io-------

Ec

Io-------

© ¹§ ·

pre d t

Ec 1

Io

---------- P pi lo t 1

L1

– Io – Ec

Io

-------

© ¹

§ · pre d

1

t= [1

P pi lo t 1

Lm1

– Io – Ec

Io

-------

© ¹

§ · pre d

1

– d

Ec 2

Io---------- P pi lo t 2

L2 – Io – Ec

Io-------

© ¹§ ·

pre d

1

t= [2 P pi lo t 2 Lm2

– Io – Ec

Io-------

© ¹§ ·

pre d

1

– d

M Shadowing Ec Io e – 2signals

P pil ot i Lmi

– Io – Ec

Io-------

© ¹§ ·

pre d

1

– =

'12 Ec

Io-------

© ¹§ ·

pre d

1 Ec

Io-------

© ¹§ ·

pre d

2

– =

'12

R LnoMRC

M Shadowing Ec Io e –

2signals 1 P L1 L2

Ec 1Io

----------Ec

Io-------

© ¹§ ·

pre d

1

Ec 2

Io----------

Ec

Io-------

© ¹§ ·

pre d

1

© ¹§ · – =

R LnoMRC

M Shadowing Ec Io e – 2signals 1 P [1 [2 [1 M Shadowing Ec Io e –

2signals! [2 M Shadowing Ec Io e – 2signals '1

2 – ! – =

[L [P 1 [P

2

P [1 [2 [1 M Shadowing Ec Io e –

2signals! [2 M Shadowing Ec Io e –

2signals'1

2 – ! [L 'L=

P [L'L P

[P 1

[P 2

u [P

1M Shadowing Ec Io e –

2signals 'L – ! [P 2

M Shadowing Ec Io e – 2signals '1

2 – 'L – ! =

P [1 [2 [1 M Shadowing Ec Io e –

2signals! [2 M Shadowing Ec Io e –

2signals'1

2 – ! [L 'L=

P [L 'L P [P u 'P 1 M Shadowing Ec Io e – 2signals 'L – ! P [P u 'P 2 M Shadowing Ec Io e – 2signals '12 – 'L – ! =

R LnoMRC


1 P [L'L P [P

u 'P 1

M Shadowing Ec Io e – 2signals 'L – ! P [P

u 'P 2

M Shadowing Ec Io e – 2signals '1

2 – 'L – ! 'Ld

f –

f

³ – =

P [P 'P

i M Shadowing Ec Io e –

2signals'L – !

1

VP 2 S------------------ e

x 2

–

2 VP 2

----------

JSHO 'L –

f

³ dx QM Shadowing Ec Io e –

2signals 'L –

VP

-----------------------------------------------------------© ¹¨ ¸§ ·

= =




If we introduce a user defined Ec/Io standard deviation and a correlation coefficient and consider that is a

Gaussian pdf:

n Available Signals

We can generalize the previous expression for n signals (n is the number of available signals - Atoll may consider up to

3 signals):

=1 dB

=5 dB

=10 dB

Figure 4.25: Margin - Probability (Case of 2 Signals)

2 signals

=5 dB

=10 dB

Figure 4.26: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 1dB)

R LnoMRC

M Shadowing Ec Io e – 2signals 1 P [L

'L QuM Shadowing Ec Io e –

2signals 'L –

VP

-----------------------------------------------------------© ¹¨ ¸§ ·

QuM Shadowing Ec Io e –

2signals '12

– 'L –

VP

------------------------------------------------------------------------© ¹¨ ¸§ ·

'Ld

f –

f

³ – =

V U P [L

R LnoMRC


11

2 S

----------- e

x L2

–

2 ---------


2signals x LVEc I e o U –

VEc I e o 1 U –

------------------------------------------------------------------------------------

© ¹

¨ ¸§ ·


2signals '12

– x LVEc I e o U –

VEc I e o 1 U –

------------------------------------------------------------------------------------------------

© ¹

¨ ¸§ ·

x Ld

f –

f

³ – =

R LnoMRC


nsignals

11

2 S----------- e

x L2

–

2 ---------


nsignals x LVEc I e o U –

VEc I e o 1 U – ------------------------------------------------------------------------------------

© ¹¨ ¸§ ·

x QM Shadowing Ec Io e –

nsignals '1i

– x LVEc I e o U –

VEc I e o 1 U – ------------------------------------------------------------------------------------------------

© ¹¨ ¸§ ·

i 2 =

n

x Ld

f –

f

³ – =

'12

'12

'12

'13

'13



© Forsk 2010 AT282_TRG_E2 131


Correlation Coefficient Determination

For further information about determination of the correlation coefficient, please see "Correlation Coefficient Determina-

tion" on page 131.

4.7.2.2.2 Downlink Macro-Diversity Gain

Atoll determines the downlink macro-diversity gain ( ) from the shadowing margins calculated in case of

one signal and n signals.

Therefore, we have:

Where n is the number of available signals.

4.8 Appendices

4.8.1 Transmitter Radio Equipment

Radio equipment such as TMA, feeder and BTS, are taken into account to evaluate:

• Total UL and DL losses of transmitter ( ) and transmitter noise figure in UMTS HSPA,

CDMA2000 1xRTT 1xEV-DO, TD-SCDMA, WiMAX 802.16d, WiMAX 802.16e, and LTE documents,

• Transmitter total losses in GSM GPRS EGPRS documents.

In Atoll, the transmitter-equipment pair is modelled a single entity. The entry to the BTS is considered the reference point

which is the location of the transmission/reception parameters.

2 signals

=5 dB

=10 dB

Figure 4.27: Margin - Probability (Case of 3 Signals with sigma = 8dB, delta1 = 2dB)

'13

'13

Gmacro diversity – DL


M Shadowing Ec Io e – nsignals

M Shadowing Ec Io e – – =

Figure 4.28: Reference Point - Location of the Transmission/Reception parameters

Notes:

• According to the book “Radio network planning and optimisation for UMTS” by Laiho J.,

Wacker A., Novosad T., the noise figure corresponds to the loss in case of passive

components. Therefore, feeder noise figure is equal to the cable uplink losses.

• Loss and gain inputs specified in .atl documents must be positive values.

Lto ta l UL – Lto ta l DL – NF Tx

LTotal

NF Feeder LFeeder UL

=




4.8.1.1 UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, and TD-SCDMA Documents

As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we

have:

where is the BTS noise figure.

Atoll calculates total UL losses as follows:

where,

are the miscellaneous reception losses (Transmitter property),

are the feeder reception losses ( , where , and

are respectively the feeder loss per metre (Feeder property), the reception feeder length in metre (Transmitter

property) and the connector reception losses,

are the losses due to BTS configuration (BTS property),

is the antenna diversity gain (Transmitter property),

is the noise rise at transmitter due to repeaters. This parameter is taken into account only i f the transmitter

has active repeater(s),

is the gain due to TMA.

The noise rise at transmitter due to repeaters is calculated as follows:

For each active repeater ( ), Atoll calculates a noise injection margin ( ). This is the difference between the donor

transmitter noise figure ( ) and the repeater noise figure received at the donor.

where,

is the repeater noise figure,

is the repeater amplification gain (repeater property),

are the losses between the donor transmitter and the repeater (repeater property).

For each active repeater ( ), Atoll converts the noise injection margin ( ) to Watt. Then, it uses the values to calcu-

late the noise r ise at the donor transmitter due to active repeaters ( ) .

The gain due to TMA is calculated as follows:

where,

and are the composite noise figures with and without TMA respectively.

Friis' equation is used to calculate the composite noise figure when there is a TMA.

And,

where,

is the feeder noise figure,

is the TMA noise figure,

NF TX NF BTS=

NF BT S

LTota l UL – LMisc UL

LFeeder UL

LBTS Conf – UL

NR Repeaters G Ant div – UL

– GTMA – + + +=

LMisc UL

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL

+u= LFeeder I Feeder UL

LConnector UL

LBTS Conf – UL

G Ant di v – UL

NR Repeaters

GTMA

k NIM Rpk

NF TX

NI M Rpr NF TX NF Rpk

Gamp

Rpk L

T X R pk – – +© ¹

§ · – =

NF Rpk

Gamp

Rpk

LTX R – pk

k NIM Rpk

NR Repeaters

NR Repeaters 10 Log 11

NI M Rp r

-------------------

r

¦+© ¹¨ ¸u=

GTMA NF CompositeWithoutTMA

NF CompositeWithTMA

– =

NF CompositeWithTMA

NF CompositeWithoutTMA

NF CompositeWithTMA

10 Log 10

NF TM A

10 --------------------

10

NF Feeder

10 --------------------------

1 –

10

GTMAUL

10 ----------------

------------------------------------10

NF BT S

10 -------------------

1 –

10

GTMAUL

10 ----------------

10

GFeeder UL

10 ----------------------

u

-------------------------------------------------+ +

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

u=


NF BT S NF Feeder +=

NF Feeder

NF TMA



© Forsk 2010 AT282_TRG_E2 133


is the BTS noise figure,

is the TMA reception gain,

is the feeder UL gain; .

is the feeder reception loss ( , where , and

are respectively the feeder loss per metre, the reception feeder length in metre and the connector reception loss),

Atoll calculates total DL losses as follows.

where,

is the TMA transmission loss,

is the feeder transmission loss ( , where , and

are respectively the feeder loss per metre, the transmission feeder length in metre and the connector trans-

mission losses),

are the miscellaneous transmission losses,

are the losses due to BTS configuration (BTS property).

4.8.1.2 GSM GPRS EGPRS DocumentsAtoll calculates DL total losses as follows:

where,




mission loss),



4.8.1.3 WiMAX 802.16d and WiMAX 802.16e Documents

As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we

have:



where,






is the gain due to TMA, which is calculated as follows:

where,

NF BTS

GTMAUL

GFeeder UL

GFeeder UL

LFeeder UL

– =

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL


LConnector UL

LTota l DL – LTMA

DL

LFeeder

DL

LMisc

DL

LBTS Conf –

DL

+ + +=

LTMADL

LFeeder DL

LFeeder DL

LFeeder I Feeder DL

LConnector DL

+u= LFeeder I Feeder DL

LConnector DL

LMisc DL

LBTS Conf – DL

LTota l DL – LTMADL

LFeeder DL

LMisc DL

LBTS Conf – DL

+ + +=

LTMADL

LFeeder DL

LFeeder DL

LFeeder I Feeder DL

LConnector DL


LConnector DL

LMisc DL

LBTS Conf – DL

NF TX NF BTS=

NF BT S


LFeeder UL

LBTS Conf – UL

GTMA – + +=

LMisc UL

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL


LConnector UL

LBTS Conf – UL

GTMA


NF CompositeWithTMA

– =






And,

where,









where,




mission losses),



4.8.1.4 LTE Documents As the reference point is the BTS entry, the transmitter noise figure corresponds to the BTS noise figure. Therefore, we

have:



where,






is the gain due to TMA, which is calculated as follows:

where,


NF CompositeWithTMA


NF CompositeWithTMA

10 Log 10

NF TM A

10 --------------------

10

NF Feeder

10 --------------------------

1 –

10

GTMAUL

10 ----------------

------------------------------------10

NF BT S

10 -------------------

1 –

10

GTMAUL

10 ----------------

10

GFeeder UL

10 ----------------------

u

-------------------------------------------------+ +

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

u=


NF BT S NF Feeder +=

NF Feeder

NF TMA

NF BT S

GTMAUL

GFeeder UL

GFeeder UL

LFeeder UL

– =

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL


LConnector UL

Lto ta l DL – LTMADL

LFeeder DL

LMisc DL

LBTS Conf – DL

+ + +=

LTMADL

LFeeder DL

LFeeder DL

LFeeder I Feeder DL

LConnector DL


LConnector DL

LMisc DL

LBTS Conf – DL

NF TX NF BTS=

NF BT S


LFeeder UL

LBTS Conf –

ULGTMA – + +=

LMisc UL

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL


LConnector UL

LBTS Conf – UL

GTMA


NF CompositeWithTMA

– =

NF CompositeWithTMA




© Forsk 2010 AT282_TRG_E2 135



And,

where,









where,




mission losses),



4.8.2 Secondary Antennas

When secondary antennas are installed on a transmitter, the signal level received from it is calculated as follows:

(not in dB2)

Where,

P Tx is the transmitter power (P pilot in UMTS HSPA and CDMA2000 1xRTT 1xEV-DO, P P-CCPCH in TD-SCDMA, P Preamble

in WiMAX 802.16d and WiMAX 802.16e, and P DLRS in LTE),

i is the secondary antenna index,

x i is the percentage of power dedicated to the secondary antenna, i ,

is the gain of the main antenna installed on the transmitter,

LTx are transmitter losses (LTx =Ltotal-DL),

is the gain of the secondary antenna, i , installed on the transmitter,

Lmodel is the path loss calculated by the propagation model,

is the attenuation due to main antenna pattern,

is the attenuation due to pattern of the secondary antenna, i .

NF CompositeWithTMA

10 Log 10

NF TM A

10 --------------------

10

NF Feeder

10 --------------------------

1 –

10

GTMAUL

10 ----------------

------------------------------------10

NF BT S

10 -------------------

1 –

10

GTM AUL

10 ----------------

10

GFeeder UL

10 ----------------------

u

-------------------------------------------------+ +

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

u=


NF BTS NF Feeder +=

NF Feeder

NF TMA

NF BTS

GTMAUL

GFeeder UL

GFeeder UL

LFeeder UL

– =

LFeeder UL

LFeeder UL

LFeeder I Feeder UL

LConnector UL


LConnector UL

Lto ta l DL – LTMADL

LFeeder DL

LMisc DL

LBTS Conf – DL

+ + +=

LTMADL

LFeeder DL

LFeeder DL

LFeeder I Feeder DL

LConnector DL


LConnector DL

LMisc DL

LBTS Conf – DL

2. Formula cannot be directly calculated from components stated in dB and must be converted in linear values.

P re c

P Tx 1 X i

i

¦ – © ¹¨ ¸§ · Ga nt mTx –

LTx

------------------------

La nt mTx – az m el m -----------------------------------------------------------------------

P Tx X i

Ga nt i Tx –

LTx

---------------------

La nt i Tx – az i el i ----------------------------------------------

i

¦+

© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

Lmodel

------------------------------------------------------------------------------------------------------------------------------------------=

Ga nt mTx –

Ga nt i Tx –

La nt mTx – az m el m

La nt i Tx – az i el i




The definition of angles, az and el, depends on the used calculation method.

• Method 1 (must be indicated in an Atoll.ini file):

- az m: the difference between the receiver antenna azimuth and azimuth of the transmitter main antenna,

- el m: the difference between the receiver antenna tilt and tilt of the transmitter main antenna,

- az i : the difference between the receiver antenna azimuth and azimuth of the transmitter secondary antenna, i ,

- el i : the difference between the receiver antenna tilt and tilt of the transmitter secondary antenna, i ,

• Method 2 (default):

- az m : the receiver azimuth in the coordinate system of the transmitter main antenna,

- el m : the receiver tilt in the coordinate system of the transmitter main antenna,

- az i : the receiver azimuth in the coordinate system of the transmitter secondary antenna, i ,

- el i : the receiver tilt in the coordinate system of the transmitter secondary antenna, i .



Chapter 5GSM/GPRS/EDGE Networks

This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations and coverage prediction available in GSM

GPRS EDGE projects.






© Forsk 2010 AT282_TRG_E2 139

Chapter 5: GSM GPRS EDGE Networks

5 GSM GPRS EDGE NetworksThis chapter describes all the calculations performed in Atoll GSM/GPRS/EDGE documents. The first four sections

describe the signal level, interference, GPRS/EDGE-specific, and CQI calculations, respectively. The following three

sections explain the traffic analysis, network dimensioning, and KPI calculation processes. The last section describes the

neighbour allocation process in GSM.

5.1 Signal Level CalculationsThree parameters can be studied in point analysis (Profile tab) and in signal level-based coverage predictions:

Where,

• EIRP is the effective isotropic radiated power of the transmitter,

• is the loss on the transmitter-receiver path (path loss) calculated by the propagation model,

• is the transmitter antenna attenuation (from antenna patterns),

• is the shadowing margin. This parameter is taken into account when the option “Shadowing

taken into account” is selected,

• are the indoor losses, taken into account when the option “Indoor coverage” is selected,

• are the receiver losses,

• is the receiver antenna gain,

• 'P is the power offset defined for the selected TRX type in the transmitter property dialog,

• tt is the TRX type (in the GSM GPRS EGPRS.mdb document template, there are three possible TRX types, BCCH,TCH and inner TCH).

5.1.1 Point Analysis

5.1.1.1 Profile Tab

For a selected transmitter, it is possible to display the signal level received from a TRX type ( ), the path loss,

, or the total losses, . Path loss and total losses are the same for all TRX types.

If the power reduction values defined for all the subcells are the same, the received signal level from the selected trans-

mitter will be the same for all TRX types.

5.1.1.2 Reception Tab Analysis provided in the Reception tab is based on path loss matrices. Therefore, it is possible to display the signal levels

received from TBC transmitters for which path loss matrices have been calculated over their calculation areas.

For each transmitter, Atoll can display the signal level received from a TRX type ( ), the path loss, , or the

total losses, . Path loss and total losses are the same for all TRX types.

If the power reduction values defined for all the subcells are the same, the received signal level from the selected trans-

mitter will be the same for all TRX types.

Reception level bars are displayed in the order of decreasing signal level. The number of displayed bars depends on the

signal level received from the best server. Bars are only displayed for transmitters whose signal level is within a 30 dB

margin from the best server signal level.

Important:

• All the calculations are performed on TBC (to be calculated) transmitters. For the definition

of TBC transmitters please refer to "Path Loss Matrices" on page 80.

• Logarithms used in this chapter (Log function) are base-10 unless stated otherwise.

Studied Parameter Formulas

Signal level ( )

Signal level received from a transmitter on a TRX type

Path loss ( )

Total losses ( )

P rec Tx i

P re c Tx i

tt EIRP tt 'P tt – L pat hTx i

– M Shadowing model – – LIndoor – Gant Rx LRx – +=

L pat hTx i L pat h

Tx i Lmodel Lant Tx

+=

Ltotal

Tx i

Ltotal

Tx i

L pat h

Tx i

M Shadowing model – LIndoor L+ Tx LRx + + + Gant Tx Gant Rx + – =

Lmodel

Lant Tx


LIndoor

LRx

Gant Rx

P re c Tx i

tt

L pat hTx i

Ltotal Txi

P re c Tx i

tt L pat hTx i

Ltotal Tx i




5.1.2 Signal Level-based Coverage Predictions

For each TBC transmitter, Txi, Atoll calculates the selected parameter on each pixel inside the Txi calculation area. In

other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.

Coverage prediction parameters to be set are:• The coverage conditions in order to determine the service area of each TBC transmitter, and

• The display settings to select the displayed parameter and its shading levels.

5.1.2.1 Service Area DeterminationAtoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas

where coverage will be displayed.

We can distinguish eight cases as below. Let us assume that:

• Each transmitter, Txi, belongs to a Hierarchical Cell Structure (HCS) layer, k, with a defined priority and a defined

reception threshold.

• No max range is set.

5.1.2.1.1 All Servers

The service area of Txi corresponds to the pixels where:

5.1.2.1.2 Best Signal Level and a Margin


And

Where M is the specified margin (dB). The Best function considers the highest value from a list of values.

• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the highest.

• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the highest or within a 2 dB

margin from the highest.

• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels

from transmitters which are 2nd best servers.

5.1.2.1.3 Second Best Signal Level and a MarginThe service area of Txi corresponds to the pixels where:

Note:

• You can use a value other than 30 dB for the margin from the best server signal level, for

example a smaller value for improving the calculation speed. For more information on

defining a different value for this margin, see the Administrator Manual .

Note:

• For pure signal level-based calculations (not C/I or C/(I+N)), can be replaced with

or .

MinimumThreshold P re c Tx i

tt d MaximumThreshold

P rec Txi

tt

Ltotal Txi

L pat hTx i

Note:


or .



P rec Txi

tt

Ltotal Txi

L pat hTx i

P rec Tx i

tt Best

j i zP rec

Txj tt M – t

Note:


or .



P rec Txi

tt

Ltotal Txi

L pat hTx i



© Forsk 2010 AT282_TRG_E2 141


And

Where M is the specified margin (dB). The 2 nd Best function considers the second highest value from a list of values.

• If M = 0 dB, Atoll considers pixels where the received signal level from Txi is the second highest.

• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or within

a 2 dB margin from the second highest.


from transmitters which are 3rd best servers.

5.1.2.1.4 Best Signal Level per HCS Layer and a MarginFor each HCS layer, k, the service area of Txi corresponds to the pixels where:

And







5.1.2.1.5 Second Best Signal Level per HCS Layer and a Margin

For each HCS layer, k, the service area of Txi corresponds to the pixels where:

And







5.1.2.1.6 HCS Servers and a Margin


And

And the received exceeds the reception threshold defined per HCS layer.

P rec Tx i

tt 2 nd

Best

j i zP rec

Txj tt M – t

Note:


or .

MinimumThreshold P rec Txi


P rec Tx i

tt

Ltotal Tx i

L pat hTxi

P rec Tx i

BCCH Best

j i zP rec

Tx j BCCH M – t

Note:


or .



P rec Tx i

tt

Ltotal Tx i

L pat hTxi

P rec Tx i

BCCH 2 nd

Best

j i zP rec

Txj BCCH M – t

Note:


or .



P rec Tx i

tt

Ltotal Tx i

L pat hTxi

P rec Tx i

BCCH Best

j i zP rec

Tx j BCCH M – t

P rec Tx i

tt










5.1.2.1.7 Highest Priority HCS Server and a Margin


And

And Txi belongs to the HCS layer with the highest priority. The highest priority is defined by the priority field (0: lowest).








5.1.2.1.8 Best Idle Mode Reselection Criterion (C2)

Such type of coverage is useful :

• To compare idle and dedicated mode best servers for voice traffic• Display the GPRS/EDGE best server map (based on GSM idle mode)

The path loss criterion C1 used for cell selection and reselection is defined by:

The path loss criterion (GSM03.22) is satisfied if .

The reselection criterion C2 is used for cell reselection only and is defined by:

Where is the Cell Reselect Offset defined for the transmitter.


And

The Best function considers the highest value from a list of values.

On each pixel, the transmitter with the highest C2 value is kept. It corresponds to the best server in idle mode. C2 is defined

as an integer in the 3GPP specifications, therefore, the C2 values in the above calculations are rounded down to the near-

est integer.

Note:


or .

Note:

• In the case two layers have the same priority, the traffic is served by the transmitter for which the difference between the received signal strength and the HCS threshold is the

highest. The way the competition is managed between layers with the same priority can be

modified. For more information, see the Administrator Manual .

MinimumThreshold P re c

Tx i tt d MaximumThreshold

P rec Txi

tt

Ltotal Txi

L pat hTx i

P rec Tx i

BCCH Best

j i zP rec

Txj BCCH M – t

P rec Txi

tt

Note:


or .

C1 P re c Tx i

BCCH MinimumThreshold BCCH – =

C1 0 !

C2 C1 CELL_RESELECT_OFFSET +=

CELL_RESELECT_OFFSET


BCCH d MaximumThreshold

P rec Txi

tt

Ltotal Txi

L pat hTx i

C2 Tx i

BCCH Best

j C2

Tx j BCCH =



© Forsk 2010 AT282_TRG_E2 143


5.1.2.2 Coverage Display

5.1.2.2.1 Coverage Resolution

The resolution of the coverage prediction does not depend on the resolutions of the path loss matrices or the geographic

data and can be defined separately for each coverage prediction. Coverage predictions are generated using a bilinear

interpolation method from multi-resolution path loss matrices (similar to the one used to calculate site altitudes, see "Path

Loss Calculations" on page 83 for more information).

5.1.2.2.2 Display Types

It is possible to display the coverage predictions with colours depending on any transmitter attribute or other criteria such

as:

Signal Level (in dBm, dBµV, dBµV/m)

Atoll calculates signal level received from the transmitter on each pixel of each transmitter service area. A pixel of a serv-

ice area is coloured if the signal level exceeds ( ) the defined minimum thresholds (pixel colour depends on signal level).

Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as many

layers as transmitter service areas. Each layer shows the different signal levels available in the transmitter service area.

Best Signal Level (in dBm, dBµV, dBµV/m)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. When other serv-

iceWhen other service areas overlap the studied one, Atoll chooses the highest value. A pixel of a service area is coloured

if the signal level exceeds ( ) the defined thresholds (the pixel colour depends on the signal level). Coverage consists of

several independent layers whose visibility in the workspace can be managed. There are as many layers as defined

thresholds. Each layer corresponds to an area where the signal level from the best server exceeds a defined minimum

threshold.

Path Loss (dB)

Atoll calculates path loss from the transmitter on each pixel of each transmitter service area. A pixel of a service area is

coloured if path loss exceeds ( ) the defined minimum thresholds (pixel colour depends on path loss). Coverage consists

of several independent layers whose visibility in the workspace can be managed. There are as many layers as service

areas. Each layer shows the different path loss levels in the transmitter service area.

Total Losses (dB)

Atoll calculates total losses from the transmitter on each pixel of each transmitter service area. A pixel of a service area

is coloured if total losses exceed ( ) the defined minimum thresholds (pixel colour depends on total losses). Coverage

consists of several independent layers whose visibility in the workspace can be managed. There are as many layers as

service areas. Each layer shows the different total losses levels in the transmitter service area.

Best Server Path Loss (dB)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. When other service

areas overlap the studied one, Atoll determines the best transmitter and evaluates path loss from the best transmitter. A

pixel of a service area is coloured if the path loss exceeds ( ) the defined thresholds (pixel colour depends on path loss).


layers as defined thresholds. Each layer corresponds to an area where the path loss from the best server exceeds a

defined minimum threshold.

Best Server Total Losses (dB)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where service

areas overlap the studied one, Atoll determines the best transmitter and evaluates total losses from the best transmitter.

A pixel of a service area is coloured if the total losses exceed ( ) the defined thresholds (pixel colour depends on total

losses). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are

as many layers as defined thresholds. Each layer corresponds to an area where the total losses from the best server exceed a defined minimum threshold.

Number of Servers

Atoll evaluates how many service areas cover a pixel in order to determine the number of servers. The pixel colour

depends on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can

be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the number of

servers exceeds ( ) a defined minimum threshold.

t

t

t

t

t

t

t




Cell Edge Coverage Probability (%)

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the signal level from this

transmitter fulfils signal conditions defined in Conditions tab with different cell edge coverage probabilities. There is one

coverage area per transmitter in the explorer.

Best Cell Edge Coverage Probability (%)

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the best signal level received

fulfils signal conditions defined in Conditions tab. There is one coverage area per cell edge coverage probability in the

explorer.

Best C2 (dBm)

Atoll calculates C2 values received from transmitters on each pixel of each transmitter service area. When other service

areas overlap the studied one, Atoll chooses the highest value. A pixel of a service area is coloured if the C2 value

exceeds ( ) the defined thresholds (the pixel colour depends on the C2 value). Coverage consists of several independent

layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer

corresponds to an area where the best C2 value exceeds a defined minimum threshold.

5.2 Interference-based CalculationsInterference-based calculations include all the calculations that involve the calculation of interference received from inter-

fering transmitters in addition to the signal level received from the server.

5.2.1 Carrier-to-Interference Ratio CalculationMSA (Mobile Station Allocation) Definition

A wide-ranging definition of an MSA, Mobile Station Allocation, can be that it is a list of channels and an associated MAIO.

More precisely, for different frequency hopping modes, this definition can be:

• Non-hopping (NH): An MSA is the channel assigned to a TRX used by a mobile.

• Baseband hopping (BBH): An MSA is the Mobile Allocation List (MAL) and the TRX index.

• Synthesised frequency hopping (SFH): An MSA is the Mobile Allocation List (MAL) and the Mobile Allocation

Index Offset (MAIO).

From the point of view of a mobile station, BBH and SFH work in the same way.

Notations and Assumptions

In the following description:

• v is a victim transmitter,

• MSAS(v) is the set of MSAs (Mobile Station Allocations) associated to v ,

The number of MSAS(v) depends on TRX types to be analysed. You may study a given TRX type tt (there will be

as many MSA(v) as TRXs allocated to the subcell (v,tt)) or all the TRX types (the number of MSA(v) will correspond

to the number of TRXs allocated to v ).

Several MSAs, m, are related to a transmitter. Therefore, Atoll calculates the C/I for each victim trans-

mitter v with MSA m (m MSAS(v)).

t

TRX index Channel list MAIO MSA

1 53 - (53,-)

2 54 - (54,-)

TRX index Channel list MAIO MSA1 53 * ([53,54,55],0)

2 54 * ([53,54,55],1)

3 55 * ([53,54,55],2)

TRX index Channel list MAIO MSA

1 53 54 55 56 2 ([53,54,55,56],2)

2 53 54 55 56 3 ([53,54,55,56],3)

C v

m

I v

m -----------------

© ¹¨ ¸§ ·



© Forsk 2010 AT282_TRG_E2 145


Atoll considers the most interfered MSA, therefore, the displayed C/I or C/(I+N) are or

, respectively. If the Detailed Results check box is selected, the C/I values for all

MSAs are displayed.

• i is any potential interfering transmitter (TBC transmitters whose calculation areas intersect the service area of v),

• MSAS(i) is the set of MSAs related to potential interferers i ,

• INT(v) is the set of transmitters that interfere v,

• is the carrier power level received from v on m,

• corresponds to the interference received from interfering transmitters i on m,

• used in the C/I calculation is based on the C/I standard deviation.

Calculations

The carrier power level is the power received from the victim transmitter at the receiver.

If the interference conditions are based on C/(I+N), Atoll takes the total noise into account. The total noise is the sum

of the thermal noise (-121 dBm by default or user-defined), the noise figure NF , and the inter-technology down-

link noise rise .

Interference can be received from interfering transmitters i on co-channel and adjacent channels. Interference may also

be received from the transmitters of another technology, and may be caused by frequency intermodulation products.

Therefore,

Here, is the average power control gain defined for the interfering transmitter i .

Each interference component is explained below.

Co- and Adjacent Channel Interference:

is the interference received at v on m on co-channel, given by:

is the interference received at v on m on adjacent channels, given by:

Here, is the carrier power level received from i on n.

T i (n) is occupancy of the MSA n:

is the traffic load defined for the MSA n or i . It can be set to 100% in the coverage prediction properties.

is the activity factor defined for the MSA n of i . If the subcell (i,tt) supports DTX, the value specified in the

coverage prediction properties is used. Otherwise, the activity factor is 1.

Note:

• The C/I shadowing margin is applied on the carrier power level. The interference levels are

not changed.

Note:

• BCCH TRXs are always on. Therefore, DTX and traffic loads do not impact the interference

from BCCH. In other words, and for the BCCH TRXs of the

interferers.

C

I ----

© ¹§ ·

v

Mi nk

C v

m

I v

m -----------------

© ¹¨ ¸§ ·

=

C

I N to t +------------------

© ¹§ ·

v

Mi nk

C v

m

I v

m N tot +-------------------------------

© ¹¨ ¸§ ·

=

C v

m I v

m

M Shadowing

C v

m P rec v

m =

N tot

N thermal

NR i nt er t e chn o y log – v DL

N tot N thermal NF NR i nt er t echn o y log – v DL

+ +=

I v

m I cov

m I adj v

m I i nt er t echn o y log – DL

I IMPx3 GPC i

– + + +=

GPC i

I cov

m

I cov

m pm nv i

P rec i

n T i n uu

n MSAS i ¦© ¹

¨ ¸§ ·

i I NT v ¦

co

=

I adj v

m

I adj v

m pm nv i P re c

i n

F ------------------- T i n uu

n MSAS i ¦© ¹

¨ ¸§ ·

i I NT v ¦

adj

=

P rec i

n

T i n Ltraffic i

n f act i

n u=

Ltraffic i

n

f act i

n

f act i

n 1= Ltraffic i

n 1=




is the probability of having a co- or adjacent channel collision between MSAs n and m, depending on the

used frequency hopping mode.

- Collision Probability for Non Hopping Mode:

- Collision Probability for BBH and SFH Modes:

MSA m of v can be defined as the pair ([f 1,f 2,….f n], MAIO) and MSA n of i as the pair ([f’1,f’2,….f’n], MAIO’)

(where f and f’ are channels).

An occurence refers to the event when a channel f of m encounters a channel f’ of n during

hopping. A collision occurs when f and f’ are co- or adjacent channels:

such that

The probability of collision is the ratio of the number of collisions to the number of occurences:

The probibility of collision depends on the correlation between m and n. There can be two cases:

i. MSAs m and n are correlated

m and n must have identical HSN and synchronisation. The number of occurrences depends on the MAL

size, MAIO, and MAIO’.

Example:

Here, the number of occurrences is 3, the number of co-channel collisions is 1, and the number of adjacent

channel collisions is 1. Therefore,

and

ii. MSAs m and n are not correlated

m and n do not have identical HSN and synchronisation. The probability of collision is the same for all the

channels.

Example:

Here, the number of occurrences is 9, the number of co-channel collisions is 1, and the number of adjacent

channel collisions is 3. Therefore,

and

Inter-technology Downlink Interference:

is the total inter-technology interference level on m due to transmitters in a linked Atoll document.

The interference from a transmitter Tx in a linked Atoll document is given as:

Schematic view of hopping

sequences

MSA m of v

([34 37 39], MAIO=0)34 37 39

MSA n of i

([38 36 34], MAIO’=2)38 36 34

Schematic view of hopping

sequences

MSA m of v

([34 37 39], MAIO=0)34 37 39

MSA n of i

([38 36 34], MAIO’=2)38 36 34

pm nv i

pm nv i

1=

OCCUR f mv f 'ni

Co ll i sion OCCUR f mv

f 'ni = f m

v f 'n

i – 0 or 1=

pm nv i ncollision

noccurence

---------------------------=

pm nv i

co1

3---= pm n

v i adj

1

3---=

pm n

v i

co1

9---= pm n

v i

adj 1

3---=

I i nt er t e chn o y log – DL

I i nt er t e chn o y log – DL P Transmitted

Tx ic i

Ltotal Tx

IC P ic i f Tx u

------------------------------------------

ni

¦=



© Forsk 2010 AT282_TRG_E2 147


Where is the frequency used by the transmitter Tx within its list of frequencies, is the total

transmitted Tx power on , are the total losses between the transmitter Tx and the receiver, and

is the inter-technology channel protection between the frequencies used by the transmitter Tx and the victim trans-

mitter v .

Intermodulation Interference:

is the third order intermodulation interference:

Where is the intermodulation interference from frequencies used by the interfering transmitter

, is the interference due to spurious emissions from the interfering transmitter

, and is the intermodulation interference received at the mobile terminal

.

is the carrier power level received from the interferer i , is the third order intermodulation loss at the

victim transmitter v , is the spurious emission power level received from the interferer i , and is the

third order intermodulation protection factor for the terminal.

For a pair of frequencies, and , two third order intermodulation products are generated at frequencies

and . If a transmitter uses or , it is interfered by transmitters using and .

All interferer frequencies are used to calculate intermodulation products. When several frequency pairs generate

intermodulation products, the IMPs are independenly calculated and added to the interference. If power received

over different frequencies is not the same for two frequencies (not the same power offset for example), the corre-

sponding intermodulation frequencies are ignored. Frequency hopping is not considered to have any impact on

the intermodulation products. IMPs for hopping and non-hopping cases are considered to be the same.

Intermodulation products generated by the adjacent frequencies of the frequencies actually being used by an inter-

ferer are not taken into account. Similarly, intermodulation interference received on the adjacent frequencies of

the frequencies used by the victim are also ignored.


Analysis provided in the Interference tab is based on path loss matrices. Therefore, it is possible to display the interference

levels received from TBC transmitters for which path loss matrices have been calculated over their calculation areas.

Atoll displays the following at the receiver:

• The carrier power level received from the victim transmitter v on the most interfered MAS m,

• Co-channel, adjacent channel, or both co- and adjacent channel interference received from interfering transmitters

i on MAS m (for further information about noise calculation, please refer to Signal to noise calculation: noise cal-

culation part),

Interferers are sorted in the order of descending carrier power levels.

5.2.3 Interference-based Coverage Predictions

Two interference-based coverage predictions are available:

• Coverage by C/I Level: Provides a global analysis of the network quality.

Atoll calculates the C/I on each pixel within the service area of studied transmitters, determines the pixels where

the calculated C/I exceeds the defined minimum threshold, and colours these pixels depending on C/I value.

• Interfered Zones: Shows the areas where a transmitter is interfered.

Notes:

• In case of frequency hopping, the ICP value is weighted according to the fractional load.

• In the ICP, the frequency gap is based on the defined base frequency for each technology

(e.g., 935 MHz in GSM 900)

ic i i th

P Transmitted Tx

ic i

ic i Ltotal Tx

IC P ic i f Tx

I IMPx3 I IMPx3 I IMPx3TX

I IMPx3SE

I IMPx3Term

+ +=

I IMPx3TX

I IMPx3TX P rec

i

LIMPx3v

-----------------= I IMPx3SE

I IMPx3SE

P re c SE i

= I IMPx3Term

I IMPx3Term 3 P rec

i u

2 F IMPx3Prot u

---------------------------=

P re c

i

LIMPx3

P rec SE i

F IMPx3Prot

f 1 f 2

f 3 2 f 1u f 2 – = f 4 2 f 2 u f 1 – = f 3 f 4 f 1 f 2

Notes:

• Neither DTX nor traffic load of TRXs are taken into account to evaluate interference levels.

Therefore, we have .

• The C/I shadowing margin is applied on the carrier power level. The interference levels are

not changed.

T i n Ltraffic i n f act

i n u 1= =




Atoll calculates the C/I on each pixel within the service area of studied transmitters, determines the pixels where

the calculated C/I is lower than the defined maximum threshold, and colours these pixels depending on colour of

the interfered transmitter.



Coverage prediction parameters to be set are:

• The coverage conditions in order to determine the service area of each TBC transmitter,

• The interference conditions to meet for a pixel to be covered, and


The thermal noise (N = -121 dBm, by default) is used in the calculations if the coverage prediction is based on C/(I+N).

This value can be modified by the user.

5.2.3.1 Service Area Determination

Atoll uses parameters entered in the Condition tab of the coverage prediction properties dialogue to determine the areas

where coverage will be displayed. Service areas are determined in the same manner as for signal level-based coverage

predictions. See "Service Area Determination" on page 140 for more information.

5.2.3.2 Coverage Area Determination

For each victim transmitter v , coverage area corresponds to pixels where or is between the lower and upper

thresholds defined in the coverage prediction properties.

The two options defining the thresholds are explained below.

5.2.3.2.1 Interference Condition Satisfied by At Least One TRXIn this case, the coverage area of a transmitter Txi corresponds to the pixels where:

or

Where, TRX j is any TRX belonging to Txi.

5.2.3.2.2 Interference Condition Satisfied by The Worst TRX

In this case, the coverage area of a transmitter Txi corresponds to the pixels where:

or

Where, TRX j is the TRX (belonging to Txi) with the worst C/I or C/(I+N) at the pixel.








It is possible to display the coverage predictions with colours depending on any transmitter attribute or other criteria such

as:

C/I Level

Each pixel of the transmitter coverage area is coloured if the calculated C/I (or C/(I+N)) level exceeds ( ) the specified

minimum thresholds (pixel colour depends on C/I (or C/(I+N)) level). Coverage consists of several independent layers

whose visibility in the workspace can be managed. There are as many layers as transmitter coverage areas. Each layer

shows the different C/I levels available in the transmitter coverage area.

Max C/I Level

Atoll compares calculated C/I (or C/(I+N)) levels received from transmitters on each pixel of each transmitter coverage

area where coverage areas overlap the studied one and chooses the highest value. A pixel of a coverage area is coloured

if the C/I (or C/(I+N)) level exceeds ( ) the specified thresholds (the pixel colour depends on the C/I (or C/(I+N)) level).


layers as defined thresholds. Each layer corresponds to an area where the highest received C/I level exceeds a defined

minimum threshold.

C

I ----

© ¹§ ·

v

C

I N +------------

© ¹§ ·

v

Minimum threshold C

I ----

© ¹§ ·

v TRX j

d Maximum threshold Minimum threshold C

I N +------------

© ¹§ ·

v TRX j

d Maximum threshold

Minimum threshold C

I ----

© ¹§ ·

v TRX j

d Maximum threshold Minimum threshold C

I N +------------

© ¹§ ·

v TRX j

d Maximum threshold

t

t



© Forsk 2010 AT282_TRG_E2 149


Min C/I Level

Atoll compares C/I (or C/(I+N)) levels received from transmitters on each pixel of each transmitter coverage area where

the coverage areas overlap the studied one and chooses the lowest value. A pixel of a coverage area is coloured if the C/

I (or C/(I+N)) level exceeds ( ) the specified thresholds (the pixel colour depends on the C/I (or C/(I+N)) level). Coverage


defined thresholds. Each layer corresponds to an area where the lowest received C/I level exceeds a defined minimum

threshold.

5.3 GPRS/EDGE CalculationsGPRS/EDGE calculations include coding scheme selection and throughput calculation. Coding schemes may be selected

using ideal link adaptation or without it. Once coding schemes have been selected, throughputs corresponding to these

coding schemes are readily determined from the look-up tables.

The following sections describe the two categories of calculations, i.e., with and without ideal link adaptations. Ideal link

adaptation implies that the selected coding scheme corresponds to the highest available throughput under the given radio

conditions.

GPRS/EDGE calculations may be based on signal levels (C) alone, on C/I, or on C/(I+N). For calculating the noise, either

the noise figure defined for the calculations or that of the selected terminal type is used.

Different GPRS/EDGE configurations may be defined for transmitter and terminals. In this case, Atoll only selects the

coding schemes that are common in the two. If no terminal type is defined for the calculation, or if the terminal type does

not have any GPRS/EDGE configuration assigned to it, Atoll only uses the GPRS/EDGE configuration of the transmitter.

Similarly, if a transmitter does not have any GPRS/EDGE configuration assigned to it, Atoll only uses the GPRS/EDGE

configuration of the terminal type. If both the transmitter and the terminal type do not have any GPRS/EDGE configuration

assigned to them, no coding scheme selection and throughput calculation is carried out.

In the following calculations, we assume that:

• is the signal level received from the selected TRX type (tt ) or on all the TRXs of Txi on each pixel of

the Txi coverage area,

• is the Power Backoff defined for the subcell for 8PSK, 16QAM, or 32QAM modulations,

• CS is the set of all available coding schemes,

• are the values of reception thresholds for the coding schemes available in the GPRS/

EDGE configuration,

• are the values of C/I thresholds for the coding schemes available in the GPRS/EDGE configu-

ration,

• are the values of C/(I+N) thresholds for the coding schemes available in the GPRS/EDGE

configuration,• The priorities of the coding scheme lists are as follows: DBS > DAS > MCS > CS.

When the calculations are based on C/I and C/(I+N):

• Atoll calculates the carrier-to-interference ratio for all the GPRS/EDGE TBC transmitters but takes into account

all the TBC transmitters (GSM and GPRS/EDGE) to evaluate the interference.

• The reception thresholds given for signal level C are internally converted to C/N thresholds (where N is the thermal

noise defined in the document database at -121 dBm by default) in order to be indexed by C/(I+N) values. C/I

thresholds are also indexed by the C/(I+N) value.

For more information on interference (I ) calculation, see "Carrier-to-Interference Ratio Calculation" on page 144.

5.3.1 Coding Scheme Selection and Throughput Calculation

Without Ideal Link Adaptation

5.3.1.1 Calculations Based on C

Coding Scheme Selection

Atoll selects a coding scheme, cs, from among the coding schemes available in the GPRS/EDGE configuration, such that:

For each TRX type, tt ,

The selected coding scheme, cs, is the coding scheme with the lowest coding scheme number from the lowest priority

coding scheme list.

t

P re c Tx i

TR X

P Backoff Tx i

TR X

Reception Threshold CS

C

I ---- Threshold © ¹

§ ·CS

C

I N +------------ Threshold © ¹

§ ·CS

c s L ow es t C SP rec

Txi TRX P Backoff

Txi TRX – Reception Threshold CS!© ¹

§ ·=




Throughput Calculation

Once the coding scheme cs is selected, Atoll reads the corresponding throughput value for the received signal level from

the Throughput=f(C) graph associated with cs.

5.3.1.2 Calculations Based on C/I


Atoll selects two coding schemes from among the coding schemes available in the GPRS/EDGE configuration, such that:


And,

csC is the coding scheme determined from the signal level, and csC/I is the coding scheme determined from the C/I level.

Both coding schemes are the coding schemes with the lowest coding scheme number from the lowest priority coding

scheme list.

The selected coding scheme, cs, is the coding scheme with the lower coding scheme number among csC and csC/I :

.

Throughput Calculation Based on the Worst Case Between C and C/I

For the coding scheme csC determined above, a throughput value, TP C , corresponding to the signal level is determinedfrom the TP = f(C) graph.

For the coding scheme csC/I determined above, a throughput value, TP C/I , corresponding to the C/I is determined from the

TP = f(C/I) graph.

The resulting throughput TP is the lower of the two values, TP C and TP C/I : .

5.3.1.3 Calculations Based on C/(I+N)


Atoll selects two coding schemes from among the coding schemes available in the GPRS/EDGE configuration, such that:


And,

csC/N is the coding scheme determined from the C/N, and csC/(I+N) is the coding scheme determined from the C/(I+N) level.

Both coding schemes are the coding schemes with the lowest coding scheme numbers from the lowest priority coding

scheme list.

The selected coding scheme, cs, is the coding scheme with the higher coding scheme number among csC/N and csC/(I+N):

.

Throughput Calculation Based on Interpolation Between C/N and C/(I+N)

For the coding scheme csC/N

determined above, the TP = f(C) graph is internally converted to TP = f(C/N) graph. A

throughput value, TP C/N , corresponding to the C/(I+N) is determined from the TP = f(C/N) graph.

For the coding scheme csC/(I+N) determined above, the TP = f(C/I) graph is internally converted to TP = f(C/(I+N)) graph.

A throughput value, TP C/(I+N), corresponding to the C/(I+N) is determined from the TP = f(C/(I+N)) graph.

The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The

throughput interpolation method consists in interpolating TP C/N and TP C/(I+N) according to the respective weights of I and

N values.

The resulting throughput TP is given by:

Where , pN is the thermal noise power (value in Watts), and p(I+N) is the interferences + thermal noise

power (value in Watts).

csC Lowest CSP rec

Txi TRX P Backoff

Txi TRX – Reception Threshold CS!© ¹§ ·=

csC I e Lowest CSP rec

Txi TRX P Backoff

Txi TRX –

I -----------------------------------------------------------------------------

C


§ ·CS

!© ¹¨ ¸¨ ¸§ ·

=

cs Mi n csC csC I e =

T P M in TP C TP C I e =

csC N e Lowest CSP

rec

Txi TRX P

Backoff

Txi TRX –

N ----------------------------------------------------------------------------- C I N +------------ Threshold © ¹§ · CS!© ¹

¨ ¸¨ ¸§ ·

=

csC I N + e Lowest CSP rec

Txi TRX P Backoff

Txi TRX –

I N +-----------------------------------------------------------------------------

C

I N +------------ Threshold © ¹

§ ·CS

!© ¹¨ ¸¨ ¸§ ·

=

cs Max csC N e csC I N + e =

TP D TP C N e u 1 D – TP C I N + e u+=

D pN

p I N + ---------------------=



© Forsk 2010 AT282_TRG_E2 151


5.3.2 Coding Scheme Selection and Throughput Calculation With

Ideal Link Adaptation

5.3.2.1 Calculations Based on C

Throughput Calculation

For the received signal level, and coding schemes whose reception thresholds are lower than the received signal level,

Atoll determines the highest throughput from the graphs available in the GPRS/EDGE configuration.


The selected coding scheme, cs, is the one corresponding to the highest throughput calculated above.

If there are more than one coding schemes providing the highest throughput at the pixel, the selected coding scheme, cs,

is the one with the lowest coding scheme number from the lowest priority coding scheme list.

5.3.2.2 Calculations Based on C/I

Throughput Calculation Based on Worst Case Between C and C/I

For the received signal level, and coding schemes whose reception thresholds are lower than the received signal level,Atoll determines the highest throughput from the graphs available in the GPRS/EDGE configuration.

For the received C/I, and coding schemes whose C/I thresholds are lower than the received C/I, Atoll determines the high-

est throughput from the graphs available in the GPRS/EDGE configuration.

The resulting throughput TP is the lower of the two values, TP C and TP C/I .


The selected coding scheme, cs, is the one corresponding to the lower of the two highest throughputs calculated above.

If there are more than one coding schemes providing the highest throughputs at the pixel, the selected coding scheme,

cs, is the one with the lowest coding scheme number from the lowest priority coding scheme list.


Throughput Calculation Based on Interpolation Between C/N and C/(I+N)

Atoll internally converts the TP = f(C) graphs into TP = f(C/N) graphs. For the received C/(I+N), and coding schemes

whose C/(I+N) thresholds are lower than the received C/(I+N), Atoll determines the highest throughput from the TP = f(C/

N) graphs available in the GPRS/EDGE configuration.

Atoll internally converts the TP = f(C/I) graphs into TP = f(C/(I+N)) graphs. For the received C/(I+N), and coding schemes

whose C/(I+N) thresholds are lower than the received C/(I+N), Atoll determines the highest throughput from the TP = f(C/

(I+N)) graphs available in the GPRS/EDGE configuration.

The final throughput is computed by interpolating between the throughput values obtained from these two graphs. The

throughput interpolation method consists in interpolating TP C/N and TP C/(I+N) according to the respective weights of I and

N values.

TP=f C

TP C Highest TP=f C P rec Tx i

TR X P Backoff Txi

TR X – = = CSP rec

Txi TRX P Backoff


§ ·

TP=f C

TP C Highest TP=f C P rec Tx i

TR X P Backoff Txi

TR X – = = CSP rec

Txi TRX P Backoff


§ ·

TP=f C I e

TP C I e Highest TP=f C I e P re c

Tx i TR X P Backoff

Tx i TR X –

I ---------------------------------------------------------------------------=

© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

= CSP rec

Txi TRX P Backoff

Txi TRX –

I -----------------------------------------------------------------------------

C


§ ·CS

!© ¹¨ ¸¨ ¸§ ·

T P M in TP C TP C I e =

TP C N e Highest TP=f C N ----

P re c

Tx i TR X P

Backoff

Txi TR X –

I N +---------------------------------------------------------------------------=

© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

= CSP rec

Txi TRX P Backoff

Txi TRX –

I N +-----------------------------------------------------------------------------

C

I N +------------ Threshold © ¹

§ ·CS

!© ¹¨ ¸¨ ¸

§ ·

TP C I N + e Highest TP=f C

I N +------------

P re c Tx i

TR X P Backoff Txi

TR X –

I N +---------------------------------------------------------------------------=

© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

= CSP rec

Txi TRX P Backoff

Txi TRX –

I N +-----------------------------------------------------------------------------

C

I N +------------ Threshold © ¹

§ ·CS

!© ¹¨ ¸¨ ¸§ ·




The resulting throughput TP is given by:

Where , pN is the thermal noise power (value in Watts), and p(I+N) is the interferences + thermal noise

power (value in Watts).


The selected coding scheme, cs, is the one corresponding to the higher of the two highest throughputs calculated above.

If there are more than one coding schemes providing the highest throughputs at the pixel, the selected coding scheme,

cs, is the one with the highest coding scheme number from the highest priority coding scheme list.

5.3.3 Application Throughput Calculation

Application throughput is calculated from the RLC/MAC throughput as follows:

Where is the RLC/MAC throughput, and and are the throughput offset (kbps) and the through-

put scaling factor (%) defined for the selected service.

5.3.4 BLER Calculation

Block error rate is calculated as follows:

Where TP is the throughput per timeslot calculated for a pixel and TP MAX is the maximum throughput per timeslot read

from the GPRS/EDGE configuration used for the calculations.

5.3.5 GPRS/EDGE Coverage Predictions

Two GPRS/EDGE coverage predictions are available:

• GPRS/EDGE Coding Schemes: Shows the areas where various coding schemes are available.

• Packet Throughput and Quality: Shows the throughputs corresponding to the coding schemes available.


other words, each pixel inside the Txi calculation area is considered a probe (non-interfering) receiver.Coverage prediction parameters to be set are:


• The interference conditions to meet for a pixel to be covered, and







We can distinguish eight cases as below. Let us assume that:


reception threshold.• Each transmitter, Txi, is GPRS/EDGE-capable.






TP D TP C N e u 1 D – TP C I N + e u+=

D pN

p I N + ---------------------=

TP App li cat io n TP RLC MAC e SF

100 ---------- TP Offset – u=

TP RLC MAC e TP Offset SF

BLER

TP

TP MAX

------------------ I f T P T P MAX d

0 If TP TP MAX ! ¯°®°-

=

SubcellReceptionThreshold P rec Txi

tt d


tt d



© Forsk 2010 AT282_TRG_E2 153


And







5.3.5.1.3 Second Best Signal Level and a Margin


And







5.3.5.1.4 Best Signal Level per HCS Layer and a MarginFor each HCS layer, k, the service area of Txi corresponds to the pixels where:

And







5.3.5.1.5 Second Best Signal Level per HCS Layer and a MarginFor each HCS layer, k, the service area of Txi corresponds to the pixels where:

And









And




P rec Tx i

tt Best

j i zP rec

Txj tt M – t

SubcellReceptionThreshold P rec Tx i

tt d

P re c Tx i

tt 2 nd

Best

j i zP re c

Tx j tt M – t


tt d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t


tt d

P re c Tx i

BCCH 2 nd

Best

j i zP rec

Txj BCCH M – t


tt d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t

P rec Tx i

tt










And






margin from the highest.• If M = -2 dB, Atoll considers pixels where the received signal level from Txi is 2 dB higher than the signal levels


5.3.5.1.8 Best Idle Mode Reselection Criterion (C2)

Such type of coverage is useful:

• To compare idle and dedicated mode best servers for voice traffic

• Display the GPRS/EDGE best server map (based on GSM idle mode)

The path loss criterion C1 used for cell selection and reselection is defined by:

The path loss criterion (GSM03.22) is satisfied if .

The reselection criterion C2 is used for cell reselection only and is defined by:

Where is the Cell Reselect Offset defined for the transmitter.


And

The Best function considers the highest value from a list of values.

On each pixel, the transmitter with the highest C2 value is kept. It corresponds to the best server in idle mode. C2 is defined

as an integer in the 3GPP specifications, therefore, the C2 values in the above calculations are rounded down to the near-

est integer.








It is possible to display the coverage predictions with colours depending on criteria such as:

Note:

• In the case two layers have the same priority, the traffic is served by the transmitter for

which the difference between the received signal strength and the HCS threshold is the




tt d

P re c

Tx i

BCCH Best

j i z P rec

Txj

BCCH M – t

P rec Txi

tt

C1 P re c Tx i

BCCH MinimumThreshold BCCH – =

C1 0 !

C2 C1 CELL_RESELECT_OFFSET +=

CELL_RESELECT_OFFSET


BCCH d

C2 Tx i

BCCH Best

j C2

Tx j BCCH =



© Forsk 2010 AT282_TRG_E2 155


GPRS/EDGE Coding Schemes: Coding Schemes

Only the pixels with a coding scheme assigned are coloured. The pixel colour depends on the assigned coding scheme.

Coverage consists of several independent layers whose visibility in the map window can be managed. There are as many

layers as transmitter coverage areas. Each layer shows the coding schemes available in the transmitter coverage area.

GPRS/EDGE Coding Schemes: Best Coding Schemes

On each pixel, Atoll chooses the highest coding scheme available from the TRXs of different transmitters covering that

pixel. Only the pixels with a coding scheme assigned are coloured. The pixel colour depends on the assigned coding

scheme. Coverage consists of several independent layers whose visibility in the map window can be managed. There are

as many layers as possible coding schemes. Each layer shows the areas where a given coding scheme can be used.

Packet Throughput and Quality: RLC/MAC Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated RLC/MAC throughput per timeslot from any transmitter covering

that pixel exceeds the defined minimum threshold. The pixel colour depends on the RLC/MAC throughput per timeslot.


layers as transmitter coverage areas and throughput display thresholds. Each layer shows the RLC/MAC throughput that

a transmitter can provide on one timeslot.

Packet Throughput and Quality: Best RLC/MAC Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated highest RLC/MAC throughput per timeslot from any transmitter

covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the highest RLC/MAC throughput

per timeslot. Coverage consists of several independent layers whose visibility in the map window can be managed. There

are as many layers as throughput display thresholds. Each layer shows the best RLC/MAC throughput that any transmitter

can provide on one timeslot.

Packet Throughput and Quality: Average RLC/MAC Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated average RLC/MAC throughput per timeslot from all the transmit-

ters covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the average RLC/MAC

throughput per timeslot. Coverage consists of several independent layers whose visibility in the map window can be

managed. There are as many layers as throughput display thresholds. Each layer shows the average RLC/MAC through-

put that all the transmitters can provide on one timeslot.

Packet Throughput and Quality: Application Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated application throughput per timeslot from any transmitter covering

that pixel exceeds the defined minimum threshold. The pixel colour depends on the application throughput per timeslot.


layers as transmitter coverage areas and throughput display thresholds. Each layer shows the application throughput that

a transmitter can provide on one timeslot.

Packet Throughput and Quality: Best Application Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated highest application throughput per timeslot from any transmitter

covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the highest application through-

put per timeslot. Coverage consists of several independent layers whose visibility in the map window can be managed.

There are as many layers as throughput display thresholds. Each layer shows the best application throughput that any

transmitter can provide on one timeslot.

Packet Throughput and Quality: Average Application Throughput/Timeslot (kbps)

A pixel of the coverage area is coloured if the calculated average application throughput per timeslot from all the transmit-

ters covering that pixel exceeds the defined minimum threshold. The pixel colour depends on the average application

throughput per timeslot. Coverage consists of several independent layers whose visibility in the map window can be

managed. There are as many layers as throughput display thresholds. Each layer shows the average application through-

put that all the transmitters can provide on one timeslot.

Packet Throughput and Quality: Max Application Throughput (kbps)

A pixel of the coverage area is coloured if the calculated application throughput from any transmitter covering that pixel

exceeds the defined minimum threshold. The pixel colour depends on the application throughput for all the timeslots

supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). Coverage consists

of several independent layers whose visibility in the map window can be managed. There are as many layers as transmitter

coverage areas and throughput display thresholds. Each layer shows the application throughput that a transmitter can

provide on all available timeslots in the terminal.

Packet Throughput and Quality: Best Max Application Throughput (kbps)

A pixel of the coverage area is coloured if the calculated highest application throughput from any transmitter covering that

pixel exceeds the defined minimum threshold. The pixel colour depends on the highest application throughput for all the

timeslots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). Coverage




consists of several independent layers whose visibility in the map window can be managed. There are as many layers as

throughput display thresholds. Each layer shows the highest application throughput that any transmitter can provide on all

available timeslots in the terminal.

Packet Throughput and Quality: Average Max Application Throughput (kbps)

A pixel of the coverage area is coloured if the calculated average application throughput from all the transmitters covering

that pixel exceeds the defined minimum threshold. The pixel colour depends on the average application throughput for all

the timeslots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots).


layers as throughput display thresholds. Each layer shows the average application throughput that all the transmitters can

provide on all available timeslots in the terminal.

Packet Throughput and Quality: User Throughput (kbps)

A pixel of the coverage area is coloured if the calculated user throughput from any transmitter covering that pixel exceeds

the defined minimum threshold. The pixel colour depends on the user throughput for all the timeslots supported by the

selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user throughput is calculated

by applying the throughput reduction factor, determined using the selected dimensioning model, to the application through-

put. Coverage consists of several independent layers whose visibility in the map window can be managed. There are as

many layers as transmitter coverage areas and throughput display thresholds. Each layer shows the user throughput that

a transmitter can provide on all available timeslots in the terminal.

Packet Throughput and Quality: Max User Throughput (kbps)

A pixel of the coverage area is coloured if the calculated highest user throughput from any transmitter covering that pixel

exceeds the defined minimum threshold. The pixel colour depends on the highest user throughput for all the timeslots

supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user through-

put is calculated by applying the throughput reduction factor, determined using the selected dimensioning model, to theapplication throughput. Coverage consists of several independent layers whose visibility in the map window can be

managed. There are as many layers as throughput display thresholds. Each layer shows the highest user throughput that

any transmitter can provide on all available timeslots in the terminal.

Packet Throughput and Quality: Average User Throughput (kbps)

A pixel of the coverage area is coloured if the calculated average user throughput from all the transmitters covering that

pixel exceeds the defined minimum threshold. The pixel colour depends on the average user throughput for all the times-

lots supported by the selected terminal type (Number of Simultaneous Carriers x Number of DL Timeslots). The user

throughput is calculated by applying the throughput reduction factor, determined using the selected dimensioning model,

to the application throughput. Coverage consists of several independent layers whose visibility in the map window can be

managed. There are as many layers as throughput display thresholds. Each layer shows the average user throughput that

all the transmitters can provide on all available timeslots in the terminal.

Packet Throughput and Quality: BLER (%)

A pixel of the coverage area is coloured if the calculated BLER from any transmitter exceeds the defined minimum thresh-

old. The pixel colour depends on the BLER. Coverage consists of several independent layers whose visibility in the map

window can be managed. There are as many layers as transmitter coverage areas and BLER display thresholds. Each

layer shows the BLERs that the covered pixels experience on one timeslot.

Packet Throughput and Quality: Max BLER (%)

A pixel of the coverage area is coloured if the calculated highest BLER from all the transmitters exceeds the defined mini-

mum threshold. The pixel colour depends on the BLER. Coverage consists of several independent layers whose visibility

in the map window can be managed. There are as many layers as BLER display thresholds. Each layer shows the BLER

that the covered pixels experience on one timeslot.

5.4 Codec Mode Selection and CQI Calculations

Atoll supports FR, HR, EFR, and AMR codec modes. A codec configuration contains codec mode adaptation thresholdsand quality graphs for circuit quality indicators. Atoll has the following circuit quality indicators included by default:

• FER or Frame Erasure Rate: The number of frames in error divided by the total number of frames. These frames

are usually discarded, in which case this can be called the Frame Erasure Rate.

• BER or Bit Error Rate: BER is a measurement of the raw bit error rate in reception before the decoding process

begins. Any factor that impacts the decoding performance, such as frequency hopping, will impact the correlation

between BER and FER, or the perceived end-user voice quality.

• MOS or Mean Opinion Score: Voice quality can be quantified using mean opinion score (MOS). MOS values can

only be measured in a test laboratory environment. MOS values range from 1 (bad) to 5 (excellent). Different voice

codecs have slightly different FER to MOS correlation since the smaller the voice codec bit rate is, the more sen-

sitive it becomes to frame erasures.



© Forsk 2010 AT282_TRG_E2 157


The default codec configurations in Atoll include default FER, BER, and MOS quality graphs with respect to the carrier to

interference ratio, and codec mode adaptation thresholds (calculated from the FER vs. C/I graphs for all codec modes at

5 % FER).

Figure 5.1: FER vs. C/I Graphs

Figure 5.2: BER vs. C/I Graphs

Figure 5.3: MOS vs. C/I Graphs




5.4.1 Circuit Quality Indicator Calculations

Circuit quality indicator calculations include codec mode selection and CQI calculation. Codec modes may be selected

using ideal link adaptation or without it. Once codec modes have been selected, CQI corresponding to these codec modes

are determined from the look-up tables.

The following sections describe the two categories of calculations, i.e., with and without ideal link adaptations. Ideal link

adaptation implies that the selected codec mode corresponds to the best value of the reference CQI under the given radio

conditions. Without ideal link adaptation, the codec mode is selected based on the codec adaptation thresholds.

CQI calculations may be based on C/N or on C/(I+N). For calculating the noise, either the noise figure defined for the calcu-

lations or that of the selected terminal type is used.

Different codec configurations may be defined for transmitter and terminals. In this case, Atoll only selects the codec

modes that are common in the two. If no terminal type is defined for the calculation, or if the terminal type does not have

any codec configuration assigned to it, Atoll only uses the codec configuration of the transmitter. Similarly, if a transmitter

does not have any codec configuration assigned to it, Atoll only uses the codec configuration of the terminal type. If both

the transmitter and the terminal type do not have any codec configuration assigned to them, no codec mode selection and

CQI calculation is carried out.

If more than one codec modes satisfy the C/N or C/I conditions, Atoll selects the higher priority codec mode.

In the following calculations, we assume that:

• is the signal level received from the selected TRX type (tt ) or on all the TRXs of Txi on each pixel of

the Txi coverage area,

• CM is the set of all available codec modes,

• are the values of adaptation thresholds for the codec modes available in the codec

configuration,

The computed noise is compared to the codec configuration reference noise . If the values are the same,

the defined graphs are used as is, otherwise the graphs are downshifted by the difference .

When the calculations are based on C/(I+N):

• Atoll calculates the carrier-to-interference ratio for all the TBC transmitters with codec configurations assigned,

but takes into account all the TBC transmitters (with and without codec configurations) to evaluate the interference.

For more information on interference (I ) calculation, see "Carrier-to-Interference Ratio Calculation" on page 144.

Ideal link adaptation for circuit quality indicator studies is defined at the codec configuration level. If the ideal link adaptation

option is checked, Atoll will select the codec mode, for the transmitter under study, according to the codec quality graphs

(CQI = f(C/N) and CQI = f(C/I)) related to the defined reference CQI, which may be different from the CQI being calculated.

Otherwise, Atoll will use the adaptation thresholds defined in the Adaptation Thresholds tab to determine the codec mode

to be used in the studies.

5.4.2 CQI Calculation Without Ideal Link Adaptation

5.4.2.1 Calculations Based on C/N

Atoll selects the highest priority codec mode, cm, from among the codec modes available in the codec configuration:


For , Atoll determines the CQI from the CQI=f(C/N) graph associated to the selected codec mode, cm.


Atoll selects the highest priority codec mode, cm, from among the codec modes available in the codec configuration:

References:The graphs are based on:

[1] T. Halonen, J. Romero, J. Melero; GSM, GPRS and EDGE performance – Evolution towards 3G/UMTS, John Wiley

and Sons Ltd.

[2] J. Wigard, P. Mogensen; A simple mapping from C/I to FER and BER for a GSM type of air interface.

[3] 3GPP Specifications TR 26.975 V6.0.0; Performance characterization of the Adaptive Multi-Rate (AMR) speech

codec (Release 6)

P rec Txi

TR X

Ad ap ta ti on Threshold CM

N N Ref

N N Ref –

cm Highest Priority CM P rec

Txi TRX

N ------------------------------ Adaptation Threshold CM !© ¹

¨ ¸¨ ¸§ ·

=

P rec Txi

TR X

N -----------------------------



© Forsk 2010 AT282_TRG_E2 159



For , Atoll determines the CQI from the CQI=f(C/I) graph associated to the selected codec mode, cm.

5.4.3 CQI Calculation With Ideal Link Adaptation

5.4.3.1 Calculations Based on C/NIdeal link adaptation is used by a codec configuration according to a defined reference CQI (MOS by default).

Atoll calculates signal level received from Txi on each pixel of Txi coverage area and converts it into C/N values as

described earlier. Then, Atoll filters all the codec modes that satisfy the C/N criterion (defined by the CQI = f(C/N) graphs

for the reference CQI) and are common between the transmitter and the terminal type codec configuration.

The selected codec mode among these filtered codec modes will be,

For each TRX type, tt , , for MOS

Or, , for BER and FER

Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI

gives the highest or the lowest value at the received C/N level, .

If more than one codec mode graphs give the same value for reference CQI, then Atoll selects the codec mode with the

highest priority.

From the CQI = f(C/N) graph associated to the selected codec mode cm, Atoll evaluates the CQI for which the study was

performed corresponding to for the selected codec mode.


Ideal link adaptation is used by a codec configuration according to a defined reference CQI (MOS by default).Atoll calculates the C/I level received from the transmitter on each pixel of Txi coverage area, for each TRX and converts

it into C/(I+N). Then, Atoll filters all the codec modes that satisfy the C/(I+N) criteria (defined by the CQI = f(C/I) graphs

for the reference CQI) and are common between the transmitter and the terminal type codec configuration.

The selected codec mode among these filtered codec modes will be,

For each TRX type, tt , , for MOS

Or, , for BER and FER

Where, cm is the codec mode with the highest priority among the set of codec modes CM for which the reference CQI

gives the highest or the lowest value at the received C/(I+N) level, .

If more than one codec mode graphs give the same value for reference CQI, then Atoll selects the codec mode with the

highest priority.

From the CQI = f(C/I) graph associated to the selected codec mode cm (indexed with the C/(I+N) values), Atoll evaluates

the CQI for which the study was performed corresponding to for the selected codec mode.

cm Highest Priority CM P rec

Txi TRX

I N +------------------------------ Adaptation Threshold CM !© ¹

¨ ¸¨ ¸§ ·

=

P rec Tx i

TR X

I N +-----------------------------

cm Highest Priority CM

CQI Ref Highest CQI=f C

N ----=

P rec Txi

TRX

N tot

------------------------------© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

=© ¹¨ ¸¨ ¸¨ ¸§ ·

=


CQI Ref Lowest CQI=f C

N ---- =

P rec Txi

TRX

N tot ------------------------------© ¹¨ ¸

§ ·

© ¹¨ ¸

§ ·=© ¹

¨ ¸¨ ¸¨ ¸§ ·

=

P rec Txi

TR X

N tot

-----------------------------

P rec Txi

TR X

N to t

-----------------------------


CQI Ref Highest CQI=f C

I ----=

P rec Txi

TRX

I N tot +------------------------------

© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

=© ¹¨ ¸¨ ¸¨ ¸§ ·

=


CQI Ref Lowest CQI=f C

I ---- =

P rec Txi

TRX

I N tot +------------------------------

© ¹¨ ¸§ ·

© ¹¨ ¸§ ·

=© ¹¨ ¸¨ ¸¨ ¸§ ·

=

P rec Txi

TR X

I N + to t

-----------------------------

P rec Txi

TR X

I N + tot

-----------------------------




5.4.4 Circuit Quality Indicators Coverage Predictions

The Circuit Quality Indicators coverage predictions show the areas BER, FER, and MOS values in the transmitter coverage

areas.



Coverage prediction parameters to be set are:


• The interference and quality indicator conditions to meet for a pixel to be covered, and







We can distinguish seven cases as below. Let us assume that:


reception threshold.

• Each transmitter, Txi, has a codec configuration assigned.






And









And



• If M = 2 dB, Atoll considers pixels where the received signal level from Txi is either the second highest or withina 2 dB margin from the second highest.



5.4.4.1.4 Best Signal Level per HCS Layer and a Margin


And


BCCH d


BCCH d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t


BCCH d

P re c Tx i

BCCH 2 nd

Best

j i zP rec

Txj BCCH M – t


BCCH d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t



© Forsk 2010 AT282_TRG_E2 161








5.4.4.1.5 Second Best Signal Level per HCS Layer and a Margin


And









And










And









SubcellReceptionThreshold P rec

Tx i BCCH d

P re c Tx i

BCCH 2 nd

Best

j i zP rec

Txj BCCH M – t

SubcellReceptionThreshold P rec Tx i BCCH d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t

P rec Tx i

BCCH

Note:

• In the case two layers have the same priority, the traffic is served by the transmitter for

which the difference between the received signal strength and the HCS threshold is the




BCCH d

P re c Tx i

BCCH Best

j i zP rec

Txj BCCH M – t

P rec Tx i

BCCH











It is possible to display the coverage predictions with colours depending on criteria such as:

BER

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the BER value. Coverage consists


coverage areas and BER display thresholds. Each layer shows the BER in the transmitter coverage area.

FER

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the FER value. Coverage consists


coverage areas and FER display thresholds. Each layer shows the FER in the transmitter coverage area.

MOS

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the MOS value. Coverage consists

of several independent layers whose visibility in the map window can be managed. There are as many layers as transmitter coverage areas and MOS display thresholds. Each layer shows the MOS in the transmitter coverage area.

Max BER

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest BER value among the

BER values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in

the map window can be managed. There are as many layers as BER display thresholds. Each layer shows the BER value.

Max FER

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest FER value among the

FER values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in

the map window can be managed. There are as many layers as FER display thresholds. Each layer shows the FER value.

Max MOS

Only the pixels with a codec mode assigned are coloured. The pixel colour depends on the highest MOS value among the

MOS values for all the transmitters covering the pixel. Coverage consists of several independent layers whose visibility in

the map window can be managed. There are as many layers as MOS display thresholds. Each layer shows the MOS

value.

5.5 Traffic AnalysisWhen starting a traffic analysis, Atoll distributes the traffic from maps to transmitters of each layer according to the

compatibility criteria defined in the transmitter, services, mobility type, terminal type properties. Transmitters considered

in traffic analysis are the active and filtered transmitters that belong to the focus zone.

5.5.1 Traffic Distribution

5.5.1.1 Normal Cells (Nonconcentric, No HCS Layer)

5.5.1.1.1 Circuit Switched Services

A user with a given circuit switched service, c , a terminal, t , and a mobility type, m, will be distributed to the BCCH and

TCH subcells of a transmitter if:

• The terminal, t , works on the frequency band used by the BCCH subcell,

Notes:

• If no focus zone exists in the .atl document, Atoll takes into account the computation zone.

• For details of the average timeslot capacity calculation, see the Network Dimensioning

section (calculation of minimum reduction factor).



© Forsk 2010 AT282_TRG_E2 163


• The terminal , t , works on the frequency band used by the TCH subcell.

5.5.1.1.2 Packet Switched Services

A user with a given packet switched service, p, a terminal, t , and a mobility type, m, will be distributed to the BCCH and

TCH subcells of a transmitter if:

• The transmitter is an GPRS/EDGE station (option specified in the transmitter property dialog),

• The terminal , t , is technologically compatible with the transmitter,

• The terminal , t , works on the frequency band used by the BCCH subcell,

• The terminal , t , works on the frequency band used by the TCH subcell.

5.5.1.2 Concentric CellsIn case of concentric cells, TCH_INNER TRX type has the highest priority to carry traffic.


A user with a given circuit switched service, c , a terminal, t , and a mobility type, m, will be distributed to the TCH_INNER,

BCCH and TCH subcells of a transmitter if:


• The terminal , t , works on the frequency band(s) used by the TCH_INNER and TCH subcells.


A user with a given packet switched service, p, a terminal, t , and a mobility type, m, will be distributed to the TCH_INNER,

BCCH and TCH subcells of a transmitter if:


• The terminal , t , is technologically compatible with the transmitter,

• The terminal , t , works on the frequency band used by the BCCH subcell,• The terminal, t , works on the frequency band(s) used by the TCH_INNER and TCH subcells.

5.5.1.3 HCS Layers

For each HCS layer, k , you may specify the maximum mobile speed supported by the transmitters of the layer.


A user with a given circuit switched service, c , a terminal, t , and a mobility type, m, will be distributed to the BCCH and

TCH subcells (and TCH_INNER in case of concentric cells) of a transmitter if:


• The terminal , t , works on the frequency band(s) used by the TCH_INNER and TCH subcells,

• The user’s mobility, m, is less than the maximum speed supported by the layer, k .


A user with a given packet switched service, p, a terminal, t , and a mobility type, m, will be distributed to the BCCH and

TCH subcells (and TCH_INNER in case of concentric cells) of a transmitter if:


• The terminal, t , is technologically compatible with the transmitter,

• The terminal, t , works on the frequency band used by the BCCH subcell,

• The terminal, t , works on the frequency band(s) used by the TCH_INNER and TCH subcells,

• The user mobili ty, m, is less than the maximum speed supported by the layer, k .

5.5.2 Calculation of the Traffic Demand per Subcell

Here we assume that:

• Users considered for evaluating the traffic demand fulfil the compatibility criteria defined in the transmitter, serv-

ices, mobility, terminal properties as explained above.

• Atoll distributes traffic on subcell service areas, which are determined using the option “Best signal level per HCS

layer” with a 0dB margin and the subcell reception threshold as lower threshold.• Same traffic is distributed to the BCCH and TCH subcells.

5.5.2.1 User Profile Traffic Maps

5.5.2.1.1 Normal Cells (Nonconcentric, No HCS Layer)

Number of subscribers ( ) for each TCH subcell (Txi, TCH), per user profile up with a given mobility m, is inferred as:

Where Sup,m is the TCH service area containing the user profile up with the mobility m and D is the user profile density.

X up m

X up m Txi TCH Sup m Txi TCH Du=




For each behaviour described in the user profile up, Atoll calculates the probability for the user to be connected with a

given service using a terminal t .

Circuit Switched Services

For a circuit switched service c , we have:

Where N call is the number of calls per hour and d is the average call duration (in seconds).

Then, Atoll evaluates the traffic demand, , in Erlangs for the subcell (Txi, TCH) service area.

Packet Switched Services (Max Rate)

For a max rate packet switched service p, we have:

Where N call is the number of calls per hour and V is the transmitted data volume per call (in Kbytes).

Then, Atoll evaluates the traffic demand, , in kbits/s for the subcell (Txi, TCH) service area.

Packet Switched Services (Constant Bit Rate)

For a constant bit packet switched service p, we have:

Where N call is the number of calls per hour and d is the average call duration (in seconds).

Then, Atoll evaluates the traffic demand, , in kbits/s for the subcell (Txi, TCH) service area.

5.5.2.1.2 Concentric Cells

In case of concentric cells, Atoll distributes a part of traffic on the TCH_INNER service area (TCH_INNER is the highest

priority traffic carrier) and the remaining traffic on the outer ring served by the TCH subcell. The traffic spread over the

TCH_INNER subcell may overflow to the TCH subcell. In this case, the traffic demand is the same on the TCH_INNER

subcell but increases on the TCH subcell.

Number of subscribers ( ) for each TCH_INNER (Txi, TCH_INNER) and TCH (Txi, TCH) subcell, per user profile up

with a given mobility m, is inferred as:

and respectively refer to the TCH_INNER and TCH subcell service areas

containing the user profile up with the mobility m. D is the user profile density.

pup c t N call d u

3600 ----------------------=

Dup c t m

Dup c t m Txi TCH X up m Txi TCH pup c t u=

pup p t N call V u 8 u

3600 --------------------------------=

Dup p t m

Dup p t m Txi TCH X up m Txi TCH pup p t u=

pup p t N call d u

3600 ----------------------=

Dup p t m

Dup p t m Txi TCH X up m Txi TCH pup p t u=

Note:

• Traffic overflowing from the TCH_INNER to the TCH is not uniformly spread over the TCH

service area. It is still located on the TCH_INNER service area.

Figure 5.4: Representation of a Concentric Cell TXi

X up m

X up m Txi,TCH_INNER Sup m Txi,TCH_INNER Du=

X up m Txi,TCH Sup m Txi,TCH Sup m Txi,TCH_INNER – > @ Du=

Sup m Txi,TCH_INNER Sup m Txi,TCH



© Forsk 2010 AT282_TRG_E2 165


Circuit Switched Services

For each user of the user profile up using a circuit switched service c with a terminal t , Atoll calculates the probability

( ) of the user being connected. Calculations are detailed in "Circuit Switched Services" on page 162.

Then, Atoll evaluates the traffic demand, , in Erlangs in the (Txi, TCH_INNER ) and (Txi, TCH) subcell service

areas.

Where is the maximum rate of traffic overflow (in %) specified for the TCH_INNER subcell.

Packet Switched Services (Max Rate)

For each user of the user profile up using a max rate packet switched service p with a terminal t , probability of the user

being connected ( ) is calculated as explained in "Packet Switched Services" on page 163.

Atoll evaluates the traffic demand, , in kbits/s in the (Txi, TCH_INNER) and (Txi, TCH) subcell service areas.


Packet Switched Services (Constant Bit Rate)

For each user of the user profile up using a constant bit packet switched service p with a terminal t , probability of the user


Atoll evaluates the traffic demand, , in kbits/s in the (Txi, TCH_INNER) and (Txi, TCH) subcell service areas.


5.5.2.1.3 HCS Layers

We assume two HCS layers: the micro layer has a higher priority than the macro layer. Txi belongs to the micro layer and

Txj to the macro. The traffic contained in the input traffic map can be assigned to all the HCS layers.

Normal Cells

Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option “Best signal level per

HCS layer” meaning that there is an overlap between HCS layers service areas. Let denote this

area (TCH service area of the macro layer overlapped by the TCH service area of the micro layer). Traffic on the overlap-

ping area is distributed to the TCH subcell of the micro layer because it has a higher priority. On this area, traffic of the

micro layer may overflow to the macro layer. In this case, the traffic demand is the same on the TCH subcell of the micro

layer but increases on the TCH subcell of the macro layer.

Atoll evaluates the traffic demand on the micro layer (higher priority) as explained above. For further details, please refer

to formulas for normal cells. Then, it proceeds with the macro layer (lower priority).

pup c t

Dup c t m

Dup c t m Txi,TCH_INNER X up m Txi,TCH_INNER pup c t u=

Dup c t m Txi,TCH X up m Txi,TCH pup c t u Dup c t m Txi,TCH_INNER Omax Txi,TCH_INNER u+=

Omax Txi,TCH_INNER

pup p t

Dup p t m

Dup p t m Txi,TCH_INNER X up m Txi,TCH_INNER pup p t u=

Dup p t m Txi,TCH X up m Txi,TCH pup p t u Dup p t m Txi,TCH_INNER Omax Txi,TCH_INNER u+=

Omax Txi,TCH_INNER

pup p t

Dup p t m

Dup p t m Txi,TCH_INNER X up m Txi,TCH_INNER pup p t u=

Dup p t m Txi,TCH X up m Txi,TCH pup p t u Dup p t m Txi,TCH_INNER Omax Txi,TCH_INNER u+=

Omax Txi,TCH_INNER

Note:

• Traffic overflowing to the macro layer is not uniformly spread over the TCH service area of

Txj. It is only located on the overlapping area.

Figure 5.5: Representation of Micro and Macro Layers

Soverlapping macro

Txj TCH




Number of subscribers ( ) for each TCH subcell (Txj, TCH) of the macro layer, per user profile up with the mobility

m, is inferred as:

Where is the TCH service area of Txj containing the user profile up with the mobility m and D is the

profile density.

For each user described in the user profile up with the circuit switched service c and the terminal t , the probability for the

user being connected ( ) is calculated as explained in "Circuit Switched Services" on page 162.

Then, Atoll evaluates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service area.

For each user described in the user profile up with the packet switched service p and the terminal t , probability for the user

to be connected ( ) is calculated as explained in "Packet Switched Services" on page 163.

Then, Atoll evaluates the traffic demand, , in kbits/s in the subcell (Txj, TCH) service area.

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro

layer) and is the TCH service area of Txi containing the user profile up with the mobility m.

Concentric Cells

Atoll evaluates the traffic demand on the micro layer (higher priority HCS layer) as explained above. For further details,

please refer to formulas given in case of concentric cells. Then, it proceeds with the macro layer (lower priority HCS layer).

The traffic capture is calculated with the option “Best signal level per HCS layer”. It means that there are overlapping areas

between HCS layers where traffic is spread according to the layer priority. On these areas, traffic of the higher priority layer

may overflow.

The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area

overlapped by the TCH service area of the micro layer and another overlapped

by the TCH_INNER service area of the micro layer .

Let us consider three areas, S1, S2 and S3.

Figure 5.6: Concentric Cells

X up mmacro

up mmacro

Txj TCH Sup mmacro

Txj TCH Sup m overlapping – macro

Txj TCH – > @ Du=

Sup mmacro

Txj TCH

pup c t

Dup c t mmacro

Dup c t mmacro

Txj TCH X up mmacro

Txj TCH pup c t u Dup c t mmicro

Txi TCH Sup m overlapping –

macroTxj TCH

Sup mmicro

Txi TCH ---------------------------------------------------------------------------u Omax Txi TCH u+=

pup p t

Dup p t mmacro

Dup p t mmacro

Txj TCH X up mmacro

Txj TCH pup p t u Dup p t mmicro

Txi TCH Sup m overlapping –

macroTxj TCH

Sup mmicro

Txi TCH ---------------------------------------------------------------------------u Omax Txi TCH u+=

Omax Txi TCH

Sup mmicro

Txi TCH

Soverlapping Txi TCH – macro

Txj,TCH_INNER

Soverlappi ng Txi,TCH_INNER – macro

Txj,TCH_INNER

1 Sup mmacro

Txj,TCH_INNER Sup m overlapping – Txi TCH – macro

Txj,TCH_INNER – =

S2 Sup m overlapping – Txi,TCH_INNER – macro

Txj,TCH_INNER =

3 Sup m overlapping – Txi TCH – macro

Txj,TCH_INNER S2 – =



© Forsk 2010 AT282_TRG_E2 167


Where is the TCH_INNER subcell service area of Txj containing the user profile up with the

mobility m. We only consider the overlapping areas containing the user profile up with the mobility m.

On S1, the number of subscribers per user profile up with a given mobility m ( ) is inferred:

Where D is the user profile density.

The traffic spread over the TCH_INNER service area of the micro layer may overflow on the TCH subcell. The traffic over-

flowing to the TCH subcell is located on the TCH_INNER service area. On S2 , the TCH subcell traffic coming from the

TCH_INNER subcell traffic overflow may overflow proportional to R 2 .

The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S3 proportional to R 3.

Where and are the TCH and TCH_INNER service areas of Txi respectively

containing the user profile up with the mobility m.

For each user described in the user profile up with a circuit switched service c and a terminal t , the probability for the user

being connected ( ) is calculated as explained in "Circuit Switched Services" on page 162. Then, Atoll evaluates

the traffic demand, , in Erlangs in the subcell (Txj, TCH_INNER) service area.

For each user described in the user profile up with a packet switched service p and a terminal t , probability for the user to

be connected ( ) is calculated as explained in "Packet Switched Services" on page 163.

Then, Atoll evaluates the traffic demand, , stated in kbits/s in the subcell (Txj, TCH_INNER) service area.

Where and are the maximum rates of traffic overflow (stated in %) specified

for the TCH and TCH_INNER subcells of Txi respectively.

The area of the TCH ring of the macro layer is overlapped by the micro layer. There are two parts: an area overlapped by

the TCH service area of the micro layer and another one by the

TCH_INNER service area of the micro layer .

Let us consider three areas, S’ 1, S’ 2 and S’ 3.

Where and are the TCH and TCH_INNER subcell service areas of Txj

respectively. We only consider the overlapping areas containing the user profile up with the mobility m.

On S’ 1, the number of subscribers per user profile up with a given mobility m ( ) is inferred:

Where D is the user profile density.

Sup mmacro

Txj,TCH_INNER

X up mmacro

X up mmacro

Txj,TCH_INNER S1 Du=

R 2

S2

Sup mmicro

Txi,TCH_INNER -----------------------------------------------------------------=

R 3S3

Sup mmicro

Txi,TCH Sup mmicro

Txi,TCH_INNER – -------------------------------------------------------------------------------------------------------------------=

Sup mmicro

Txi,TCH Sup mmicro

Txi,TCH_INNER

pup c t

Dup c t m

macro

Dup c t mmacro

Txj,TCH_INNER

X up mmacro

Txj,TCH_INNER pup c t u +

R 2 Dup c t mmicro

Txi,TCH_INNER u Oma x Txi,TCH_INNER u Oma x Txi,TCH u +

R 3 X up mmicro

Txi TCH u pup c t u Omax Txi TCH u

=

pup p t

Dup p t mmacro

Dup p t mmacro

Txj,TCH_INNER

X up mmacro

Txj,TCH_INNER pup p t u +

R 2 Dup p t mmicro

Txi,TCH_INNER u Omax Txi,TCH_INNER u Omax Txi,TCH u +

R 3 X up mmicro Txi TCH u pup p t u Omax Txi TCH u

=

Omax Txi TCH Omax Txi,TCH_INNER


Txj,TCH -- TCH_INNER



'1 Sup mmacro

Txj,TCH Sup mmacro

Txj,TCH_INNER Sup m overlapping – Txi TCH – macro

Txj,TCH -- TCH_INNER – – =

S'2 Sup m overlapping – Txi,TCH_INNER – macro

Txj,TCH -- TCH_INNER =

S'3 Sup m overlapping – Txi TCH – macro

Txj,TCH -- TCH_INNER S'2 – =

Sup mmacro

Txj,TCH Sup mmacro

Txj,TCH_INNER

X up mmacro

X up mmacro

Txj,TCH S'1 Du=




The traffic spread over the TCH_INNER service area of the micro layer may overflow on the TCH subcell. The traffic over-

flowing on the TCH subcell is located on the TCH_INNER service area. On S’ 2 , the TCH subcell traffic coming from the

TCH_INNER subcell traffic overflow may overflow proportionally to R’ 2 .

The traffic spread over the ring served by the TCH subcell of the micro layer only may overflow on S’ 3 proportional to R’ 3.

Where and are the TCH and TCH_INNER service areas of Txi respectively

containing the user profile up with the mobility m.

For each user described in the user profile up with a circuit switched service c and a terminal t , the probability for the user

being connected ( ) is calculated as explained in "Circuit Switched Services" on page 162.

Then, Atoll evaluates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service area.

For each user described in the user profile up with a packet switched service p and a terminal t , the probability for the user


Then, Atoll evaluates the traffic demand, , in kbits/s in the subcell (Txj, TCH) service area.

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi (micro

layer), the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txi (macro

layer), the maximum rate of traffic overflow indicated for the TCH_INNER subcell of Txj (macro

layer) and the number of subscribers with the user profile up and mobility m on the TCH service area

of Txi (as explained in "Concentric Cells" on page 163).

5.5.2.2 Sector Traffic Maps

We assume that the traffic map is built from a coverage by transmitter prediction calculated for the TCH subcells with

options:

• “HCS Servers” and no margin if the network only consists of normal cells and concentric cells,

• “Highest Priority HCS Server” and no margin in case of HCS layers.

When creating the traffic map, you have to specify the traffic demand per transmitter and per service (throughput for a max

rate packet switched service and Erlangs for a circuit switched or constant bit rate packet switched service) and the global

distribution of terminals and mobility types.

Let denote the Erlangs for the circuit switched service, c , on the TCH subcell of Txi.

Let denote the throughput of the packet switched service (Max Bit Rate), p, on the TCH subcell of Txi.

Let denote the Erlangs for the packet switched service (Constant Bit Rate), p, on the TCH subcell of Txi.

We assume that 100% of users have the terminal, t , and the mobility type, m.

5.5.2.2.1 Normal Cells (Nonconcentric, No HCS Layer)

For each circuit switched service, c , Atoll evaluates the traffic demand, Dc,t,m, in Erlangs in the subcell (Txi, TCH) service

area.

R '2

S'2

Sup mmicro

Txi,TCH_INNER -----------------------------------------------------------------=

R '3S'3

Sup mmicro

Txi,TCH Sup mmicro

Txi,TCH_INNER – -------------------------------------------------------------------------------------------------------------------=

Sup mmicro

Txi,TCH Sup mmicro

Txi,TCH_INNER

pup c t

Dup c t mmacro

Dup c t mmacro

Txj TCH

X up mmacro

Txj TCH pup c t u +

Dup c t mmacro

Txj,TCH_INNER Omax Txj,TCH_INNER u +

R '2 Dup c t mmicro


R '3 X up mmicro

Txi TCH u pup c t mu Oma x Txi TCH u

=

pup p t

Dup p t mmacro

Dup p t mmacro

Txj TCH

X up mmacro

Txj TCH pup p t u +

Dup p t mmacro


R '2 Dup p t mmicro


R '3 X up mmicro

Txi TCH u pup p t mu Omax Txi TCH u

=

Omax Txi,TCH

Omax Txi,TCH_INNER Omax Txj,TCH_INNER

X up mmicro

Txi TCH

E c Txi TCH

T p Txi TCH

E p Txi TCH

Dc t m Txi TCH E c Txi TCH =



© Forsk 2010 AT282_TRG_E2 169


For each packet switched service (Max Bit Rate), p, Atoll evaluates the traffic demand, D p,t,m, in kbits/s in the subcell (Txi,

TCH) service area.

For each packet switched service (Constant Bit Rate), p, Atoll evaluates the traffic demand, D p,t,m, in kbits/s in the subcell

(Txi, TCH) service area.

where is the guaranteed bit rate of the constant bit rate packet switched service p.

5.5.2.2.2 Concentric CellsIn case of concentric cells, Atoll distributes a part of traffic on the TCH_INNER service area (TCH_INNER is the highest

priority traffic carrier) and the remaining traffic, on the ring served by the TCH subcell only. The traffic spread over the

TCH_INNER subcell may overflow to the TCH subcell. In this case, the traffic demand is the same on the TCH_INNER

subcell and rises on the TCH subcell.

For each circuit switched service, c , Atoll evaluates the traffic demand, Dc,t,m, in Erlangs in the subcell, (Txi, TCH_INNER)

and (Txi, TCH), service areas.

and

For each packet switched service (Max Bit Rate), p, Atoll evaluates the traffic demand, D p,t,m, in kbits/s in the subcell,

(Txi, TCH_INNER) and (Txi, TCH), service areas.

and

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell,

and are the TCH and TCH_INNER service areas of Txi respectively.

For each packet switched service (Constant Bit Rate), p, Atoll evaluates the traffic demand, D p,t,m, in kbits/s in the subcell,

(Txi, TCH_INNER) and (Txi, TCH), service areas.

and

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell,and are the TCH and TCH_INNER service areas of Txi respectively.

5.5.2.2.3 HCS Layers

We assume we have two HCS layers: the micro layer has a higher priority and the macro layer has a lower one. Txi belongs

to the micro layer and Txj to the macro one. The traffic contained in the input traffic map can be assigned to all the HCS

layers.

Normal Cells

Atoll distributes traffic on the TCH service areas. The traffic capture is calculated with the option “HCS Servers”. It means

that there is an overlapping area between HCS layers. Let denote the TCH service area of the

macro layer overlapped by the TCH service area of the micro layer. Traffic on the overlapping area is distributed to the

D p t m Txi TCH T p Txi TCH =

D p t m Txi TCH E p Txi TCH TP p GBR u=

TP p GBR

Note:

• Traffic overflowing from the TCH_INNER to the TCH is not uniformly spread over the TCH

service area. It is only located on the TCH_INNER service area.

Dc t m Txi,TCH_INNER S Txi,TCH_INNER

S Txi TCH ----------------------------------------------------- E c Txi TCH u=

Dc t m Txi,TCH S Txi,TCH S Txi,TCH_INNER –

S Txi TCH ----------------------------------------------------------------------------------------------- E c Txi TCH u +

Dc t m Txi,TCH_INNER Omax Txi,TCH_INNER u

=

D p t m Txi,TCH_INNER S Txi,TCH_INNER

S Txi TCH ----------------------------------------------------- T p Txi TCH u=

D p t m Txi,TCH S Txi,TCH S Txi,TCH_INNER –

S Txi TCH ----------------------------------------------------------------------------------------------- T p Txi TCH u +

D p t m Txi,TCH_INNER Oma x Txi,TCH_INNER u

=

Omax Txi,TCH_INNER

S Txi,TCH S Txi,TCH_INNER

D p t m Txi,TCH_INNER S Txi,TCH_INNER

S Txi TCH ----------------------------------------------------- E p Txi TCH TP p GBR uu=

D p t m Txi,TCH S Txi,TCH S Txi,TCH_INNER –

S Txi TCH ----------------------------------------------------------------------------------------------- E p Txi TCH TP p GBR uu +

D p t m Txi,TCH_INNER Oma x Txi,TCH_INNER u=

Omax Txi,TCH_INNER S Txi,TCH S Txi,TCH_INNER

Soverlapping macro

Txj TCH




TCH subcell of the micro layer (higher priority layer). On this area, traffic of the micro layer may overflow to the macro layer.

In this case, the traffic demand is the same on the TCH subcell of the micro layer but rises on the TCH subcell of the macro

layer.

Atoll starts evaluating the traffic demand on the micro layer (highest priority HCS layer).

For each circuit switched service, c , Atoll calculates the traffic demand, , in Erlangs in the subcell (Txi, TCH) service

area.

For each packet switched service (Max Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the subcell

(Txi, TCH) service area.

For each packet switched service (Constant Bit Rate), p, Atoll calculates the traffic demand, , in kbits/s in the

subcell (Txi, TCH) service area.

Then, Atoll proceeds with the macro layer (lower priority HCS layer).

For each circuit switched service, c , Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj, TCH) service

area.


(Txj, TCH) service area.

Where is the maximum rate of traffic overflow (in %) specified for the TCH subcell of Txi (micro cell) and

the TCH service area of Txi.


subcell (Txj, TCH) service area.

Where is the maximum rate of traffic overflow (in %) specified for the TCH subcell of Txi (micro cell) and

the TCH service area of Txi.

Concentric Cells

Atoll evaluates the traffic demand on the micro layer as explained above in case of concentric cells and then proceeds

with the macro layer (lower priority layer).

The traffic capture is calculated with the option “HCS Servers”. It means that there is overlapping areas between HCS

layers where traffic is spread over according to the layer priority. On these areas, traffic of the higher priority layer may

overflow.

Note:

• Traffic overflowing on the macro layer is not uniformly spread over the TCH service area of

Txj. It is only located on the overlapping area.

Note:

• You can restrict the traffic assignement of each traffic map to a specific HCS layer in the

running options of the traffic capture. If you do so, no overflow occurs between HCS layers

and the only overflow which is considered occurs within concentric cells (See "Concentric

Cells" on page 163).

Dc t m micro

Dc t m micro

Txi TCH E c Txi TCH =

D p t m micro

D p t m micro

Txi TCH T p Txi TCH =

D p t m micro

D p t m micro

Txi TCH E p Txi TCH TP p GBR u=

Dc t m macro

Dc t m macro

Txj TCH E c Txj TCH Dc t m micro

Txi TCH Soverlapping

macroTxj TCH

Smicro

Txi TCH -------------------------------------------------------------u Omax Txi TCH u+=

D p t m macro

D p t m macro

Txj TCH T p Txj TCH D p t m micro


macroTxj TCH

Smicro


Omax Txi TCH

Smicro

Txi TCH

D p t m macro

D p t m macro

Txj TCH E p Txi TCH TP p GBR u D p t m micro


macroTxj TCH

Smicro


Omax Txi TCH

Smicro

Txi TCH



© Forsk 2010 AT282_TRG_E2 171


The TCH_INNER service area of the macro layer is overlapped by the micro layer. This area consists of two parts: an area

overlapped by the TCH service area of the micro layer and another overlapped

by the TCH_INNER service area of the micro layer .

Let us consider three areas, S1, S2 and S3.

Where is the TCH_INNER subcell service area of Txj.

The traffic specified for Txj in the map description ( ) is spread over S1 proportionally to R 1.

is the TCH service area of Txj in the traffic map with the option “Best signal level of the highest priority

layer”.

The traffic spread over the TCH_INNER service area of the micro layer may overflow to the TCH subcell. The traffic over-

flowing to the TCH subcell is located on the TCH_INNER service area. On S2 , the TCH subcell traffic coming from the

TCH_INNER subcell traffic overflow may overflow proportional to R 2 .

The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S3 proportional to R 3.

For each circuit switched service, c , Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj,

TCH_INNER) service area.


(Txj, TCH_INNER) service area.

Figure 5.7: Concentric Cells

Soverlapping Txi TCH –

macroTxj,TCH_INNER


Txj,TCH_INNER

S1 Smacro

Txj,TCH_INNER Soverlapping Txi TCH – macro

Txj,TCH_INNER – =

S2 Soverlappi ng Txi,TCH_INNER – macro

Txj,TCH_INNER =

S3 Soverlapping Txi TCH –

macroTxj,TCH_INNER S2 – =

Smacro

Txj,TCH_INNER

E c Txj TCH

R 1S1

Smap

Txj TCH --------------------------------------------=

Sma p

Txj TCH

R 2

S2

Smicro

Txi,TCH_INNER -----------------------------------------------------------------=

R 3S3

Smicro

Txi,TCH Smicro

Txi,TCH_INNER – -------------------------------------------------------------------------------------------------------------------=

Dc t m macro

Dc t m macro

Txj,TCH_INNER

R 1 E c Txj TCH u +

R 2 Dc t m micro

Txi,TCH_INNER u Omax Txi,TCH_INNER u Omax Txi TCH u +

R 3S

microTxi TCH S

microTxi,TCH_INNER –

Smicro

Txi TCH ---------------------------------------------------------------------------------------------------------------------------u E c Txi TCH u Omax Txi TCH u

=

D p t m macro




Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,

is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi

and is the TCH subcell service area of Txi.


subcell (Txj, TCH_INNER) service area.

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,

is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi

and is the TCH subcell service area of Txi.

The area of the TCH ring of the macro layer is overlapped by the micro layer. There are two parts: an area overlapped by

the TCH service area of the micro layer and another overlapped by the

TCH_INNER service area of the micro layer .

Let us consider three areas, S’ 1, S’ 2 and S’ 3.

Where and are the TCH and TCH_INNER subcell service areas of Txjrespectively.

The traffic specified for Txj in the map description ( ) is spread over S’ 1 proportional to R’ 1.

is the TCH service area of Txj in the traffic map with the option “Best signal level of the highest priority

layer”.

The traffic spread over the TCH_INNER service area of the micro layer may overflow to the TCH subcell. The traffic over-

flowing to the TCH subcell is located on the TCH_INNER service area. On S’ 2 , the TCH subcell traffic coming from the

TCH_INNER subcell traffic overflow may overflow proportional to R’ 2 .

The traffic spread over the ring only served by the TCH subcell of the micro layer may overflow on S’ 3 proportional to R’ 3.

For each circuit switched service, c , Atoll calculates the traffic demand, , in Erlangs in the subcell (Txj, TCH) serv-

ice area.

D p t m macro

Txj,TCH_INNER

R 1 T p Txj TCH u +

R 2 D p t m micro


R 3S

microTxi TCH S


Smicro

Txi TCH ---------------------------------------------------------------------------------------------------------------------------u T p Txi TCH u Omax Txi TCH u

=

Omax Txi TCH

Omax Txi,TCH_INNER

Smicro

Txi TCH

D p t m macro

D p t m macro

Txj,TCH_INNER

R 1 E p Txi TCH TP p GBR uu +

R 2 D p t m micro

Txi,TCH_INNER u Oma x Txi,TCH_INNER u Oma x Txi TCH u +

R 3

Smicro

Txi TCH Smicro

Txi,TCH_INNER –

Smicro

Txi TCH ---------------------------------------------------------------------------------------------------------------------------

uE p Txi TCH TP p GBR u Omax Txi TCH u© ¹

¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

u

=

Omax Txi TCH

Omax Txi,TCH_INNER

Smicro

Txi TCH





'1 Smacro

Txj TCH Smacro

Txj,TCH_INNER – Soverlapping Txi TCH – macro

Txj,TCH -- TCH_INNER – =

S'2 Soverlappi ng Txi,TCH_INNER –

macroTxj,TCH -- TCH_INNER =

S'3 Soverlapping Txi TCH – macro

Txj,TCH -- TCH_INNER S'2 – =

S

macro

Txj TCH S

macro

Txj,TCH_INNER

E c Txj TCH

R '1S'1

Smap

Txj TCH --------------------------------------------=

Sma p

Txj TCH

R '2

S'2

S

micro

Txi,TCH_INNER

-----------------------------------------------------------------=

R '3S'3

Smicro

Txi,TCH Smicro

Txi,TCH_INNER – -------------------------------------------------------------------------------------------------------------------=

Dc t m macro



© Forsk 2010 AT282_TRG_E2 173


For each packet switched service (Max Bit Rate), p,Atoll calculates the traffic demand, , in kbits/s in the subcell

(Txj, TCH) service area.

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell

of Txj, is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,

is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi,

is the TCH subcell service area of Txi and is the TCH_INNER subcell service

area of Txi.

For each packet switched service (Constant Bit Rate), p,Atoll calculates the traffic demand, , in kbits/s in the

subcell (Txj, TCH) service area.

Where is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell

of Txj, is the maximum rate of traffic overflow (stated in %) specified for the TCH subcell of Txi,

is the maximum rate of traffic overflow (stated in %) specified for the TCH_INNER subcell of Txi,

is the TCH subcell service area of Txi and is the TCH_INNER subcell service

area of Txi.

5.6 Network DimensioningAtoll is capable of dimensioning a GSM GPRS EDGE network with a mixture of circuit and package switched services.

This section describes the technical details of Atoll’s dimensioning engine.

5.6.1 Dimensioning Models and Quality Graphs

In Atoll, a dimensioning model is an entity utilized by the dimensioning engine along with other inputs (traffic, limitations,

criteria, etc.) in the process of dimensioning. A dimensioning model defines the QoS KPIs to be taken into account when

dimensioning a network for both circuit and packet switched traffic. The user can define either to use Erlang B or Erlang

C queuing model for circuit switched traffic and can define which KPIs to consider when dimensioning the network for

packet switched traffic. The dimensioning engine will only utilize the quality curves of the KPI selected. The KPIs not

selected are supposed to be either already satisfactory or not relatively important.

Dc t m macro

Txj TCH

R '1 E c u Txj TCH +

Dc t m macro


R '2 Dc t m micro


R '3S

microTxi,TCH S


Smicro

Txi,TCH ------------------------------------------------------------------------------------------------------------------------u E c Txi TCH u Omax Txi TCH u

=

D p t m macro

D p t m macro

Txj TCH

R '1 T pu Txj TCH +

Dc t m macro


R '2 D p t m micro


R '3S

microTxi,TCH S


Smicro

Txi,TCH ------------------------------------------------------------------------------------------------------------------------u T p Txi TCH u Omax Txi TCH u

=

Omax Txj,TCH_INNER

Omax Txi TCH

Omax Txi,TCH_INNER

Smicro

Txi,TCH Smicro

Txi,TCH_INNER

D p t m macro

D p t m macro

Txj TCH

R '1 E p Txi TCH TP p GBR uu +

Dc t m macro


R '2 D p t m micro


R '3

Smicro

Txi,TCH Smicro

Txi,TCH_INNER –

Smicro

Txi,TCH ------------------------------------------------------------------------------------------------------------------------

uE p Txi TCH TP p GBR u Omax Txi TCH u© ¹

¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

u

=

Omax Txj,TCH_INNER

Omax Txi TCH

Omax Txi,TCH_INNER

Smicro

Txi,TCH Smicro

Txi,TCH_INNER




5.6.1.1 Circuit Switched Traffic

The network dimensioning for circuit switched traffic is performed using the universally accepted and adopted Erlang B

and Erlang C formulas. The dimensioning criterion in these formulas is the Grade of Service or the allowed blocking prob-

ability of the circuit switched traffic.

In the Erlang B approach, this Grade of Service is defined as the percentage of incoming circuit switched calls that are

blocked due to lack of resources or timeslots. This formula implies a loss system. The blocked calls are supposed to be

lost and the caller has to reinitiate it.

In the Erlang C approach, the Grade of Service is the percentage of incoming calls that are placed in a waiting queue when

there are no resources available, until some resources or timeslots are liberated. This queuing system has no lost calls.

As the load on the system increases, the average waiting time in the queue also increases.

These formulas and their details are available in many books. For example, Wireless Communications Principles and

Practice by Theodore S. Rappaport, Prentice Hall.

Following the common practice, network dimensioning in Atoll is based on the principle that a voice or GSM call has prior-

ity over data transmission. Therefore, as explained later in the network dimensioning steps, Atoll first performs network

dimensioning according to the circuit switched traffic present in the subcell in order to ensure the higher priority service

availability before performing the same for the packet switched traffic.

5.6.1.2 Packet Switched Traffic

Since packet switched traffic does not occupy an entire timeslot the whole time, it is much more complicated to study than

circuit switched traffic. Packet traffic is intermittent and bursty. Whenever there is packet data to be transferred, a Tempo-

rary Block Flow (TBF) is initiated for transferring these packets. Multiple TBFs can be multiplexed on the same timeslot.

This implies that there can be many packet switched service users that have the same timeslots assigned for packet data

transfer but at different intervals of time.

This multiplexing of a number of packet switched service users over the same timeslots incurs a certain reduction in thethroughput (data transfer rate) for each multiplexed user. This reduction in the throughput is more perceivable when the

system traffic load is high. The following parts describe the three most important Key Performance Indicators in GPRS/

EDGE networks and how they are modelled in Atoll.

5.6.1.2.1 Throughput

Throughput is defined as the amount of data delivered to the Logical Link Control Layer in a given unit of time. Each tempo-

rary block flow (TBF), and hence each user, has an associated measured throughput sample in a given network. Each

network will have a different throughput probability distribution depending on the load and network configuration. Instead

of using the precise probability distributions, it is more practical to compute the average and percentile throughput values.

In GPRS, the resources are shared between the users being served, and consequently, the throughput is reduced as the

number of active users increases. This reduction in user perceived throughput is modelled through a reduction factor. The

throughput experienced by a user accessing a particular service can be calculated as:

User throughput = Number of allocated timeslots x Timeslot capacity x Reduction Factor

Or

User throughput per allocated timeslot = Timeslot capacity x Reduction Factor

Timeslot Capacity

The timeslot capacity is the average throughput per fully utilized timeslot. It represents the average throughput from the

network point of view. It mainly depends on the network’s propagation conditions and criteria in the coverage area of a

transmitter (carrier power, carrier-to-interference distribution, etc.). It is a measure of how much data the network is able

to transfer with 1 data Erlang, or in other words, how efficiently the hardware resources are being utilized by the network.

It may also depend on the RLC protocol efficiency.

Atoll computes the average timeslot capacity during the traffic analysis and is used to determine the minimum throughput

reduction factor. But since this information is displayed in the network dimensioning results (only due to relevance), this

information has been considered as a part of the network dimensioning process in this document.

Timeslot Utilisation

Timeslot utilization takes into account the average number of timeslots that are available for packet switched traffic. It is ameasure of how much the network is loaded with data services. Networks with timeslot utilisation close to 100% are close

to saturation and the end-user performance is likely to be very poor.

In Atoll this parameter is termed as the Load (Traffic load for circuit switched traffic and packet switched traffic load for

packet switched traffic). It is described in more detail in the Network dimensioning steps section.

Reduction Factor

Reduction factor takes into account the user throughput reduction due to timeslot sharing among many users. The figure

below shows how the peak throughput available per timeslot is reduced by interference and sharing.Reduction factor is a

function of the number of timeslots assigned to a user (Nu), number of timeslots available in the system (Ns) and the aver-

age system packet switched traffic load (Lp) (utilization of resources in the system). Data Erlangs or data traffic is given by:

Data Erlangs LP N Su=



© Forsk 2010 AT282_TRG_E2 175


More precisely, the reduction factor is a function of the ratio Ns/Nu (Np). Np models the equivalent timeslots that are avail-

able for the packet switched traffic in the system. For example, a 24-timeslot system with each user assigned 3 timeslots

per connection can be modelled by a single timeslot connection system with 8 timeslots in total.

The formula for reduction factor can be derived following the same hypotheses followed by Erlang in the derivation of the

blocking probability formulas (Erlang B and Erlang C).

Let X be a random variable that measures the reduction factor in a certain system state:

Where n is the instantaneous number of connections in the system. The throughput reduction factor is defined as:

Or,

Here, P(X=n) is the probability function of having n connections in the system. Under the same assumptions as those of

the Erlang formulas, the probability function can be written as:

Hence the reduction factor can finally be written as:

This formula is not directly applicable in any software application due to the summations up to infinity. Atoll uses the follow-

ing version of this formula that is exactly the same formula without the summation overflow problem.

Figure 5.8: Reduction of Throughput per Timeslot

X

0 if n = 0

1 if 0 < n N P d

N P

n------- if n > N P

{

RF X P X n= P X 0 = -----------------------

n 0 =

f

¦{

RF X P X n=

P

i 0 =

f

¦ X i =

-------------------------------

n 0 =

f

¦=

P X n=

LP N P n

n!--------------------------

LP N P i

i !------------------------

i 0 =

N P

¦LP N P i

N P ! N P

i N P –

---------------------------------

i N P 1+=

f

¦+

-----------------------------------------------------------------------------------------------= if 0 n N P d d

P X n=

LP N P n

N P ! N P

i N P –

---------------------------------

LP N P i

i !------------------------

i 0 =

N P

¦LP N P i

N P ! N P

i N P –

---------------------------------

i N P 1+=

f

¦+

-----------------------------------------------------------------------------------------------= if n > N P

RF

LP N P i

i !------------------------

i 1=

N P

¦LP N P i

N P ! N P

i N P –

---------------------------------

i N P 1+=

f

¦N P

i -------

© ¹§ ·+

LP N P i

i !------------------------

i 1=

N P

¦LP N P i

N P ! N P

i N P –

---------------------------------

i N P 1+=

f

¦+

-----------------------------------------------------------------------------------------------------------------=




The default quality curves for the Reduction Factor have been derived using the above formula. Each curve is for a fixed

number of timeslots available for packet switched traffic (Np) describing the reduction factor at different values of packet

switched traffic load (Lp). The figure below contains all the reduction factor quality curves in Atoll. The Maximum reduction

factor can be 1, implying a maximum throughput, and the minimum can be 0, implying a saturated system with no datathroughput.

Each curve in the above figure represents an equivalent number of packet switched timeslots, NP.

5.6.1.2.2 Delay

Delay is the time required for an LLC PDU to be completely transferred from the SGSN to the MS, or vice versa. As the

delay is a function of the delays and the losses incurred at the packet level, the network parameters, such as the packet

queue length, and different protocol properties, such as the size of the LLC PDU, become important. It is also quite

dependent upon the radio access round trip time (RA RTT) and has a considerable impact on the application level perform-

ance viewed by the user.

The delay parameter is a user level parameter rather than being a network level quantity, like throughput per cell, timeslot

capacity, TBF blocking and reduction factor, hence it is difficult to model and is currently under study. Hence, no default

curve is presently available for delay in Atoll.

5.6.1.2.3 Blocking Probability

In GPRS, there is no blocking as in circuit switched connections. If a new temporary block flow (TBF) establishment is

requested and there are already M users per timeslot, M being the maximum limit of multiplexing per timeslot (Multiplexing

factor), the request is queued in the system to be established later when resources become available.

Supposing that M number of users can be multiplexed over a single timeslot (PDCH), we can have a maximum of M * Np

users in the system. This implies that if a new TBF is requested when there are already M * N p users active, it will be

blocked and placed in a queue. So the blocking probability is the probability of having M * Np + 1 users in the system or

more, meaning,

as in this case n is always greater than Np, we have,

Figure 5.9: Reduction Factor for Different Packet Switched Traffic Loads (L p, X-axis)

RF

LP N P n

n!--------------------------

n 1=

N P

¦N P

N P 1+

N P !--------------------- 1 LP – ln

LP n

n------

n 1=

N P

¦+

© ¹¨ ¸¨ ¸§ ·

–

LP N P n

n!--------------------------

n 1=

N P

¦LP N P

N P

N P !-----------------------------

LP

1 LP – ---------------+

-----------------------------------------------------------------------------------------------------------------------------------=

P X n= for n M N P 1+=






5.6.2 Network Dimensioning Process

The network dimensioning process is described below in detail. As the whole dimensioning process is in fact a chain of

small processes that have there respective inputs and outputs, with outputs of a preceding one being the inputs to the

next, the best method is to detail each process individually in form of steps of the global dimensioning process.

5.6.2.1 Network Dimensioning Engine

During the dimensioning process, Atoll first computes the number of timeslots required to accommodate the circuit

switched traffic. Then it calculates the number of timeslots to add in order to satisfy the demand of packet switched traffic.

This is performed using the quality curves entered in the dimensioning model used. If the dimensioning model has been

indicated to take all three KPIs in to account (throughput reduction factor, delay and blocking probability), the number of timeslots to be added is calculated such that:

1. The throughput reduction factor is greater than the minimum throughput reduction factor,

2. Delay is less than the maximum permissible delay defined in the service properties, and

3. The blocking probability is less than the maximum allowable blocking probability defined in the service properties.

The figure below depicts a simplified flowchart of the dimensioning engine in Atoll.

On the whole, following are the inputs and outputs of the network dimensioning process:

5.6.2.1.1 Inputs

• Circuit switched traffic demand

• Packet switched traffic demand

• Timeslot configurations defined for each subcell

• Target traffic overflow rate and Half-rate traffic ratio for each subcell

• Service availability criteria: minimum required throughput per user, maximum permissible delay, maximum allow-

able blocking probability etc.

• Dimensioning model parameters: Maximum number of TRXs per transmitter, dimensioning model for circuit

switched traffic, number of minimum dedicated packet switched timeslots per transmitter, maximum number of

TRXs added for packet switched services, KPIs to consider, and their quality curves.

5.6.2.1.2 Outputs

• Number of required TRXs per transmitter

• Number of required shared, circuit switched and packet switched timeslots

• Traffic load

• Served circuit switched traffic

• Served packet switched traffic

• Effective rate of traffic overflow

• Actual KPI values: throughput reduction factor, delay and blocking probability

5.6.2.2 Network Dimensioning Steps

This section describes the entire process step by step as it is actually performed in Atoll. Details of the calculations of the

parameters that are calculated during each step are described as well.

5.6.2.2.1 Step 1: Timeslots Required for CS Traffic

Atoll computes the number of timeslots required to accommodate the circuit switched traffic assigned to each subcell.

Atoll takes the circuit switched traffic demand (Erlangs) either user-defined or calculated in the traffic analysis and

assigned to the current subcell and the maximum blocking probability defined for the circuit switched service, and

computes the required number of timeslots to satisfy this demand using the Erlang B or Erlang C formula (as defined by

the user).

If the user-defined target rate of traffic overflow per subcell, OTarget, is greater than the maximum blocking rate defined in

the services properties, it is going to be taken as the Grade of Service required for that subcell instead of the maximum

blocking rate of the service.

For the blocking probability GoS and circuit switched traffic demand TDC, Atoll determines the required number of times-

lots TSreq. C for each subcell using formulas described below. In fact, Atoll searches for TSreq. C value until the defined

grade of service is reached.

Figure 5.11: Network Dimensioning Process



© Forsk 2010 AT282_TRG_E2 179


For Erlang B, we have:

For Erlang C, we have:

Atoll considers the effect of half-rate circuit switched traffic by taking into account a user-defined percentage of half-rate

traffic. Atoll computes the effective equivalent number of full-rate timeslots that will be required to carry the total traffic with

the defined percentage of half-rate traffic.

If the number of timeslots required to accommodate the full-rate circuit switched traffic is TSreq. FR, and the percentage of

half-rate traffic within the subcell is defined by HR, then the effective number of equivalent full-rate circuit switched times-

lots TSeff. that can carry this traffic mix is calculated by:

Atoll employs this simplified approach to integrating half-rate circuit switched traffic, which provides approximately the

same results as obtained by using the half-rate traffic charts.

5.6.2.2.2 Step 2: TRXs Required for CS Traffic and Dedicated PS Timeslots

This stage of the network dimensioning process computes the number of TRXs required to carry the circuit switched traffic

demand through the number of required timeslots calculated above and the timeslot configuration defined by the user in

the network settings. Atoll distributes the number of required circuit switched timeslots calculated in Step 1 taking into

account the presence of dedicated packet switched timeslots in each TRX according to the timeslot configurations.

If a timeslot configuration defines a certain number of dedicated packet switched timeslots pre-allocated in certain TRXs,

those timeslots will not be considered capable of carrying circuit switched traffic and hence will not be allocated. For exam-

ple, if 4 timeslots have been marked as packet switched timeslots in the first TRX and Atoll computes 8 timeslots for carry-

ing a certain circuit switched traffic demand, then the number of TRXs to be allocated cannot be 1 even if there is no packet

switched traffic considered yet.

The total numbers of timeslots that carry circuit switched and packet switched traffic respectively are the sums of respec-

tive dedicated and shared timeslots:

and

5.6.2.2.3 Step 3: Effective CS Blocking, Effective CS Traffic Overflow and Served CS Traffic

In this step, the previously calculated number of required TRXs is used to compute the effective blocking rate for the circuit

switched traffic. This is performed by using the Erlang B or Erlang C formula with the circuit switched traffic demand and

the number of required TRXs as inputs and computing the Grade of Service (or blocking probability). It then calculates the

effective traffic overflow rate, Oeff..

In case of Erlang B formula, the effective rate of traffic overflow for the circuit switched traffic is the same as the circuit

switched blocking rate. While in case of the Erlang C model, the circuit switched traffic is supposed to be placed in an

infinite-length waiting queue. This implies that there is no overflow in this case.

From this data, it also computes the served circuit switched traffic. This is the difference of the circuit switched traffic

demand and the percentage of traffic that overflows from the subcell to other subcells calculated above. Hence, for an

effective traffic overflow rate of Oeff. and the circuit switched traffic demand of TDC, the served circuit switched traffic STC

is computed as:

5.6.2.2.4 Step 4: TRXs to Add for PS Traffic

This step is the core of the dimensioning process for packet switched services. First of all, Atoll computes the number of

TRXs to be added to carry the packet switched traffic demand. This is the number of TRXs that contain dedicated packet

switched and shared timeslots.

To determine this number of TRXs, Atoll calculates the equivalent average packet switched traffic demand in timeslots by

studying each pixel covered by the transmitter. This calculation is in fact performed in the traffic analysis process or is user-

defined in the subcells table. Knowing the traffic demand per pixel of the covered area in terms of kbps and the maximum

attainable throughput per pixel (according to the C and/or C/I conditions and the coding scheme curves in the GPRS/EDGE

configuration), Atoll calculates the average traffic demand in packet switched timeslots by:

Go S

TDC TSreqC

TS reqC !--------------------------------

TDC k

k !-------------------

k 0 =

TS reqC

¦

-----------------------------------=

Go S

TD C TS reqC

TDC TS reqC

TS reqC ! 1TDC

TS reqC

------------------- – © ¹§ · TDC k

k !-------------------

k 0 =

TS reqC 1 –

¦ +

------------------------------------------------------------------------------------------------------------------------------------------------------=

TS eff TS reqFR 1HR

2 --------- – © ¹

§ ·u=

TSP TSS TSP dedicated += TSC TSS TSC dedicated +=

ST C TDC 1 Oeff – =




The average timeslot capacity of a transmitter is calculated by dividing the packet switched traffic demand over the entire

coverage area (in kbps) by the packet switched traffic demand in timeslots calculated above.

With the number of timeslots required to serve the circuit switched traffic, the timeslots required for packet switched traffic

and their respective distributions according to the timeslot configurations being known, Atoll calculates the number of

timeslots available for carrying the packet switched traffic demand. These timeslots can be dedicated packet switched

timeslots and the shared ones. So, following the principle that shared timeslots are potential carriers of both traffic types,

The packet switched traffic load is calculated by the formula:

The second important parameter for the calculation of Reduction Factor, Delay and Blocking Probability is the equivalent

number of available timeslots for packet switched traffic, i.e. NP. This is computed by dividing the total number of timeslots

available for carrying packet switched traffic by the number of downlink timeslots defined in the mobile terminal properties.

So, NP is calculated at this stage as:

Where, TSTerminal is the number of timeslots that a terminal will use in packet switched calls.

The number timeslots that a terminal can use in packet switched calls is the product of the number of available DL timeslots

for packet-switched services (on a frame) and the number of simultaneous carriers (in case of EDGE evolution).

The number of timeslots that a terminal will use in packet switched calls is determined by taking the lower of the maximum

number of timeslots for packet switched service defined in the service properties and the maximum number of timeslots

that a mobile terminal can use for packet switched services (see above).

and

Here, the min(X,Y) function yields the lower value among X and Y as result.

Now, knowing the packet switched traffic load, LP, and the equivalent number of available timeslots, NP, Atoll finds out

the KPIs that have been selected before launching the dimensioning process using the quality curves stored in the dimen-

sioning model.

This particular part of this step can be iterative if the KPIs to consider in dimensioning are not satisfied in the first try. If the

KPIs calculated above are within acceptable limits as defined by the user, it means that the dimensioning process hasacceptable results. If these KPIs are not satisfied, then Atoll increases the number of TRXs calculated for carrying packet

switched traffic by 1 (each increment adding 8 more timeslots for carrying packet switched traffic as the least unit that can

be physically added or removed is a TRX) and resumes the computations from Step 3. It then recalculates the packet

switched traffic load, LP, and the equivalent number of available timeslots, NP. Then it recomputes the KPIs with these

new values of LP and NP. If the KPIs are within satisfactory limits the results are considered to be acceptable. Otherwise,

Atoll performs another iteration to find the best possible results.

The calculated values of all the KPIs are compared with the ones defined in the service properties. The values for maxi-

mum Delay and Blocking probability are defined directly in the properties but the minimum throughput reduction factor is

calculated by Atoll using the user’s inputs: minimum throughput per user and required availability. This calculation is in

fact performed during the traffic analysis process, but since it is relevant to the d imensioning procedure, it is displayed in

a column in the dimensioning results so that the user can easily compare the minimum requirement on the reduction factor

KPI with the resulting one. If dimensioning is not based on a traffic analysis, the minimum throughput reduction factor is a

user-defined parameter.

Minimum Throughput Reduction Factor Calculation

The minimum throughput reduction factor is computed using the input data: minimum required throughput per user defined

in the service properties, the average throughput per timeslot deduced from the throughput curves stored in the GPRS/

EDGE configuration properties for each coding scheme, the total number of downlink timeslots defined in the properties

of the mobile terminal (See defintion above) and the required availability defined in the service prop-

erties.

It is at the stage of calculating the average timeslot capacity per transmitter that Atoll studies each covered pixel for carrier

power or carrier-to-interference ratio. According to the measured carrier power or carrier-to-interference ratio, Atoll

deduces the maximum throughput available on that pixel through the throughput vs. C or throughput vs. C/I curves of the

GPRS/EDGE configuration.

The throughput per timeslot per pixel TPTS, Pixel can be either a function of carrier power C, or carrier power C and the

carrier-to-interference ratio C/I, depending on the user-defined traffic analysis RF conditions criteria. Therefore,

TDP Timeslots

Traffic demand per pixel (kbps)

Throughput per pixel (kbps)----------------------------------------------------------------------------------

pi xel

¦=

TSP TSS TSP dedicated +=

TSC TSS TSC dedicated +=

LP

ST C TS C dedicated – TD P Timeslots+

TS P

-----------------------------------------------------------------------------------------------=

N P

TS P

TS Terminal

----------------------------=

TSTerminal min TSMax Service TSMax TerminalType =

TSMax TerminalType TS DL TerminalType CarriersDL TerminalTypeu=

TSMax TerminalType



© Forsk 2010 AT282_TRG_E2 181


Or

and

The required availability parameter defines the percentage of pixels within the coverage area of the transmitter that must

satisfy the minimum throughput condition. This parameter renders user-manageable flexibility to the throughput require-

ment constraint.

To calculate the minimum throughput reduction factor for the transmitter, Atoll computes the minimum throughput reduc-

tion factor for each pixel using the formula:

Once the minimum reduction factor for each pixel is known, Atoll calculates the global minimum reduction factor that is

satisfied by the percentage of covered pixels defined in the required availability. The following example may help in under-

standing the concept and calculation method.

Example: Let the total number of pixels, covered by a subcell S, be 1050. The reliability level set to 90%. This implies that

the required minimum throughput for the given service will be available at 90% of the pixels covered. This, in turn, implies

that there will be a certain limit on the reduction factor, i.e. if the actual reduction factor in that subcell becomes less than

a minimum required, the service will not be satisfactory.

Atoll computes the minimum reduction factor at each pixel using the formula mentioned above, and outputs the following

results:

So for a reliability level of 90%, the corresponding RFmin will be the one provided at least 90% of the pixels covered, i.e.

945 pixels. The corresponding value of the resulting RFmin in this example hence turns out to be 0.9, since this value

covers 962 pixels in total. Only 87 of the covered pixels imply an RFmin of 0.98. These will be the pixels that do not provide

satisfactory service.

This calculation is performed for each service type available in the subcell coverage area. The final minimum throughput

reduction factor is the highest one amongst all calculated for each service separately.

The minimum throughput reduction factor RFmin value is a minimum requirement that must be fulfilled by the network

dimensioning process when the Reduction Factor KPI is selected in the dimensioning model.

5.6.2.2.5 Step 5: Served PS Traffic

Atoll calculates the served packet switched traffic using the number of timeslots available to carry the packet switched

traffic demand. As the result of the above iterative step, Atoll always finds the best possible answer in terms of number of

timeslots required to carry the packet switched traffic demand unless the requirement exceeds the maximum limit on the

RFmin Number of pixels

0.3 189

0.36 57

0.5 20

0.6 200

0.72 473

0.9 23

0.98 87

Figure 5.12: Minimum Throughput Reduction Factor

TP TS Pixel f C =

TP TS Pixel f C = TP TS Pixel f C

i ----

© ¹§ ·=

RF min Pixel TP user min

TP TS Pixel TSTerminal u--------------------------------------------------------------=




number of the packet switched traffic timeslots defined in the dimensioning model properties. Hence, there is no packet

traffic overflow unless the packet switched traffic demand requires more TRXs than the maximum allowed

5.6.2.2.6 Step 6: Total Traffic Load

This step calculates the final result of the dimensioning process, i.e. the total traffic load. The total traffic load L is calculated

as:

Where,

• STC is the served circuit switched traffic• STP is the served packet switched traffic

• TSC, dedicated is the number of dedicated circuit switched timeslots

• TSP, dedicated is the number of dedicated packet switched timeslots

• TSS is the number of shared timeslots

5.7 Key Performance Indicators CalculationThis feature calculates the current values for all circuit switched and packet switched Key Performance Indicators as a

measure of the current performance of the network. It can be used to evaluate an already dimensioned network in which

recent traffic changes have been made in limited regions to infer the possible problematic areas and then to improve the

network dimensioning with respect to these changes.

The concept of this computation is the inverse of that of the dimensioning process. In this case, Atoll has the results of

the dimensioning process already committed and known. Atoll then computes the current values for all the KPIs knowing

the number of required TRXs, the respective numbers of shared and dedicated timeslots and the circuit switched and

packet switched traffic demands.

The computation algorithm utilizes the parameters set in the dimensioning model properties and the quality curves for the

throughput reduction factor, delay and the blocking probability.

The following conventional relations apply:

If,

• TSC, dedicated is the number of timeslots dedicated to the circuit switched traffic,

• TSP, dedicated is the number of timeslots dedicated to the packet switched traffic,

• TSS is the number of shared timeslots for a transmitter,

Then, the number of timeslots available for the circuit switched traffic, TSC, is defined as:

And the number of timeslots available for the packet switched traffic, TSP, is given by:

5.7.1 Circuit Switched Traffic

For each subcell, Atoll has already calculated the effective traffic overflow rate and the blocking rate during the dimen-

sioning process. Also knowing the circuit switched traffic demand, TDC, and the number of timeslots available for circuit

switched traffic, TSC, the blocking probability can be easily computed using the Erlang formulas or tables.

5.7.1.1 Erlang B

Under the current conditions of circuit switched traffic demand, TDC, and the number of timeslots available for the circuit

switched traffic, TSC, the percentage of blocked circuit switched traffic can be computed through:

In a network dimensioning based on Erlang B model, the circuit switched traffic overflow rate, OC, is the same as the

percentage of traffic blocked by the subcell calculated above.

5.7.1.2 Erlang C

Similarly, under the current conditions of circuit switched traffic demand, TDC, and the number of timeslots available for

the circuit switched traffic, TSC, the percentage of delayed circuit switched traffic can be computed through:

LST C ST P +

TS C dedicated TSP dedicated TSS+ +------------------------------------------------------------------------------------------------=

TSC TSS TSC dedicated +=

TSP TSS TSP dedicated +=

% of blocked traffic

TDC

TSC

TSC !--------------------------

TDC k

k !-------------------

k 0 =

TSC

¦

------------------------------=



© Forsk 2010 AT282_TRG_E2 183


If the circuit switched traffic demand, TDC, is higher than the number of timeslots available to accommodate circuit

switched traffic, the column for this result will be empty signifying that there is a percentage of circuit switched traffic actu-

ally being rejected rather than just being delayed under the principle of Erlang C model.

The circuit switched traffic overflow rate, OC, will be 0 if the circuit switched traffic demand, TDC, is less than the number

of timeslots available for the circuit switched traffic, TSC.

If, on the other hand, the circuit switched traffic demand, TDC, is higher than the number of timeslots available to carry the

circuit switched traffic, TSC, then there will be a certain percentage of circuit switched traffic that will overflow from the

subcell. This circuit switched traffic overflow rate, OC, is calculated as:

5.7.1.3 Served Circuit Switched Traffic

The result of the above two processes will be a traffic overflow rate for the circuit switched traffic for each subcell, OC. The

served circuit switched traffic, STC, is calculated as:

5.7.2 Packet Switched TrafficIdentifying the total traffic demand, TDT, (circuit switched traffic demand + packet switched traffic demand) as:

The following two cases can be considered.

5.7.2.1 Case 1: Total Traffic Demand > Dedicated + Shared Timeslots

In the case where the total number of timeslots available is less than the total traffic demand, there will be packet switched

data traffic that will be rejected by the subcell as it will not be able to accommodate it. The following results are expected

in this case:

5.7.2.1.1 Traffic Load

The traffic load will be 100%, as the subcell will have more traffic to carry than it can. This implies that the system will be

loaded to the maximum and even saturated. Hence the user level quality of service is bound to be very unsatisfactory.

5.7.2.1.2 Packet Switched Traffic Overflow

In a 100% loaded, or even saturated subcell, the packet switched data calls will start being rejected because of shortage

of available resources. Hence there will be a perceptible packet switched traffic overflow in this subcell, OP. This overflow

rate is calculated as show below:

5.7.2.1.3 Throughput Reduction Factor

The resulting throughput reduction factor for a 100% loaded or saturated subcell will be 0. Hence, the throughput perceived

by the packet switched service user will be 0, implying a very bad quality of service.

5.7.2.1.4 Delay

Again for a 100% loaded or saturated subcell, the delay at the packet switched service user end will be infinite as there is

no data transfer (throughput = 0).


All the data packets will be rejected by the system since it is saturated and has no free resources to allocate to incoming

data packets. Hence, the blocking probability will be 100%.

5.7.2.1.6 Served Packet Switched Traffic

With the packet switched data traffic overflowing from the subcell, there wi ll be a part of that traffic that is not served. The

served packet switched data traffic, STP, is calculated on the same principle as the served circuit switched traffic:

% of traffic delayed TDC

TSC

TD C TSC

TS C ! 1TDC

TSC

----------- – © ¹§ · TDC

k

k !-------------------

k 0 =

TSC 1 –

¦ +

--------------------------------------------------------------------------------------------------------------------------=

OC

TD C TSC –

TDC

-----------------------------=

ST C TDC 1 OC – =

TDT TDC TDP +=

OP 1TSC dedicated TS P dedicated TSS+ + ST C – ^ `

TDP

----------------------------------------------------------------------------------------------------------------------------- – 100 u=

ST P TDP 1 OP – =




5.7.2.2 Case 2: Total Traffic Demand < Dedicated + Shared Timeslots

In the case where the total traffic demand is less than the number of timeslots available to carry the traffic, the subcell will

not be saturated and there will be some deducible values for all the data KPIs. In a normally loaded subcell, the packet

switched data traffic will have no overflow percentage. This is due to the fact that the packet switched data traffic is rather

placed in a waiting queue than be rejected.

Therefore, there will be a within limits packet switched traffic load, LP, calculated as under:

The second parameter for computing the KPIs from the quality curves of the dimensioning model is the number of equiv-alent timeslots available for the packet switched data traffic, NP, which is calculated in the same manner as in the dimen-

sioning process as well:

These parameters calculated, now Atoll can compute the required KPIs through their respective quality curves.

5.7.2.2.1 Traffic Load

The traffic load is computed knowing the total traffic demand and the total number of timeslots available to carry the entire

traffic demand:

5.7.2.2.2 Packet Switched Traffic OverflowIn a normally loaded subcell, no packet switched data calls will be rejected. The packet switched traffic overflow will, there-

fore, be 0.

5.7.2.2.3 Throughput Reduction Factor

The resulting throughput reduction factor for a normally loaded subcell is calculated through the throughput reduction

factor quality curve for given packet switched traffic load, LP, and number of equivalent timeslots, NP.

5.7.2.2.4 Delay

The resulting delay the subcell is calculated through the delay quality curve for given packet switched traffic load, LP, and

number of equivalent timeslots, NP.


The resulting blocking probability for a normally loaded subcell is calculated through the blocking probability quality curve

for given packet switched traffic load, LP, and number of equivalent timeslots, NP.

5.7.2.2.6 Served Packet Switched Traffic

As there is no overflow of the packet switched traffic demand from the subcell under consideration, the served packet

switched traffic will be the same as the packet switched traffic demand:

5.8 Neighbour AllocationThe intra-technology neighbour allocation algorithm takes into account all the TBC transmitters. It means that all the TBC

transmitters of the .atl document are potential neighbours.

The transmitters to be allocated will be called TBA transmitters. They must fulfil the following conditions:

• They are act ive,

• They satisfy the filter criteria applied to the Transmitters folder,

• They are located inside the focus zone,

• They belong to the folder on which allocation has been executed. This folder can be either the Transmitters folder

or a group of transmitters or a single transmitter.

Only TBA transmitters may be assigned neighbours.

LP

ST C TS C dedicated – TD P Timeslots+

TS P

-----------------------------------------------------------------------------------------------=

N P

TS P

TS Terminal

----------------------------=

Traffic Load TDT

TSC dedicated TSP dedicated TSS+ +------------------------------------------------------------------------------------------------=

ST P TD P =

Note:




© Forsk 2010 AT282_TRG_E2 185


5.8.1 Global Allocation for All Transmitters

We assume a reference transmitter A and a candidate neighbour, transmitter B.

When automatic allocation starts, Atoll checks following conditions:

1. The distance between both transmitters must be less than the user-definable maximum inter-site distance. If the

distance between the reference transmitter and the candidate neighbour is greater than this value, then the

candidate neighbour is discarded.

2. The calculation options,

Force co-site transmitters as neighbours: This option enables you to force transmitters located on the reference transmitter

site in the candidate neighbour list. This constraints can be weighted among the others and ranks the neighbours through

the importance field (see after).

Force adjacent transmitters as neighbours: This option enables you to force transmitters geographically adjacent to the

reference transmitter in the candidate neighbour list. This constraints can be weighted among the others and ranks the

neighbours through the importance field (see after).

Force neighbour symmetry: This option enables user to force the reciprocity of a neighbourhood link. Therefore, if the refer-

ence transmitter is a candidate neighbour of another transmitter, the later will be considered as candidate neighbour of the

reference transmitter.

Force exceptional pairs: This option enables you to force/forbid some neighbourhood relationships. Therefore, you may

force/forbid a transmitter to be candidate neighbour of the reference transmitter.

Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current neighboursand carries out a new neighbour allocation. If not selected, the existing neighbours are kept.

3. There must be an overlapping zone ( ) with a given cell edge coverage probabil ity where:

• S A is the area where the received signal level from the transmitter A is greater than a minimum signal level. S A is

the coverage area of reference transmitter A restricted between two boundaries; the first boundary represents the

start of the handover area (best server area of A plus the handover margin named “handover start”) and the

second boundary shows the end of the handover area (best server area of A plus the margin called “handover

end”)

• SB is the coverage area where the candidate transmitter B is the best server.

Notes:

• Adjacence criterion: Geographically adjacent transmitters are determined on the basis of

their Best Server coverages in 2G (GSM GPRS EDGE) projects. More precisely, a

transmitter TXi is considered adjacent to another transmitter TXj if there exists at least one

pixel of TXi Best Server coverage area where TXj is the 2nd Best Server. The ranking of

the adjacent neighbour transmitter increases with the number of these pixels. The figure

below shows the above concept.

• When this option is checked, adjacent cells are sorted and listed from the most adjacent to

the least, depending on the above criterion. Adjacence is relative to the number of pixels

satisfying the criterion.

• This criteria is only applicable to transmitters belonging to the same HCS layer. The

geographic adjacency criteria is not the same in 3G (UMTS HSPA, CDMA2000) projects.

S A SB




Atoll calculates either the percentage of covered area ( ) if the option “Take into account Covered Area” is

selected, or the percentage of traffic covered on the overlapping area for the option “Take into account Covered

Traffic”. Then, it compares this value to the % minimum covered area (minimum percentage of covered area for the option

“Take into account Covered Area” or minimum percentage of covered traffic for the option “Take into account Covered

Traffic”). If this percentage is not exceeded, the candidate neighbour B is discarded.

The coverage condition can be weighted among the others and ranks the neighbours through the importance field (see

number 4 below).

4. The importance values are used by the allocation algorithm to rank the neighbours according to the allocation

reason, and to quantify the neighbour importance.

Atoll lists all neighbours and sorts them by importance value so as to eliminate some of them from the neighbour list i f the

maximum number of neighbours to be allocated to each transmitter is exceeded. If we consider the case for which thereare 15 candidate neighbours and the maximum number of neighbours to be allocated to the reference transmitter is 8.

Among these 15 candidate neighbours, only 8 (having the highest importances) will be allocated to the reference trans-

mitter.

As indicated in the table below, the neighbour importance depends on the neighbourhood cause; this value goes from 0

to 100%.

Except forced neighbour case (importance = 100%), priority assigned to each neighbourhood cause is now linked to the

(IF) Importance Function evaluation. The importance is evaluated through a function (IF), taking into account the following

3 factors:

• Co-site factor (C) which is a Boolean factor,

• Adjacency factor (A) which deals with the percentage of adjacency,

• Overlapping factor (O) meaning the percentage of overlapping

Figure 5.13: Overlapping Zones

Neighbourhood cause WhenImportance

value

Existing neighbour Only if the Delete existing neighbours option is not selected

and in case of a new allocation

Existing

importance

Exceptional pair Only if the Force exceptional pairs option is selected 100 %

Co-site transmitter Only if the Force co-site transmitters as neighbours option is

selected(IF) function

Adjacent transmitter Only if the Force adjacent transmitters as neighbours option isselected

(IF) function

Neighbourhood relationship that fulfils

coverage conditionsOnly if the % minimum covered area is exceeded (IF) function

Symmetric neighbourhood

relationshipOnly if the Force neighbour symmetry option is selected (IF) function

S A SB

S A

---------------------- 100 u

S A SB



© Forsk 2010 AT282_TRG_E2 187


The (IF) function is user-definable using the Min importance and Max importance fields.

The (IF) function is evaluated as follows:

Where Delta(x) = Max(x) - Min(x)

In the Results part, Atoll provides the list of neighbours, the number of neighbours and the maximum number of neigh-

bours allowed for each cell. In addition, it indicates the importance (in %) of each neighbour and the allocation reason.

Therefore, a neighbour may be marked as exceptional pair, co-site, adjacent, coverage or symmetric. For neighbours

accepted for co-site, adjacency and coverage reasons, Atoll displays the percentage of area meeting the coverage condi-

tions (or the percentage of covered traffic on this area) and the corresponding surface area (km2) (or the traffic covered

on the area in Erlangs), the percentage of area meeting the adjacency conditions and the corresponding surface area

(km

2

). Finally, if cells have previous allocations in the list, neighbours are marked as existing.

Factor Min

importanceDefault value

Max

importanceDefault value

Overlapping factor (O) Min(O) 1% Max(O) 30%

Adjacency factor (A) Min(A) 30% Max(A) 60%

Co-site factor (C) Min(C) 60% Max(C) 100%

Neighbourhood cause

(IF) function(IF) function with default Min

and Max default valuesCo-site Adjacent

no no Min(O) + Delta(O)(O) 1% + 29%(O)

no yesMin(A)+Delta(A){Max(O)(O)+(100%-

Max(O))(A)}30% + 30%{30%(O) + 70%(A)}

yes yesMin(C)+Delta(C){Max(O)(O)+(100%-

Max(O))(A)}60% + 40%{30%(O )+ 70%(A)}

Notes:

• If there is no overlapping between the range of each factor, the neighbours will be ranked

by neighbourhood cause. Using the default values for minimum and maximum importance

fields, neighbours will be ranked in this order: first co-site neighbours, then adjacent

neighbours, and finally neighbours found on overlapping criterion.

• If ranges of (IF) factors overlap each other, the neighbours may not be ranked by

neighbourhood cause.

• The ranking between neighbours from the same category will depend on (A) and (O)

factors.

• The default value of Min(O)= 1%, ensures that neighbours selected for symmetry will have

an importance greater than 0%. With a value of Min(O)= 0%, neighbours selected for

symmetry, will have an importance field greater than 0% only if there is some overlapping.

Notes:

• No coverage prediction is needed to perform an automatic neighbour allocation. When

starting an automatic neighbour allocation, Atoll automatically calculates the path loss

matrices if not found.

• Atoll uses traffic map(s) selected in the default traffic analysis in order to determine the

percentage of traffic covered in the overlapping area.

• When the option “Force adjacent transmitters as neighbours” is used, the margin

“handover start” is not taken into account. Atoll considers a fixed value of 0 dB.

• A forbidden neighbour must not be listed as neighbour except if the neighbourhood

relationship already exists and the Delete existing neighbours option is unchecked when

you start the new allocation. In this case, Atoll displays a warning in the Event viewer

indicating that the constraint on the forbidden neighbour will be ignored by algorithm

because the neighbour already exists.

• The force neighbour symmetry option enables the users to consider the reciprocity of aneighbourhood link. This reciprocity is allowed only if the neighbour list is not already full.

Thus, if transmitter B is a neighbour of the transmitter A while transmitter A is not a

neighbour of the transmitter B, two cases are possible:

1st case: There is space in the transmitter B neighbour list: the transmitter A will be added

to the list. It will be the last one.

2nd case: The transmitter B neighbour list is full: Atoll will not include transmitter A in the

list and will cancel the link by deleting transmitter B from the transmitter A neighbour list.

• When the options “Force exceptional pairs” and “Force symmetry” are selected, Atoll

considers the constraints between exceptional pairs in both directions so as to respect

symmetry condition. On the other hand, if neighbourhood relationship is forced in one

direction and forbidden in the other one, symmetry cannot be respected. In this case, Atoll

displays a warning in the Event viewer.




5.8.2 Allocation for a Group of Transmitters or One Transmitter

In this case, Atoll allocates neighbours to:

• TBA transmitters,

• Neighbours of TBA transmitters marked as exceptional pair, adjacent and symmetric,

• Neighbours of TBA transmitters that satisfy coverage conditions.

Automatic neighbour allocation parameters are described in "Global Allocation for All Transmitters" on page 185.

5.9 AFP Appendices

5.9.1 The AFP Cost Function

The notations listed hereafter are used to describe the cost function:

• TRG: Group of TRXs

• TRGs: Set of all the TRGs

• : If and only if

• : Size of any group g

• ARFCN: Set of al l the frequencies

• : Set of all the subsets of frequencies

• : The largest integer

• : Number of times a group is assigned to TRGi in the assignment A

For example:

- When i is NH, g is a single member group containing one of the frequencies assigned at TRGi.

If |g| is not 1 or if g does not contain a frequency assigned at i, then .

- When i is BBH, can be either 0 or equal to the number of TRXs in TRGi.

= Number of TRXs in TRGi g is the set of frequencies assigned to TRXs of TRG i. (|g| = number of

TRXs in TRGi).

When we talk about "TRXs of i using g", and in the case of BBH, then there are |g| such virtual TRXs, eachusing the entire group g and having a virtual MAIO [0, |g| - 1].

- When i is SFH, must be less than or equal to the umber of TRXs in TRGi. g is the set of

frequencies assigned to n TRXs of TRGi.

We assume all the groups assigned to TRGi to have the same length.

• TSi : Number of timeslots available for each TRX in TRGi

• TLi : Traffic load of TRGi (calculated or user-defined)

of a single TRX in TRGi divided by TSi

• TSU i : Downlink timeslot use ratio (due to DTX) at TRGi

• CF i : Cost factor of TRGi (AFP Weight)

• QMIN i : Minimum required quality (in C/I) at TRGi

• PMAX i : Percentage permitted to have quality lower than QMINi at TRGi

• REQi : Required number of TRXs at TRGi

A communication uses the group g in TRGi if its mobile allocation is g. The probability to be interfered is denoted by

(i’ is the TRX index). Different TRX indexes may have different MAIOs. is a function of the whole

frequency assignment. The precise definition of the term “to be interfered” is provided afterwards. The probability penalty

due to violating a separation constraint is . It is a function of the whole frequency assignment as well.

The term “Atom” will be used in the following context:

For two TRGs, i and k,

i and k are synchronised, have the same HSN, the same MAL length and the same hopping mode.

• In the Results, Atoll displays only the transmitters for which it finds new neighbours.

Therefore, if a transmitter has already reached its maximum number of neighbours before

starting the new allocation, it will not appear in the Results table.

g

2 ARF CN

x x d

A i g g 2 AR FCN

Ai g 1=

Ai g 0 =

Ai g

Ai g

Ai g Ai g n=

TL i #Erlangs=

P i i ' g A P i i ' g A

P i i ' g A

AT OM i ATOM k {



© Forsk 2010 AT282_TRG_E2 189


NH TRGs or BBH TRGs are always in separate atoms. If two TRGs interfere but are not in the same atom, these can be

taken as unsynchronised. The quality of unsynchronised TRGs is a function of all possible frequency combinations. For

synchronised TRGs, pairs of frequencies emitted at the same time are known.

5.9.1.1 Cost Function

The Atoll AFP cost function is a TRX based cost and not an interference matrix entry based cost. It counts the impaired

traffic of the network TRXs in weighted Erlangs.

The cost function is reported to the user during the AFP progress with the help of its 5 components: , ,

, and .

= + + + +

where,

represents the missing TRX cost component

represents the separation component

represents the additional cost component (interference, cost of changing a TRX)

represents the corrupted TRX cost component

represents the out-of-domain frequency assignment cost component

In the above equations,

• i’ is the TRX index belonging to .

• is the number of missing TRXs for the subcell i .

• is the cost value for a missing TRX. This value can vary between 0 and 10. The default cost value is set to 1

and can be modified in the AFP module properties dialog.• is the number of corrupted TRXs for the subcell i .

• is the cost value of a corrupted TRX. This value can vary between 0 and 10. The default cost value is set to 10

and can be modified in the AFP module properties dialog.

• is the number of TRXs, for the subcell i , having out-of-domain frequencies assigned.

• is the cost value of a TRX with out-of-domain frequencies assigned. This value can vary between 0 and 1. The

default cost value is set to 0.5 and can be modified in the AFP module properties dialog.

And, as mentioned earlier, a virtual TRX is considered in case of BBH.

If i’ is valid, the algorithm evaluates the cost of a valid TRX. This cost has two components, and

.

< <mis <sep

<comp <corr <dom

< <mi s <sep <comp <corr <dom

<mis

<sep

<comp

<corr

<dom

<mis MIS_TRX i Ou TL i CF i TS i uuu

i T RG s¦=

<corr CORR_TRX i :u TL i CF i TS i uuu

i T RG s¦=

<dom DOM_TRX i Zu TL i CF i TS i uuu

i T RG s¦=

<sep G'i i ' g A

g 2 AR FCN

i ' TRXs of i using g

¦

© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

TL i CF i TS i uuu

i T RG s¦=

<comp G''i i ' g A

g 2 ARF CN

i ' TRXs of i using g

¦

© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸

¨ ¸§ ·

TL i CF i TS i uuu

i T RG s¦=

0 1 ... A i g 1 – ^ `

MIS_TRX i

MIS_TRX i MAX 0 REQ i A i g

g 2 ARF CN

¦ – =

O

CORR_TRX i

:

DOM_TRX i

Z

G'i i ' g A

G''i i ' g A




• is the separation violation probability penalty.

• is complementary probability penalty due to interference and the cost of modifying a TRX.

If the option “Take into account the cost of all the TRXs” available in the AFP module properties dialog is selected,

then,

and

Or if the option “Do not include the cost of TRXs having reached their quality target” available in the AFP module

properties dialog is selected, the algorithm compares with the quality target specified for i ,

:

If ,

Then and .

Otherwise,

Both and will be equal 0.

is the same as (separation violation probability penalty) and the same as

(complementary probability penalty due to interference and the cost of modifying a TRX) in most cases. These are

explained in detail in the next sections.

5.9.1.2 Cost Components

Separation violation and interference cost components are described hereafter. Parameters considered in the cost func-

tion components can be fully controlled by the user. Some of these parameters are part of the general data model (qualityrequirements, percentage of interference allowed per subcell), while others (such as separation costs and diversity gains)

can be managed through the properties dialog of the Atoll AFP module.

5.9.1.2.1 Separation Violation Cost Component

The separation violation cost component is evaluated for each TRX. Estimation is based on costs specified for the required

separations.

Let denote the required separation constraint between TRGi and TRGk. Let denote the user

defined separation penalty for a required separation “s” and actual separation “z”. is used instead of

as abbreviation.

is considered to be the effect of a separation violation on the th TRX of TRGi assigned the group g, caused by

the th TRX of TRGk assigned the group .

denotes the overall weight of the separation violation cost component. This value can be between 0 and 1, set to 1 by

default. It can be modified in the AFP module properties dialog.

represents the weight of the specific separation constraint between i and k. This specific weight depends on the type

of separation violation and follows the following priority rule:

1. Exceptional pairs

2. Co-transmitters

3. Co-site

4. Neighbours

For example, if a pair of subcells are co-site and neighbours at the same time, they will be considered as co-site because

higher priority. Hence, of these subcells will be the weight of co-site relations. If only a neighbour relation exists

between two subcells, then will be further weighted by the neighbour relation importance. The value of remains

between 0 and 1. The default weights of each type of separation are available in the Separation cost tab.

If

Then , which is same for all values of k.

If

G'i i ' g A

G''i i ' g A

G'i i ' g A P 'i i ' g A = G''i i ' g A P ''i i ' g A =

P 'i i ' g A P ''i i ' g A +

P MAX

P ' i i ' g A P ''i i ' g A + P MAX !

G'i i ' g A P 'i i ' g A = G''i i ' g A P ''i i ' g A =

G'i i ' g A G''i i ' g A

P 'i i ' g A G'i i ' g A P ''i i ' g A G''i i ' g A

Note:

• The AFP module properties dialog takes probability percentages as inputs while this

document deals in probability values.

SEP_CONSTR i k Cost s z

SE P i k v

Cost SEP_CONSTR i k z

[ii 'kgg 'k ' i '

k ' g '

J

J ik

J ik

J ik J ik

AT OM i ATOM k z

[ii 'kg g 'k ' J J ik

SE P i k f f ' –

f g f ' g '

¦

g g 'u----------------------------------------------uu=

AT OM i ATOM k =



© Forsk 2010 AT282_TRG_E2 191


Then

In the above equations, is the number of frames in the MAL g . .

Let denote the instantaneous frame number from 0 to .

While modulo and is the frequency in g,

And modulo and is the frequency in g’.

In addition, frequencies belonging to a MAL with a low fractional load, and breaking a separation constraint, should not be

weighted equally as in a non-hopping separation breaking case. Therefore, the cost is weighted by an interferer diversity

gain.

The separation gain, denoted by is basically a function of the MAL length (and, of course, of the

hopping mode). With frequency hopping, the effects of DTX and traffic load become more significant (due to the consid-

eration of the average case instead of the worst case). For this reason, it is possible to consider these effects in

through the relevant option available in the Advanced tab of the AFP module properties dialog.

Without this option, the is:

is the user defined interferer diversity gain (dB) for a given MAL length. It is used in definition as well.

On the other hand, if this option is selected, the becomes,

Where ,

And

More than one separation violations may exist for a TRX. Many “small” and have to be combined to form

one cost element, the . This is done through iterating over all violating assignments and by summing up an equiv-

alent to the probability of not being violated while considering each separation violation as an independent probability

event. This sum is naturally limited to 100% of the TRX traffic, and is given by,

In the above formula, if , then , so that interference with itself is not taken into account.

5.9.1.2.2 Interference Cost Component

The interference cost component is evaluated for each TRX. Its estimation is based on interference histograms calculated

for pairs of subcells. In addition, it takes into account frequency and interferer diversity gains and models frequency

hopping and gain due to DTX.

When estimating , the following problems are encountered:

Note:

• Since , we shortly denote the two as .

[ii 'kgg 'k ' J J ik

SE P i k g X g 'W –

f_n 0 1 .. . F_N 1 – ^ `¦

F_N -------------------------------------------------------------------------------------uu=

F_N g F_N g g =

F_N g F_N g ' = F_N

f_n F_N

X f_n MAIO Ai g i ' + = F_N g X Xth

W f_n MAIO Ak g ' k ' + = F_N g 'W Wth

Gi k g g ' 1

10 0.1 SEP_GAIN i k g g ' u

--------------------------------------------------------------------=

SEP_GAIN i k g g '

SEP_GAIN i k g g '

SEP_GAIN i k g g '

SEP_GAIN i k g g ' I_DIV g =

I_DIV g P i i ' g A

SEP_GAIN i k g g '

SEP_GAIN i k g g ' I_DIV g 0.5 TSU_GAIN k u m in 10 4 2 I_DIV g + 2 ASYN_GAIN i k g ' +

4-----------------------------------------------------------------------u© ¹

§ ·+© ¹§ ·u© ¹

§ ·+=

TSU_GAIN k log 10

1

TLk TS U k u-------------------------------

© ¹§ ·=

ASYN_GAIN i k g ' 0 if ATOM(i) = ATOM(k)

I_DIV( g ' Otherwise=

Gi k g g ' ['ii 'kgg '

P 'i i ' g A

P 'i i ' g A

1 1 [ii 'kgg 'k ' G i k g g ' u –

k T RG s

g ' 2 ARF CN

k ' TRXs of k using g '

–

© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

=

k i = k ' i 'z

Note:

• Interference histograms are described in User Manual (GSM GPRS EGPRS project

management, GSM GPRS EGPRS network optimisation, GSM GPRS EGPRS generic

AFP management). Interference histograms can also be exported to files. For further

description, refer to "Interferences" on page 196.

P ''i i ' g A




• The QMINi C/I quality indicator corresponds to the accumulated interference level of all interferers while the C/I

interference histograms correspond to pair-wise interferences.

• Both QMINi and the histograms correspond to a single frequency. In case of a MAL containing more than one fre-

quencies, interferences on several different frequencies of a MAL must be combined.

This estimation, presented below, is the simplest possible as it solves the first problem by linear summation and

truncation at the value of 1 and it solves the second problem by averaging and adding the two diversity gains:

• , the frequency diversity gain, and

• , the interferer diversity gain.

Hereafter, denotes the global weight of interference cost component. This value can vary between 0 and 1 and is set

to 0.35 by default, which can be modified in the AFP module properties dialog.Let be the number of frames in the MAL g . .

Let denote the instantaneous frame number from 0 to .

Let be the j ’th MAIO of , where j is one of the TRXs.

The value of is one of

If TRGk is NH, then .

If TRGk is BBH, then .

As said earlier, in case of BBH, we consider virtual TRXs, the j th TRX has the MAIO j.

Let be the i th frequency in the group g .

Similar to the definition of , is defined as an interference event. is the effect interference on the

th TRX of TRGi assigned the group g, caused by the th TRX of TRGk assigned the group .

If

Then

Where

If

Then,

Since , these are both represented by .

Where,

,

,

modulo ,

modulo ,

Therefore, we have,

In the above formula, if , then , so that interference with itself is not taken into account.

The sum is limited to 100% of the TRX traffic. is quite similar to . The

only difference is the frequency diversity gain, , added to .

F_DIV g

I_DIV g

D

F_N g F_N g g =

f_n F_N

MAIO Ak g ' j Ak g ' 0 1 ... Ak g ' 1 – ^ `

MAIO Ak g ' j 0 1 ... g ' ^ `

MAIO Ak g ' j 0 =

MAIO Ak g ' j j =

g '

g i

[ii 'kgg 'k ' ['ii 'kgg 'k ' ['ii 'kgg 'k '

i ' k ' g '

AT OM i ATOM k z

['ii 'kg g 'k '

Probability C

I ik

----- Q_UBi k f f ' © ¹§ ·

g g 'u--------------------------------------------------------------------------------

f g f ' g '¦=

Q_UBi k f f ' QMIN i f f ' – ADJ_SUP INTERF_GAIN i k g g ' +u – =

AT OM i ATOM k =

F_N g F_N g ' = F_N

[ii 'kgg 'k '

Probability C

I ik

----- Q_UBi k f f ' © ¹§ ·

F_N --------------------------------------------------------------------------------

¯ ¿° °® ¾° °- ½

f_n 0 1 . .. F_N 1 – ^ `¦=

f g X=

f ' g 'W=

X f_n MAIO Ai g i ' + = F_N

W f_n MAIO Ak g ' k ' + = F_N

Q_UBi k f f ' QMIN i f f ' – ADJ_SUP INTERF_GAIN i k g g ' +u – =

P ''i i ' g A 1 1 P 'i i ' g A – 1 [ii 'kg g 'k ' –

k T RG s

g ' 2 ARF CN

k ' TRXs of k using g '

u

¯ ¿° °° °° °® ¾° °° °° °- ½

– P 'i i ' g A – =

i k = k ' i 'z

INTERF_GAIN i k g g ' SEP_GAIN i k g g '

F_DIV g SEP_GAIN i k g g '



© Forsk 2010 AT282_TRG_E2 193


5.9.1.2.3 I_DIV, F_DIV and Other Advanced Cost Parameters

When combining interference effects (or separation violation effects) on different frequencies belonging to a MAL, the

following considerations should be taken into account:

1. Non-linearity of Frame Error Rate (FER) with respect to average C/I conditions and MAL length.

2. Interference Diversity Gain. This factor represents that the effect of average negative effects over user geographic

location are directly proportional to the MAL length.

3. Frequency Diversity Gain. This factor models the gain due to diversity of multi-path effects and should be applied

to the interference cost component only.

4. The fact that long MALs with synthesized hopping permit discarding the worst case estimation and include a gain

due to DTX and low traffic load at the interferer end.

The Advanced properties tab shown in the figure below facilitates modelling these effects.

The Interference Diversity Gain table lists the values of I_DIV provided as a functions of MAL length. This gain is applied

to the interference cost component and to the separation constraint violation cost component. Therefore, it provides a

means to model the non-linear FER effects and interference diversity both. The default values in this table correspond to

the curve . This equation generates values somewhat lower than empirical best-found values (this is

because we prefer a slightly pessimistic cost function to be on the safe side).

The other table contains the F_DIV values, which are the same as the I_DIV values by default.

5.9.2 The AFP Blocked Traffic Cost

This section provides additional information on the AFP cost components used for the optimisation of the number of TRXs.

This optimisation is performed for each traffic pool in the network. In most cases, the traffic pool is equivalent to a trans-mitter and corresponds to the BCCH and TCH subcells. In more complex cases, a traffic pool may include additional

subcells, and more than one traffic pools may exist per transmitter.

The cost component described below, and the recalculation of traffic loads, is only used when the AFP performs the oprim-

isation of the number of TRXs.

The notations listed hereafter are used for the description.

• {BCCH, TCH(1), TCH(2), …, TCH(n)}: Subcells of a traffic pool.

For concentric cells, at least two traffic pools exist per transmitter. The BCCH subcell may not always be part of

the pool’s TRX types.

• {d(0), d(1), d(2), …, d(n)}: Number of required TRXs of each TRX type in the pool

• {ts(0), ts(1), ts(2), …, ts(n)}: Numbers of traffic timeslots

• {L(0), L(1), L(2), …, L(n)}: Traffic loads

Figure 5.14: The Advanced tab of the AFP module Properties dialogue

y 2 log 10 x u=




• {CF(0), CF(1), CF(2), …, CF(n)}: AFP cost factors

• CS (Erlangs): Overall circuit-switched traffic demand of the traffic pool (Subcells table

or traffic analysis results)

• PS (Data Timeslots): Overall packet-switched traffic demand of the traffic pool (Subcells table

or traffic analysis results)

If CS or PS is less than 1, its value is set to 1 in order to avoid working with transmitters carrying no traffic.

• {nb(0), nb(1), nb(2), …, nb(n)}: Number of TRXs in the frequency plan

• {HR(0), HR(1), HR(2), …, HR(n)}: TCH HR use ratios

5.9.2.1 Calculation of New Traffic Loads Including Blocked Traffic Loads

During the optimisation of the number of TRXs, traffic loads are calculated in order to determine the blocked traffic loads

. The blocked traffic load is then multiplied by the AFP cost weight and the number of timeslots to calculate the

blocked traffic cost.

Without the optimisation of the number of required TRXs, the network’s weighted Erlangs are calculated as follows:

With the optimisation of the number of TRXs, the network’s weighted Erlangs are calculated as follows:

and represent the load estimation and the blocked load estimation of the AFP. They are calculated at traf-fic pool level for the vector {nb(0), nb(1), nb(2), …, nb(n)} as follows:

Where

is determined from the above equation once is known. is obtained from the Erlang B equation

applied to the traffic pool demand and the total number of timeslots (TTS):

The Max() function above gives 1 timeslot when there is no TRX.

The above equations give the number of served circuit-switched timeslots (SCS):

The number of served packet-switched timeslots (SPS) is obtained as follows:

is given by:

is given by:

BL nb

WE d i ts i u L i u CF i u

i 0 =

n

¦=

WE nb i ts i u BL nb L nb +^ `u CF i u

i 0 =

n

¦=

BL nb L nb

BL nb L nb +

PS CS 1HR

2 --------------- – © ¹

§ ·u¯ ¿® ¾- ½

+

Max 1 nb i ts i u

i 0 =

n

¦© ¹¨ ¸¨ ¸§ ·

-------------------------------------------------------------------=

HR Max i 0 =n

HR i =

BL nb L nb L nb

TT S M ax 1 nb i ts i u1

HR

2 --------------- – © ¹

§ ·-------------------------------

i 0 =

n

¦

© ¹¨ ¸¨ ¸¨ ¸

§ ·

=

P Blocking ErlangB CS TTS =

SCS 1HR

2 --------------- – © ¹

§ · CS 1 P Blocking – uu=

SPS Min PS Max 1 nb i ts i u

i 0 =

n

¦

© ¹

¨ ¸¨ ¸§ ·

SC S –

¯ ¿° °® ¾° °- ½

=

L nb

L nb S CS S PS+

Max 1 nb i ts i u

i 0 =

n

¦© ¹¨ ¸¨ ¸§ ·

------------------------------------------------------------------=

BL nb



© Forsk 2010 AT282_TRG_E2 195


Once and are known, replaces TLi in the cost function (See "The AFP Cost Function" on

page 188), and is used to generate a new cost component, the blocked Erlangs of the pool:

5.9.2.2 Recalculation of CS and PS From Traffic Loads

In earlier versions, the detailed traffic demand information is not available. In order to guide the AFP to generate it from

the loads, the following two equations with three variables must be solved. The equations are solvable due to the mono-

tone nature of the Erlang B function.

Inputs for a given traffic pool:

• {d(0), d(1), d(2), …, d(n)}: Number of required TRXs of each TRX type in the pool

• L: Traffic load

• :

• MB: Maximum blocking rate (between 0 and 1).

The ratio of packet-switched demand is given by:

Here, we assume that a traffic load of 1 is generated by a demand of (1+MB)*TTS’ which generates a blocking rate of MB.

In other words, the ratio is calculated so that the worst case blocking rate is BM, giving a load of 1.

The following equations are solved to find PS’, CS’, and R’, which are calculated for a traffic load of 1.

When the traffic load of a pool is not 1, PS is different from PS’ and CS is different from CS’. Here, however, we assume

that R’ = R. This assumption implies that R is more or less the same as MB for big traffic pools and considerably larger

than MB for smaller pools.

The following equations are solved to find PS, CS, and R, which are calculated for the actual traffic loads.

The above five equations are solved to get the values of the five variables PS, PC, , SCS, SPS, and calculate

the cost.

BL nb PS CS 1

HR

2 --------------- – © ¹

§ ·u+

Max 1 nb i ts i u

i 0 =

n

¦© ¹¨ ¸¨ ¸§ ·

------------------------------------------------------------------ L nb – =

L nb BL nb L nb

BL nb

nb i ts i u BL nb u CF i ui 0 =

n

¦

TT S' TT S' Max 1d i ts i u

1HR

2 --------------- – © ¹

§ ·------------------------------

i 0 =

n

¦

© ¹¨ ¸¨ ¸¨ ¸§ ·

=

R PS

PS CS 1HR

2 --------------- – © ¹

§ ·u+

------------------------------------------------------------=

M B E rl an gB C S' TT S' =

R 'PS '

PS ' CS ' 1HR

2 --------------- – © ¹

§ ·u+

--------------------------------------------------------------=

1 MB+ TT S'u PS '

1HR

2 --------------- – © ¹

§ ·------------------------------ CS '+=

R PS

PS CS 1HR

2 --------------- – © ¹

§ ·u+

------------------------------------------------------------=

P Blocking ErlangB CS TTS' =

SCS 1 HR 2

--------------- – © ¹§ · CS 1 P Blocking – uu=

SPS Min PS Max 1 d i ts i u

i 0 =

n

¦© ¹¨ ¸¨ ¸§ ·

SC S –

¯ ¿° °® ¾° °- ½

=

S CS S PS+ d i ts i u L i u

i 0 =

n

¦=

P Blocking




5.9.2.3 Testing the Blocked Cost Using Traffic Analysis

As long as the conditions below hold truw, the blocked cost calculation in the AFP and the effective overflow calculation

in the KPI calculation and dimensioning use the same algorithm. The conditions are:

• The AFP cost factors are 1,

• The HR ratios are the same within the subcells of a traffic pool,

• The dimensioning model is based on Erlang B,

• The timeslot configurations are the default ones,

• There exists at least one TRX in the traffic pool (and at least one Erlang of traffic),

• All transmitters belong to the same HCS Layer.

Effective Overflow rate =

Output: New values for CS and PS.

5.9.3 Interferences

This appendix provides a high-level overview of interferences taken into account by the AFP.

5.9.3.1 Using Interferences

If interferences are to be taken into account by the AFP, they must be calculated or imported beforehand. In order to do

this, the user should previously decide to take interferences into account (enabling the loading of all the potential inter-

ferers). Otherwise, Atoll does not allow performing their computation by disabling the histogram part in the corresponding

dialog.

5.9.3.2 Cumulative Density Function of C/I LevelsFor each [interfered subcell, interfering subcell] pair, Atoll calculates a C/I value on each pixel of the interfered subcell

service area (as if the two subcells share the same channel). Then, Atoll integrates these C/I values to determine a C/I

distribution and transforms this distribution function into a cumulative density function in the normal way.

In Atoll, both the IMco and IMadj are represented by this Cumulative Density function This implies that each query for the

probability to have C/I conditions worse than X dB requires a single memory access: the co-channel interference proba-

bility at X dB. In order to deduce the adjacent interference probability value, Atoll looks up the cumulative density function

at the value corresponding to X - Y dB, Y dB being the adjacency suppression value. The following example may be helpful

in further clarifying this concept:

Example: Let [TX1, BCCH] and [TX2, BCCH] be the interfered and interfering subcells respectively. The service

areas for both have been defined by Best Server with 0 dB margin. The interference probability is stated in percent-

age of interfered area.

In this case, we observe that the probability for C/I (BCCH of TX2 effecting the BCCH of TX1) being greater than

0 is 100% (which is normal because TX1 is the Best Server). The probability of having a C/I value at least equal

to 31 dB is 31.1%. For a required C/I level of 12 dB on the BCCH of TX1, the interference probability is 6.5% (as

this requirement is fulfilled with a probability of 93.5%).

1 L nb L nb BL nb +------------------------------------------ –

Figure 5.15: The cumulative density of C/I levels between [TX1, BCCH] and [TX2, BCCH]

Note:

• The subcell power offset does not enter the calculation results in the .clc file. It is added

later by the AFP interface. On the other hand, its influence on the subcell service zone is

taken into account in the .clc file.



© Forsk 2010 AT282_TRG_E2 197


5.9.3.3 Precise Definition

is defined to be the probability of a communication (call) occupying a timeslot in subcell v (victim) to have

C/I conditions of C_I with respect to a co-channel interference from the BCCH TRX of cell n (neighbour). We assume C_I

values to be discrete and in dB. CDF(Pci) is the cumulative density function of Pci:

5.9.3.4 Precise Interference Distribution Strategy

Why does Atoll calculate and maintain precise interference distributions, while the most common solution (used by most

other tools) is rather to compress the information into two values: the co-channel and adjacent-channel interference prob-

abilities?

The reason is simply that it,

• improves the AFP result,

• introduces very little (or no) overhead, and

• creates more generic interference information.

5.9.3.4.1 Direct Availability of Precise Interference Distribution to the AFP

In the presence of frequency hopping, and when one or more frequencies are common (or adjacent) in two interfering MAL

sequences, the hopping gain depends on following factors:

• the MAL length,

• the traffic load on the interferer TRX,

• DTX level , and

• the number of common (and adjacent) frequencies in the two MALs.

All these factors cannot be pre-calculated since it is the AFP that determines the MAL length and the MAL frequencies.

5.9.3.4.2 Efficient Calculation and Storage of Interference Distribution

In the innermost loop of the calculation process Atoll increments a counter each time a C/I level has a certain value. In

the case of a two-entry IM, there are only two counters for each [interfered, interferer] pair. In the case of precise distribu-

tion information, there are about 40 counters per pair. In both cases, the number of operations is the same: one increment

of an integer value. Once Atoll finishes the counting for an [interfered, interferer] pair, it compresses the information from

the counters to a Cumulative Density Function (CDF) representation. In this way, access to interference probability at a

certain level is instantaneous. Thus, the only overheads are the read / write times to the files and the memory occupation

at running time. These two overheads are negligible and do not affect the calculations, the heaviest part of the task.

5.9.3.4.3 Robustness of the IM

By having precise C/I distributions calculated and exported, the user is free to change the following settings without the

need for recalculating their interference distributions:

1. Quality requirements of network elements (required C/I, % Probability Max, …),

2. C/I weighting (the interference levels above and below the C/I target),

3. Separation requirements and/or neighbour relations,

4. Hopping gain values, DTX activities, traffic load levels, HSNs, synchronisation information,

5. Any frequency assignment setting (MAL length directives, frequency domains, assignment strategies, number of

required TRXs, cost function parameters, …), or

6. Remove equipment

By not mixing any of the elements above, the interference information keeps its original probability units and is easier to

check and validate. Therefore, the user spends less time on interference recalculations than in the case of a two-entry

matrix (where “everything” is included).

5.9.3.5 Traffic Load and Interference Information Discrimination

Atoll maintains the traffic load separate from the interference information. The reasons for implementing this strategy are

explained here.

Let us look at the possible alternatives to this strategy:

1. The mixed option: The interference information contains the traffic information as well. In this way, each IM entry

will contain the quantity of traffic interfered if a co-channel / adjacent channel reuse exists.

2. The separated option: The AFP has separate access to traffic load information and to interference probabilities

(As in Atoll).

Knowing the difference between the two alternative solutions explains why the second strategy has been opted for for

Atoll. However, in detail, this has been done because:

• Option 2 is a superset that contains option 1. But option 1, being a subset, does not contain option 2 (i.e. once the

information are mixed they cannot be separated).

• It does not create any overhead (the size of the additional information is negligible compared to the size of the IM).

Pci v n C_I

CDF Pci v n C_I Pci v n x

x C_I t¦=




• It helps keeping the unit definitions simpler.

• It is facilitates merging IMs with different traffic units.

• The traffic information can be used for weighting the separation violation component.

• The traffic load can be used in deciding whether a TRX can be left uncreated.

For example, if there are too many TRXs at a site and the user wishes that the AFP remove one of them, in order

to be able to not violate site constraints, the AFP must know the traffic loads in order to choose a low load TRX to

be removed.

• The gain introduced by the traffic load of the interferer depends on the hopping mode and the MAL size. Incorpo-

rating this gain in the IM (as a result of the mixed option) means that the IMs become hopping-mode and MAL-

size dependent. This is a bad idea since the AFP should be able to change the MAL. And the user should be able

to change the hopping mode without recalculating the IM. In addition, an IM calculated externally to Atoll, with a

non-hopping BCCH can be used for the hopping TCH.

A third option also exists. Though, this option is so practically useless due to its inefficiency. It consists in mixing IM and

traffic but still keeping the traffic in its isolated form. This is again a bad idea because of the unit definition and the variety

of IM sources. It involves less benefits than the option chosen in Atoll.



Chapter 6UMTS HSPA Networks

This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations, simulations and prediction studies available

in UMTS HSPA projects.






© Forsk 2010 AT282_TRG_E2 201

Chapter 6: UMTS HSPA Networks

6 UMTS HSPA Networks

6.1 General Prediction Studies

6.1.1 Calculation Criteria

Three criteria can be studied in point analysis (Profile tab) and in common coverage studies. Study criteria are detailed in

the table below:

where,

EIRP is the effective isotropic radiated power of the transmitter,

ic is a carrier number,

is the loss on the transmitter-receiver path (path loss) calculated by the propagation model,

is the transmitter antenna attenuation (from antenna patterns),

is the shadowing margin. This parameter is taken into account when the option “Shadowing taken into

account” is selected,

are the indoor losses, taken into account when the option “Indoor coverage” is selected,

are the receiver losses,

is the receiver antenna gain,

is the transmitter antenna gain,

is the transmitter loss ( ). For information on calculating transmitter loss, see "UMTS HSPA,

CDMA2000 1xRTT 1xEV-DO, and TD-SCDMA Documents" on page 132.


6.1.2.1 Profile Tab

Atoll displays either the signal level received from the selected transmitter on a carrier ( ), or the highest signal

level received from the selected transmitter on all the carriers.

6.1.2.2 Reception Tab

Analysis provided in the Reception tab is based on path loss matrices. So, you can study reception from TBC transmitters

for which path loss matrices have been computed on their calculation areas.

Study criteria Formulas

Signal level ( ) in dBmSignal level received from a transmitter on a carrier (cell)

Path loss ( ) in dBm

Total losses ( ) in dBm

Notes:

• ( is the cell pilot power).

• It is possible to analyse all the carriers. In this case, Atoll takes the highest pilot power of

cells to calculate the signal level received from a transmitter.

• Atoll considers that and equal zero.

P re c P rec ic EIRP ic L pat h – M Shadowing model – – LIndoor – Gterm Lterm – +=

L pat hL pat h Lmodel Lant Tx

+=

Ltotal Ltotal L pat h LTx Lterm Lindoor M Shadowing model – + + + + GTx Gterm+ – =

Lmodel

Lant Tx


LIndoor

Lterm

Gterm

GTx

LTx LTx Lto ta l DL – =

EIRP ic P pil ot ic GTx LTx – += P pi lo t ic

Gterm Lterm

Note:

• For a selected transmitter, it is also possible to study the path loss, , or the total

losses, . Path loss and total losses are the same on any carrier.

P rec ic

L pat h

Ltotal




For each transmitter, Atoll displays either the signal level received on a carrier, ( ), or the highest signal level

received on all the carriers.

Reception bars are displayed in a decreasing signal level order. The maximum number of reception bars depends on the

signal level received from the best server. Only reception bars of transmitters whose signal level is within a 30 dB margin

from the best server can be displayed.

6.1.3 Coverage Studies

For each TBC transmitter, Txi, Atoll determines the selected criterion on each pixel inside the Txi calculation area. In fact,

each pixel within the Txi calculation area is considered as a potential (fixed or mobile) receiver.

Coverage study parameters to be set are:

• The study conditions in order to determine the service area of each TBC transmitter,

• The display settings to select how to colour service areas.


Atoll uses parameters entered in the Condition tab of the coverage study property dialogue to predetermine areas where

it will display coverage.

We can distinguish three cases:


The service area of Txi corresponds to the bins where:



And

M is the specified margin (dB).

Best function: considers the highest value.



And

M is the specified margin (dB).

2 nd Best function: considers the second highest value.

Note:

• For a selected transmitter, it is also possible to study the path loss, , or the total

losses, . Path loss and total losses are the same on any carrier.

• You can use a value other than 30 dB for the margin from the best server signal level, for

example a smaller value for improving the calculation speed. For more information on

defining a different value for this margin, see the Administrator Manual .

P re c ic

L pat h

Ltotal


ic or Ltotal Tx i

or L pat hTxi d MaximumThreshold

Notes:

• If the margin equals 0 dB, Atoll will consider bins where the signal level received from Txi

is the highest.

• If the margin is set to 2 dB, Atoll will consider bins where the signal level received from Txi

is either the highest or 2dB lower than the highest.

• If the margin is set to -2 dB, Atoll will consider bins where the signal level received from Txi

is 2dB higher than the signal levels from transmitters, which are 2nd best servers.


ic or Ltotal Tx i


P re c Tx i

ic Best

j i zP re c

Tx j ic M – t


ic or Ltotal Tx i


P re c Tx i

ic 2 nd

Best

j i zP re c

Tx j ic M – t



© Forsk 2010 AT282_TRG_E2 203



6.1.3.2.1 Plot Resolution

Prediction plot resolution is independent of the matrix resolutions and can be defined on a per study basis. Prediction plots

are generated from multi-resolution path loss matrices using bilinear interpolation method (similar to the one used to eval-

uate site altitude).


It is possible to display the transmitter service area with colours depending on any transmitter attribute or other criteria

such as:

Signal Level (in dBm, dBµV, dBµV/m)

Atoll calculates signal level received from the transmitter on each pixel of each transmitter service area. A pixel of a serv-

ice area is coloured if the signal level exceeds ( ) the defined minimum thresholds (pixel colour depends on signal level).


layers as transmitter service areas. Each layer shows the different signal levels available in the transmitter service area.

Best Signal Level (in dBm, dBµV, dBµV/m)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where other serv-

ice areas overlap the studied one, Atoll chooses the highest value. A pixel of a service area is coloured if the signal level

exceeds ( ) the defined thresholds (the pixel colour depends on the signal level). Coverage consists of several independ-

ent layers whose visibility in the workspace can be managed. There are as many layers as defined thresholds. Each layer

corresponds to an area where the signal level from the best server exceeds a defined minimum threshold.

Path Loss (dB)

Atoll calculates path loss from the transmitter on each pixel of each transmitter service area. A pixel of a service area is

coloured if path loss exceeds ( ) the defined minimum thresholds (pixel colour depends on path loss). Coverage consists

of several independent layers whose visibility in the workspace can be managed. There are as many layers as service

areas. Each layer shows the different path loss levels in the transmitter service area.

Total Losses (dB)

Atoll calculates total losses from the transmitter on each pixel of each transmitter service area. A pixel of a service area

is coloured if total losses exceed ( ) the defined minimum thresholds (pixel colour depends on total losses). Coverage


service areas. Each layer shows the different total losses levels in the transmitter service area.

Best Server Path Loss (dB)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where other serv-

ice areas overlap the studied one, Atoll determines the best transmitter and evaluates path loss from the best transmitter.

A pixel of a service area is coloured if the path loss exceeds ( ) the defined thresholds (pixel colour depends on path

loss). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are as

many layers as defined thresholds. Each layer corresponds to an area where the path loss from the best server exceeds

a defined minimum threshold.

Best Server Total Losses (dB)

Atoll calculates signal levels received from transmitters on each pixel of each transmitter service area. Where service

areas overlap the studied one, Atoll determines the best transmitter and evaluates total losses from the best transmitter.

A pixel of a service area is coloured if the total losses exceed ( ) the defined thresholds (pixel colour depends on total

losses). Coverage consists of several independent layers whose visibility in the workspace can be managed. There are

as many layers as defined thresholds. Each layer corresponds to an area where the total losses from the best server

exceed a defined minimum threshold.

Notes:

• If the margin equals 0 dB, Atoll will consider bins where the signal level received from Txi

is the second highest.

• If the margin is set to 2 dB, Atoll will consider bins where the signal level received from Txi

is either the second highest or 2dB lower than the second highest.

• If the margin is set to -2 dB, Atoll will consider bins where the signal level received from Txi

is 2dB higher than the signal levels from transmitters, which are 3rd best servers.

t

t

t

t

t

t




Number of Servers

Atoll evaluates how many service areas cover a pixel in order to determine the number of servers. The pixel colour

depends on the number of servers. Coverage consists of several independent layers whose visibility in the workspace can

be managed. There are as many layers as defined thresholds. Each layer corresponds to an area where the number of

servers exceeds ( ) a defined minimum threshold.

Cell Edge Coverage Probability (%)

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the signal level from this

transmitter fulfils signal conditions defined in Conditions tab with different Cell edge coverage probabilities. There is one

coverage area per transmitter in the explorer.

Best Cell Edge Coverage Probability (%)

On each pixel of each transmitter service area, the coverage corresponds to the pixels where the best signal level received

fulfils signal conditions defined in Conditions tab. There is one coverage area per cell edge coverage probability in the

explorer.

6.2 Definitions and FormulasInput parameters and formulas used in simulations and predictions (coverage predictions and point analysis) are detailed

in the tables below.

6.2.1 Inputs

This table lists simulation and prediction inputs (calculation options, quality targets, active set management conditions,

etc.).

t

Name Value Unit Description

Clutter parameter None Orthogonality factor

Site equipment parameter None MUD factor

Terminal parameter - HSDPA properties None MUD factor

Frequency band parameter None Carrier number

Cell parameter None

Threshold for macro diversity

specified for a transmitter on a given

carrier ic

Mobility parameter NoneEc/I0 target on downlink for the best

server

Global parameter NonePilot RSCP threshold for compressed

mode activation

Global parameter NoneEc/I0 threshold for compressed mode

activation

(Reception equipment, R99 bearer, Mobility) parameter None Eb/Nt target on downlink

Global parameter NoneDownlink Eb/Nt target increase due to

compressed mode activation

(Reception equipment, R99 bearer, Mobility) parameter None Eb/Nt target on uplink

Global parameter NoneUplink Eb/Nt target increase due to

compressed mode activation

Site parameter NoneNumber of channel elements available

for a site on uplink

Site parameter NoneNumber of channel elements available

for a site on downlink

Simulation result NoneNumber of channel elements of a site

consumed by users on uplink

Simulation result NoneNumber of channel elements of a site

consumed by users on downlink

F ortho

F MU DTx

F MU DTerm

ic

AS_Th Tx i ic

Q pil ot req

E c

I 0

------© ¹§ ·

threshold

RSCP pi lo t CM act ivat ion –

Q pi lo t CM act ivat ion –

Qre qDL

E bN t ------

© ¹§ ·

req

DL

QreqDL

'

Qre qUL

E bN t ------

© ¹§ ·

req

UL

QreqUL'

N max C E U L –

N I

N max C E D L –

N I

N C E U L –

N I

N C E D L –

N I



© Forsk 2010 AT282_TRG_E2 205


Site equipment parameter - UL overhead resources for common

channels/cellNone

Number of channel elements used by

the cell for common channels on

uplink

Site equipment parameter - DL overhead resources for common

channels/cellNone

Number of channel elements used by

the cell for common channels on

downlink

(R99 Bearer, site equipment) parameter NoneNumber of channel elements used for

R99 traffic channels on uplink

(R99 Bearer, site equipment) parameter NoneNumber of channel elements used for

R99 traffic channels on downlink

(HSUPA Bearer, site equipment) parameter NoneNumber of channel elements

consumed by the HSUPA bearer on

uplink

Site parameter kbpsMaximum Iub backhaul throughput for

a site in the uplink

Site parameter kbpsMaximum Iub backhaul throughput for

a site in the downlink

Simulation result kbpsIub backhaul throughput for a site in

the uplink

Simulation result kbpsIub backhaul throughput for a site in

the downlink

Site equipment parameter kbpsIub throughput required by the cell for

common channels in the downlink

Site equipment parameter % HSDPA Iub backhaul overhead

Site equipment parameter kbpsThroughput carried by an E1/T1/

Ethernet link

(R99 Bearer, site equipment) parameter kbpsIub backhaul throughput consumed by

the R99 bearer in the uplink

(R99 Bearer, site equipment) parameter kbpsIub backhaul throughput consumed by

the R99 bearer in the downlink

(HSUPA Bearer, site equipment) parameter kbpsIub backhaul throughput consumed by

the HSUPA bearer in the uplink

Simulation constraint NoneMaximum number of 512 bit-length

OVSF codes available per cell (512)

Simulation result NoneNumber of 512 bit-length OVSF codes

used by the cell

Site equipment parameter - DL overhead resources for commonchannels/cell None Number of 256 bit-length OVSF codesused by the cell for common channels

Cell parameter (for HSDPA only) None

Maximum number of 16 bit-length

OVSF codes available per cell for HS-

PDSCH

Cell parameter (for HSDPA only) None

Minimum number of 16 bit-length

OVSF codes available per cell for HS-

PDSCH

Terminal parameter None Terminal Noise Figure

Transmitter parameter (user-defined or calculated from transmitter

equipment characteristics)None Transmitter Noise Figure

K 1.38 10-23 J/K Boltzman constant

T 293 K Ambient temperature

W 3.84 MHz Hz Spreading Bandwidth

Cell parameter None Inter-technology downlink noise rise

Cell parameter only used as input of the Monte-Carlo simulation None Inter-technology uplink noise rise

Network parameter

If not defined, it is assumed that there is no inter-carrier interferenceNone

Interference reduction factor between

two adjacent carriers and

Network parameter

If not defined, it is assumed that there is no inter-technology downlink

interferences due to external transmitters

None

Inter-technology Channel Protection

between the signal transmitted by Tx

and received by m assuming the

frequency gap between (external

network) and

N Overhead C – E UL –

N Overhead C – E DL –

N R99 T – CH C – E UL –

N R99 T – CH C – E DL –

N HSUPA C – E

T Iub m – ax UL

N I

T Iub m – ax DL

N I

T IubUL

N I

T IubDL

N I

T IubOverhead DL –

Overhead Iub

HSDPA

T E1 T1 e Ethernet e

T IubR99 T – C H U L –

T IubR99 T – C H D L –

T IubHSUPA

N max Codes

Txi ic

N Codes

Txi ic

N Overhead C – odes

N max Codes HS – PDSCH

Txi ic

N minCodes HS – PDSCH

Txi ic

NF term

NF Tx

NR i nt er t echn o y log – Tx DL

NR i nt er t echn o y log – Tx UL

RF ic ic adj ic ic adj

IC P ic i ic Tx m

ic i

ic




Simulation constraint (g lobal parameter or cel l parameter) % Maximum uplink load factor

Simulation constraint (global parameter or cell parameter) % Maximum percentage of used power

W Thermal noise at transmitter

W Thermal noise at terminal

W bps Chip rate

Site equipment parameter None Uplink rake receiver efficiency factor

Terminal parameter NoneDownlink rake receiver efficiency

factor

R99 bearer parameter kbps R99 bearer downlink nominal bit rate

R99 bearer parameter NoneDownlink spreading factor for active

users

R99 bearer parameter NoneDownlink spreading factor for inactive

users

R99 bearer parameter None

ratio between DPCCH and DPCH

transmission duration on downlink

DPCCH and DPCH respectively refer

to the Dedicated Physical Control

Channel and Dedicated Physical

Channel

Cell parameter kbpsMaximum connection rate per user on

downlink

R99 bearer parameter kbps R99 bearer uplink nominal bit rate

Service parameter kbpsUplink activity factor on E-DPCCH

channels

Service parameter kbpsDownlink Activity factor on A-DPCH

channel

Service parameter kbps

Minimum required bit rate that the

service should have in order to be

available in the uplink

Service parameter kbps

Minimum required bit rate that the

service should have in order to be

available in the downlink

R99 bearer parameter None

ratio between the DPCCH and DPCH

powers transmitted on uplink

DPCCH and DPCH respectively refer

to the Dedicated Physical Control

Channel and Dedicated Physical

Channel

Cell parameter kbpsMaximum connection rate per user on

uplink

None Service downlink processing gain

None Service uplink processing gain

HSDPA study result kbps User application throughput ondownlink

HSDPA Bearer parameter kbpsRLC peak rate supported by the

HSDPA bearer

HSDPA study result

Without MIMO:

With MIMO (transmit diversity):

With MIMO (spatial multiplexing):

kbpsRLC peak rate provided in the

downlink

X max UL

%Power max DL

N 0 Tx

NF Tx K T u W uu NR i nt er t e chn o y log – Tx ULu

N 0 Term

NF Term K T u W NR i nt er t e chn o y log – Tx DL

uuu

R c W 10 3 –

f rake efficiency

UL

f rake efficiency DL

R nominal DL

F spreading DL

Active user

F spreading DL

Inactive user

r c DL

R max DL

R nominal UL

f act ADPCH – UL

f act ADPCH –

DL

R Guaranteed UL

R Guaranteed DL

r c UL

R max UL

G pDL

W

R nominal DL

----------------------

G pUL

W

R nominal UL

----------------------

T applicationDL

R RLC peak – DL

I HSDPABearer

R RLC peak – DL

R RLC peak –

DLIndex HSDPABearer

R RLC peak – DL

Index HSDPABearer

R RLC peak – DL

Index HSDPABearer 1 f SM Ga in – GSM Max

1 – u+ u



© Forsk 2010 AT282_TRG_E2 207


%

HSDPA bearer consumption for a

packet (HSPA - Constant Bit Rate)

service user

HSDPA study result kbpsRLC peak throughput supported by

the HSDPA bearer

HSDPA study result kbps Average RLC throughput supported

by the HSDPA bearer

HSDPA study result kbpsMAC rate supported by the HSDPA

bearer

HSDPA study result kbpsMAC throughput supported by the

HSDPA bearer

HSUPA study result kbps User application throughput on uplink

HSUPA study result kbpsUser average application throughput

on uplink

HSUPA Bearer parameter kbpsRLC peak rate supported by the

HSUPA bearer

HSUPA study resultkbps RLC peak rate provided in the uplink

%

HSUPA bearer consumption for a

packet (HSPA - Constant Bit Rate)

service user

HSUPA study result kbpsMinimum RLC throughput supported

by the HSUPA bearer

HSUPA study result kbps Average RLC throughput supported

by the HSUPA bearer

HSUPA study result kbpsMAC rate supported by the HSUPA

bearer

Service parameter (for HSDPA only) kbps Throughput offset

Service parameter (for HSDPA only) % Scaling factor

Transmitter parameter W

Maximum shared power

Available only if the inter-carrier power

sharing option is activated

Cell parameter W Cell synchronisation channel power

Cell parameter W

Cell other common channels (except

CPICH and SCH) power a

Cell parameter W Cell pilot power

Cell parameter (user-defined or simulation result) (for HSDPA only)W

Available cell HSDPA power

HSDPA: High Speed Downlink Packet

Access

Simulation result (for HSDPA only) W

Cell HS-PDSCH power

HS-PDSCH: High Speed Physical

Downlink Shared Channel

Cell parameter (for HSDPA only) W

Cell HS-SCCH power

HS-SCCH: High Speed Shared

Control Channel

Cell parameter (user-defined or simulation result) (for HSDPA only)number of HS-SCCH channels

managed by the cell

Cell parameter (for HSDPA only) W Cell headroom power

Cell parameter W Maximum Cell power

Simulation result WR99 traffic channel power transmitted

on carrier ic

R99 bearer parameter WMinimum power allowed on R99 traffic

data channel

R99 bearer parameter WMaximum power allowed on R99

traffic data channel

Cell parameter W

Cell HSUPA power

HSUPA: High Speed Uplink Packet

Access

C HSDPABearer

R Guaranteed DL

R RLC peak –

DLI HSDPABearer

---------------------------------------------------------------------

T RLC peak – DL

T RLC Av – DL

R MAC DL

T MAC DL

T applicationUL

T applicat ion Av – UL

R RLC peak – UL

I HSUPABearer

R RLC peak –

UL

R RLC peak – UL

I HSUPABearer

C HSUPABearer

R Guaranteed UL

R RLC peak – UL

I HSUPABearer ---------------------------------------------------------------------

T RLC Min – UL

T RLC Av – UL

R MAC UL

'R

SF Rate

P ma x Tx i

P SC H Txi ic

P OtherCCH Txi ic

P pi lot Txi ic

P HSDPA Txi ic P HS PDSCH – Txi ic nHS SCCH – P u

HS SCCH – Txi ic +

P HS PDSCH – Txi ic

P HS SCCH – Txi ic

nHS SCCH –

P Headroom Txi ic

P max Txi ic

P tch Txi ic

P tchmin

P tchma x

P HSUPA Txi ic





© Forsk 2010 AT282_TRG_E2 209


6.2.2 Ec/I0 Calculation

This table details the pilot quality ( or ) calculations.

Result calculated from cell edge coverage probability and UL Eb/Nt

standard deviationNone

UL Eb/Nt Shadowing margin

Only used in prediction studies

n=2 or 3

Global parameter (default value)

NoneUL quality gain due to signal diversity

in soft handoff d.

Simulation result None

Random shadowing error drawn

during Monte-Carlo simulation

Only used in simulations

In prediction studiese

For Ec/I0 calculation

For DL Eb/Nt calculation

For UL Eb/Nt calculation

In simulations

None Transmitter-terminal total loss

W Chip power received at terminal

WBit power received at terminal on

carrier ic

WTotal power received at terminal from

a transmitter on carrier ic

W

Total power received at terminal from

traffic channels of a transmitter on

carrier ic

WBit power received at transmitter on

carrier ic used by terminal

W Bit power received at transmitter oncarrier ic used by terminal

WBit power received at transmitter on

DPDCH from a terminal on carrier ic

a. For the calculation of interference, also includes the MBMS SCCPCH channel power

when the optional MBMS feature is activated. You must modify the data structure for activating the optional MBMS feature.

For more information, see the Administrator Manual.

b. on uplink and on downlink. For information on calculating transmitter

losses on uplink and downlink, see "UMTS HSPA, CDMA2000 1xRTT 1xEV-DO, and TD-SCDMA Documents" on

page 132.

c. corresponds to the shadowing margin evaluated from the shadowing error probabil ity

density function (n paths) in case of downlink Ec/I0 modelling.

d. corresponds to the shadowing margin evaluated from the shadowing error probabil ity

density function (n paths) in case of uplink soft handoff modelling.

e. In uplink prediction studies, only carrier power level is downgraded by the shadowing margin

( ). In downlink prediction studies, carrier power level and intra-cell interference are downgraded by

the shadowing model ( or ) while extra-cell interference level is not. Therefore,

or is set to 1 in downlink extra-cell interference calculation.




ULM Shadowing Eb Nt e UL –

npathsM Shadowing Eb Nt e UL – – =

E Shadowing

LT

L pat h LTx Ltermu Lbody Lindoor M Shadowing Ec Io e – uuuu

GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------

L pat h LTx Ltermu Lbody Lindoor M Shadowing Eb Nt e DL – uuuu

GTx Gtermu------------------------------------------------------------------------------------------------------------------------------------------------------------ -

L pat h LTx Ltermu Lbody Lindoor M Shadowing Eb Nt e UL – uuuu

GTx Gtermu------------------------------------------------------------------------------------------------------------------------------------------------------------ -

L pat h LTx Ltermu Lbody Lindoor E Shadowing uuuu

GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------

P c Txi ic P pil ot Txi ic

LT

-----------------------------------

P bDL

Txi ic P tch Txi ic

LT

--------------------------------

P tot DL

Txi ic P tx Txi ic

LT

------------------------------

P traf DL

Txi ic P tch Txi ic

LT

--------------------------------

tch ic ¦

P bUL

ic P term

LT

--------------

P b R99 – UL ic P term R99 –

LT

---------------------------

P b DPDCH – UL

ic P b R99 – UL

ic 1 r c UL

– u

P OtherCCH Txi ic

LTx Lto ta l UL – = LTx Lto ta l DL – =

M Shadowing Ec Io e – npaths

M Shadowing Eb Nt e UL

– npaths


M Shadowing Eb Nt e DL – M Shadowing Ec Io e –

M Shadowing Eb Nt e DL – M Shadowing Ec Io e –

Q pi lo t Ec Io e




6.2.3 DL Eb/Nt Calculation

This table details calculations of downlink traffic channel quality ( or ). When the optional MBMS feature is

activated, the MBMS Eb/Nt is also calculated in the same manner. You must modify the data structure for activating the

optional MBMS feature. For more information, see the Administrator Manual .


WDownlink intra-cell interference at

terminal on carrier

WDownlink extra-cell interference at

terminal on carrier

WDownlink inter-carrier interference at

terminal on carrier

WDownlink inter-technology interference

at terminal on carrier ic a

Without Pilot:

Total noise:

WTotal received noise at terminal on

carrier

None

Quality level at terminal on pilot for

carrier

a. In the case of an interfering GSM external network in frequency hopping, the ICP value is weighted according to the

fractional load.

I intraDL

tx i ic P tot DL

tx i ic UBT S Du – P tot DL

tx i ic P SCH tx i ic

LT

---------------------------------- – © ¹§ ·u

ic

I extraDL

ic P tot DL

tx j ic

t xj j i z¦ ic

I in ter carr ier – DL

ic P to t

DLtx j ic adj

txj j ¦

RF ic ic adj

------------------------------------------------- ic

I i nt er t echn o y log –

DLic

P Transmitted Tx

ic i

Ltotal Tx

IC P ic i ic Tx m

u------------------------------------------

ni

¦

I 0 DL

ic

I intraDL

tx i ic I extraDL

ic I in ter carr ier –

DLic I i nt er t e chn o y log –

DLic + + +

N 0 Term

1 D – UBT S P c tx i ic uu – +

P tot DL

tx i ic I extraDL

ic I in ter carr ier – DL

ic + +

I i nt er t e chn o y log – DL

ic N 0 Term

+ +

ic

Q pi lo t

tx i ic E c

I 0

------

© ¹

§ ·UBTS Du P c tx i ic u

I 0 DL ic

-------------------------------------------------------

ic

QtchDL Eb

Nt -------

© ¹§ ·

DL


WDownlink intra-cell interference at

terminal on carrier


terminal on carrier


terminal on carrier

WDownlink inter-technology interference

at terminal on carrier ic a


carrier ic

Without useful signal:

Total Noise:

None

Quality level at terminal on a traffic

channel from one transmitter on

carrier b

None

Quality level at terminal using carrier

due to combination of all

transmitters of the active set (Macro-

diversity conditions).

I intraDL tx i ic P tot DL tx i ic UBT S F orthou – P tot DL tx i ic P

SCH

tx i ic

LT ---------------------------------- – © ¹§ ·u ic

I extraDL

ic P tot DL

tx j ic

t xj j i z¦ ic


ic P to t

DLtx j ic adj

txj j ¦

RF ic ic adj ------------------------------------------------- ic

I i nt er t echn o y log – DL

ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦

N tot DL

ic I intraDL

ic I extraDL


ic I i nt er t e chn o y log – DL

ic N + + 0

Term+ +

QtchDL

tx i ic E bN t ------

© ¹§ ·

DL

UBT S P bDL

tx i ic u

N tot DL

ic 1 F ortho – UBTS P bDL

tx i ic uu – ---------------------------------------------------------------------------------------------------------------- GDiv

DLG p

DLuu

UBT S P bDL

tx i ic u

N tot DL

ic ------------------------------------------------- GDiv

DLu G pDLu

ic

QDL

ic f rake efficiency DL

QtchDL

tx k ic

tx k ActiveSet ¦u ic



© Forsk 2010 AT282_TRG_E2 211


6.2.4 UL Eb/Nt Calculation

This table details calculations of uplink traffic channel quality ( or ).

None Soft handover gain on downlink

W

Required transmitter traffic channel

power to achieve Eb/Nt target at

terminal on carrier


fractional load.

b. Calculation option may be selected in the Global parameters tab. The chosen option will be taken into account only in

simulations. In point analysis and coverage studies, Atoll uses the option “Total noise” to evaluate DL and UL Eb/Nt.

GSHODL Q

DLic

QtchDL

BestServer ic ----------------------------------------------------------

P tchre q

tx i ic Qreq

DL

QDL

ic --------------------- P tch tx i ic u

ic

QtchUL Eb

Nt -------

© ¹§ ·

UL


W

Total power received at transmitter

from intra-cell terminals using carrier

W


from extra-cell terminals using carrier

WUplink inter-carrier interference at

terminal on carrier

WTotal received interference at

transmitter on carrier

WTotal noise at transmitter on carrier

(Uplink interference)

Without useful signal:

Total noise:

None

Quality level at transmitter on a traffic

channel for carrier a

No HO:

Softer HO:

Soft, softer/soft HO (No MRC):

Softer/soft HO (MRC):None

Quality level at site using carrier ic due

to combination of all transmitters of

the active set located at the same site

and taking into account increasing of

the quality due to macro-diversity

(macro-diversity gain).

In simulations .

None Soft handover gain on uplink

I tot

ULintratx i ic

P bUL

ic

term

tx i

¦ic

I tot

ULextratx i ic

P bUL

ic

term

t xj j i z

¦ic

I in ter carr ier –

ULtx i ic

P bUL

ic adj

term

txj j

¦

RF ic ic adj ---------------------------------------

ic

I tot UL

tx i ic I tot

ULextratx i ic 1 F MUD

Tx Utermu – + I tot

UL intratx i ic u I in ter carr ier –

ULtx i ic + ic

N tot UL

tx i ic I tot UL

tx i ic N 0 tx

+ic

Qtch

UL

tx i ic

E b

N t ------© ¹§ ·

UL

Uterm P b DPDCH –

ULic u

N tot UL

tx i ic 1 F MUDTx

– Utermu P bUL

ic u – ---------------------------------------------------------------------------------------------------------------- GDiv

ULu G p

ULu

Uterm P b DPDCH – UL

ic u

N tot UL

tx i ic ---------------------------------------------------------- GDiv

ULu G pULu

ic

QUL

ic

QtchUL

tx i ic

f rake efficiency UL

QtchUL

tx k ic

tx k ActiveSet samesite

¦u

Ma x tx

k ActiveSet

QtchUL

tx k ic Gmacro diversity – ULu

Ma x

tx k

tx l

, ActiveSet

tx k

samesite

tx l

othersite

f rake efficiency UL

QtchUL

tx k ic

tx k

¦u QtchUL

tx l ic

© ¹¨ ¸¨ ¸¨ ¸§ ·

Gmacro diversity – ULu


UL1=

GSHOUL Q

ULic

QtchUL

BestServer ic ----------------------------------------------------------




6.3 Active Set Management

The mobile’s active set (AS) is the list of the transmitters to which the mobile is connected. The active set may consist of one or more transmitters; depending on whether the service supports soft handover and on the terminal active set size.

The terminal frequency bands are taken into account and transmitters in the mobile’s active set must use a frequency band

supported by the terminal. Finally, the quality of the pilot (Ec » I0) is what determines whether or not a transmitter can belong

to the active set. The active set management is detailed hereafter. Cells entering a mobile’s active set must satisfy the

following conditions:

• The best server (first cell entering active set)

The pilot quality from the best serving cell must exceed the Ec/I0 threshold. Best server cell is the one with the

highest pilot quality.

• Other cells in the active set

- Must use the same carrier as the best server,

- The pilot quality difference between other candidate cells and the best server must be less than the AS

threshold specified for the best server,

- Other candidate cells must belong to the neighbour list of the best server if it is located on a site where the

equipment imposes this restriction (the “restricted to neighbours” option selected in the equipment properties).

6.4 SimulationsThe simulation process consists of two steps:

1. Obtaining a realistic user distribution

Atoll generates a user distribution using a Monte-Carlo algorithm, which requires traffic maps and data as input.

The resulting user distribution complies with the traffic database and maps provided to the algorithm.

Each user is assigned a service, a mobility type, and an activity status by random trial, according to a probability

law that uses the traffic database.

The user activity status is an important output of the random trial and has direct consequences on the next step

of the simulation and on the network interferences. A user may be either active or inactive. Both active and inactive

users consume radio resources and create interference.

Then, Atoll randomly assigns a shadowing error to each user using the probability distribution that describes the

shadowing effect.

Finally, another random trial determines user positions in their respective traffic zone and whether they are indoors

or outdoors (according to the clutter weighting and the indoor ratio per clutter class defined for the traffic maps).

2. Power control simulation

6.4.1 Generating a Realistic User Distribution

During the simulation, a first random trial is performed to determine the number of users and their activity status. Four activ-

ity status are modelled:

• Active UL: the user is active on UL and inactive on DL

• Active DL: the user is active on DL and inactive on UL

• Active UL+DL: the user is active on UL and on DL

• Inactive: the user is inactive on UL and on DL

The determination of the number of users and the activity status allocation depend on the type of traffic cartography used.

W

Required terminal power to achieve

Eb/Nt target at transmitter on carrier

a. Calculation option may be selected in the Global parameters tab. The chosen option will be taken into account

only in simulations. In point analysis and coverage studies, Atoll uses the option “Total noise” to evaluate DL and UL Eb/

Nt.

P termre q

ic Qre q

UL

QUL

ic --------------------- P termu

ic

Note:

• Atoll follows a Poisson distribution to determine the total number of users attempting a

connection in each simulation. In order for Atoll to use a constant total number of users

attempting a connection, the following lines must be added to the Atoll.ini file:

[CDMA]

RandomTotalUsers=0



© Forsk 2010 AT282_TRG_E2 213


6.4.1.1 Simulations Based on User Profile Traffic Maps

User profile environment based traffic maps: Each pixel of the map is assigned an environment class which contains a list

of user profiles with an associated mobility type and a given density (number of subscribers with the same profile per km²).

User profile traffic maps: Each polygon and line of the map is assigned a density of subscribers with given user profile and

mobility type. If the map is composed of points, each point is assigned a number of subscribers with given user profile and

mobility type.

The user profile models the behaviour of the different subscriber categories. Each user profile contains a list of services

and their associated parameters describing how these services are accessed by the subscriber.

From environment (or polygon) surface (S) and user profile density (D), a number of subscribers (X) per user profile is

inferred.

For each behaviour described in a user profile, according to the service, frequency use and exchange volume, Atoll calcu-

lates the probability for the user being active in uplink and in downlink at an instant t.

6.4.1.1.1 Circuit Switched Service (i)

User profile parameters for circuit switched services are:

• The used terminal (equipment used for the service (from the Terminals table)),

• The average number of calls per hour ,

• The average duration of a call (seconds) .

The number of users and their distribution per activity status is determined as follows:

1. Calculation of the service usage duration per hour ( : probability of a connection):

2. Calculation of the number of users trying to access the service i ( ):

Next, we can take into account activity periods during the connection in order to determine the activity status of each user.

3. Calculation of activity probabilit ies:

Probability of being inactive on UL and DL:

Probability of being active on UL only:

Probability of being active on DL only:

Probability of being active both on UL and DL:

Where, and are respectively the UL and DL activity factors defined for the circuit switched service i.

4. Calculation of number of users per activity status:

Number of inactive users on UL and DL:

Number of users active on UL and inactive on DL:

Number of users active on DL and inactive on UL:

Number of users active on UL and DL both:

Therefore, a user when he is connected can have four different activity status: either active on both links, or inactive on

both links, or active on UL only, or active on DL only.

6.4.1.1.2 Packet Switched Service (j)

User profile parameters for packet switched services are:

• The used terminal (equipment used for the service (from the Terminals table)),

Notes:

• When user profile traffic maps are composed of lines, the number of subscribers (X) per

user profile is calculated from the line length (L) and the user profile density (D) (nb of

subscribers per km) as follows:

• The number of subscribers (X) is an input when a user profile traffic map is composed of

points.

X S Du=

X L Du=

N call

d

p0

po

N call d u

3600 ----------------------=

ni

ni X p0 u=

pinactive 1 f ac t UL

– 1 f act DL

– u=

pUL f act UL

1 f act DL

– u=

pDL f act DL

1 f act UL

– u=

pU L D L+ f ac t UL

f act DLu=

f act UL

f act DL

ni inactive

ni pinactiveu=

ni UL ni pULu=

ni DL ni pDLu=

ni UL D L+ ni pU L D L+u=




• The average number of packet sessions per hour ,

• The volume (in kbytes) which is transferred on the downlink and the uplink during a session.

A packet session consists of several packet calls separated by a reading time. Each packet call is defined by its size and

may be divided in packets of fixed size (1500 Bytes) separated by an inter arrival time.

In Atoll, a packet session is described by following parameters:

: Average number of packet calls on the uplink during a session,

: Average number of packet calls on the downlink during a session,

: Average time (millisecond) between two packets calls on the uplink ,

: Average time (millisecond) between two packets calls on the downlink ,

: Average time (millisecond) between two packets on the uplink ,

: Average time (millisecond) between two packets on the downlink ,

: Packet size (Bytes) on uplink,

: Packet size (Bytes) on downlink.

The number of users and their distribution per activity status is determined as follows:

1. Calculation of the average packet call size (kBytes):

and

Where and are the UL and DL efficiency factors defined for the packet switched service j.

2. Calculation of the average number of packets per packet call:

and

3. Calculation of the average duration of inactivity within a packet call (s):

and

4. Calculation of the average duration of inactivity in a session (s):

and

Figure 6.1: Description of a Packet Session

Note:

• For packet (HSDPA) and packet (HSPA) services, and are set to 1.

Note:

• 1kBytes = 1024Bytes.

N sess

V DL V UL

N pac ket c – al l UL

N pac ket c – al l DL

'T pac ket ca ll – UL

'T pac ket ca ll –

DL

'T pac ket UL

'T pac ket DL

S pac ke t UL

S pac ke t DL

S pac ke t c – al l UL V UL

N pac ket c – al l UL

f eff ULu

--------------------------------------------= S pac ket c – al l DL V DL

N pac ket c – al l DL

f eff DLu

--------------------------------------------=

f eff UL

f eff DL

f eff UL

f eff DL

N pac ket UL in t S pac ket c – all

UL

S pac ket UL

1024 e ------------------------------------

© ¹¨ ¸§ · 1+= N pac ket

DL in t S pac ket c – al l

DL

S pac ke t DL

1024 e ------------------------------------

© ¹¨ ¸§ · 1+=

DInactivity UL pac ket ca ll –

N pac ke t UL

1 – 'T pac ket ULu

1000 ---------------------------------------------------------------= DInactivity

DL pac ket cal l –

N pac ket DL

1 – 'T pac ket DLu

1000 ---------------------------------------------------------------=

DInactivity UL session N pac ke t c – all

ULDInactivity

UL pac ket cal l – u= DInactivity DL session N pac ket c – al l

DLDInactivity

DL pac ke t ca ll – u=



© Forsk 2010 AT282_TRG_E2 215


5. Calculation of the average duration of activity in a session (s):

and

Where and are the uplink and downlink average requested rates defined for the service j.

Therefore, the average duration of a connection (in s) is:

and

6. Calculation of the service usage duration per hour (probability of a connection):

and

7. Calculation of the probability of being connected:

Therefore, the number of users who want to get the service j is:

As you can see on the picture above, we have to consider three possible cases when a user is connected:

• 1st case: At a given time, packets are downloaded and uploaded.

In this case, the probability of being connected is:

• 2nd case: At a given time, packet are uploaded (no packet is downloaded).

Here, the probability of being connected is:

• 3rd case: At a given time, packet are downloaded (no packet is uploaded).

In this case, the probability of being connected is:

Now, we have to take into account activity periods during the connection in order to determine the activity status of each

user.

8. Calculation of the probability of being active:

and

Therefore, we have:

• 1st case: At a given time, packets are downloaded and uploaded.

The user can be active on UL and inactive on DL; this probability is:

The user can be active on DL and inactive on UL; this probability is:

The user can be active on both links; this probability is:

The user can be inactive on both links; this probability is:

• 2nd case: At a given time, packet are uploaded (no packet is downloaded).

The user can be active on UL and inactive on DL; this probability is:

D Ac ti vi ty UL session N pac ket c – al l

UL N pac ket UL

S pac ke t UL

8 uu

R averageUL

1000 u------------------------------------------------------u= D Ac ti vi ty

DL session N pac ke t c – al l DL N pac ket

DLS pac ket

DL8 uu

R averageDL

1000 u------------------------------------------------------u=

R averageUL

R averageDL

DConnectionUL

D Ac ti vi ty UL session DInactivity

UL session+= DConnectionDL

D Act iv it y DL session DInactivity

DL session+=

pConnectionUL N sess

3600 -------------- DConnection

ULu= pConnectionDL N sess

3600 -------------- DConnection

DLu=

pConnected 1 1 pConnectionUL

– 1 pConnectionDL

– u – =

n j X pConnected u=

pConnected U L D L+ pConnection

UL

pConnection

DL

u pConnected

-----------------------------------------------------------------=

pConnected UL pConnection

UL1 pConnection

DL – u

pConnected

-------------------------------------------------------------------------------=

pConnected DL pConnection

DL1 pConnection

UL – u

pConnected

-------------------------------------------------------------------------------=

f UL D Ac ti vi ty

UL session

DInactivity UL session D Ac ti vi ty

UL session+ ------------------------------------------------------------------------------------------------------= f

DL D Ac ti vi ty DL session

DInactivity DL session D Ac ti vi ty

DL session+ ------------------------------------------------------------------------------------------------------=

pUL1

f UL

1 f DL

– u pConnected U L D L+

u=

pDL

1f DL

1 f UL

– u pConnected

U L D L+u=

pU L D L+1

f UL

f DLu pConnected

U L D L+u=

pinactive1

1 f – UL 1 f

DL – u pConnected

U L D L+u=

pUL2

f UL

pConnected ULu=





• 3rd case: At a given time, packet are downloaded (no packet is uploaded).

The user can be active on DL and inactive on UL; this probability is:


9. Calculation of number of users per activity status

Number of inactive users on UL and DL:

Number of users active on UL and inactive on DL:

Number of users active on DL and inactive on UL:

Number of users active on UL and DL:

Therefore, a user when he is connected can have four different activity status: either active on both links, or inactive on

both links, or active on UL only, or active on DL only.

6.4.1.2 Simulations Based on Sector Traffic Maps

Sector traffic maps can be based on live traffic data from OMC (Operation and Maintenance Centre). Traffic is spread over

the best server coverage area of each transmitter and each coverage area is assigned either the throughputs in the uplink

and in the downlink or the number of users per activity status or the total number of users (including all activity statuses).

6.4.1.2.1 Throughputs in Uplink and Downlink

When selecting Throughputs in Uplink and Downlink, you can input the throughput demands in the uplink and downlink

for each sector and for each listed service.

Atoll calculates the number of users active in uplink and in downlink in the Txi cell using the service ( N UL and N DL) as

follows:

and

is the kbits per second transmitted in UL in the Txi cell to supply the service.

is the kbits per second transmitted in DL in the Txi cell to supply the service.

is the downlink average requested rate defined for the service,

is the uplink average requested rate defined for the service.

N UL and N DL values include:

• Users active in uplink and inactive in downlink (ni (UL)),

• Users active in downlink and inactive in uplink (ni (DL)),

• And users active in both links (ni (UL+DL)).

Atoll takes into account activity periods during the connection in order to determine the activity status of each user.

Activity probabilities are calculated as follows:

Probability of being inactive in UL and DL:

Probability of being active in UL only:

Note:

• The user distribution per service and the activity status distribution between the users are

average distributions. And the service and the activity status of each user are randomlydrawn in each simulation. Therefore, if you compute several simulations at once, the

average number of users per service and average numbers of inactive, active on UL, active

on DL and active on UL and DL users, respectively, will correspond to calculated

distributions. But if you check each simulation, the user distribution between services as

well as the activity status distribution between users is different in each of them.

pinactive2

1 f – UL pConnected

ULu=

pDL3

f DL

pConnected DLu=

pinactive3

1 f – DL pConnected

DLu=

n j inactive

n j pinactive1

pinactive2

pinactive3

+ + u=

n j UL n j pUL1

pUL2

+ u=

n j DL n j pDL1

pDL3

+ u=

n j UL DL+ n j pU L D L+

1u=

N UL

R t UL

R averageUL

----------------------= N DL

R t DL

R averageDL

----------------------=

R t UL

R t DL

R averageDL

R averageUL

pinactive 1 f ac t UL

– 1 f act DL

– u=

pUL f act UL

1 f act DL

– u=



© Forsk 2010 AT282_TRG_E2 217


Probability of being active in DL only:

Probability of being active both in UL and DL:

Where, and are respectively the UL and DL activity factors defined for the service i.

Then, Atoll calculates the number of users per activity status:We have:

Therefore, we have:

Number of users active in UL and DL both:

Number of users active in UL and inactive in DL:

Number of users active in DL and inactive in UL:

Number of inactive users in UL and DL:

Therefore, a connected user can have four different activity status: either active in both links, or inactive in both links, or

active in UL only, or active in DL only.

6.4.1.2.2 Total Number of Users (All Activity Statuses)

When selecting Total Number of Users (All Activity Statuses), you can input the number of connected users for each

sector and for each listed service ( ).

Atoll takes into account activity periods during the connection in order to determine the activity status of each user.

Activity probabilities are calculated as follows:

Probability of being inactive in UL and DL:

Probability of being active in UL only:

Probability of being active in DL only:

Probability of being active both in UL and DL:

Where, and are respectively the UL and DL activity factors defined for the service i.

Then, Atoll calculates the number of users per activity status:

Number of inactive users in UL and DL:

Number of users active in UL and inactive in DL:

Number of users active in DL and inactive in UL:

Number of users active in UL and DL both:

Therefore, a connected user can have four different activity status: either active in both links, or inactive in both links, or

active in UL only, or active in DL only.

6.4.1.2.3 Number of Users per Activity Status

When selecting Number of Users per Activity Status, you can directly input the number of inactive users ( ), the

number of users active in the uplink ( ), in the downlink ( ) and in the uplink and downlink ( ), for

each sector and for each service.

Note:


pDL f act DL

1 f act UL

– u=

pU L D L+ f act UL

f ac t DL

u=

f act UL

f act DL

f act UL

f act DL

pUL pU L D L++ n j UL n j DL n j U L DL+ + + u N UL=

pDL pU L D L++ n j UL n j DL n j U L DL+ + + u N DL=

ni UL DL+ mi nN UL pU L D L+u

pUL pU L D L++-------------------------------------

N DL pU L D L+u

pDL pU L D L++-------------------------------------© ¹

§ ·=

ni UL N UL ni U L DL+ – =

ni DL N DL ni U L DL+ – =

ni inactive n j UL n j DL n j UL DL+ + +

1 pinactive –

-------------------------------------------------------------------------------------- pinactiveu=

Note:


ni

pinactive 1 f act UL

– 1 f act DL

– u=

pUL f act UL

1 f act DL

– u=

pDL f act DL

1 f act UL

– u=

pU L D L+ f act UL

f ac t DLu=

f act UL

f act DL

f act UL

f act DL

ni inactive

ni pinactiveu=

ni UL ni pULu=

ni DL ni pDLu=

ni UL DL+ ni pU L D L+u=

ni inactive

ni UL ni DL ni UL DL+




6.4.2 Power Control Simulation

The power control algorithm simulates the way a UMTS network regulates itself by using uplink and downlink power

controls in order to minimize interference and maximize capacity.

HSDPA users (i.e., Packet (HSDPA), Packet (HSPA) and Packet (HSPA - Constant Bit Rate) service users) are linked to

the A-DPCH radio bearer (an R99 radio bearer). Therefore, the network uses a A-DPCH power control on UL and DL and

then it performs fast link adaptation on DL in order to select an HSDPA radio bearer. For HSUPA users (i.e., Packet

(HSPA) and Packet (HSPA - Constant Bit Rate) service users), the network first uses a E-DPCCH/A-DPCH power control

on UL and DL, checks that there is an HSDPA connection on downlink and then carries out noise rise scheduling in order

to select an HSUPA radio bearer on uplink. Atoll simulates these network regulation mechanisms with an iterative algo-

rithm and calculates, for each user distribution, network parameters such as cell power, mobile terminal power, active set

and handoff status for each terminal. During each iteration of the algorithm, all the users (i.e., Circuit (R99), Packet (R99),

Packet (HSDPA), Packet (HSPA) and Packet (HSPA - Constant Bit Rate) service users) selected during the user distribu-

tion generation (1st step) attempt to connect one by one to network transmitters. The process is repeated until the network

is balanced, i.e., until the convergence criteria (on UL and DL) are satisfied.

Note:

• The activity status distribution between users is an average distribution. In fact, in each

simulation, the activity status of each user is randomly drawn. Therefore, if you compute

several simulations at once, average numbers of inactive, active on UL, active on DL and

active on UL and DL users correspond to the calculated distribution. But if you check each

simulation, the activity status distribution between users is different in each of them.

Figure 6.2: UMTS HSPA Power Control Algorithm

R99 part

Initialisation

Mi Best Server Determination

Mi Active Set Determination

UL and DL Interference Update

For each R99, HSDPAand HSUPA mobile, Mi

DL Power Control

UL Power Control

Congestion and Radio Resource Control

For HSDPA users, this part of the algorithm is performed for the A-DPCH bearer (R99 bearer)For HSUPA users, this part isperformed for the E-DPCCH/A-DPCH bearer (R99 bearer)

HSDPA part

Fast Link Adaptation

Mobile Scheduling

Radio Resource Control

For each HSDPA andHSUPA mobile, Mi

HSUPA part

Admission Control

For each HSUPAmobile, Mi

Radio Resource Control

Convergence Study

Noise Rise Scheduling



© Forsk 2010 AT282_TRG_E2 219


As shown in Figure 6.2: on page 218, the simulation algorithm is divided in three parts. All users are evaluated by the R99

part of the algorithm. HSDPA and HSUPA bearer users, unless they have been rejected during the R99 part of the algo-

rithm, are then evaluated by the HSDPA part of the algorithm. Finally, HSUPA bearer users, unless they have been

rejected during the R99 or HSDPA parts of the algorithm, are then evaluated by the HSUPA part of the algorithm.

The steps of this algorithm are detailed below.

6.4.2.1 Algorithm Initialization

The total power transmitted by the base station txi on the carrier , , is initialised to

. Uplink received powers by

the base station txi on carrier , , and are initialised to 0 W (i.e.

no connected mobile).

6.4.2.2 R99 Part of the Algorithm

The algorithm is detailed for any iteration k . X k is the value of the X (variable) at the iteration k . In the algorithm, all

and thresholds depend on the user mobility type and are defined in the R99 bearer property dialogue. All variables

are described in Definitions and formulas part.

Here, the rate downgrading is not taken into account.

The algorithm applies to single frequency band networks and to dual-band networks. Dual-band terminals can have thefollowing configurations:

- Configuration 1: The terminal can work on f1 and f2 without any priority (select "All" as main frequency band

in the terminal property dialogue).

- Configuration 2: The terminal can work on f1 and f2 but f1 has a higher priority (select "f1" as main frequency

band and "f2 " as secondary frequency band in the terminal property dialogue).

For each mobile M b

Determination of M b’s Best Server

For each transmitter txi containing M b in its calculation area and working on the main frequency band supported by the

M b’s terminal (i.e. either f1 for a single frequency band network, or f1 or f2 for a dual-band terminal with the configuration

1, or f1 for a dual-band terminal with the configuration 2).

Calculation of

If user selects “without Pilot”

Determination of the best transmitter, tx BS , for each carrier ic .

For each carrier ic , selection of the transmitter with the highest , .

Analysis of candidate cells, (tx BS ,ic).

For each pair (tx BS,ic), calculation of the uplink load factor:

Rejection of bad candidate cells if the pilot is not received or if the uplink load factor is exceeded during the admission load

control (if simulation respects a loading factor constraint and M b was not connected in previous iteration)

If then (tx BS,ic) is rejected by M b

If , then (tx BS,ic) is rejected by M b

Else

Keep (tx BS,ic) as good candidate cell

ic m P Tx tx i ic m

P pi lot tx i ic m P SC H tx i ic m P otherCCH tx i ic m P HSDPA tx i ic m P HSUPA tx i ic + + + +

ic m I tot UL intra

tx i ic m I tot ULextra

tx i ic m I in ter carr ier – UL tx i ic m

X k UL

tx i ic m I tot UL

tx i ic m

N tot UL

tx i ic m ---------------------------------- 0 = =

Qre qUL

Qre qDL

Q pil ot k

tx i ic Mb D UBT Su P u

c tx i M b ic

P tot DL tx i ic I extra

DL ic I in ter carr ier – DL ic I i nt er t echn o y log –

DL ic N + + 0 Term+ +

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pil ot k tx i ic Mb

D UBT Su P uc

tx i M b ic

I intraDL

ic I extraDL



ic + + +

N 0 Term

1 D – UBTSu P c tx i M b ic u – +© ¹¨ ¸¨ ¸§ ·-----------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pi lot k tx i M b ic tx BS M b

X k UL

tx BS ic I tot UL

tx BS ic

N tot UL

tx BS ic ----------------------------------- ' X

UL+=

Q pil ot k tx BS M b ic Qre q

pi lo t Mobil i ty M b

X k UL

tx BS ic X max UL!




For dual band terminals with the configuration 1 or terminals working on one frequency band only, if no good candidate

cell has been selected, M b has failed to be connected to the network and is rejected.

For dual band terminals with the configuration 2, if no good candidate cell has been selected, try to connect M b to trans-

mitters txi containing M b in their calculation area and working on the secondary frequency band supported by the M b’s

terminal (i.e. f2). If no good candidate cell has been selected, M b has failed to be connected to the network and is rejected.

For each NodeB having candidate cells, determination of the best carrier, ic BS , within the set of candidate cells

of the NodeB.

If a given carrier is specified for the service requested by M b

is the carrier specified for the service

Else the carrier selection mode defined for the site equipment is considered.

If carrier selection mode is “Min. UL Load Factor”

is the cell with the lowest

Else if carrier selection mode is “Min. DL Total Power”

is the cell with the lowest

Else if carrier selection mode is “Random”

is randomly selected

Else if carrier selection mode is "Sequential"

is the first carrier where

Endif

is the best serving cell ( ) and its pilot quality is

In the following lines, we will consider as the carrier used by the best serving cell

Active Set Determination

For each station txi containing M b in its calculation area, using , and, if neighbours are used, neighbour of

Calculation of


Rejection of txi from the active set if difference with the best server is too high

If then txi is rejected

Else txi is included in the M b active set

Rejection of a station if the mobile active set is full

Station with the lowest in the active set is rejected

EndFor

Uplink Power Control

Calculation of the terminal power required by M b to obtain the R99 radio bearer:

For each cell (txi,ic) of the M b active set

Calculation of quality level on M b traffic channel at (txi,ic), with the minimum power allowed on traffic channel for the M b

service

ic BS M b

ic BS M b X k UL

tx BS ic

ic BS M b P tx tx BS ic k

ic BS M b

ic BS M b X k UL

tx BS ic X max ULd

tx BS ic BS( , )k

M b BestCell k M b Q pi lot k

max M b

ic

ic

BestCell k M b

Q pil ot k tx i M b ic

D UBTSu P uc

tx i M b ic

P tot DL

tx i ic I extraDL



ic N 0 Term

+ + + +-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pil ot k tx i M b ic

D UBT Su P uc

tx i M b ic

I intraDL

tx i ic I extraDL


ic I i nt er t echn o y log – DL

ic + + +

N 0 Term

1 D – UBTSu P c tx i M b ic u – +© ¹¨ ¸¨ ¸§ ·---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pil ot k

max M b Q pil ot k

tx i M b ic – AS_Th BestCell k M b !

Q pi lo t k

P termR99 req –

M b ic k

P b R99 – UL

tx i M b ic P term R99 –

reqM b ic

k 1 –

LT tx i M b ---------------------------------------------------------=

P b DPDCH – UL

tx i M b ic P b R99 – UL

tx i M b ic 1 r c UL

– u=



© Forsk 2010 AT282_TRG_E2 221


if the user is active,

if the user is inactive,

If user selects "Total noise",

End For

If (M b is in not in handoff)

Else if (M i is in softer handoff)

Else if (M b is in soft, or softer/soft without MRC)

Else if (M b is in soft/soft)

Else if (M b is in softer/soft with MRC)

End If

If compressed mode is operated,

If then

If then M b cannot select any cell and its active set is cleared

If then M b cannot be connected

Endif

Note:

• Compressed mode is operated if:

- M i and S j support compressed mode,

And

- Either if the Ec/I0 Active option is selected,

- Or if the RSCP Active option is selected.

P b DPCCH –

ULtx i M b ic P b R99 –

ULtx i M b ic r c

ULu=

P b R99 –

ULtx i M b ic P b DPCCH –

ULtx i M b ic P b DPDCH –

ULtx i M b ic +=

P b R99 – UL

tx i M b ic P b DPCCH – UL

tx i M b ic =

QtchUL

tx i M b ic k

Uterm P ub DPDCH –

ULtx i M b ic

k

N to t UL

tx i ic 1 F MUDTx

– Utermu P b R99 – UL

tx i M b ic k 1 – u – -------------------------------------------------------------------------------------------------------------------------------------------------------- G p

ULService M b Gdiv

ULuu=

QtchUL

tx i M b ic k

Uterm P ub DPDCH –

ULtx i M b ic

k

N tot UL

tx i ic --------------------------------------------------------------------------------- G p

ULService M b Gdiv

ULuu=

Qk UL

M b QtchUL

tx i M b ic k =

Qk UL

M b f rake efficiency UL

QtchUL

tx i M b ic k

txi Act iveSet ¦u=

Qk UL

M b Ma x

tx i Act iveSet

QtchUL

tx i M b ic k

Gmacro diversity – UL 2 linksu=

Qk UL

M b Ma x

tx i Act iveSet

QtchUL

tx i M b ic k


Qk UL

M b Max f rake efficiency UL

QtchUL

ic

txi Act iveSet samesite

¦u Qother site

tch

ULic

© ¹¨ ¸¨ ¸¨ ¸§ ·


P term R99 – re q

M b ic k

QreqUL

Service M b Mobil i ty M b

Qk UL

M b -------------------------------------------------------------------------------------------- P term R99 –

reqM b ic k 1 – u=

Q pi lot k

Resulting tx i M b ic Q pi lot

CM act ivat ion – d

P c tx i M b ic RSCP pi lo t CM act ivat ion – d

P term R99 – re q

M b ic k

QreqUL

Service M b Mobil i ty M b 'Qre qUL

Service M b Mobil i ty M b u

Qk UL

M b --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- P term R99 –

reqM b ic k 1 – u=

P term R99 –

reqM b ic k P term

minM b P term R99 –

re qM b ic k P term

mintx i M b =

P term R99 – req

M b ic k P termmax

M b !

R nominal UL

M b R ma x UL

tx i ic t




Downlink Power Control

If (mobile does not use a packet switched service that is inactive on the downlink)

For each cell (tx i ,ic) in M b active set

Calculation of quality level on (txi,ic) traffic channel at M b with the minimum power allowed on traffic channel for the M b

service

If the user selects the option "Total noise"

End For

Do

For each cell (txi,ic) in M b active set

Calculation of the required power for DL traffic channel between (txi,ic) and M b:

If compressed mode is operated.

If then is set to

Recalculation of a decreased (a part of the required quality is managed by the cells set to )

If the user is inactive, then his contribution to interference in the calculation of is .

EndFor

While and M b active set is not empty

If then M b cannot be connected

Endif

Note:

• Compressed mode is operated i f:

- M i and S j support compressed mode,

And

- Either if the Ec/I0 Active option is selected,

- Or if the RSCP Active option is selected.

P bDL

tx i M b ic P tch

mi nService M b

LT tx i M b -----------------------------------------------------=

QtchDL tx i M b ic k UBTS P u b

DL

tx i M b ic k

N tot DL

ic 1 F ortho – UBT Su P bDL

tx i M b ic k 1 –

u – ------------------------------------------------------------------------------------------------------------------------------------ G p

DL Service M b u Gdiv DLu=

QtchDL

tx i M b ic k

UBTS P ub

DLtx i M b ic

k

N tot DL

ic -------------------------------------------------------------- G p

DLService M b Gdiv

DLuu=

Qk DL

M b f rake efficiency DL

QtchDL

tx i M b ic k

txi Act iveSet ¦u=

P tchre q

tx i M b ic k

QreqDL


Qk DL

M b -------------------------------------------------------------------------------------------- P tch

minService M b u=

P tchre q

tx i M b ic k

QreqDL

Service M b Mobil i ty M b 'QreqDL

Service M b Mobil i ty M b u

Qk DL

M b --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- P tch

minService M b u=

Q pi lo t k

Resulting

tx i M b ic Q pil ot

CM act ivat ion –

d

P c tx i M b ic RSCP pil ot CM act ivat ion –

d

P tc hre q

tx i M b ic k P tchmax

Service M b ! tx i ic P tchma x

Qre qDL

P tchmax

P bDL

tx i M b ic P tch

re qService M b

LT tx i M b ----------------------------------------------------=

QtchDL

tx i M b ic k

UBT S P ub

DLtx i M b ic

N tot DL

ic 1 F ortho – UBT Su P bDL

tx i M b ic u – --------------------------------------------------------------------------------------------------------------------------- G p

DLService M b u Gdiv

DLu=

N tot DL

ic P bDL

tx i M b ic r c DLu

Qk DL

M b f rake efficiency DL

QtchDL

tx i M b ic k

txi Act iveSet ¦u=

Qk DL

M b Qre qDL


R nominal DL

M b R max DL

tx i ic t



© Forsk 2010 AT282_TRG_E2 223


Uplink and Downlink Interference Update

Update of interference on active mobiles only (old contributions of mobiles and stations are replaced by the new ones).

For each cell (txi,ic)

Update of

EndFor

For each mobile M i

Update of

EndFor

EndFor

Control of Radio Resource Limits (OVSF Codes, Cell Power, Channel Elements, Iub BackhaulThroughput)


While

Rejection of the mobile with the lowest service priority starting from the last admitted

EndFor


While


EndFor

For each NodeB, Ni

While


While


EndFor

For each NodeB, Ni

While


While


EndFor

Uplink Load Factor Control

For each cell (txi,ic) with


EndFor

While at least one cell with exists

6.4.2.3 HSDPA Part of the Algorithm

Packet (HSDPA) and packet (HSPA) service users active on DL as well as all packet (HSPA - Constant Bit Rate) service

users (i.e., active and incative), unless they have been rejected during the R99 part of the algorithm, are then evaluated

by the HSDPA part of the algorithm.

6.4.2.3.1 HSDPA Power Allocation

The total transmitted power of the cell ( ) is the sum of the transmitted R99 power, the HSUPA power and the trans-

mitted HSDPA power.

N tot UL

tx i ic

N tot DL

ic

P tx tx i ic k

P max

------------------------------ %Power max DL!

N Codes tx i ic k N ma x Codes tx i ic !

N C E D L –

N i k N max C E D L –

N i !

N C E U L –

N i k N max C E U L –

N i !

T Iu bDL

N I k T Iub m – ax DL

N I !

T Iu bUL

N I k T Iub m – ax UL

N I !

X UL

tx i ic X max UL!

X UL

tx i ic X ma x UL!

P tx ic




• In case of a static HSDPA power allocation strategy, Atoll checks in the simulation that:

where:

is the maximum DL load allowed.

Therefore, if the maximum DL load is set to 100%, we have:

• In case of dynamic HSDPA power allocation strategy, Atoll checks in the simulation that:

And it calculates the available HSDPA power as follows:

6.4.2.3.2 Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users

The number of HS-SCCH channels ( ) is the maximum number of HS-SCCH channels that the cell can manage.

This parameter is used to manage the number of packet (HSDPA) and packet (HSPA) service users simultaneously

connected to an HSDPA bearer. This parameter is not taken into account for packet (HSPA - Constant Bit Rate) service

users as HS-SCCH-less operation (i.e., HS-DSCH transmissions without any accompanying HS-SCCH) is performed.

Each packet (HSDPA) and packet (HSPA) service user consumes one HS-SCCH channel. Therefore, at a time (over a

transmission time interval), the number of these users connected to an HSDPA bearer cannot exceed the number of HS-SCCH channels per cell.

The maximum number of HSDPA users ( ) corresponds to the maximum number of HSDPA bearer users that the cell

can support. Here, all HSDPA bearer users, i.e., packet (HSDPA) service users, packet (HSPA) service users and packet

(HSPA - Constant Bit Rate) service users, are taken into consideration.

Let us assume there are 30 HSDPA bearer users in the cell:

• 10 packet (HSPA - Constant Bit Rate) service users with any activity status.

• 20 packet (HSDPA) and packet (HSPA) service users active on DL.

All users are connected to the A-DCH R99 bearer. Finally, the number of HS-SCCH channels and the maximum number

of HSDPA users respectively equal 4 and 25.

The scheduler manages the maximum number of users within each cell. Packet (HSPA - Constant Bit Rate) service users

have the highest priority and are processed first, in the order established during the generation of the user distribution.

After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler ranks the remaining HSDPA bearer

users (i.e., packet (HSDPA) and packet (HSPA) service users) according to the selected scheduling technique. Users are

treated as described in the figure below.

• All packet (HSPA - Constant Bit Rate) service users may be served if there are enough HSDPA power, Iub back-

haul throughput and OVSF codes available in order for them to obtain the lowest HSDPA bearer that provides a

RLC peak rate higher or equal to the guaranted bit rate defined for the service. In this case, they will be connected.

Else, they will be rejected.

• Then, among the packet (HSDPA) and packet (HSPA) service users:

- The first four users may be simultaneously served if there are enough HSDPA power, Iub backhaul throughputand OVSF codes available in order for them to obtain an HSDPA bearer. In this case, they will be connected.

Else, they will be delayed.

- The next eleven ones will be delayed since there are no longer HS-SCCH channels available. Their connec-

tion status will be "HS-SCCH Channels Saturation".

- Finally, the last five users will be rejected beacuse the maximum number of HSDPA user has been fixed to

25. Their connection status will be "HSDPA Scheduler Saturation".

6.4.2.3.3 HSDPA Bearer Allocation Process

The HSDPA bearer allocation process depends on the type of service requested by the user. As explained before, packet

(HSPA - Constant Bit Rate) service users have the highest priority and are processed first, in the order established during

the generation of the user distribution. After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler

ranks the remaining HSDPA bearer users (i.e., packet (HSDPA) and packet (HSPA) service users) and shares the cell

radio resources between them.

P tx ic P t x R99 – ic P tx H – SDPA ic P HSUPA ic + +=

P tx ic P max ic %Power max DL

ud

%Power max DL

P tx ic P max ic d

P t x R99 – ic P HSUPA ic + P max ic %Power max DLud

P HSDPA ic P max ic P Headroom ic – P t x R99 – ic – P HSUPA ic – =

Figure 6.3: Connection status of HSDPA bearer users

nHS SCCH –

nmax



© Forsk 2010 AT282_TRG_E2 225


Packet (HSPA - Constant Bit Rate) Service Users

Let us focus on the ten packet (HSPA - Constant Bit Rate) service users mentionned in the example of the previous para-

graph "Number of HS-SCCH Channels and Maximum Number of HSDPA Bearer Users" on page 224. Fast link adaptation

is carried out on these users in order to determine if they can obtain an HSDPA bearer that provides a RLC peak rate

higher or equal to the service guaranteed bit rate. As HS-SCCH less operation is performed, only HSDPA bearers using

the QPSK modulation and two HS-PDSCH channels at the maximum can be selected and allocated to the users. The

users are processed in the order established during the generation of the user distribution and the cell’s available HSDPA

power is shared between them as explained below. Several Packet (HSPA - Constant Bit Rate) service users can share

the same HSDPA bearer. Then, Atoll calculates the HSDPA bearer consumption ( in %) for each user and takes into

account this parameter when it determines the resources consumed by the user (i.e., the HSDPA power used, the number

of OVSF codes and the Iub backhaul throughput).In the bearer allocation process shown below, the 10 packet (HSPA - Constant Bit Rate) service users are represented by

M j, with j = 1 to 10. And, the initial values of their respective HSDPA powers is 0, i.e. PHSDPA(B(MX)) = 0, where X = 0 to

10. These power values are assigned one by one by the scheduler, so that with their allocated values, looped back to the

starting point, are used in successive steps.

Packet (HSDPA) and Packet (HSPA) Service Users

After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler share the cell’s remaining resources

between packet (HSDPA) and packet (HSPA) service users. Let us focus on the packet (HSDPA) and packet (HSPA) serv-

For the user, M j, with j varying from 1 to 10:

Figure 6.4: HSDPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users

C




ice users, especially on the first four users mentionned in the example of the previous paragraph, "Number of HS-SCCH

Channels and Maximum Number of HSDPA Bearer Users" on page 224. A new fast link adaptation is carried out on these

users in order to determine if they can obtain an HSDPA bearer. They are processed in the order defined by the scheduler

and the cell’s HSDPA power available after all Packet (HSPA - Constant Bit Rate) service users have been served is

shared between them as explained below.

In the bearer allocation process shown below, the 4 packet (HSDPA) and packet (HSPA) service users are represented

by M j, with j = 1 to 4. And, the initial values of their respective HSDPA powers is 0, i.e. PHSDPA(B(MX)) = 0, where X = 0

to 4. These power values are assigned one by one by the scheduler, so that with their allocated values, looped back to

the starting point, are used in successive steps.

6.4.2.3.4 Fast Link Adaptation Modelling

Fast link adaptation (or Adaptive Modulation and Coding) is used in HSDPA. The power on the HS-DSCH channel is trans-

mitted at a constant power while the modulation, the coding and the number of codes are changed to adapt to the radio

conditions variations. Based on the reported channel quality indicator (CQI), the node-B may change every 2ms the modu-

lation (QPSK, 16QAM, 64QAM), the coding and the number of codes during a communication.

Atoll calculates for each user either the best pilot quality (CPICH Ec/Nt) or the best HS-PDSCH quality (HS-PDSCH Ec/

Nt); this depends on the option selected in Global parameters (HSDPA part): CQI based on CPICH quality or CQI based


Figure 6.5: HSDPA Bearer Allocation Process for Packet (HSDPA) and Packet (HSPA) Service Users



© Forsk 2010 AT282_TRG_E2 227


on HS-PDSCH quality (CQI means channel quality indicator). Then, it determines the HS-PDSCH CQI, calculates the best

bearer that can be used and selects the suitable bearer so as to comply with cell and terminal user equipment HSDPA

capabilities. Once the bearer selected, Atoll finds the highest downlink rate that can be provided to the user and may

calculate the application throughput.

CQI Based on CPICH Quality

When the option “CQI based on CPICH quality” is selected, Atoll proceeds as follows.

1. CPICH Qual ity Calculation

Let us assume the following notation: corresponds to the CPICH quality.

Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total

noise.

Therefore, we have:

for the total noise option,

And

for the without useful signal option.

With

ic adj is a carrier adjacent to ic .

is the interference reduction factor, defined between ic and ic adj

and set to a value different from 0.

is the inter-technology interference at the receiver on ic .

is the interfering carrier of an external transmitter

is the inter-technology Channel Protection between the signal transmitted by Tx and received by m assuming

the frequency gap between (external network) and .

(1)

, and are defined in "Inputs" on page 204.

1. In the HSDPA coverage prediction, is calculated as follows:

)

Ec

Nt ------- ic © ¹

§ · pi lo t

Ec

Nt ------- ic © ¹

§ · pi lo t

UBTS Du P c i ic u

N tot DL

ic ----------------------------------------------=

Ec

Nt ------- ic © ¹

§ · pi lo t

UBT S Du P c i ic u

N tot DL

ic 1 D – UBT S P c i ic uu –

---------------------------------------------------------------------------------------=

N tot

DL

ic I intra

DL

ic I extra

DL

ic I in ter carr ier –

DL

ic I i nt er t e chn o y log –

DL

ic N + + 0

term

+ +=

I intraDL

ic P tot DL

ic

tx i

UBTS 1 F – MUDterm 1 D – uu+ P to t

DLic

tx i

P SC H ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

UBT S – u P to t DL

ic

tx i

P SC H ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

u=

I extraDL

ic P to t DL

ic

t xj j i z¦=


ic

P tot DL

ic adj

t xj j ¦RF ic ic adj

---------------------------------------=

RF ic ic adj


ic


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic

P c i ic

P pil ot ic

LT i

-----------------------=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=

UBTS D N 0 term




2. CPICH CQI Determination

Let us assume the following notation: corresponds to the CPICH CQI. is read in the table

. This table is defined for the terminal reception equipment and the selected mobility.

3. HS-PDSCH Quality Calculat ion

Atoll proceeds as follows:

1st step: Atoll calculates the HS-SCCH power ( ).

is the HS-SCCH power on carrier ic . It is either fixed by the user (when the option “HS-SCCH Power

Dynamic Allocation”in the cell property dialogue is unchecked) or dynamically calculated (when the option “HS-SCCH

Power Dynamic Allocation” is selected).

In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ). It

is specified in mobility properties.

We have:


And


With


is the interference reduction factor, defined between ic and ic adj and set to a value different from 0.





Note:

• Atoll performs intra-cell interference computations based on the total power. You can

instruct Atoll to use maximum power by adding the following lines in the Atoll.ini file:

[CDMA]

PmaxInIntraItf = 1

In this case, Atoll considers the following formula:

I intraDL

ic P max ic

LT

----------------------- UBTS 1 F – MUDterm

1 D – uu+P max ic P –

SCH ic

LT

------------------------------------------------------© ¹§ · UBT S – u

P max ic P – SCH

ic

LT

------------------------------------------------------© ¹§ ·u=

CQ I pi lot CQ I pil ot

CQ I pi lot f Ec

Nt ------- ic © ¹

§ · pil ot

© ¹§ ·=

P HS SCCH –

P HS SCCH – ic

Ec

Nt ------- ic © ¹§ · HS SCCH –

re q

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBT S P c i ic u

N to t DL

ic ------------------------------------=

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBT S P c i ic u

N tot DL

ic 1 F ortho – 1 F MUDterm

– u UBT S P c i ic uu –

---------------------------------------------------------------------------------------------------------------------------------------=

N tot

DLic I

intra

DLic I

extra

DLic I

in ter carr ier –

DLic I

i nt er t echn o y log –

DLic N

0

term+ + + +=

I intraDL

ic P tot DL

ic

tx i

UBT S 1 F – MUDterm 1 F ortho – uu+ P tot

DLic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

UBT S – u P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

u=

I extraDL

ic P tot DL

ic

t xj j i z¦=


ic

P tot DL

ic adj


---------------------------------------=

RF ic ic adj


ic


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx m

u------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic



© Forsk 2010 AT282_TRG_E2 229


and

(2)

, , and are defined in "Inputs" on page 204.

Therefore,


And


2nd step: Atoll calculates the HS-PDSCH power ( ).

is the power available for HSDPA on the carrier ic . This parameter is either a simulation output, or a user-

defined cell input.

Therefore, we have:

is the number of HS-SCCH channels.

3rd step: Then, Atoll evaluates the HS-PDSCH quality

Let us assume the following notation: corresponds to the HS-PDSCH quality.

We have:


And


Here, Atoll works on the assumption that five HS-PDSCH channels are used (n=5).

With


)

P c i ic

P HS SCCH – ic

LT i

---------------------------------------=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=

UBTS F ortho F MUDterm

N 0 term

P HS SCCH – ic

Ec Nt ------- ic © ¹

§ ·HS SCCH –

reqN tot

DLic u

UBTS

---------------------------------------------------------------------------

© ¹¨ ¸¨ ¸¨ ¸§ ·

LT i u=

P HS SCCH – ic

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

re q

N tot DL

ic u

UBT S 1 1 F ortho – 1 F MUDterm

– uEc

Nt ------- ic © ¹

§ ·HS SCCH –

re q

u+© ¹§ ·u

----------------------------------------------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸¨ ¸§ ·

LT i u=

P HS PDSCH –

P HSDPA ic

P HSDPA ic P HS PDSCH – ic nHS SCCH – P uHS SCCH –

ic +=

P HS PDSCH – ic P HSDPA ic nHS SCCH – P uHS SCCH –

ic – =

nHS SCCH –

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

UBT S P c i ic u

N to t

DLic

------------------------------------=

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

UBTS P c i ic u

N tot DL

ic 1 F ortho – 1 F MU Dterm

– u UBTS

P c i ic

n-----------------uu –

---------------------------------------------------------------------------------------------------------------------------------------=

N tot DL

ic I intraDL

ic I extraDL



ic N + + 0

term+ +=

I intraDL

ic P tot DL

ic

tx i

UBTS 1 F – MUDterm

1 F ortho – uu+ P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

UBTS – u P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

u=

I extraDL

ic P tot DL

ic

t xj j i z¦=


ic

P tot DL

ic adj


---------------------------------------=










And

(3)


4. HS-PDSCH CQI Determination

The best bearer that can be used depends on the HS-PDSCH CQI. Let us assume the following notation:

corresponds to the HS-PDSCH CQI. Atoll calculates as follows:

5. HSDPA Bearer Selection

Atoll selects the HSDPA bearer associated to this CQI (in the table Best Bearer=f(HS-PDSCH CQI) defined for the termi-

nal reception equipment and the user mobility) and compatible with the user equipment and cell capabilities.

HSDPA bearers can be classified into two categories:

• HSDPA bearers using QPSK and 16QAM modulations: They can be selected for all users connected to HSPA and

HSPA+ capable cells. The number of HS-PDSCH channels required by the bearer must not exceed the maximum

number of HS-PDSCH codes available for the cell.

For packet (HSPA - Constant Bit Rate) service users, HS-SCCH-less operation (i.e., HS-DSCH transmissions

without any accompanying HS-SCCH) is performed. In this case, the UE is not informed about the transmission

format and has to revert to blind decoding of the transport format used on the HS-DSCH. Complexity of blind

detections in the UE is decreased by limiting the transmission formats that can be used (i.e., the HSDPA bearers

available). Therefore, only HSDPA bearers using the QPSK modulation and two HS-PDSCH channels at the maxi-

mum can be selected and allocated to these users. Additionally, the selected HSDPA bearer must provide a RLC

peak rate higher or equal to the guaranted bit rate defined for the service.

• HSDPA bearers using 64QAM modulation (improvement introduced by the release 7 of the 3GPP UTRA specifi-

cations, referred to as HSPA+): These HSDPA bearers can be allocated to packet (HSDPA) and packet (HSPA)

users connected to cells with HSPA+ capabilities only. The number of HS-PDSCH channels required by the bearer

must not exceed the maximum number of HS-PDSCH codes available for the cell. These HSDPA bearers cannot

be allocated to packet (HSPA - Constant Bit Rate) service users.

Atoll considers an HSDPA bearer as compatible with the user equipment if:

• The transport block size does not exceed the maximum transport block size supported by the user equipment.


)

Note:



[CDMA]

PmaxInIntraItf = 1


RF ic ic adj


DLic


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic

P c i ic

P HS PDSCH – ic

LT i

------------------------------------------=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=


N 0 term

I intraDL

ic P ma x ic

LT

----------------------- UBTS 1 F – MUDterm 1 F ortho – uu+

P max ic P – SCH

ic

LT

------------------------------------------------------© ¹§ · UBTS – u

P max ic P – SC H

ic

LT

------------------------------------------------------© ¹§ ·u=

CQ I HS PDSCH – CQ I HS PDSCH –

CQ I HS PDSCH – CQ I pi lot P pi lot – P HS PDSCH – +=



© Forsk 2010 AT282_TRG_E2 231


• The number of HS-PDSCH channels required by the bearer does not exceed the maximum number of HS-PDSCH

channels that the terminal can use.

• The modulation is supported by the user equipment.

When there are several HSDPA bearers compatible, Atoll selects the HSDPA bearer that provides the highest RLC peak

rate. When several HSDPA bearers can supply the same RLC peak rate, Atoll chooses the HSDPA bearer with the high-

est modulation scheme. Finally, if no HSDPA bearer is compatible, Atoll allocates a lower HSDPA bearer compatible with

the user equipment and cell capabilities which needs fewer resources.

Let’s consider the following examples.

Example1: One packet (HSDPA) user with category 13 user equipment and a 50km/h mobility.

The user equipment capabilities are:

• Maximum transport block size: 35280 bits

• Maximum number of HS-PDSCH channels: 15

• Highest modulation supported: 64QAM

• MIMO Support: No

The cell to which the user is connected supports HSPA+ functionalities (i.e. 64QAM modulation in the DL and MIMO

systems) and the maximum number of HS-PDSCH channels is 15.

1st case: The CQI experienced by the user equals 26. Therefore, Atoll can choose between two HSDPA bearers, the

bearer indexes 26 and 31.

Characteristics of the bearer index 26 are:

• Transport block size: 17237 bits

• Number of HS-PDSCH channels used: 12

• 16QAM modulation is used

• RLC Peak Rate: 8.32 Mb/s






Both HSDPA bearers are compatible with the user equipment and cell capabilities. Atoll selects the HSDPA bearer that

provides the highest RLC peak rate, i.e. the bearer index 26.

Figure 6.6: HSDPA UE Categories Table




2nd case: The CQI experienced by the user equals 27. Therefore, Atoll can choose between two HSDPA bearers, the

bearer indexes 27 and 32.











Both HSDPA bearers are compatible with the user equipment and cell capabilities and the RLC peak rate they provide is

the same. Atoll selects the HSDPA bearer using the highest modulation scheme, i.e. the bearer index 32.

Example 2: One packet (HSDPA) user experiencing a CQI of 26.

Therefore, Atoll can choose between two HSDPA bearers, the bearer indexes 26 and 31.









• 64QAM modulation is used• RLC Peak Rate: 7.36 Mb/s

1st case: The user equipment category is 9. The cell to which the user is connected supports HSPA+ functionalities (i.e.

64QAM modulation in the DL and MIMO systems) and the maximum number of HS-PDSCH channels is 15.

The user equipment characteristics are the following:





The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the terminal. Only the

bearer index 26 is compatible with the user equipment capabilities. Atoll selects it.

Figure 6.7: HSDPA Radio Bearers Table



© Forsk 2010 AT282_TRG_E2 233


2nd case: The user equipment category is 8. The cell to which the user is connected supports HSPA+ functionalities (i.e.

64QAM modulation in the DL and MIMO systems) and the maximum number of HS-PDSCH channels is 15.

The user equipment characteristics are the following:





Here, none of HSDPA bearers are compatible with the user equipment capabilities.

The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the terminal. With the

bearer index 26, the number of HS-PDSCH channels (12) exceeds the maximum number of HS-PDSCH channels the

terminal can use (10), and the transport block size (17237 bits) exceeds the maximum transport block size (14411 bits)

the terminal can carried.

In the HSDPA Radio Bearer table, Atoll selects a lower HSDPA bearer compatible with cell and UE category capabilities.

It selects the bearer index 25.

• The number of HS-PDSCH channels (10) does not exceed the maximum number of HS-PDSCH channels the ter-

minal can use (10) and the maximum number of HS-PDSCH channels available at the cell level (15),

• The transport block size (14411 bits) does not exceed the maximum transport block size (14411 bits) the terminal

can carried.

• 16QAM modulation is supported by the terminal and the cell.

3rd case: The user equipment category is 13. The cell to which the user is connected supports HSPA functionalities and

the maximum number of HS-PDSCH channels is 15.

The user equipment capabilities are:


• Maximum number of HS-PDSCH channels:15



The bearer index 31 cannot be selected because it requires a modulation scheme not supported by the cell. On the other

hand, the bearer index 26 is compatible with cell and UE category capabilities. Therefore, it is allocated.

6. HS-PDSCH Quality Update

Once the bearer selected, Atoll exactly knows the number of HS-PDSCH channels. Therefore, when the method “Without

useful signal” is used, it may recalculate the HS-PDSCH quality with the real number of HS-PDSCH channels (A default

value (5) was taken into account in the first HS-PDSCH quality calculation).

CQI Based on HS-PDSCH Quality

When the option “CQI based on HS-PDSCH quality” is selected, Atoll proceeds as follows.

1. HS-PDSCH Quality Calculation

Atoll proceeds as follows:

1st step: Atoll calculates the HS-SCCH power ( ).

is the HS-SCCH power on carrier ic . It is either fixed by the user (when the option “HS-SCCH Power

Dynamic Allocation”in the cell property dialogue is unchecked) or dynamically calculated (when the option “HS-SCCH

Power Dynamic Allocation” is selected).

In this case, the HS-SCCH power is controlled so as to reach the required HS-SCCH Ec/Nt (noted ). It

is specified in mobility properties.

We have:


And


With

P HS SCCH –

P HS SCCH – ic

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

re q

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBTS P c i ic u

N tot DL

ic ------------------------------------=

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBTS P c i ic u

N tot DL


– u UBTS P c i ic uu –

---------------------------------------------------------------------------------------------------------------------------------------=

N tot DL

ic I intraDL

ic I extraDL



ic N + + 0

term+ +=

I intraDL

ic P tot DL

ic

tx i

UBTS 1 F – MUDterm 1 F ortho – uu+ P tot

DLic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

UBTS – u P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

u=










And

(4)


Therefore,


And


2nd step: Atoll calculates the HS-PDSCH power ( )

is the power available for HSDPA on the carrier ic . This parameter is either a simulation output, or a user-

defined cell input.

Therefore, we have:

is the number of HS-SCCH channels.

3rd step: Then, Atoll evaluates the HS-PDSCH quality

Let us assume the following notation: corresponds to the HS-PDSCH quality.

Two options, available in Global parameters, may be used to calculate Nt: option Without useful signal or option Total

noise.

We have:


)

I extraDL

ic P tot DL

ic

t xj j i z¦=


ic

P tot DL

ic adj


---------------------------------------=

RF ic ic adj

I i nt er t echn o y log – DL ic


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx m

u------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic

P c i ic

P HS SCCH – ic

LT i

---------------------------------------=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=


N 0 term

P HS SCCH – ic

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

req

N to t DL

ic u

UBTS

---------------------------------------------------------------------------

© ¹¨ ¸¨ ¸¨ ¸§ ·

LT i u=

P HS SCCH –

ic

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

req

N to t DL

ic u

UBT S 1 1 F ortho – 1 F MUDterm

– u Ec Nt ------- ic © ¹

§ ·HS SCCH –

requ+© ¹

§ ·u

----------------------------------------------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸

¨ ¸§ ·

LT i

u=

P HS PDSCH –

P HSDPA ic

P HSDPA ic P HS PDSCH – ic nHS SCCH – P uHS SCCH –

ic +=

P HS PDSCH – ic P HSDPA ic nHS SCCH – P uHS SCCH –

ic – =

nHS SCCH –

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –



© Forsk 2010 AT282_TRG_E2 235



And


Here, Atoll works on the assumption that five HS-PDSCH channels are used (n=5). Then, it calculates the HS-PDSCH

CQI and the bearer to be used. Once the bearer selected, Atoll exactly knows the number of HS-PDSCH channels and

recalculates the HS-PDSCH quality with the real number of HS-PDSCH channels.With







And

(5)


2. HS-PDSCH CQI Determination


)

Note:



[CDMA]

PmaxInIntraItf = 1


Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

UBT S P c i ic u

N to t DL

ic ------------------------------------=

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

UBTS P c i ic u

N tot DL

ic 1 F ortho – 1 F MU Dterm

– u UBTS

P c i ic

n-----------------uu –

---------------------------------------------------------------------------------------------------------------------------------------=

N tot DL

ic I intraDL

ic I extraDL



ic N + + 0

term+ +=

I intraDL

ic P tot DL

ic

tx i


DLic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

UBTS – u P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹¨ ¸¨ ¸§ ·

u=

I extraDL

ic P tot DL

ic

t xj j i z¦=


ic

P tot DL

ic adj

t xj j ¦RF ic ic

adj

---------------------------------------=

RF ic ic adj


ic


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic

P c i ic

P HS PDSCH – ic

LT i

------------------------------------------=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=


N 0 term

I intraDL

ic P max ic

LT

----------------------- UBTS 1 F – MUDterm 1 F ortho – uu+

P max ic P – SCH

ic

LT

------------------------------------------------------© ¹§ · UBT S – u

P max ic P – SCH

ic

LT

------------------------------------------------------© ¹§ ·u=




Let us assume the following notation: corresponds to the HS-PDSCH CQI. is read in

the table . This table is defined for the terminal reception equipment and the

specified mobility.

3. HSDPA Bearer Selection

The bearer is selected as described in "HSDPA Bearer Selection" on page 230.

6.4.2.3.5 MIMO Modelling

MIMO - Transmit Diversity

If the cell to which the user is connected supports HSPA+ with transmit diversity and if the user’s terminal HSDPA UE

category supports MIMO, we have:

in dB

Where

is the downlink transmit diversity gain (in dB) corresponding to the numbers of transmission and reception antenna

ports (respectively defined in the transmitter and terminal properties).

is the additional diversity gain in downlink (in dB). It is defined for the clutter class of the user.

MIMO - Spatial Multiplexing

is the RLC peak rate that the selected HSDPA bearer ( ) can provide.

It is read in the HSDPA Radio Bearer table.

If the cell to which the user is connected supports HSPA+ with spatial multiplexing and if the user’s terminal HSDPA UE

category supports MIMO, we have:

Where

is the maximum spatial multiplexing gain (in dB) for a given number of transmission and reception antennas (respec-

tively defined in the transmitter and terminal properties).

is the spatial multiplexing gain factor defined for the clutter class of the user.

6.4.2.3.6 Scheduling Algorithms

The scheduler manages the maximum number of users within each cell. Packet (HSPA - Constant Bit Rate) service users

have the highest priority and are processed first, in the order established during the generation of the user distribution.

After processing the packet (HSPA - Constant Bit Rate) service users, the scheduler ranks the remaining HSDPA bearer

users (i.e., packet (HSDPA) and packet (HSPA) service users) according to the selected scheduling technique.Three

scheduling algorithms are available , Max C/I, Round Robin and Proportional Fair. Impact they have on the simulation

result is described in the tables below.

Let us consider a cell with 16 packet (HSDPA) and packet (HSPA) service users. All of them are active on DL and

connected to the A-DCH R99 bearer. There is no packet (HSPA - Constant Bit Rate) service user in the cell and the

number of HS-SCCH channels and the maximum number of HSDPA users have been respectively set to 4 and 15.

• Max C/I: 15 users (where 15 corresponds to the maximum number of HSDPA users defined) enters the scheduler

in the same order as in the simulation. Then, they are sorted in descending order by the channel quality indicator

(CQI), i.e. in a best bearer descending order.

CQ I HS PDSCH – CQ I HS PDSCH –

CQ I HS PDSCH – f Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

© ¹§ ·=

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

Ec

Nt ------- ic © ¹

§ ·HS PDSCH –

GTDDL

'GTDDL

+ +=

GTDDL

'GTDDL

R RLC peak – DL Index HSDPABearer Index HSDPABearer

R RLC peak – DL

R RLC peak – DL

Index HSDPABearer 1 f SM Ga in – GSM Max

1 – u+ u=

GSM Max

f SM Ga in –

Mobiles Simulation RankBest Bearer

(kbps)

DL Obtained Rate

(kbps)

Connection

StatusM1 2 2400 2400 Connected

M2 15 2400 1440 Connected


M4 9 2080 3.4 Delayed

M5 10 2080 3.4 Delayed

M6 12 2080 3.4 Delayed

M7 13 2080 3.4 Delayed

M8 14 2080 3.4 Delayed

M9 7 1920 3.4 Delayed



© Forsk 2010 AT282_TRG_E2 237


• Round Robin: Users are taken into account in the same order than the one in the simulation (random order).

• Proportional Fair: 15 users (where 15 corresponds to the maximum number of HSDPA users defined) enters the

scheduler in the same order as in the simulation. Then, they are sorted in an ascending order according to a new

random parameter which corresponds to a combination of the user rank in the simulation and the channel quality

indicator (CQI).

For a user i , the random parameter is calculated as follows:

Where,

is the user rank in the simulation.

is the user rank according to the CQI.

M10 1 1600 3.4 Delayed

M11 3 1600 3.4 Delayed

M12 4 1600 3.4 Delayed

M13 5 1600 3.4 Delayed

M14 6 1600 3.4 Delayed

M15 11 1440 3.4 Delayed

M16 16 2080 0 Scheduler Saturation

Mobiles Simulation RankBest Bearer

(kbps)

DL Obtained Rate

(kbps)

Connection

Status



M3 3 1600 3.4 Delayed

M4 4 1600 3.4 Delayed

M5 5 1600 3.4 Delayed

M6 6 1600 3.4 Delayed

M7 7 1920 3.4 Delayed

M8 8 2080 3.4 Delayed

M9 9 2080 3.4 Delayed

M10 10 2080 3.4 Delayed

M11 11 1440 3.4 Delayed

M12 12 2080 3.4 Delayed

M13 13 2080 3.4 Delayed

M14 14 2080 3.4 Delayed

M15 15 2400 3.4 Delayed

M16 16 2080 0 Scheduler Saturation

Note:

• You can change the default weights by editing the atoll.ini file. For more information, see

the Administrator Manual.

MobilesSimulation

RankCQI Rank RP

Best Bearer

(kbps)

DL Obtained

Rate (kbps)

Connection

Status

M1 2 1 150 2400 2400 Connected

M2 1 10 550 1600 960 Connected

M3 8 3 550 2080 160 Connected

M4 9 4 650 2080 3.4 Delayed

M5 3 11 700 1600 3.4 Delayed

M6 10 5 750 2080 3.4 Delayed

M7 4 12 800 1600 3.4 Delayed

RP i

RP i 50 R i Simu

u 50 R i CQI

u+=

R i Simu

R i CQI




6.4.2.4 HSUPA Part of the Algorithm

Packet (HSPA) service users active on UL as well as all packet (HSPA - Constant Bit Rate) service users (i.e., active and

incative), unless they have been rejected during the R99 or HSDPA parts of the algorithm, are then evaluated by the

HSUPA part of the algorithm. Atoll manages the maximum number of users within each cell. Packet (HSPA - Constant

Bit Rate) service users have the highest priority and are processed first, in the order established during the generation of

the user distribution. Then, Atoll considers packet (HSPA) service users in the order established during the generation of

the user distribution.

Let us assume there are 12 HSUPA bearer users in the cell:

• 3 packet (HSPA - Constant Bit Rate) service users with any activity status. All of them have been connected to an

HSDPA bearer.

• 9 packet (HSPA) service users active on UL. The first four packet (HSPA) have been connected to an HSDPAbearer, the last one has been rejected and the remaining four have been delayed in the HSDPA part.

Finally, the maximum number of HSUPA users equals 10.

In this case, Atoll will consider the first ten HSUPA bearer users only and will reject the last two users in order not to

exceed the maximum number of HSUPA users allowed in the cell (their connection status is "HSUPA scheduler satura-

tion").

6.4.2.4.1 Admission Control

During admission control, Atoll selects a list of HSUPA bearers for each user. The selected HSUPA bearers have to becompatible with the user equipment and capabilities of each HSUPA cell of the active set. For packet (HSPA - Constant

Bit Rate) service users, the list is restricted to HSUPA bearers that provide a RLC peak rate higher than the guaranteed

bit rate.

Let us focus on one packet (HSPA) service user with category 3 user equipment and a 50km/h mobility. This user is

connected to one cell only. The cell supports HSPA+ functionalities, i.e the cell supports QPSK and 16QAM modulations

in the UL.

HSUPA user equipment categories are provided in the HSUPA User Equipment Categories table. The capabilities of the

category 3 user equipment are:

• Maximum Number of E-DPDCH codes: 2

• TTI 2 ms: No so it supports 10 ms TTI

• Minimum Spreading Factor: 4

• Maximum Block Size for a 2ms TTI: no value

M8 7 9 800 1920 3.4 Delayed

M9 15 2 850 2400 3.4 Delayed

M10 5 13 900 1600 3.4 Delayed

M11 12 6 900 2080 3.4 Delayed

M12 6 14 1000 1600 3.4 Delayed

M13 13 7 1000 2080 3.4 Delayed

M14 14 8 1100 2080 3.4 Delayed

M15 11 15 1300 1440 3.4 Delayed

M16 16 - - 2080 0Scheduler

Saturation

Mobiles ServiceSimulation

Rank

HSDPA

Connection

Status

Evaluation by

the HSUPA

part of the

algorithm

M1 Packet (HSPA - Constant Bit Rate) 4 Connected Yes



M4 Packet (HSPA) 1 Connected Yes




M8 Packet (HSPA) 6 Delayed Yes



M11 Packet (HSPA) 11 Delayed No

M12 Packet (HSPA) 12 Rejected No



© Forsk 2010 AT282_TRG_E2 239


• Maximum Block Size for a 10ms TTI: 14484 bits

• Highest Modulation Supported: QPSK

HSUPA bearer characteristics are provided in the HSUPA Bearer table. An HSUPA bearer is described with following char-

acteristics:

• Radio Bearer Index: The bearer index number.

• TTI Duration (ms): The TTI duration in ms. The TTI can be 2 or 10 ms.

• Transport Block Size (Bits): The transport block size in bits.

• Number of E-DPDCH Codes: The number of E-DPDCH channels used.

• Minimum Spreading Factor: The smallest spreading factor used.

• Modulation: the modulation used (QPSK or 16QAM)

• RLC Peak Rate (bps): The RLC peak rate represents the peak rate without coding (redundancy, overhead,

addressing, etc.).

HSUPA bearers can be classified into two categories:

• HSUPA bearers using QPSK modulation: They can be selected for users connected to HSPA and HSPA+ capable

cells.

• HSUPA bearers using 16QAM modulation (improvement introduced by the release 7 of the 3GPP UTRA specifi-

cations, referred to as HSPA+). These HSUPA bearers can be allocated to users connected to cells with HSPA+

capabilities only.

Atoll considers an HSUPA bearer as compatible with the category 3 user equipment if:

• The TTI duration used by the bearer is supported by the user equipment (10 ms).

• The transport block size does not exceed the maximum transport block size supported by the user equipment

(14484 bits):

• The number of E-DPDCH channels required by the bearer does not exceed the maximum number of E-DPDCH

channels that the terminal can use (2).

• The minimum spreading factor used by the bearer is not less than the smallest spreading factor supported by the

terminal (4).

• The modulation required by the bearer is supported by the terminal.

The HSUPA bearers compatible with category 3 user equiment are framed in red:

Then, during admission control, Atoll checks that the lowest compatible bearer in terms of the required E-DPDCH Ec » Nt

does not require a terminal power higher than the maximum terminal power allowed.

Atoll uses the HSUPA Bearer Selection table. Among the compatible HSUPA bearers, Atoll chooses the one with the

lowest required Ec/Nt threshold.

Here, this is the index 1 HSUPA bearer; the required Ec/Nt threshold to obtain this bearer is -21.7dB.

Figure 6.8: HSUPA UE Categories Table

Figure 6.9: HSUPA Radio Bearers Table




Then, from the required Ec/Nt threshold, , Atoll calculates the required terminal power, .

With

(6)

, , , , and are defined in "Inputs" on page 204.

Atoll rejects the user if the terminal power required to obtain the lowest compatible HSUPA bearer ( )

exceeds the maximum terminal power (his connection status is "HSUPA Admission Rejection").

At the end of this step, the number of non-rejected HSUPA bearer users is . All of them will be connected to an

HSUPA bearer at the end.

6.4.2.4.2 HSUPA Bearer Allocation Process

The HSUPA bearer allocation process depends on the type of service requested by the user. As explained before, packet

(HSPA - Constant Bit Rate) service users have the highest priority and are processed first, in the order established during

the generation of the user distribution. After the admission control on packet (HSPA - Constant Bit Rate) service users,

Atoll performs a noise rise scheduling, followed by a radio resource control. Then, it repeats the same steps on packet

(HSPA) service users.

Packet (HSPA - Constant Bit Rate) Service Users

Let us focus on the three packet (HSPA - Constant Bit Rate) service users mentionned in the example of the previous

paragraph "HSUPA Part of the Algorithm" on page 238. We assume that all of them have been admitted. Noise rise sched-

uling and radio resource control are carried out on each user in order to determine the best HSUPA bearer that the user

can obtain. Several Packet (HSPA - Constant Bit Rate) service users can share the same HSUPA bearer. Then, Atoll

6. In the HSUPA coverage prediction, is calculated as follows:

)

Figure 6.10: HSUPA Bearer SelectionTable

Ec

Nt -------

© ¹§ ·

E DPDCH –

req

P term HSUPA – req

P term HSUPA – re q Ec

Nt -------

© ¹§ ·=

E DPDCH –

re q

LT N tot ULuu

N tot UL

ic 1 F – MUDtx Utermu I u tot

UL intraic I to t

ULextraic I in ter carr ier –

ULic N + 0

tx + +=

LT


GTx Gtermu

--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu------------------------------------------------------------------------------------------------------------------------------------------------------------ -=

Uterm F MUDtx

I tot

UL intraI tot

ULextraI in ter carr ier –

ULN 0

tx

P term HSUPA – re q

nHSUPA



© Forsk 2010 AT282_TRG_E2 241


calculates the HSUPA bearer consumption ( in %) for each user and takes into account this parameter when it deter-

mines the resources consumed by the user (i.e., the terminal power used, the number of channel elements and the Iub

backhaul throughput).

In the bearer allocation process shown below, the 3 packet (HSPA - Constant Bit Rate) service users are represented by

M j, with j = 1 to 3.

Packet (HSPA) Service Users

Let us focus on the seven packet (HSPA) service users mentionned in the example of the previous paragraph "HSUPA

Part of the Algorithm" on page 238. We assume that all of them have been admitted. Noise rise scheduling and radio

resource control are carried out on each user in order to determine the best HSUPA bearer that the user can obtain.

In the bearer allocation process shown below, the 7 packet (HSPA) service users are represented by M j, with j = 1 to 7.


Figure 6.11: HSUPA Bearer Allocation Process for Packet (HSPA - Constant Bit Rate) Service Users


Figure 6.12: HSUPA Bearer Allocation Process for Packet (HSPA) Service Users

C




6.4.2.4.3 Noise Rise Scheduling

Determination of the Obtained HSUPA Bearer

The obtained HSUPA radio bearer is the bearer that the user obtains after noise rise scheduling and radio resource control.

Packet (HSPA - Constant Bit Rate) service users have the highest priority and are processed first. Therefore, after the

admission control, the noise rise scheduling algorithm attempts to evenly share the remaining cell load between the packet

(HSPA - Constant Bit Rate) service users admitted in admission control; in terms of HSUPA, each user is allocated a right

to produce interference. The remaining cell load factor on uplink ( ) depends on the maximum load

factor allowed on uplink and how much uplink load is produced by the served R99 traffic. It can be expressed as follows:

Then, Atoll evenly shares the remaining cell load factor between the packet (HSPA - Constant Bit Rate) service users

admitted during the previous step ( ).

From this value, Atoll calculates the maximum E-DPDCH Ec » Nt allowed ( ) for each packet (HSPA -

Constant Bit Rate) service user. For further information on the calculation, see "Uplink Load Factor Due to One User" on

page 256.

for the Without useful signal option

for the Total noise option

Then, it selects an HSUPA bearer. The allocation depends on the maximum E-DPDCH Ec » Nt allowed and on UE and cell

capabilities. Atoll selects the best HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer

( ) with the highest potential throughput ( ) where:

•

• And

When several HSUPA bearers are available, Atoll selects the one with the lowest .

After the noise rise scheduling,Atoll carries out radio resource control, verifying if enough channel elements and Iub back-

haul throughput are available for the HSUPA bearer assigned to the user. For information on radio resource control, see

"Radio Resource Control" on page 245.

After processing all packet (HSPA - Constant bit rate) service users, Atoll carries out noise rise scheduling and radio

resource control on packet (HSPA) service users. During the noise rise scheduling, Atoll distributes the remaining cell

load factor available after all packet (HSPA - Constant Bit Rate) service users have been served. It can be expressed as

follows:

The remaining cell load factor is shared equally between the admitted packet (HSPA) service users ( ).

From this value, Atoll calculates the maximum E-DPDCH Ec » Nt allowed ( ) as explained above and selects

an HSUPA bearer for each packet (HSPA) service user. After the noise rise scheduling, Atoll carries out radio resource

control on packet (HSPA) service users. For information on radio resource control, see "Radio Resource Control" on

page 245.

Example: We have a cell with six packet (HSPA) service users and no packet (HSPA - Constant Bit Rate) user. All packet

(HSPA) service users have been admitted.

The remaining cell load factor equal to 0.6 is shared between the packet (HSPA) service users. Therefore, the UL load

factor alloted to each user is 0.1. Let’s take the cell UL reuse factor equal to 1.5. Atoll calculates the maximum E-DPDCH

Ec » Nt allowed (the Without useful signal option is selected).

' X HSPA CBR – UL

tx i ic

' X HSPA CBR –

ULtx i ic X ma x

ULtx i ic X R99

ULtx i ic – =

nHSPA CBR –

' X user UL

tx i ic ' X HSPA CBR –

ULtx i ic

nHSPA CBR –

--------------------------------------------------------=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ec

Nt -------

© ¹§ ·

E DPDCH –

max 1

F UL

tx i ic

' X user UL

tx i ic -------------------------------------- 1 –

-----------------------------------------------=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max ' X user UL

F UL

------------------=

Index HSUPABearer

R RLC peak – UL

Index HSUPABearer

N Rt x Index HSUPABearer ------------------------------------------------------------------------------------

Ec

Nt -------

© ¹§ ·

E DPDCH –

re q Ec

Nt -------

© ¹§ ·

E DPDCH –

max

d

P term HSUPA –

reqP d term

max

Ec

Nt -------

© ¹§ ·

E DPDCH –

re q

' X HSPAUL

tx i ic X max UL

tx i ic X R99UL

tx i ic – X HSPA CBR – UL

tx i ic – =

nHSPA

' X user UL

tx i ic ' X

HSPA

ULtx i ic

nHSPA-----------------------------------------=

Ec

Nt -------© ¹

§ ·E DPDCH –

max



© Forsk 2010 AT282_TRG_E2 243


We have:

Here, the obtained HSUPA bearer is the index 5 HSUPA bearer. It provides a potential throughput of 128 kbps and

requires E-DPDCH Ec » Nt of -13 dB (lower than -11.5 dB) and a terminal power lower than the maximum terminal power

allowed.

.

Noise Rise Scheduling in Soft Handover

With HSUPA, uplink soft handover impacts the scheduling operation. While HSDPA sends data from one cell only, with

HSUPA all cells in the active set receive the transmission from the terminal. Therefore, all the cells are impacted by the

transmission in terms of noise rise.

For each HSUPA capable cell of the active set , Atoll calculates the maximum E-DPDCH Ec » Nt allowed

( ) as explained in "HSUPA Bearer Allocation Process" on page 240.

For each cell of the active set , Atoll calculates the maximum terminal power allowed to obtain an HSUPA radio

bearer ( ).

With

(7)


As HSUPA bearer users in soft handover use the lowest granted noise rise, Atoll chooses the lowest of maximum terminal

power allowed for each cell of the active set .

Once Atoll knows the selected maximum terminal power ( ) , i t recalculates the maximum E-DPDCH Ec » Nt

allowed ( ) for each HSUPA capable cell of the active set.

HSUPA Bearers

Index

Required Ec/Nt

Threshold (dB)

Nb of

Retransmissions

RLC Peak Rate

(kbps)

Potential

Throughput

(kbps)

1 -21.7 2 32 16

2 -19 2 64 32

3 -16.1 2 128 64

4 -13.9 2 192 96

5 -13 2 256 128

6 -10.1 2 512 256

7 -8 2 768 384

8 -7 2 1024 512

7. In the HSUPA coverage prediction, is calculated as follows:

)

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

-11.5 dB=

tx k ic

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

tx k ic

tx k ic

P term HSUPA – max

tx k ic

P term HSUPA – ma x

tx k ic mi nEc

Nt -------

© ¹§ ·

E DPDCH –

max

tx k ic LT N tot ULuu© ¹

§ · P termmax © ¹

§ ·=

N tot UL

ic 1 F – MUDtx

Utermu I u tot

UL intraic I tot


ULic N + 0

tx + +=

LT


GTx Gtermu--------------------------------------------------------------------------------------------------------------------------------=

LT

LT


GTx Gtermu------------------------------------------------------------------------------------------------------------------------------------------------------------ -=

Uterm F MUDtx

I tot

UL intraI tot

ULextraI in ter carr ier – UL

N 0 tx

tx k ic


mi ntx k AS

P term HSUPA – max

tx k ic =


Ec

Nt

-------

© ¹

§ ·

E DPDCH –

max

tx k ic

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

tx k ic P term HSUPA –

max

LT N tot ULu

-----------------------------------=




Then, Atoll calculates the maximum E-DPDCH Ec » Nt allowed ( ) after signal recombination of all HSUPA

capable cells of the active set 8.

For softer (1/2) and softer-softer (1/3) handovers, we have:

For soft (2/2) and soft-soft (3/3) handovers, we have:

For softer-soft handover (2/3), it depends on if the MRC option is selected (option available in Global parameters). If se-

lected, we have:

Else, we have:

Then, Atoll selects an HSUPA bearer as previously explained in "HSUPA Bearer Allocation Process" on page 240. The

allocation depends on the maximum E-DPDCH Ec » Nt allowed and on UE and cell capabilities. Atoll selects the best

8. In HSUPA coverage predictions, we have the following:

For softer (1/2) and softer-softer (1/3) handovers:

For soft handover (2/2):

For soft-soft handover (3/3):

For softer-soft handover (2/3), it depends on if the MRC option is selected (option available in Global parameters). If se-

lected, we have:

Else, we have:

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

f rake efficiency UL Ec

Nt -------

© ¹§ ·

E DPDCH –

max

tx k ic


¦u=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ma x

tx k

ActiveSet

Ec

Nt -------

© ¹§ ·

E DPDCH –

ma x

tx k ic © ¹§ · Gmacro diversity –

UL 2linksu=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ma x

tx k

ActiveSet

Ec

Nt -------

© ¹§ ·

E DPDCH –

ma x


UL 3linksu=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max Ma x

tx k

tx l

, ActiveSet

tx k

samesite

tx l

othersite


Nt -------

© ¹§ ·

E DPDCH –

max tx k ic

tx k

¦uEc

Nt -------

© ¹§ ·

E DPDCH –

max tx l ic

© ¹¨ ¸¨ ¸¨ ¸§ ·

Gmacro diversity – UL 2linksu

=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ma x

tx k

ActiveSet

Ec

Nt -------

© ¹§ ·

E DPDCH –

ma x


UL 2linksu=

Ec

Nt -------© ¹

§ ·E DPDCH –

max


Nt -------© ¹

§ ·E DPDCH –

max

tx k ic


¦u=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ma x

tx k

ActiveSet

Ec

Nt -------

© ¹§ ·

E DPDCH –

ma x

tx k ic © ¹§ ·=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max Ma x

tx k

tx l

, ActiveSet

tx k

samesite

tx l

othersite


Nt -------

© ¹§ ·

E DPDCH –

max

tx k ic

tx k

¦uEc

Nt -------

© ¹§ ·

E DPDCH –

max

tx l ic

© ¹¨ ¸¨ ¸¨ ¸§ ·

=

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

Ma x

tx k

ActiveSet

Ec

Nt -------

© ¹§ ·

E DPDCH –

ma x

tx k ic © ¹§ ·=



© Forsk 2010 AT282_TRG_E2 245


HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer ( ) with the highest po-

tential throughput ( ) where:

•

When several HSUPA bearers are available, Atoll selects the one with the lowest .

Determination of the Requested HSUPA Bearer

The requested HSUPA radio bearer is selected from the HSUPA bearers compatible with the user equipment. Atoll deter-

mines the HSUPA bearer the user would obtain by considering the entire remaining load of the cell. The user is treated as

if he is the only user in the cell. Therefore, if we go on with the previous example, the maximum E-DPDCH Ec » Nt allowed

is equal to -1.8 dB and the requested HSUPA bearer is the index 7 HSUPA bearer. It requires E-DPDCH Ec » Nt of -8 dB

(lower than -1.8 dB) and a terminal power lower than the maximum terminal power allowed.

6.4.2.4.4 Radio Resource Control

Atoll checks to see if enough channel elements are available and if the Iub backhaul throughput is sufficient for the HSUPA

bearer assigned to the user (taking into account the maximum number of channel elements defined for the site and the

maximum Iub backhaul throughput allowed on the site in the uplink). If not, Atoll allocates a lower HSUPA bearer ("down-

grading") which needs fewer channel elements and consumes lower Iub backhaul throughput. If no channel elements are

available, the user is rejected. On the same hand, if the maximum Iub backhaul throughput allowed on the site in the uplink

is still exceeded even by using the lowest HSDPA bearer, the user is rejected.

6.4.2.5 Convergence CriteriaThe convergence criteria are evaluated for each iteration, and can be written as follow:

Atoll stops the algorithm if:

1st case: Between two successive iterations, and are lower ( ) than their respective thresholds (defined when

creating a simulation).

The simulation has reached convergence.

Example: Let us assume that the maximum number of iterations is 100, UL and DL convergence thresholds are set to 5.

If and between the 4th and the 5th iteration, Atoll stops the algorithm after the 5th iteration. Convergence

has been reached.

2nd case: After 30 iterations, and/or are still higher than their respective thresholds and from the 30th iteration,

and/or do not decrease during the next 15 successive iterations.

The simulation has not reached convergence (specific divergence symbol).

Examples: Let us assume that the maximum number of iterations is 100, UL and DL convergence thresholds are set to 5.

1. After the 30th iteration, and/or equal 100 and do not decrease during the next 15 successive iterations:

Atoll stops the algorithm at the 46th iteration. Convergence has not been reached.

2. After the 30th iteration, and/or equal 80, they start decreasing slowly until the 40th iteration (without going

under the thresholds) and then, do not change during 15 successive iterations: Atoll stops the algorithm at the

56th iteration without reaching convergence.

3rd case: After the last iteration.

If and/or are still strictly higher than their respective thresholds, the simulation has not reached convergence

(specific divergence symbol).

If and are lower than their respective thresholds, the simulation has reached convergence.

Index HSUPABearer

R RLC peak – UL

Index HSUPABearer

N Rtx Index HSUPABearer ------------------------------------------------------------------------------------

Ec

Nt -------

© ¹§ ·

E DPDCH –

req Ec

Nt -------

© ¹§ ·

E DPDCH –

max

d

Ec

Nt -------© ¹

§ ·E DPDCH –

re q

'DL max int ma x

StationsP tx ic k P tx ic k 1 – –

P tx ic k

---------------------------------------------------------------------------------------- 100 u© ¹¨ ¸§ ·

in t ma x

StationsN user

DLic k N user

DLic k 1 – –

N user DL

ic k

---------------------------------------------------------------------------------------------------- 100 u© ¹¨ ¸¨ ¸§ ·

© ¹¨ ¸¨ ¸§ ·

=

'UL max int ma x

StationsI to t UL

ic k I tot UL

ic k 1 – –

I tot UL

ic k

---------------------------------------------------------------------------------------- 100 u© ¹¨ ¸¨ ¸§ ·

in t ma x

StationsN user

ULic k N user

ULic k 1 – –

N user UL

ic k

---------------------------------------------------------------------------------------------------- 100 u© ¹¨ ¸¨ ¸§ ·

© ¹¨ ¸¨ ¸§ ·

=

'UL 'DL d

'UL 5 d 'DL 5 d

'UL 'DL

'UL 'DL

'UL 'DL

'UL 'DL

'UL 'DL

'UL 'DL




6.4.3 Results

6.4.3.1 R99 Related Results

This table contains some R99 specific simulation results provided in the Cel ls and Mobiles tabs of the simulation property

dialogue.


NoneNumber of E1/T1/Ethernet links

required by the site

NoneDownlink intra-cell interference at

terminal on carrier


terminal on carrier


terminal on carrier

W Downlink inter-technology interferenceat terminal on carrier ic a

WTotal effective interference at terminal

on carrier (after unscrambling)


carrier

W


from intra-cell terminals using carrier

W


from extra-cell terminals using carrier

WUplink inter-carrier interference at

terminal on carrier

WTotal received interference at

transmitter on carrier

WTotal noise at transmitter on carrier

(Uplink interference)

None Cell uplink load factor on carrier

None Cell uplink reuse factor on carrier

NoneCell uplink reuse efficiency factor on

carrier

NbE1 T1 e Ethernet e RoundUp Max T IubDL

N I T E1 T1 e Ethernet e e T IubUL

N I T E1 T1 e Ethernet e e

I intraDL

tx i ic P tot

DLtx i ic F ortho UBT Su – P tot

DLtx i ic

tx i

P SC H tx i ic

LT

---------------------------------- –

© ¹¨ ¸¨ ¸§ ·

u

1 F ortho – UBT Su – P bDL

tx i ic u

ic

I extraDL

ic P tot DL

tx j ic

t xj j i z¦ ic


DLic

P to t DL

tx j ic adj

txj j ¦

RF ic ic adj ------------------------------------------------- ic

I i nt er t echn o y log – DL ic

P Transmitted

Tx ic

i

Ltotal Tx

IC P ni ic Tx m

u------------------------------------------

ni ¦

I tot DL

ic I intraDL

ic I extraDL



ic + + +ic

N tot DL

ic I tot DL

ic N 0 Term

+ic

I tot

UL intratx i ic

P bUL

ic

term

tx i

¦ic

I tot

ULextratx i ic

P bUL

ic

term

t xj j i z

¦ic


ULtx i ic

P bUL

ic adj

term

t xj j

¦

RF ic ic adj ---------------------------------------

ic

I tot UL

tx i ic I to t

ULextratx i ic 1 F MU D

Tx Utermu – + I tot

UL intratx i ic u I in ter carr ier –

ULtx i ic + ic

N tot UL

tx i ic I tot UL

tx i ic N 0 tx

+ic

X UL

tx i ic I tot UL

tx i ic

N tot UL

tx i ic ------------------------------ ic

F UL tx i ic I tot

ULtx i ic

I tot

ULintratx i ic 1 F MUD

Tx – Utermu u

------------------------------------------------------------------------------------------- ic

E UL

tx i ic 1

F UL

tx i ic ------------------------------

ic



© Forsk 2010 AT282_TRG_E2 247


6.4.3.2 HSPA Related Results

At the end of the R99 part, packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users can be:

• Either connected and in this case, they obtain the requested R99 bearer,

• Or rejected exactly for the same reasons as R99 users.

Only connected packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users are considered in

the HSDPA part. At the end of the HSDPA part, packet (HSDPA) and packet (HSPA) service users can be:

• Either connected if they obtain an HSDPA bearer,

• Or rejected if the maximum number of HSDPA users per cell is exceeded,

• Or delayed in case of lack of resources (HSDPA power, HS-SCCH power, HS-SCCH channels, OVSF codes).

Packet (HSPA - Constant Bit Rate) service users can be:

• Either connected if they obtain an HSDPA bearer,

• Or rejected for the following reasons: the maximum number of HSDPA users per cell is exceeded, the lowest

HSDPA bearer the user can obtain does not provide a RLC peak rate higher than the guaranted bit rate, the HS-

SCCH signal quality is not sufficient, there are no more OVSF codes available, the maximum Iub backhaul

throughput allowed on the site in the downlink is exceeded.

In the HSUPA part, Atoll processes packet (HSPA) service users and packet (HSPA - Constant Bit Rate) service users

who are connected to an HSDPA bearer or were delayed in the previous step. At the end, they can be:

• Either connected if they obtain an HSUPA bearer,

• Or rejected for the following reasons: the maximum number of HSUPA users per cell is exceeded, the terminal

power required to obtain the lowest compatible HSUPA bearer exceeds the maximum terminal power, there are

no more channel elements available, the maximum Iub backhaul throughput allowed on the site in the uplink is

exceeded, the lowest compatible HSUPA bearer they can obtain does not provide a RLC peak rate higher than

the guaranted bit rate (only for packet (HSPA - Constant Bit Rate) service users).

6.4.3.2.1 Statistics Tab

In the Statistics tab, Atoll displays as results:

• The number of rejected users.

• The number of delayed users.

• The number of R99 bearer users connected to a cell (result of the R99 part). This figure includes R99 users as

well as HSDPA and HSUPA bearer users since all of them request an R99 bearer.

- The number of R99 bearer users per frequency band.

- The number of R99 bearer users per activity status.

- The downlink and uplink rates ( and ) generated by their connection to R99 bearers. Only active

users are considered.

and

is the downlink nominal rate of the user R99 radio bearer and is the uplink

nominal rate of the user R99 radio bearer.

Simulation result available per cell

with

Simulation result available per mobile

None Downlink load factor on carrier

None Downlink reuse factor on a carrier

dB Noise rise on downlink

dB Noise rise on uplink


fractional load.

X DL

tx i ic

I extraDL


ic + LT u

P Tx DL

tx i ic ---------------------------------------------------------------------------------------- 1 F ortho UBT Su – +

1

CI reqDL

------------- 1 F ortho UBT Su – +

-------------------------------------------------------------------------------------------------------------------------------------------

tch

¦

CI re qDL Qreq

DL

G pDL

------------=

I tot DL

ic N to t

DLic

--------------------

ic

F DL

tx i ic I tot DL

ic

I intraDL

tx i ic -------------------------------- ic

NR DL

tx i ic 10 1 X DL

tx i ic – log –

NR UL

tx i ic 10 1 X UL

tx i ic – log –

R R99DL

R R99UL

R R99DL

R nominal DL

R99 Bearer

Ac ti ve

users

¦= R R99UL

R nominal UL

R99 Bearer

Ac ti ve

users

¦=

R nominal DL

R99 Bearer R nominal UL

R99 Bearer




• The number of connected users with an HSDPA bearer (result of the HSDPA part) and the downlink rate they gen-

erate. Packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users are considered since

they all request an HSDPA bearer. On the other hand, only active users are taken into consideration in the down-

link rate calculation ( ).

is the RLC peak rate provided in the downlink.

• The number of connected HSUPA bearer users (result of the HSUPA part). Only packet (HSPA) and packet

(HSPA - Constant Bit Rate) service users are considered.

In addition, Atoll indicates the uplink data rate generated by active users connected with an HSUPA bearer

( ):

is the RLC peak rate provided in the uplink.

6.4.3.2.2 Mobiles Tab

In the Mobiles tab, Atoll indicates for each user:

• The uplink and downlink total requested rates in kbps (respectively, and )

For circuit and packet (R99) service users, the DL and UL total requested rates correspond to the DL and UL nominal rates

of the R99 bearer associated to the service.

For packet (HSDPA) service users, the uplink requested rate corresponds to the nominal rate of ADPCH R99 radio bearer

and the downlink requested rate is the sum of the ADPCH radio bearer nominal rate and the RLC peak rate that the

selected HSDPA radio bearer can provide. Here, the user is treated as if he is the only user in the cell and then, Atoll

determines the HSDPA bearer the user would obtain by considering the entire HSDPA power available of the cell.

For HSUPA bearer users (i.e., packet (HSPA) and packet (HSPA - Constant Bit Rate) service users), the uplink requested

rate is equal to the sum of the ADPCH-EDPCCH radio bearer nominal rate and the RLC peak rate of the requested HSUPA

radio bearer. The requested HSUPA radio bearer is selected from the HSUPA bearers compatible with the user equip-

ment. Here, the user is treated as if he is the only user in the cell and then, Atoll determines the HSUPA bearer the user

would obtain by considering the entire remaining load of the cell. The downlink requested rate is the sum of the ADPCH-

EDPCCH radio bearer nominal rate and the RLC peak rate that the requested HSDPA radio bearer can provide. The

requested HSDPA radio bearer is determined as explained in the previous paragraph.

• The uplink and downlink total obtained rates in kbps (respectively, and )

For circuit and packet (R99) service users, the obtained rate is the same as the requested rate if he is connected without

being downgraded. Otherwise, the obtained rate is lower (it corresponds to the nominal rate of the selected R99 bearer).

If the user is rejected, the obtained rate is zero.

For a packet (HSDPA) service connected to an HSDPA bearer, the uplink obtained rate equals the requested one and the

downlink obtained rate corresponds to the instantaneous rate; this is the sum of the A-DPCH radio bearer nominal rate

and the RLC peak rate provided by the selected HSDPA radio bearer after scheduling and radio resource control. If the

user is delayed (he is only connected to an R99 radio bearer), uplink and downlink obtained rates correspond to the uplink

and downlink nominal rates of ADPCH radio bearer. Finally, if the user is rejected either in the R99 part or in the HSDPA

part (i.e., because the HSDPA scheduler is saturated), the uplink and downlink obtained rates are zero.

For a connected packet (HSPA) service user, on uplink, if the user is connected to an HSUPA bearer, the obtained uplink

rate is the sum of the ADPCH-EDPCCH radio bearer nominal rate and the RLC peak rate provided by the selected HSUPA

radio bearer after noise rise scheduling. On downlink, if the user is connected to an HSDPA bearer, the obtained downlink

rate corresponds to the instantaneous rate. The instantaneous rate is the sum of the ADPCH-EDPCCH radio bearer nomi-

R HSDPADL

R HSDPADL

R RLC peak – DL

Ac ti ve

users

¦=

R RLC peak –

DL

R HSUPAUL

R HSUPAUL

R RLC peak –

UL

Ac ti ve

users

¦=

R RLC peak – UL

R requested UL

M b R requested DL

M b

R requested DL

M b R nominal DL

R99 Bearer =

R requested UL

M b R nominal UL

R99 Bearer =

R requested DL

M b R nominal DL

AD PC H R99 Bearer R RLC peak –

DL+=

R requested UL

M b R nominal UL

AD PC H R99 Bearer =

R requested DL

M b R nominal DL

AD PC H ED PC CH R99 Bearer – R RLC peak – DL

+=

R requested UL

M b R nominal UL

AD PC H ED PC CH R99 Bearer – R RLC peak – UL

+=

R obtained UL

M b R obtained DL

M b



© Forsk 2010 AT282_TRG_E2 249


nal rate and the RLC peak rate provided by the selected HSDPA radio bearer after scheduling and radio resource control.

If the user is delayed, the obtained downlink rate corresponds to the downlink nominal rate of ADPCH-EDPCCH radio

bearer. If the user is rejected, the obtained uplink and downlink rates are "0."

For a connected packet (HSPA - Constant Bit Rate) service user, the uplink and downlink total obtained rates are the sum

of the ADPCH-EDPCCH radio bearer nominal rate and the guaranteed bit rate defined for the service. If the user is

rejected, the uplink and downlink total obtained rates are "0".

• The mobile total power ( )

for packet (HSPA) service users

for packet (HSPA - Constant Bit Rate) service users

And

for circuit and packet (R99) service users and packet (HSDPA) service users

• The HSDPA application throughput in kbps ( )

This is the net HSDPA throughput without coding (redundancy, overhead, addressing, etc.).

Where:

is read in the quality graph defined for the triplet “reception equipment-selected bearer-mobility” (HSDPA

Quality Graphs tab in the Reception equipment properties). This graph describes the variation of BLER as a function of

the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll calculates the corresponding BLER.

and respectively represent the scaling factor between the application throughput and the RLC (Radio Link

Control) throughput and the throughput offset. These two parameters model the header information and other supplemen-

tary data that does not appear at the application level. They are defined in the service properties.

is the minimum number of TTI (Transmission Time Interval) between two TTI used; it is defined in the terminal user

equipment category properties.

• The number of OVSF codes

This is the number of 512-bit length OVSF codes consumed by the user.

• The required HSDPA power in dBm ( )

It corresponds to the HSDPA power required to provide the HSDPA bearer user with the downlink requested rate. The

downlink requested rate is the data rate the user would obtain if he was the only user in the cell. In this case, Atoll deter-

mines the HSDPA bearer the user would obtain by considering the entire HSDPA power available of the cell.

is the HS-PDSCH power required to obtain the selected HSDPA bearer (in dBm). If the HSDPA bearer

allocated to the user is the best one, corresponds to the available HS-PDSCH power of the cell. On

the other hand, if the HSDPA bearer has been downgraded in order to be compliant with cell and UE capabilities or for

another reason, will be lower than the available HS-PDSCH power of the cell.

• The served HSDPA power in dBm ( )

This is the HSDPA power required to provide the HSDPA bearer user with the downlink obtained rate. The downlink

obtained rateis the data rate experienced by the user after scheduling and radio resource control.

for packet (HSDPA) and packet (HSPA) service users

And

for packet (HSPA - Constant Bit Rate) service users

Where

is the HS-PDSCH power required to obtain the selected HSDPA bearer.

• The No. of HSUPA Retransmissions (Required)

The maximum number of retransmissions in order to have the requested HSUPA radio bearer with a given BLER.

• The No. of HSUPA Retransmissions (Obtained)

Note:

• For packet (HSPA - Constant Bit Rate) service users, .

P term

P term P term R99 – f act EDPCCH – ULu= P term HSUPA – +

P term P term R99 – f act EDPCCH –

UL

u= P term HSUPA – C HSDPABearer u+

f act EDPCCH – UL

0.1=

P term P term R99 – =

T applicationDL

M b

T applicationDL

M b R obtained

DLM b 1 B LE R HSDPA – SF Rateuu 'R –

'TT I -----------------------------------------------------------------------------------------------------------------------------------=

BLER HSDPA

SF Rate 'R

'TT I

P HSDPA required

P HSDPA required

P HS PDSCH – used

nHS SCCH – P HS SCCH – u+=

P HS PDSCH – used

P HS PDSCH – used

P HS PDSCH – used

P HSDPA served

P HSDPA served

P HS PDSCH – used

nHS SCCH – P HS SCCH – u+=

P HSDPA served

P HS PDSCH – used

C HSDPABearer u=

P HS PDSCH – used




The maximum number of retransmissions in order to have the obtained HSUPA radio bearer with a given BLER.

• The HSUPA application throughput in kbps ( )

This is the net HSUPA throughput without coding (redundancy, overhead, addressing, etc.).

Where:

is the residual BLER after retransmissions. It is read in the quality graph defined for the quartet “recep-

tion equipment-selected bearer-number of retransmissions-mobility” (HSUPA Quality Graphs tab in the Reception equip-ment properties). This graph describes the variation of BLER as a function of the measured quality (E-DPDCH Ec/Nt).

Knowing the E-DPDCH Ec/Nt, Atoll calculates the corresponding BLER.




is the maximum number of retransmissions for the obtained HSUPA bearer. This figure is read in the HSUPA Bearer

Selection table.

The following columns appear if, when creating the simulation, you select "Detailed information about mobiles":

• The uplink and downlink requested RLC peak rates (kbps)

Downlink and uplink requested RLC peak rates are not calculated for circuit and packet (R99) service users.

For packet (HSDPA) service users, the uplink RLC peak rate is not calculated and the downlink requested RLC peak rate

is the data rate that the selected HSDPA radio bearer can provide. Here, the user is treated as if he is the only user in the

cell and then, Atoll determines the HSDPA bearer he would obtain by considering the entire HSDPA power available of the cell.

For HSUPA bearer users (i.e., packet (HSPA) and packet (HSPA - Constant Bit Rate) service users), the requested uplink

RLC peak rate is the data rate of the requested HSUPA radio bearer. The requested HSUPA radio bearer is selected from

the HSUPA bearers compatible with the user equipment. Here, the user is treated as if he is the only user in the cell and

then, Atoll determines the HSUPA bearer the user would obtain by considering the entire remaining load of the cell. If the

user is connected to an HSDPA bearer in the downlink, the downlink requested RLC peak rate is the rate that the

requested HSDPA radio bearer can provide. The requested HSDPA radio bearer is determined as explained in the previ-

ous paragraph.

• The uplink and downlink obtained RLC peak rate (kbps)

Downlink and uplink obtained RLC peak rates are not calculated for circuit and packet (R99) service users.

For a packet (HSDPA) service user connected to an HSDPA bearer, the uplink obtained RLC peak rate is not calculated,

and the downlink obtained RLC peak rate is the rate provided by the selected HSDPA radio bearer after scheduling and

radio resource control.

For a connected packet (HSPA) service user, on uplink, if the user is connected to an HSUPA bearer, the obtained uplinkRLC peak rate is the rate provided by the selected HSUPA radio bearer after noise rise scheduling. On downlink, if the

user is connected to an HSDPA bearer, the downlink obtained RLC peak rate is the rate provided by the selected HSDPA

radio bearer after scheduling and radio resource control.

For a connected packet (HSPA - Constant Bit Rate) service user, the uplink and downlink obtained RLC peak rates are

the uplink and downlink guaranteed bit rates defined for the service.

6.4.3.2.3 Cells Tab

In the Cells tab, Atoll gives:

• The available HSDPA power in dBm ( )

This is:

- Either a fixed value in case of a static HSDPA power allocation strategy,

- Or a simulation result when the option "HSDPA Power Dynamic Allocation" is selected. We have:

with

• The transmitted HSDPA power in dBm ( )

It corresponds to the HSDPA power used to serve HSDPA bearer users.

• The number of HSDPA users in the cell

T applicationUL

M b

T applicationUL

M b R obtained

ULM b 1 B LE R HSUPA – SF Rateuu 'R –

N Rtx

-----------------------------------------------------------------------------------------------------------------------------------=

BLER HSUPA N Rtx

SF Rate 'R

N Rtx

P HSDPA ic

P HSDPA ic P max ic P Headroom ic – P t x R99 – ic – P HSUPA ic – =

P t x R99 – ic P pil ot ic P SCH ic P OtherCCH ic P tch ic P tch ic f act ADPCH – DLu

tch used for

HSPA users

¦+

tch used for

R99 users

¦+ + +=

P tx H – SDPA ic

P tx H – SDPA cell P HSDPA M b served

M b cell ¦=



© Forsk 2010 AT282_TRG_E2 251


They are the connected and delayed HSDPA bearer users. This figure includes packet (HSDPA), packet (HSPA) and

packet (HSPA - Constant Bit Rate) users.

• The number of simultaneous HSDPA users in the cell ( )

It corresponds to the number of connected HSDPA bearer users that the cell supports at a time, i.e. within one transmis-

sion time interval. All these users are connected to the cell at the end of the HSDPA part of the simulation; they have a

connection with the R99 bearer and an HSDPA bearer.

• The instantaneous HSDPA rate in kbps ( )

This is the number of kilobits per second that the cell supports on downlink to provide simultaneous connected HSDPA

bearer users with an HSDPA bearer.

• The instantaneous HSDPA MAC Throughput in kbps ( )

Where,

is the transport block size (in kbits) of the HSDPA bearer selected by the user; it is defined for each HSDPA

bearer in the HSDPA Radio Bearers table.

is the minimum number of TTI (Transmission Time Interval) between two TTI used; it is defined in the terminal

user equipment category properties.

is the TTI duration, i.e. (2000 TTI in one second). This value is specified by the 3GPP.

• The average instantaneous HSDPA rate in kbps ( )

• The HSDPA application throughput in kbps ( )

Either if the scheduling algorithm is Round Robin or Proportional Fair,

Or if the scheduling algorithm is Max C/I.

is the user with the highest in the cell.

• The minimum HSDPA RLC peak rate in kbps ( )

It corresponds to the lowest of RLC peak rates obtained by HSDPA bearer users connected to the cell.

• The maximum HSDPA RLC peak rate in kbps ( )

It corresponds to the highest of RLC peak rates obtained by HSDPA bearer users connected to the cell.

• The number of HSUPA users in the cell ( ):

They are the HSDPA bearer users connected to the cell.

• The HSUPA application throughput in kbps ( )

• The uplink cell load factor due to HSUPA traffic ( ):

Where

nM b

R Inst DL

cell

R Inst DL

cell R obtained DL

M b

M b cell ¦=

T MAC DL

cell

T MAC DL

cell Sblock M b

T TT I 'u TTI M b --------------------------------------------

M b cell ¦=

Sblock M b

'TTI M b

T TTI 2 3 –

u10 s

R Av Ins t –

DLcell

R Av In st – DL

cell

R obtained DL

M b

M b cell ¦

nM b

---------------------------------------------------------=

T applicationDL

cell

T applicationDL

cell T applicationDL

M b

M b cell ¦=

T applicationDL

cell T applicationDL

M b maxC I e =

M b maxC I e C I e

mM b cell

in R obtained DL

M b

mM b cell

ax R obtained DL

M b

nM c

T applicationUL

cell

T applicationUL

cell T applicationUL

M b

M b cell ¦=

X HSUPAUL

cell

X HSUPAUL

cell I tot UL

cell HSUPA

N tot UL

cell ---------------------------------------------=




is the total interference at transmitter received from HSUPA bearer users.

6.4.3.2.4 Sites Tab

In the Sites tab, Atoll displays:

• The instantaneous HSDPA rate carried by the site in kbps ( )

• The instantaneous HSDPA MAC Throughput carried by the site in kbps ( in kbps)

• The HSUPA rate carried by the site in kbps ( )

6.4.4 Appendices

6.4.4.1 Admission Control in the R99 Part

During admission control of the R99 part of the simulation, Atoll calculates the uplink load factor of a considered cell

assuming the mobile concerned is connected with it. Here, activity status assigned to users is not taken into account. So

even if the mobile is not active on UL, it can be rejected due to cell load saturation. To calculate the cell UL load factor,

either Atoll takes into account the mobile power determined during power control if mobile was connected in previous iter-

ation, or it estimates a load rise due to the mobile and adds it to the current load. The load rise ( ) is calculated as

follows:

6.4.4.2 Resources Management

6.4.4.2.1 OVSF Codes Management

OVSF codes are managed on the downlink during the simulation since this resource is downlink limited only. Atoll checks

the availability of this resource during the simulation, first in the R99 part and then in the HSDPA part. It determines the

number of codes that will be consumed by each cell.

OVSF codes form a binary tree. Codes of longer lengths are generated from codes of a shorter length. Length-k OVSF

codes are generated from length-k/2 OVSF codes. Therefore, if one channel needs 1 length-k/2 OVSF code, it is equiva-

lent to use 2 length-k OVSF codes, or 4 length-2k OVSF codes and so on.

512 512-bit-length codes per cell are available in UMTS HSPA projects.

In the R99 part, during the resource control, Atoll determines the number of 512 bit-length codes that will be consumed

for each cell.

If the cell supports HSUPA, Atoll allocates codes for the DL channels used for HSUPA:

• A 128 bit-length code for the E-HICH and E-RGCH channels (i.e. four 512 bit-length OVSF codes), for each cell.

Therefore, Atoll will take four 512-bit-length codes,

• A 256 bit-length code for the E-AGCH channel (i.e. two 512 bit-length OVSF codes), for each cell. Therefore, Atoll

will take two 512-bit-length codes,If the cell supports HSDPA, Atoll reserves for potential HSDPA bearer users:

• The minimum number of HS-PDSCH codes defined for the cell, . They are 16-bit length OVSF

codes (i.e. thirty-two 512 bit-length OVSF codes). Therefore, Atoll will take 512-bit-

length codes,

• A 128 bit-length code per HS-SCCH channel (i.e. four 512 bit-length OVSF codes), for each cell. Therefore, Atoll

will take 512-bit-length codes,

Then, it allocates to the cell OVSF codes to support R99 bearers required by users:

• A 256 bit-length code per common channel (i.e. two 512 bit-length OVSF codes), for each cell. Therefore, Atoll

will take 512-bit-length codes,

I to t UL

cell HSUPA

R Inst DL

site

R Inst DL

site R Inst DL

cell

ce ll s i te¦=

T MAC DL site

T MAC DL

site T MAC DL

cell

ce ll s i te¦=

R UL

site

R UL

site R obtained UL

M c

M c site¦=

' X UL

' X UL 1

1W

QreqUL

R nominal ULu

----------------------------------------+

--------------------------------------------------=

N mi nCodes HS – PDSCH –

32 N mi nCodes HS – PDSCH –

u

4 nHS SCCH – u

2 N u Overhead Codes –



© Forsk 2010 AT282_TRG_E2 253


• A code per cell-receiver link, for TCH (traffic channels). The length of code to be allocated,Code_Length, depends

on the user activity. We have:

Either when the user is active,

Or if the user is inactive.

The number of 512 bit-length OVSF codes needed is calculated from the length of the code to be allocated

as follows:

The OVSF code allocation follows the “Buddy” algorithm, which guarantees that:

• If a k-length OVSF code is used, all of its children with lengths 2k, 4k, …, cannot be used as they will not be orthog-

onal.

• If a k-length OVSF code is used, all of its ancestors with lengths k/2, k/4, …, cannot be used as they will not be

orthogonal.

Example: We consider a user with a service requiring the UDD64 R99 radio bearer. This user is active on DL while

connected to a cell (which does not support HSDPA). The spreading factor for active users has been set to 64 and site

equipment requires four overhead downlink channel elements per cell. Atoll will consume four 256 bit-length OVSF codes

for common channels (i.e. eight 512 bit-length OVSF codes) and a 64 bit-length OVSF code for traffic channels (i.e. eight

additional 512 bit-length OVSF codes).

In the HSDPA part, packet (HSDPA), packet (HSPA) and packet (HSPA - Constant Bit Rate) service users) are assigned

an HSDPA bearer (Fast link adaptation). Therefore, Atoll allocates to the cell:

• 16-bit length OVSF codes per cell-HSDPA receiver, for HS-PDSCH. This figure depends on the HSDPA bearer

assigned to the user and on the type of service.

For packet (HSDPA) and packet (HSPA) service users, Atoll needs 512-bit-length

codes for each user connected to the cell. is the number of HS-PDSCH channels required by

the HSDPA bearer.

For packet (HSPA - Constant Bit Rate) service users, Atoll needs 512-

bit-length codes for each user connected to the cell. is the number of HS-PDSCH channels

required by the HSDPA bearer.

Figure 6.13: OVSF Code Tree Indices (Not OVSF Code Numbers)

Notes:

• In the R99 part, the OVSF code allocation follows the mobile connection order (mobileorder in the Mobiles tab).

• The OVSF code and channel element management is differently dealt with in case of

“softer” handover. Atoll allocates OVSF codes for each cell-mobile link while it globally

assigns channel elements to a site.

Note:

• When HSDPA bearer users (at least one) are connected to the cell, Atoll gives the cell

back the minimum number of OVSF codes reserved for HS-PDSCH ( ).

On the other hand, if no HSDPA bearer user is connected, Atoll still keeps these codes

and the codes for HS-SCCH too. This is the same with HSUPA bearer users. Even if no

HSUPA bearer user is connected to the cell, Atoll still keeps the codes for E-HICH, E-

RGCH and E-AGCH channels.

Code_Length F spreading DL

Active user =

Code_Length F spreading DL

Inactive user =

N Codes-TCH

N Codes-TCH 512

Code_Length

------------------------------------=

32 N Codes HS – PDSCH –

u

N Codes HS – PDSCH –

32 N Codes HS – PDSCH –

u C HSDPABearer u

N Codes HS – PDSCH –

N mi nCodes HS – PDSCH –




6.4.4.2.2 Channel Elements Management

Channel elements are controlled in the R99 and the HSUPA parts of the simulation. Atoll checks the availability of this

resource in the uplink and downlink.

In the R99 part, during the resource control, Atoll determines the number of channel elements required by each site for

R99 bearer users in the uplink and downlink. Then, in the HSUPA part, Atoll carries out another resource control after

allocating HSUPA bearers. It takes into account the channel elements consumed by HSUPA bearer users in the uplink

and recalculates the number of channel elements required by each site in the uplink.

In the uplink, Atoll consumes channel elements for each cell j on a site N I . This figure includes:

• Channel elements for R99 bearers:

- channel elements for control channels,

- per cell-receiver link, for R99 TCH (traffic channels).

• Channel elements for HSUPA bearers:

- per cell-receiver link, for packet (HSPA) service users.

- per cell-receiver link, for packet (HSPA - Constant Bit Rate) service users.

Therefore, the number of channel elements required on uplink at the site level, , is:

On downlink, Atoll consumes channel elements for each cell j on a site N I . This figure includes:

• Channel elements for R99 bearers

- channel elements for control channels (Pilot channel, Synchronisation channel, common

channels),


Therefore, the number of channel elements required on downlink at the site level, , is:

6.4.4.2.3 Iub Backhaul Throughput

The Iub backhaul throughput is controlled in the R99, the HSDPA and the HSUPA parts of the simulation. Atoll checks

the availability of this resource in the uplink and downlink.

In the R99 part, during the resource control, Atoll determines the Iub throughput required by each site for R99 bearer users

in the uplink and downlink. Then, in the HSDPA part, Atoll performs a resource control in the downlink after allocating

HSDPA bearers. It takes into account the Iub backhaul throughput consumed by HSDPA bearer users in the downlink and

recalculates the Iub backhaul throughput required by each site in the downlink. Finally, in the HSUPA part, Atoll carries

out a resource control in the uplink after allocating HSUPA bearers. It takes into account the Iub backhaul throughput

consumed by HSUPA bearer users in the uplink and updates the Iub backhaul throughput required by each site in the

uplink.

In the uplink, the Iub backhaul throughput consumed by each cell j on a site N I , , includes:

• The Iub backhaul throughput required for R99 bearers:


• The Iub backhaul throughput required for HSUPA bearers:

- per cell-receiver link, for packet (HSPA) service users.


Therefore, the Iub backhaul throughput required on uplink at the site level, , is:

Note:

• In case of “softer” handover (the mobile has several links with co-site cells), Atoll allocates

channel elements for the best serving cell-mobile link only.

N C E U L –

j

N Overhead CE – UL –

N R99 T – C H C E – UL –

N HSUPA CE –

N HSUPA CE –

C HSUPABearer u

N C E U L –

N I

N C E U L –

N I N C E U L –

j

j N I ¦=

N C E D L –

j

N Overhead CE – DL –

N R99 T – C H C E – DL –

N C E D L –

N I

N C E D L –

N I N C E D L –

j

j N I ¦=

T IubUL

j

T IubR99 T – C H U L –

T IubHSUPA

T IubHSUPA

C HSUPABearer u

T Iu bUL

N I

T IubUL

N I T IubUL

j

j N I ¦=



© Forsk 2010 AT282_TRG_E2 255


In the downlink, the Iub backhaul throughput consumed by each cell j on a site N I , , includes:

• The Iub backhaul throughput required for R99 bearers:

- for R99 control channels (Pilot channel, Synchronisation channel, common channels).


• The Iub backhaul throughput required for HSUPA bearers:

- per cell-receiver link, for packet (HSDPA) and packet (HSPA) service users.


With

Therefore, the Iub backhaul throughput required on downlink at the site level, , is:

6.4.4.3 Downlink Load Factor Calculation

Atoll calculates a downlink load factor for each cell (available in the Cells tab of any simulation result) and each connected

mobile (available in the Mobiles tab of any given simulation result).

6.4.4.3.1 Downlink Load Factor per Cell

Approach for downlink load factor evaluation is highly inspired by the downlink load factor defined in the book “WCDMA

for UMTS by Harry Holma and Antti Toskala”.

Let be the required quality.

and are the processing gain on downlink and the Eb/Nt target on downlink respectively.

In case of soft-handoff, required quality is limited to the effective contribution of the transmitter.

where

At mobile level, we have a required power, P tch:

With when the user is active on the downlink and when the user is inactive. In case of an HSDPA bearer

user, .

Note:

• In case of “softer” handover (the mobile has several links with co-site cells), Iub backhaul

throughput is consumed by the best serving cell-mobile link only.

T Iu bDL

j

T IubOverhead DL –

T IubR99 T – C H D L –

T IubHSDPA

T IubHSDPA

C HSDPABearer u

T IubHSDPA

R RLC peak – DL

Overhead IubHSDPA

R RLC peak – DLu+=

T IubDL

N I

T IubDL

N I T Iu bDL

j

j N I ¦=

CI req

Qre qDL

G pDL

------------=

G pDL

QreqDL

P tx DL

ic P pil ot ic P SC H ic P otherCCH ic P tch ic

tch

¦+ + +=

P tx DL

ic P CCH ortho

ic P CCH nonOrtho

ic P tch ic

tch

¦+ +=

P CCH ortho

ic P pi lot ic P otherCCH ic +=

P CCH nonOrtho

ic P SCH ic =

P tc h ic CI req I extra ic I in ter carr ier – ic I i nt er t echn o y log – ic I intra ic N 0 term

+ + + + LT uu r u=

r 1= r r c DL

=

r f act ADPCH – DL=

P tc h ic CI req

I extra ic I in ter carr ier – ic I i nt er t echn o y log – ic + +

1 F ortho UBT Su – +P tx

DLic P CCH

nonOrthoic – P tch ic –

LT

---------------------------------------------------------------------------------------------uP CCH

nonOrthoic

LT

----------------------------------- N 0 term

+ +© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

LT uu r u=

P tc h ic

I extra ic I in ter carr ier – ic I i nt er t e chn o y log – ic + + LT r 1 F ortho UBT Su – P tx DL

ic u r u+uu

F ortho UBTSu P CCH nonOrtho

ic r uu N 0 term

LT r uu+

+

1

CI req r u---------------------- 1 F ortho UBTSu – +

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=




is the total power received at the receiver from the cell with which it is connected.

is the total power received at the receiver from other cells.

is the inter-carrier interference received at the terminal.

is the inter-technology interference received at the terminal from an external transmitter.

We have:

Therefore, the downlink load factor can be expressed as:

The downlink load factor represents the signal degradation in relation to the reference interference (thermal noise plus

synchronisation channel power).

6.4.4.3.2 Downlink Load Factor per Mobile

Atoll evaluates the downlink load factor for any connected mobile as follows:

6.4.4.4 Uplink Load Factor Due to One User

This part details how Atoll calculates the contribution of one user to the UL load factor ( ).

In this calculation, we assume that the cell UL reuse factor ( ) is constant.

The result depends on the option used to calculate Nt (Without useful signal or Total noise that you may select in Global

parameters).

I intra ic

I extra ic

I in ter carr ier – ic

I i nt er t echn o y log – ic

P tx DL

ic

P CCH ortho

ic P CCH nonOrtho

ic +

I extra ic I in ter carr ier – ic I i nt er t echn o y log – ic + + LT r uu

1 F ortho UBT Su – P tx DL

ic u r u F ortho UBTSu P CCH nonOrtho

ic r uu N 0 term

LT r uu+ + +© ¹¨ ¸¨ ¸§ ·

1

CI re q r u---------------------- 1 F ortho UBT Su – +

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- -

© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·

tch

¦+

=

P tx DL

ic P CCH ortho

ic P CCH nonOrtho

ic

I extra ic I in ter carr ier – ic I i nt er t e chn o y log – ic + + LT r uu

P tx DL

ic ---------------------------------------------------------------------------------------------------------------------------------------------------------- P tx

DLic u

1

CI req r u---------------------- 1 F ortho UBT Su – +

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

tch

¦+ + +=

1 F ortho UBT Su – P tx DL

ic r uu

1

CI re q r u

---------------------- 1 F ortho UBT Su – +

------------------------------------------------------------------------------------

tch

¦F ortho UBTSu P CCH

nonOrthoic r u N 0

termLT r uu+u

1

CI req r u

---------------------- 1 F ortho UBT Su – +

-----------------------------------------------------------------------------------------------------------------------------

tc h

¦+

P tx DL

ic

I extra ic I in ter carr ier – ic I i nt er t e chn o y log – ic + + LT r uu

P tx DL

ic ---------------------------------------------------------------------------------------------------------------------------------------------------------- 1 F ortho UBTS r uu – +

1


---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

tch

¦ P tx DL

ic –

P CCH ortho

ic P CCH nonOrtho

ic F ortho UBTSu P CCH

nonOrthoic r u N 0

termLT r uu+u

1


-----------------------------------------------------------------------------------------------------------------------------

tch

¦+ +=

P tx DL

ic

P CCH ortho

ic P CCH nonOrtho

ic F ortho UBT Su P CCH

nonOrthoic r u N 0

termLT r uu+u

1

CI re q r u---------------------- 1 F ortho UBTSu – +

-----------------------------------------------------------------------------------------------------------------------------

tch

¦+ +

1

I extra ic I in ter carr ier – ic I i nt er t e chn o y log – ic + + LT r uuP tx

DLic

---------------------------------------------------------------------------------------------------------------------------------------------------------- 1 F ortho UBTS r uu – +

1


---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

© ¹¨ ¸¨ ¸¨ ¸¨ ¸§ ·

tch

¦ –

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

X DL

I extra ic I in ter carr ier – ic I i nt er t echn o y log – ic + + LT r uu

P tx DL

ic ---------------------------------------------------------------------------------------------------------------------------------------------------------- 1 F ortho UBTS r uu – +

1


---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

tch

¦=

X DL I tot

DLic

N tot DL

ic --------------------=

' X k UL

F UL

tx i ic



© Forsk 2010 AT282_TRG_E2 257


Without Useful Signal Option

We note

As , we have:

Therefore, we have:

So, we can conclude that the contribution of one user to the UL load is defined as:

QreqUL

k W

R nominal UL

k ------------------------------

P bUL

k re q

I intra P bUL

k re q – I + extra I in ter carr ier – N 0 tx

+ +-------------------------------------------------------------------------------------------------------------------------u=

QreqUL

k W

R nominal UL

k ------------------------------

P bUL

k req

I intra F ULu P b

ULk req – N 0

tx +

--------------------------------------------------------------------------------u=

P bUL

k req 1 Qre qUL

k R nominal

ULk

W ------------------------------u+

© ¹

¨ ¸§ ·

u QreqUL

k R nominal

ULk

W ------------------------------u I intra F

ULu N 0

tx + u=

P bUL

k req

QreqUL

k R nominal

ULk

W ------------------------------u I intra F

ULuu

1 QreqUL

k R nominal

ULk

W ------------------------------u+

--------------------------------------------------------------------------------------------

Qre qUL

k R nominal

ULk

W ------------------------------u N 0

tx u

1 QreqUL

k R nominal

ULk

W ------------------------------u+

------------------------------------------------------------------------+=

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q

Qre qUL

k R nominal

ULk

W ------------------------------u=

P bUL

k req

I intra F ULu

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

N 0 tx

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------+=

I intra P bUL

k re q

K

¦=

I intra I intra F UL 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦uu N 0 tx 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦u+=

I intra

N 0 tx 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦u

1 F UL 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦u – ------------------------------------------------------------------------------------------------=

I intra

N 0 tx

F UL e

1

F UL 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦u

-------------------------------------------------------------------------------------- 1 –

------------------------------------------------------------------------------------------------=

X

UL I intra I +extra

I in ter carr ier – +

I intra I +extra

I in ter carr ier – N 0 tx + +------------------------------------------------------------------------------------

I intra F ULu

I intra F ULu N 0 tx +-------------------------------------------

1

1N 0

tx

I intra F UL

u----------------------------+

--------------------------------------= = =

X UL

F UL 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------

K

¦u=




Total Noise Option

We note

As , we have:

Therefore, we have:

So, we can conclude that the contribution of one user to the UL load is defined as:

6.4.4.5 Inter-carrier Power Sharing Modelling

Inter-carrier power sharing enables the network to dynamically allocate available power from R99-only and HSDPA carri-

ers among HSDPA carriers.

In this part, we will consider the most common scenario, a network consisting of an R99-only carrier (c 1) and an HSDPA

carrier with dynamic power allocation (c 2 ) (c

2 does not support HSUPA).

As explained in The User Manual , the maximum power of the HSDPA cell must be set to the same value as the maximum

shared power in order to use power sharing efficiently. In this case, the HSDPA cell can use 100% of the available power,

i.e, all of the R99-only cell’s unused power can be allocated to the HSDPA cell.

Let’s take the following example to measure the impact of the inter-carrier power sharing.

• 1st case: Inter-carrier power sharing is not activated

On c 1, we have: and .

On c 2 , we have: , and .

Therefore,

' X k UL

F UL 1

1

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req---------------------------------------------- 1+

© ¹¨ ¸¨ ¸¨ ¸§ ·--------------------------------------------------------------u=

QreqUL

k W

R nominal UL

k ------------------------------

P bUL

k req

I intra I +extra

I in ter carr ier – N 0 tx

+ +------------------------------------------------------------------------------------u=

QreqUL

k W

R nominal UL

k ------------------------------

P bUL

k req

I intra F UL

u N 0 tx

+-------------------------------------------u=

P bUL

k re q QreqUL

k R nominal

ULk

W ------------------------------u I intra F

ULu N 0

tx + u=

Ec

Nt ------- k © ¹

§ ·E DPDCH –

re q

Qre qUL

k R nominal

ULk

W ------------------------------u=

P bUL

k re qEc

Nt ------- k © ¹

§ ·E DPDCH –

re q

I intra F ULu N 0

tx + u=

I intra P bUL

k re q

K

¦=

I intra I intra F ULu N 0

tx +

Ec

Nt ------- k © ¹

§ ·E DPDCH –

req

K

¦u=

I intra

N 0 tx Ec

Nt ------- k © ¹

§ ·E DPDCH –

req

K

¦u

1 F UL

– ---------------------------------------------------------------------=

X UL I intra I +

extraI in ter carr ier – +

I intra I +extra

I in ter carr ier – N 0 tx

+ +------------------------------------------------------------------------------------

I intra F ULu

I intra F ULu N 0

tx +

-------------------------------------------1

1N 0

tx

I intra F ULu

----------------------------+

--------------------------------------= = =

X UL

F UL Ec

Nt ------- k © ¹

§ ·E DPDCH –

req

K

¦u=

' X k UL

F UL Ec

Nt ------- k © ¹

§ ·E DPDCH –

req

u=

P max Tx c 1 43dBm= P t x R99 – Tx c 1 39.1dBm=

P max Tx c 2 43dBm= P t x R99 – Tx c 2 36.1dBm= P Headroom Tx c 2 0d B=

P HSDPA Tx c 2 P max Tx c 2 P t x R99 – Tx c 2 – P Headroom Tx c 2 – = 42dBm=



© Forsk 2010 AT282_TRG_E2 259


• 2nd case: Inter-carrier power sharing is activated and

On c 1, we have: and .

On c 2 , we have: , and .

Therefore,

6.4.4.6 Best Server Determination in Monte Carlo Simulations - Old Method

Before Atoll 2.8.0, best server determination used to be performed by selecting the best carrier within transmitters accord-

ing to the selected method (site equipment) and then the best transmitter using the best carrier. To switch back to thismethod, add the following lines in the Atoll.ini file:

The method is described below:

For each station txi containing M b in its calculation area and using the main frequency band supported by the M b’s terminal

(i.e. either f1 for a single frequency band network, or f1 or f2 for a dual-band terminal without any priority on frequency

bands, or f1 for a dual-band terminal with f1 as main frequency band).

Determination of .

If a given carrier is specified for the service requested by M b and if it is used by txi

is the carrier specified for the service.

Else the carrier selection mode defined for txi is considered.

If carrier selection mode is “Min. UL Load Factor”

For each carrier ic used by txi , we calculate current loading factor:

EndFor

is the carrier with the lowest

Else if carrier selection mode is “Min. DL Total Power”

is the carrier with the lowest

Else if carrier selection mode is “Random”

is randomly selected

Else if carrier selection mode is "Sequential"

is the first carrier so that

Calculation of


Rejection of station txi if the pilot is not received

If then txi is rejected by M b

If

Admission control (If simulation respects a loading factor constraint and M b was not connected in previous iteration).

P max Tx 46dBm=

P ma x Tx c 1 43dBm= P t x R99 – Tx c 1 39.1dBm=

P ma x Tx c 2 46dBm= P t x R99 – Tx c 2 36.1dBm= P Headroom Tx c 2 0d B=

P HSDPA Tx c 2 P max Tx P t x R99 – Tx c 1 – P t x R99 – Tx c 2 – P Headroom Tx c 2 – = 44.4dBm=

[CDMA]

MultiBandSimu = 0

BestCarrier k tx i M b


X k UL

tx i ic I to t UL

tx i ic

N tot UL

tx i ic ------------------------------ ' X

UL+=

BestCarrier k tx i M b X k UL

tx i ic

BestCarrier k tx i M b P tx tx i ic k


BestCarrier k tx i M b X k UL

tx i ic X ma x UL

d

Q pil ot k tx i BestCarrier

D UBT Su P uc

tx i M b BestCarrier

P tot DL

tx i BestCarrier k tx i M b I extraDL

BestCarrier k tx i M b + +


BestCarrier k tx i M b I i nt er t e chn o y log – DL

BestCarrier k tx i M b N 0 Term

+ +© ¹¨ ¸¨ ¸§ ·--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pil ot k tx i BestCarrier

D UBT Su P uc

tx i M b BestCarrier

P tot DL

txi BestCarrier k tx i M b I extraDL

BestCarrier k tx i M b +

I + in ter carr ier – DL

BestCarrier k tx i M b I i nt er t echn o y log – DL

BestCarrier k tx i M b +

N 0 Term

1 D – UBTSu P c tx i M b BestCarrier u – +© ¹¨ ¸¨ ¸¨ ¸¨ ¸¨ ¸§ ·-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------=

Q pil ot k tx i M b BestCarrier Qreq

pi lo t Mobil i ty M b

Q pil ot k tx i M b BestCarrier Q pi lo t k

max M b !





© Forsk 2010 AT282_TRG_E2 261


Endif

EndFor

If no Tx BS has been selected, M b has failed to be connected to the network and is rejected.

6.5 UMTS HSPA Prediction Studies

6.5.1 Point Analysis6.5.1.1 AS Analysis Tab

Let us suppose a receiver with a terminal, a service and a mobility type. This receiver does not create any interference.

You can make the prediction for a specific carrier or for all carriers of the main frequency band for the selected terminal.

If you have selected a dual-band terminal, you can make the coverage prediction on a specific carrier or on all carriers of

any frequency band for the selected terminal, or for all the carriers of all the frequency bands. The analysis is based on

the following parameters:

• The uplink load factor and the downlink total power of cells,

• The available HSDPA power of the cell in case of an HSDPA bearer user,

• The cell UL reuse factor, the cell UL load factor due to HSUPA and the maximum cell UL load factor for HSUPA

bearer users.

These parameters can be results of a given simulation, average values calculated from a group of simulations, or user-

defined cell inputs. In the last case, when no value is defined in the Cells table, Atoll uses the following default values:

• Total transmitted power = 50% of the maximum power (i.e, 40 dBm if the maximum power is set to 43 dBm)

• Uplink load factor = 50%.

• Uplink reuse factor = 1

• Uplink load factor due to HSUPA = 0%

• Maximum uplink load factor = 75%

On the other hand, no default value is used for the HSDPA power; this parameter must be defined by the user.

6.5.1.1.1 Bar Graph and Pilot Sub-Menu

We can consider the following cases:

1st case: Analysis based on a specific carrier

The carrier that can be used by transmitters is fixed. In this case, for each transmitter i containing the receiver in its calcu-

lation area and using the selected carrier, Atoll calculates the pilot quality at the receiver on this carrier. Then, it deter-

mines the best serving transmitter using the selected carrier ic .

2nd case: Analysis based on all carriers

Atoll determines the best carrier for each transmitter i which contains the receiver in its calculation area and uses a

frequency band supported by the receiver’s terminal. The best carrier selection depends on the option selected for the site

equipment (UL minimum noise, DL minimum power, random, sequential). Then, Atoll calculates the pilot quality at the

receiver from these transmitters on their best carriers ( ic) and defines the best server (on its best carrier).

3rd case: Analysis based on all carriers of any frequency band (for dual-band terminals with priority defined on frequency

bands only)

The frequency band that can be used is fixed. Atoll determines the best carrier for each transmitter i containing the

receiver in its calculation area and using the selected frequency band. The best carrier selection depends on the option

selected for the site equipment (UL minimum noise, DL minimum power, random, sequential). Then, Atoll calculates the

pilot quality at the receiver from these transmitters on their best carriers (ic) and defines the best server (on its best carrier).

Ec/I0 (or ) Evaluation

Let us assume that ic is either the best carrier or the selected carrier of a transmitter i containing the receiver in its radiuscalculation and ic adj is another carrier adjacent to ic . An interference reduction factor, , is defined between ic

and ic adj and set to a value different from 0.

Two ways may be used to calculate I0.

Option Total noise: Atoll considers the noise generated by all the transmitters and the thermal noise.

Option Without pilot: Atoll considers the total noise deducting the pilot signal.

Calculation option may be selected in Global parameters.

Therefore, we have:

Tx BS M b tx i =

Q pi lot ic

RF ic ic adj

Q pil ot i ic UBT S Du P c i ic u

I 0 DL

ic --------------------------------------------------=




With,


And

for the without

pilot option.

1st step: calculation for each cell (i,ic )

is the pilot power of a transmitter i on carrier ic at the receiver.

is the total loss between transmitter i and receiver.

2nd step: , and calculations

We have:

and

For each transmitter of the network, is the total power received at the receiver from the transmitter on the best

carrier ic of the transmitter i .

is the total power transmitted by the transmitter on the best carrier. Total power transmitted by each cell is either

a simulation result (provided in Simulation properties (Cells tab)) or a value user-defined in Cell properties.

For each transmitter of the network, is the total power received at the receiver from the transmitter on the

carrier ic adj . This carrier is adjacent to ic .

is the total power transmitted by the transmitter on the carrier ic adj . Total power transmitted by each cell is either

a simulation result (provided in Simulation properties (Cells tab)) or a value user-defined in Cell properties.

3rd step: calculation

4th step: and evaluation using formulas described above

5th step: calculation

The macro-diversity gain, , models the decrease in shadowing margin due to the fact there are several

available pilot signals at the mobile.

I 0 DL

ic P tot DL

i ic I extraDL



ic N + + 0

term+ +=

I 0 DL

ic I intraDL

ic I extraDL

ic + I in ter carr ier –

DLic I i nt er t e chn o y log –

DLic N + + 0

term1 D – UBT S P c i ic uu – +=

P c i ic

P c i ic

P c i ic P pi lo t i ic

LT I

----------------------------=

LT I

LT I

LTx L pat h Ltermu Lbody u LIndoor M Shadowing Ec Io e – uuu

GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=

P tot DL

j ic P tot DL

i ic P to t DL

j ic adj

I extraDL

ic P to t DL

j ic

t xj j i z¦=

I intraDL

ic P tot DL

i ic UBT S Du P tot DL

i ic P SCH ic

LT

------------------------ – © ¹§ ·u – =


ic

P tot DL

j ic adj


-------------------------------------------=


ic P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦=

P tot DL

ic

P tot DL

ic P Tx ic

LT

-------------------=

P Tx ic

P tot DL

ic adj

P tot DL

ic adj P Tx ic adj

LT

--------------------------=

P Tx ic adj

N 0 term

N 0 term

NF Term K T u W NR i nt er t e chn o y log – Tx DLuuu=

I 0 DL

ic Q pi lo t i ic




M Shadowing Ec Io e – npaths

M Shadowing Ec Io e – – =



© Forsk 2010 AT282_TRG_E2 263


is the shadowing margin when the mobile receives n pilot signals (not necessarily from transmitters

belonging to the mobile active set).

6th step: Determination of active-set

Atoll takes the transmitter i with the highest and calculates the best pilot quality received with a fixed cell edge

coverage probability, .

If , it means pilot quality at the receiver exceeds x% of time (x is the fixed cell edge cover-

age probability). The cell whose is the highest one enters the active set as best server ( ) and

the best carrier (ic BS) of the best server, BS, will be the carrier used by other transmitters of the active set (when active

set size is greater than 1). Pilot is available.

If , no cell (i,ic) can enter the active set. Pilot is unavailable.

Then, pilot qualities at the receiver from transmitters i (except the best server) on the best carrier of the best server, ic BS,

are recalculated to determine the entire receiver active set (when active set size is greater than 1). Same formulas and

calculation method are used to update value and determine .

We have:

With,


And

for the without

pilot option.

Other cells (i,ic BS ) in the active set must satisfy the following criteria:

(optionally)

Number of Cells in Active Set

This is a user-specified input in the Terminal properties. It corresponds to the active set size.

Thermal Noise

This parameter is calculated as described above (3rd step).

I0 (Best Server)

I0 (Best server) is the total noise received at the receiver on ic BS. The notation “Best server” refers to the best server of

active set. This is relevant when using the calculation option “Without pilot”. In this case, it informs that the pilot signal of

the best server (BS,ic BS

) is deducted from the total noise.

Downlink Macro-Diversity Gain

This parameter is calculated as described above (5th step).

6.5.1.1.2 Downlink Sub-Menu

The Downlink sub-menu may contain R99-related results and HSDPA-related results when an HSPA bearer user is

modelled.

• R99-related Results

Atoll calculates the traffic channel quality from each cell (k,ic BS ) of the receiver’s active set at the receiver. No power

control is performed as in simulations. Here, Atoll determines the downlink traffic channel quality at the receiver for the

Note:

• This parameter is determined from cell edge coverage probability and Ec/I0 standard

deviation. When the Ec/I0 standard deviation is set to 0, the macro-diversity gain equals 0.


npaths

Q pi lot i ic

Q pil ot Resulting

ic

Q pil ot Resulting

ic Gmacro diversity – DL

max Q pi lo t i ic u=

Q pi lot Resulting

Q pi lo t re qt Q pi lot

Resulting ic

Q pil ot i ic Q pi lot BS ic

Q pil ot Resulting

Q pi lo t req

I 0 DL

ic BS Q pi lo t i ic BS

Q pil ot i ic UBT S Du P c i ic u

I 0 DL

ic --------------------------------------------------=

I 0 DL

ic P tot DL

i ic I extraDL



ic N 0 term

+ + + +=

I 0 DL

ic I intraDL

ic I extraDL

ic + I in ter carr ier – DL


ic N + + 0

term1 D – UBT S P c i ic uu – +=

Q pi lo t i ic BS Q pi lot BS ic BS – AS_threshold i BS ic BS t

i ic BS neighbour list i BS ic BS




maximum allowed traffic channel power per transmitter. Then, after combination, the total downlink traffic channel quality

is evaluated and compared with the specified target quality.

Eb/Nt Target

Eb/Nt target ( ) is defined for a given R99 bearer, a mobility type and a reception equipment. This parameter is avail-

able in the R99 Bearer Selection table.

Required transmitter power on traffic channels

The calculation of the required transmitter power on traffic channels ( ) may be divided into three steps.

1st step: evaluation for each cell

Let us assume the following notation: Eb/Nt max corresponds to

Therefore, for each cell (k,ic BS ), we have:

With

and

Where

is the maximum power allowed on traffic channels. This parameter is user-defined in the R99 Radio Bearers table.

is the total noise at the receiver on the best carrier of the best server.

is the intra-cell interference at the receiver on the best carrier of the best server.

is the extra-cell interference at the receiver on the best carrier of the best server.

is the inter-carrier interference at the receiver on the best carrier of the best server.

ic adj is a carrier adjacent to ic BS.


is the inter-technology interference at the receiver on the best carrier of the best server.

Notes:

• Compressed mode is operated when:

- A mobile supporting compressed mode is connected to a cell located on a site with a

compressed-mode-capable equipment

And

- Either the received Ec/I0 is lower than the Ec/I0 activation threshold (Global parameters):

,

- Or the pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):

.

• When compressed mode is activated, the downlink Eb/Nt target is increased by the value

user-defined for the DL Eb/Nt target increase field (Global parameters), .

QreqDL

Q pi lo t Resulting

Q pi lot CM act ivat ion –

d

P c RSCP pil ot CM act ivat ion –

d

'QreqDL

P tc h

re q

Qmax DL

k ic BS

Qmax DL

Qmax DL

k ic BS UBT S P b m ax –

DLk ic BS u

N tot DL

ic BS -------------------------------------------------------------- G p

DLGDiv

DLuu=

P b m ax –

DLk ic BS

P tchma x

LT k

-------------=

N tot DL

ic BS I intraDL

ic BS I extraDL

ic BS I in ter carr ier – DL

ic BS I i nt er t echn o y log – DL

ic BS N 0 term

+ + + +=

P tchma x

N tot DL

ic BS

I intraDL

ic BS

I intraDL

ic BS P tot DL

k ic BS UBTS F orthou – P tot DL

k ic BS P SC H k ic BS

LT

------------------------------------- – © ¹§ ·u=

I extraDL

ic BS

I extraDL

ic BS P tot DL

j ic BS

j j k z

¦=


ic BS


ic BS

P tot DL

j ic adj

tx j j ¦RF ic BS ic adj

-------------------------------------------=

RF ic BS ic adj


DLic BS



© Forsk 2010 AT282_TRG_E2 265





2nd step: Calculation of the total traffic channel quality

is the traffic channel quality at the receiver on ic BS after signal combination of all the transmitters k of the active set.

On downlink, if there is no handoff, we have:

For any other handoff status, we have:

Where

is the downlink rake efficiency factor defined in Terminal properties.


Eb/Nt Max for Each Cell of Active Set

For each cell (k,ic BS ), we have:

With

Notes:



compressed-mode-capable equipment

And

- Either the received Ec/I0 is lower than the Ec/I0 activation threshold (Global parameters):

.

- Or the pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):

• When compressed mode is activated, the downlink Eb/Nt target is increased by the value

user-defined for the DL Eb/Nt target increase field (Global parameters), . In this

case, we have:


ic BS P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic BSTx m

u----------------------------------------------

ni

¦=

ic i i th

IC P ic i ic BSTx m

ic i ic BS

QMAX DL

QMAX DL

ic BS Qmax DL

k ic BS =

QMAX DL

ic BS f rake efficiency DL

Qmax DL

k ic BS

k

¦u=

f rake efficiency DL

P tchreq

P tc hre q Qre q

DL

QMAX DL

ic BS -------------------------------- P tch

max u=

Q pil ot Resulting


d

P c RSCP pi lo t CM act ivat ion –

d

'QreqDL

P tchreq Qreq

DL 'QreqDLu

QMAX DL

ic BS ---------------------------------- P tch

max u=

Qmax DL

k ic BS UBT S P b max –

DLk ic BS u

N tot DL

ic BS -------------------------------------------------------------- G p

DLGDiv

DLuu=

P b max – DL k ic BS P tch

max

LT k

-------------=

N tot DL

ic BS I intraDL

ic BS I extraDL

ic BS I in ter carr ier – DL

ic BS I i nt er t e chn o y log – DL

ic BS N 0 term

+ + + +=

I intraDL

ic BS P = tot DL

k ic BS UBTS F orthou – P tot DL

k ic BS P SC H k ic BS

LT

------------------------------------- – © ¹§ ·u 1 UBT S – ma x

P tchmax

P tchre q

–

LT k

------------------------------ 0 ( , )u –

I extraDL

ic BS P tot DL

j ic BS

j j k z¦=




Where

is the required transmitter power on traffic channels.

Eb/Nt Max

is the traffic channel quality at the receiver on ic BS after signal combination of all the transmitters k of the active set.

On downlink, if there is no handoff, we have:

For any other handoff status, we have:

Where

is the downlink rake efficiency factor defined in Terminal properties.

Therefore, the service on the downlink traffic channel is available if (or

when compressed mode is activated).

Effective Eb/Nt

is the effective traffic channel quality at the receiver on ic BS.

(or when compressed mode is activated).

Downlink Soft Handover Gain

corresponds to the DL soft handover gain.

corresponds to the highest value.

• HSDPA-related Results

Atoll determines the best HSDPA bearer that the user can obtain. The HSDPA bearer user is processed as if he is the

only user in the cell i.e. he uses the entire HSDPA power available in the cell.

For further information on the fast link adaptation modelling, see "Fast Link Adaptation Modelling" on page 226.

HS-PDSCH Ec/Nt

Atoll calculates the best HS-PDSCH quality (HS-PDSCH Ec/Nt). The way of calculating it depends on the selected option

in the transmitters global parameters (HSDPA part): CQI based on CPICH quality or CQI based on HS-PDSCH quality.

For further details on the HS-PDSCH quality calculation, see either "HS-PDSCH Quality Calculation" on page 228 if theselected option is "CQI based on CPICH quality" or "HS-PDSCH Quality Calculation" on page 233 if the selected option

is "CQI based on HS-PDSCH quality".

HS-SCCH Ec/Nt

When the HS-SCCH power allocation strategy is dynamic, this parameter corresponds to the HS-SCCH Ec/Nt threshold

defined for the selected mobility type.

When the HS-SCCH power allocation strategy is static, the HS-SCCH Ec/Nt is calculated from the fixed HS-SCCH power.

We have:



ic BS

P tot DL

j ic adj

tx j j ¦RF ic BS ic adj

-------------------------------------------=


ic BS P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic BSTx m

u----------------------------------------------

ni

¦=

P tchre q

QMAX DL

QMAX DL

ic BS Qmax DL

k ic BS =

QMAX DL

ic BS f rake efficiency DL

Qmax DL

k ic BS

k

¦u=

f rake efficiency

DL

QMAX DL

ic BS Qre qDLt QMAX

DLic BS Qre q

DL 'Qre qDLut

Qeff DL

Qeff DL

min QMAX DL

QreqDL = Qeff

DLmin QMAX

DLQre q

DL 'QreqDLu =

GSHODL

GSHODL QMAX

DL ic BS

max Qmax DL

k ic BS ------------------------------------------------------=

max Qmax DL

k ic BS Qmax DL

k ic BS

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBT S P c i ic u

N to t DL

ic ------------------------------------=



© Forsk 2010 AT282_TRG_E2 267


And


With







And


CQI

It corresponds to the HS-PDSCH CQI. The way of calculating it depends on the selected option in the transmitters global

parameters (HSDPA part): CQI based on CPICH quality or CQI based on HS-PDSCH quality.

For further details on the HS-PDSCH quality calculation, see either "HS-PDSCH CQI Determination" on page 230 if the

selected option is "CQI based on CPICH quality" or "HS-PDSCH CQI Determination" on page 235 if the selected option is

"CQI based on HS-PDSCH quality".

RLC Peak Rate

Knowing the HS-PDSCH CQI, Atoll calculates the best HSDPA bearer that can be used and selects a bearer compatible

with cell and terminal user equipment HSDPA capabilities. Once the bearer selected, Atoll determines the RLC peak rate

that can be provided to the user, .

For further details of the HSDPA bearer selection, see "HSDPA Bearer Selection" on page 230.

Bearer Consumption

Atoll provides this result for packet (HSPA - Constant Bit Rate) service users only. The minimum bit rate required by the

service is allocated to these users. Therefore, they parly consume the HSDPA bearer. The bearer consumption expressed

in %, , is calculated as follows:

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

UBTS P c i ic u

N tot DL


– u UBTS P c i ic uu –

---------------------------------------------------------------------------------------------------------------------------------------=

N tot DL

ic I intraDL

ic I extraDL



ic N + + 0

term+ +=

I intraDL

ic P tot DL

ic

tx i


DLic

tx i

P SCH ic

LT

------------------------ –

© ¹

¨ ¸¨ ¸§ ·

UBTS – u P tot DL

ic

tx i

P SCH ic

LT

------------------------ –

© ¹

¨ ¸¨ ¸§ ·

u=

I extraDL

ic P to t DL

ic

t xj j i z¦=


ic

P tot DL

ic adj


---------------------------------------=

RF ic ic adj


ic

I i nt er t echn o y log – DL ic P Transmitted

Tx

ic i Ltotal

Tx IC P ic i ic

Tx mu------------------------------------------

ni

¦=

ic i i th

IC P ic i ic Tx m

ic i ic

P c i ic

P HS SCCH – ic

LT i

---------------------------------------=

LT


GTx Gtermu---------------------------------------------------------------------------------------------------------------------------------------------------=


N 0 term

R RLC peak – DL

C HSDPABearer

C HSDPABearer

R Guaranteed DL

R RLC peak – DL

I HSDPABearer ---------------------------------------------------------------------=





© Forsk 2010 AT282_TRG_E2 269


Else,


is the required terminal power.

is the uplink traffic quality target defined by the user for a given reception equipment, a given R99 bearer and a given -

mobility type. This parameter is available in the R99 Bearer Selection table.

Therefore, the service on the uplink traffic channel is available if .

Eb/Nt Max

For each cell (k,ic BS ) in the receiver’s active set, we have:

With

is the total noise at the transmitter on the best carrier of the best server. This value is calculated from the

cell uplink load factor .

is the transmitter thermal noise.

is the traffic channel quality at the transmitter on ic BS after signal combination of all the transmitters k of the

active set.

If there is no handoff (1/1):

For soft handoff (2/2):

is the uplink macro-diversity gain. This parameter is determined from the fixed cell edge coverage

probability and the uplink Eb/Nt standard deviation. When the option “Shadowing taken into account” is not selected

(Prediction properties), Atoll considers the uplink macro-diversity gain defined by the user in Global parameters.

Notes:



compressed-mode-capable equipment, and

- The received Ec/I0 is lower than the Ec/I0 activation threshold (Global parameters):

.

- The pilot RSCP is lower than the pilot RSCP activation threshold (Global parameters):

• When compressed mode is activated, the uplink Eb/Nt target is increased by the value

user-defined for the UL Eb/Nt target increase field (Global parameters), . In this

case, we have:

QMAX UL

ic BS Gmacro diversity – UL 2 links max f rake efficiency

ULQmax

ULk ic BS

k on the same site

¦u Qmax k on the same site

ULk ic BS

© ¹¨ ¸§ ·

u=

QMAX UL

ic BS Gmacro diversity – UL 2 links max Qmax

ULk ic BS u=

P term R99 – req

P term R99 –

re q

P term R99 – re q Qre q

UL

QMAX UL

ic BS -------------------------------- P term

max u=

QreqUL

Q pil ot Resulting


d

P c RSCP pi lo t CM act ivat ion – d

'QreqUL

P term R99 – req Qre q

UL 'Qre qULu

QMAX UL

ic BS ---------------------------------- P term

max u=

P term R99 – req

P termmax d

Qmax UL

k ic BS Uterm P b m ax – UL k ic BS uN tot

ULk ic BS

--------------------------------------------------------------- G pUL

GDiv UL

uu=

P b max – UL

k ic BS P term

max 1 r c

UL – u

LT k

-------------------------------------------=

N tot UL

k ic BS

X UL

k ic BS

N tot UL

k ic BS N 0

tx

1 X UL

k ic BS – ------------------------------------------ 1 Uterm – ma x

P termmax

P term R99 – req

–

LT k

----------------------------------------------- 0 ( , )u+=

N 0 tx

QMAX UL

ic BS

QMAX UL

ic BS Qmax UL

k ic BS =

QMAX UL


ULk ic BS u=

Gmacro diversity – UL 2 links





For soft-soft handoffs (3/3):

is the uplink macro-diversity gain. This parameter is determined from the fixed cell edge coverage

probability and the uplink Eb/Nt standard deviation. When the option “Shadowing taken into account” is not selected

(Prediction properties), Atoll considers the uplink macro-diversity gain defined by the user in Global parameters.

For softer and softer-softer handoffs (1/2 and 1/3):

For softer-soft handoffs (2/3), there are two possibilities. If the MRC option is selected (option available in Global param-

eters), we have:

Else,

Effective Eb/Nt

is the effective traffic channel quality at the transmitter on ic BS.


Uplink Soft Handover Gain

corresponds to the uplink soft handover gain.


• HSUPA-related Results

Atoll determines the best HSUPA bearer that the user can obtain. The HSUPA bearer user is processed as if he is theonly user in the cell i.e. he uses the entire remaining load of the cell.

For further information on the HSUPA bearer selection, see "HSUPA Bearer Allocation Process" on page 240.

Required E-DPDCH Ec/Nt

It corresponds to the E-DPDCH Ec/Nt required to obtain the HSUPA bearer ( ). This value is defined for an

HSUPA bearer ( ) and a number of retransmissions ( ) in the HSUPA Bearer Selection table.

Required Terminal Power

From , Atoll calculates the terminal power required to obtain the HSUPA bearer, .

With


max Qmax UL

k ic BS Qmax UL

k ic BS

QMAX UL


ULk ic BS u=


3 links

QMAX UL

ic BS f rake efficiency UL

Qmax UL

k ic BS

k

¦u=

QMAX UL

ic BS Gmacro diversity –

UL 2 links max f rake efficiency

ULQmax

ULk ic BS

k on the same site

¦u Qmax k on the same site

ULk ic BS

© ¹¨ ¸§ ·

u=

QMAX UL


ULk ic BS u=

Qeff UL

Qeff UL

min QMAX UL

QreqUL = Qeff

ULmin QMAX

ULQre q

ULQ' req

ULu =

GSHOUL

GSHOUL QMAX

ULic BS

max Qmax UL

k ic BS ------------------------------------------------------=

max Qmax UL

k ic BS Qmax UL

k ic BS

Ec

Nt -------

© ¹§ ·

E DPDCH –

req

Index HSUPABearer N Rtx

Ec

Nt -------

© ¹§ ·

E DPDCH –

re q


P term HSUPA – re q Ec

Nt -------

© ¹§ ·=

E DPDCH –

re q

LT N tot ULuu

N tot UL

ic 1 F – MUDtx

Utermu I u tot

UL intraic I to t


ULic N + 0

tx + +=

LT


GTx Gtermu------------------------------------------------------------------------------------------------------------------------------------------------------------ -=

Uterm F MUDtx

I tot

UL intraI tot

ULextraI in ter carr ier –

ULN 0

tx



© Forsk 2010 AT282_TRG_E2 271


RLC Peak Rate

Atoll selects the best HSUPA bearer from the HSUPA compatible bearers. This is the HSUPA bearer with the highest

potential throughput ( ) where:

•

• And

With

: the maximum E-DPDCH Ec » Nt allowed.

: the maximum terminal power allowed.

After selecting the HSUPA bearer, Atoll determines the corresponding RLC peak rate, .

Application Throughput

Atoll displays the provided application throughput ( ). The application throughput represents the net throughput

after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:

Where:

is the residual BLER after retransmissions. It is read in the quality graph defined for the quartet “recep-

tion equipment-selected bearer-number of retransmissions-mobility” (HSUPA Quality Graphs tab in the Reception equip-

ment properties). This graph describes the variation of BLER as a function of the measured quality (E-DPDCH Ec/Nt).

Knowing the E-DPDCH Ec/Nt, Atoll finds the corresponding BLER.




Bearer Consumption

Atoll provides this result for packet (HSPA - Constant Bit Rate) service users only. The minimum bit rate required by the

service is allocated to these users. Therefore, they parly consume the HSUPA bearer. The bearer consumption expressed

in %, , is calculated as follows:

6.5.2 Coverage Studies

Let us assume each pixel on the map corresponds to a probe receiver with a terminal, a mobility type and a service. This

receiver does not create any interference. You can make the coverage prediction for a specific carrier or for all carriers of

the main frequency band for the selected terminal. If you have selected a dual-band terminal, you can make the coverage

prediction on a specific carrier or on all carriers of any frequency band for the selected terminal, or for all the carriers of all

the frequency bands. Coverage predictions are based on parameters that can be either simulation results, or user-defined

cell inputs.

6.5.2.1 Pilot Reception AnalysisFor further details of calculation formulas and methods, please refer to Definitions and formulas part, and Point analysis

– AS analysis tab – Pilot sub-menu part.

We consider the following cases:

1st case: Analysis Based on a Specific Carrier

The carrier that can be used by transmitters is fixed. In this case, for each transmitter i containing the receiver in its calcu-

lation area and using the selected carrier, Atoll calculates the pilot quality at the receiver on this carrier ic given. Then, it

determines the best serving transmitter BS using the carrier ic given ( ) and calculates the best pilot quality

received with a fixed cell edge coverage probability, .

Atoll displays the best pilot quality received with a fixed cell edge coverage probability.

R RLC peak – UL

Index HSUPABearer

N Rt x Index HSUPABearer ------------------------------------------------------------------------------------

Ec

Nt -------

© ¹§ ·

E DPDCH –

req Ec

Nt -------

© ¹§ ·

E DPDCH –

max

d


P d term

max

Ec

Nt -------

© ¹§ ·

E DPDCH –

max

P termma x

R RLC peak – UL

T applicationUL

T applicationUL

M b R RLC p – eak

UL1 B LE R HSUPA – SF Rateuu 'R –

N Rtx

--------------------------------------------------------------------------------------------------------------------------=

BLER HSUPA N Rtx

SF Rate 'R

C HSUPABearer

C HSUPABearer

R Guaranteed UL

R RLC peak – UL

I HSUPABearer ---------------------------------------------------------------------=

Q pil ot BSic given

Q pi lot Resulting

ic given




2nd case: Analysis Based on All Carriers

Atoll proceeds as in point predictions. It determines the best carrier of each transmitter i containing the receiver in its calcu-

lation area and using a frequency band supported by the receiver’s terminal. The best carrier selection depends on the

option selected for the site equipment (UL minimum noise, DL minimum power, random, sequential) and is based on the

UL load percentage and the downlink total power of cells (simulation results or cell properties). Atoll calculates the pilot

quality at the receiver from these transmitters on their best carriers and determines the best serving transmitter BS on its

best carrier ic BS ( ). Then, it calculates the best pilot quality received with a fixed cell edge coverage proba-

bility, .


3rd case: Analysis based on all carriers of any frequency band (for dual-band terminals with priority defined on frequency

bands only)

The frequency band that can be used is fixed. Atoll determines the best carrier of each transmitter i containing the receiver

in its calculation area and using the selected frequency band. The best carrier selection depends on the option selected

for the site equipment (UL minimum noise, DL minimum power, random, sequential) and is based on the UL load percent-

age and the downlink total power of cells (simulation results or cell properties). Then, Atoll calculates the pilot quality at

the receiver from these transmitters on their best carriers and determines the best serving transmitter BS on its best carrier

ic BS ( ). Then, it calculates the best pilot quality received with a fixed cell edge coverage probability,

.


6.5.2.1.1 Prediction Study Inputs

The Pilot Reception Analysis depends on the downlink total transmitted power of cells. This parameter can be either a

simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the totaltransmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power is set

to 43 dBm).

6.5.2.1.2 Study Display Options

Single colour

Atoll displays a coverage if . Coverage consists of a single layer with a unique colour

( ). is a target value defined in the Mobility table by the user.

Colour per transmitter

Atoll displays a coverage if ( ). Coverage consists of several layers with asso-

ciated colours. There is a layer per transmitter with no intersection between layers. Layer colour is the colour assigned to

the best serving transmitter BS.

Colour per mobility

In this case, receiver is not completely defined and no mobility is assigned.

Coverage consists of several layers with a layer per user-defined mobility defined in Mobility sub-folder. For each layer,

area is covered if ( ). Each layer is assigned a colour and displayed with inter-

sections between layers.

Colour per probability

This display option is available only if analysis is based on all simulations in a group (i.e. if you select a group of simulations

and the “All” option in the Simulation tab of prediction properties). Coverage consists of several layers with a layer per

user-defined probability level defined in the Display tab (Prediction properties). For each layer, area is covered if

( ) in the required number of simulations. Each layer is assigned a colour and

displayed with intersections between layers.

Colour per cell edge coverage probability

Coverage consists of several layers with a layer per user-defined cell edge coverage probability, p, defined in the Display

tab (Prediction properties). For each layer, area is covered if ( ). Each layer

is assigned a colour and displayed with intersections between layers.

Q pi lot BSic BS

Q pi lot Resulting

ic BS

Q pi lot BSic BS

Q pil ot Resulting

ic BS

Q pi lot Resulting

ic Q pil ot reqt

ic ic BS or ic given= Q pi lot req

Q pi lot Resulting

ic Q pi lo t re qt ic ic BS or ic given=

Q pil ot Resulting


Q pil ot Resulting


Q pi lot Resulting

ic p Q pi lot reqt ic ic BS or ic given=



© Forsk 2010 AT282_TRG_E2 273


Colour per quality level (Ec/I0)

Coverage consists of several layers with a layer per user-defined quality threshold defined in the Display tab (Prediction

properties). For each layer, area is covered if ( ). Each layer is

assigned a colour and displayed with intersections between layers.

Colour per quality margin (Ec/I0 margin)

Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction prop-

erties). For each layer, area is covered if ( ). Each layer is


6.5.2.2 Downlink Service Area Analysis

As in point predictions, Atoll calculates traffic channel quality at the receiver for each cell (k,ic) (with ic=ic BS or ic given) in

the receiver’s active set. No power control is performed as in simulations. Here, Atoll determines downlink traffic channel

quality at the receiver for a maximum allowed traffic channel power for transmitters. Then, the total downlink traffic channel

quality ( ) is evaluated after recombination.

Atoll displays traffic channel quality at the receiver for transmitters in active set on the carrier ic ( or ).

For further details of calculation formulas and methods, see "Downlink Sub-Menu" on page 263.


The Downlink Service Area Analysis depends on the downlink total transmitted power of cells. This parameter can be

either a simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the

total transmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power

is set to 43 dBm).


Single colour

Atoll displays a coverage with a unique colour if (or if compressed mode is

activated).

is the downlink traffic quality target defined by the user for a given reception equipment, a given R99 bearer and a

given mobility type. This parameter is available in the R99 Bearer Selection table.

is the DL Eb/Nt target increase; this parameter is user-defined in the Global parameters.


Atoll displays a coverage if (or if compressed mode is activated). Coverage

consists of several layers with associated colours. There is a layer per transmitter with no intersection between layers.

Layer colour is the colour assigned to best serving transmitter.

Colour per mobility

In this case, receiver is not completely defined and no mobility is assigned. Coverage consists of several layers with a

layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if (or

if compressed mode is activated). Each layer is assigned a colour and displayed with intersec-

tions between layers.

Colour per service

In this case, receiver is not completely defined and no service is assigned. Coverage consists of several layers with a layer

per user-defined service defined in Services sub-folder. For each layer, area is covered if (or



Q pi lot Resulting

ic Q pi lot threshold

t ic ic BS or ic given=

Q pi lot Resulting

ic Q pil ot req

– Q pil ot m inarg

t ic ic BS or ic given=

Note:

• Best server and active set determination is performed as in point prediction (AS analysis).

QMAX DL

ic

ic BS ic given

QMAX DL

ic QreqDLt QMAX

DLic Qre q

DLQ' req

DLut

QreqDL

'QreqDL

QMAX DL

ic QreqDLt QMAX

DLic Qre q

DLQ' req

DLut

QMAX DL

ic QreqDLt

QMAX DL ic Qreq

DL Q' reqDLut

QMAX DL

ic QreqDLt

QMAX DL

ic QreqDL

Q' reqDLut





© Forsk 2010 AT282_TRG_E2 275


mitter for the maximum terminal power allowed. Then, the total uplink traffic channel quality ( ) is evaluated with

respect to receiver handover status.

Atoll displays traffic channel quality at transmitters in active set on the carrier ic ( or ) received from the

receiver.

For further details of calculations formulas and methods, see "Uplink Sub-Menu" on page 268.

6.5.2.3.1 Prediction Study InputsThe Uplink Service Area Analysis depends on the UL load factor of cells. This parameter can be either a simulation output,

or a user-defined cell input. In the last case, when no value is defined in the Cells table for the uplink load factor, Atoll

uses 50% as default value.


Single colour


colour is unique.

is defined for a reception equipment, a R99 bearer and a mobility type. This parameter is available in the R99 Bearer

Selection table.

is the UL Eb/Nt target increase; this parameter is user-defined in the Global parameters.



consists of several layers with associated colours. There is a layer per transmitter with no intersection between layers.

Layer colour is the colour assigned to best server transmitter.

Colour per mobility

In this case, receiver is not completely defined and no mobility is assigned. Coverage consists of several layers with a

layer per user-defined mobility defined in Mobility sub-folder. For each layer, area is covered if (or



Colour per service

In this case, receiver is not completely defined and no service is assigned. Coverage consists of several layers with a layer

per user-defined service defined in Services sub-folder. For each layer, area is covered if (or



Colour per probability

This display option is available only if analysis is based on all simulations in a group (i.e. if you select a group of simulations

and the “All” option in the Simulation tab of prediction properties). Coverage consists of several layers with a layer per

user-defined probability level defined in the Display tab (Prediction properties). For each layer, area is covered if

(or if compressed mode is activated) in the required number of simulations.Each layer is assigned a colour and displayed with intersections between layers.

Colour per maximum quality level (Max Eb/Nt)


properties). For each layer, area is covered if . Each layer is assigned a colour and displayed with

intersections between layers.

Note:

• Best server and active set determination is performed as in point prediction (AS analysis).

QMAX UL

ic

ic BS ic given

QMAX UL

ic QreqULt QMAX

ULic Qre q

ULQ' req

ULut

QreqUL

'QreqUL

QMAX UL

ic QreqULt QMAX

ULic Qre q

ULQ' req

ULut

QMAX UL

ic QreqULt

QMAX UL

ic QreqUL

Q' reqULut

QMAX UL

ic QreqULt

QMAX UL

ic QreqUL

Q' reqULut

QMAX

UL

ic Qreq

UL

t QMAX

UL

ic Qreq

UL

Q' req

UL

ut

QMAX UL

ic Threshold t




Colour per effective quality level (Effective Eb/Nt)


properties). For each layer, area is covered if . Each layer is assigned a colour and displayed

with intersections between layers.


Colour per quality margin (Eb/Nt margin)

Coverage consists of several layers with a layer per user-defined quality margin defined in the Display tab (Prediction prop-

erties). For each layer, area is covered if (or if

compressed mode is activated). Each layer is assigned a colour and displayed with intersections between layers.

Colour per required power

Coverage consists of several layers with a layer per user-defined power threshold defined in the Display tab (Prediction

properties). For each layer, area is covered if . Each layer is assigned a colour and displayed


Colour per required power margin

Coverage consists of several layers with a layer per user-defined power margin defined in the Display tab (Prediction prop-

erties). For each layer, area is covered if . Each layer is assigned a colour and displayed


Colour per soft handover gain

Coverage consists of several layers with a layer per soft handover gain value defined in the Display tab (Prediction prop-

erties). For each layer, area is covered if . Each layer is assigned a colour and displayed with inter-

sections between layers.

6.5.2.4 Downlink Total Noise Analysis

Atoll determines the downlink total noise generated by cells.

Downlink noise rise, , is calculated from the downlink total noise, , as follows:

6.5.2.4.1 Study Inputs

The Downlink Total Noise Analysis depends on the downlink total transmitted power of cells. This parameter can be either

a simulation output, or a user-defined cell input. In the last case, when no value is defined in the Cells table for the total

transmitted power, Atoll considers 50% of the maximum power as default value (i.e. 40 dBm if the maximum power is set

to 43 dBm).

6.5.2.4.2 Analysis on All Carriers

If all the carriers are selected, Atoll determines DL total noise for all the carriers. Then, allows the user to choose different

colours.

Colour per minimum noise level

Coverage consists of several layers with a layer per user-defined noise level defined in the Display tab (Prediction prop-

erties). For each layer, area is covered if . Each layer is assigned a colour and displayed with


QeffectiveUL

ic Threshold t

Qeff UL

ic min QMAX UL

ic Qre qUL = Qeff

ULic min QMA X

ULic Qre q

UL 'Qre qULu =

QMAX UL

ic Qre qUL

– M inarg t QMAX UL

ic QreqUL

Q' re qULu – M inarg t

P term R99 – re q

ic Threshold t

P term R99 –

re qic P term

max – M inarg t

GSHOUL

Threshold t

N tot DL

ic P tot DL

ic

txj j ¦

P tot DL

ic adj


---------------------------------------P Transmitted

Tx ic i

Ltotal Tx

IC P ic i ic Tx mu

------------------------------------------

ni

¦ N 0 term

+ ++=

NR DL ic N tot DL

NR DL ic 10 N 0

term

N tot DL

--------------© ¹¨ ¸§ ·

log – =

minN tot ic

DLic Threshold t






These parameters can be either simulation outputs, or user-defined cell inputs. In the last case, when no value is defined

in the Cells table for the total transmitted power and the number of HSDPA users, Atoll uses the following default values:

• Total transmitted power = 50% of the maximum power (i.e, 40 dBm if the maximum power is set to 43 dBm)

• Number of HSDPA users = 1

On the other hand, no default value is used for the available HSDPA power; this parameter must be defined by the user.


When considering all the HSDPA radio bearers, several display options are available in the study properties dialogue.

They can be regrouped in four categories according to the objective of the study:

• To analyse the uplink and downlink A-DPCH qualities on the map,

• To analyse the HS-SCCH quality/power,• To model fast link adaptation for a single HSDPA bearer user

• To model fast link adaptation for a defined number of HSDPA bearer users.

When studying a certain HSDPA radio bearer, only one display option is available. It allows you to display where a certain

RLC peak rate is available with different cell edge coverage probabilities.

Analysis of UL And DL A-DPCH Qualities

• Colour per Max A-DPCH Eb/Nt DL

Atoll displays the A-DPCH quality at the receiver ( ) for the best server on the carrier ic ( or ). No

power control is performed as in simulations. Here, Atoll determines downlink traffic channel quality at the receiver for a

maximum traffic channel power allowed for the best server.

For further details of calculation formulas and methods, please refer to Prediction studies: Point analysis – AS analysis tab

– Downlink sub-menu part.




• Colour per Max A-DPCH Eb/Nt UL

Atoll displays the A-DPCH quality at the best server ( ) on the carr ier ic ( or ). No power control is

performed as in simulations. Here, Atoll determines uplink traffic channel quality at the receiver for a maximum terminal

power allowed.

For further details of calculations formulas and methods, please refer to Point analysis – AS analysis tab – Uplink sub-

menu part.




Analysis of The HS-SCCH Quality/Power

• Colour per HS-SCCH Power

This display option is relevant in case of dynamic HS-SCCH power allocation only. In this case, Atoll displays on each

pixel the HS-SCCH power per HS-SCCH channel. Coverage consists of several layers with a layer per threshold. For each

layer, area is covered if . Each layer is assigned a colour and displayed with intersections

between layers.

• Colour per HS-SCCH Ec/Nt

This display option is relevant in case of static HS-SCCH power allocation only. In this case, Atoll displays on each pixel

the HS-SCCH quality per HS-SCCH channel. Coverage consists of several layers with a layer per threshold. For each

layer, area is covered if . Each layer is assigned a colour and displayed with intersections

between layers.

Fast Link Adaptation Modelling For A Single User

When you calculate the study with the following display options, Atoll considers one user on each pixel and determines

the best HSDPA bearer that the user can obtain. On each pixel, the user is processed as if he is the only user in the cell

i.e. he uses the entire HSDPA power available in the cell.

For further information on the fast link adaptation modelling, see "Fast Link Adaptation Modelling" on page 226.

QMAX DL

ic ic BS ic given

QMAX DL

ic Threshold t

QMAX UL

ic ic BS ic given

QMAX UL

ic Threshold t

P HS SCCH – ic Threshold t

Ec

Nt ------- ic © ¹

§ ·HS SCCH –

Threshold t






Coverage consists of several layers with a layer per possible RLC peak throughput ( ). For each layer, area is

covered if the RLC peak throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed


• Colour per Average RLC Throughput

Atoll displays the average RLC throughput ( ) provided on each pixel. The average RLC throughput is calculated

as follows:

Where,



the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll finds the corresponding BLER.



Coverage consists of several layers with a layer per possible average RLC throughput ( ). For each layer, area

is covered if the average RLC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and


• Colour per Application Throughput

Atoll displays the application throughput ( ) provided on each pixel. The application throughput represents the

net throughput after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:

Where:



the measured quality (HS-PDSCH Ec/Nt). Knowing the HS-PDSCH Ec/Nt, Atoll finds the corresponding BLER.






Coverage consists of several layers with a layer per possible application throughput ( ). For each layer, area is

covered if the application throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed


Fast Link Adaptation Modelling For Several Users

When you calculate the study with the following display options, Atoll considers several users per pixel and determines

the best HSDPA bearer that each user can obtain. In this case, the cell available HSDPA power is shared between HSDPA

bearer users. When the coverage prediction is not based on a simulation, the number of HSDPA bearer users is taken

from the cell properties. The displayed results of the coverage prediction will be an average result for one user.

For further information on the HSDPA bearer allocation process when there are several users, see "HSDPA Bearer Allo-

cation Process" on page 224 For further information on the fast link adaptation modelling, see "Fast Link Adaptation

Modelling" on page 226.

• Colour per MAC Throughput Per Mobile

Atoll displays the average MAC throughput per mobile ( ) provided on each pixel. The average MAC

throughput per mobile is calculated as follows:

Where,

is the number of HSDPA users within the cell.

T RLC p – eak DL

T RLC Av – DL

T RLC Av –

DL R RLC p – eak DL

1 B LE R HSDPA – u

'TT I ----------------------------------------------------------------------------------=

BLER HSDPA

'TT I

T RLC Av – DL

T applicationDL

T applicationDL R RLC p – eak

DL1 B LE R HSDPA – SF Rateuu 'R –

'TT I --------------------------------------------------------------------------------------------------------------------------=

BLER HSDPA

SF Rate 'R

'TT I

T applicationDL

T MAC DL average

T MAC DL average

T MAC DL

x

x 1=

nHSDPA

¦nHSDPA

----------------------------------------=

nHSDPA



© Forsk 2010 AT282_TRG_E2 281


is the MAC throughput of each HSDPA bearer user. For further information on the calculation of the MAC

throughput, see "Colour per MAC Throughput" on page 279.

Coverage consists of several layers with a layer per possible average MAC throughput per mobile ( ). For

each layer, area is covered if the average MAC throughput per mobile exceeds the user-defined thresholds. Each layer is


• Colour per RLC Throughput Per Mobile

Atoll displays the average RLC throughput per mobile ( ) provided on each pixel. The average RLC through-

put per mobile is calculated as follows:

Where,


is the RLC peak throughput of each HSDPA bearer user. For further information on the calculation of the

RLC peak throughput, see "Colour per RLC Peak Throughput" on page 279.

Coverage consists of several layers with a layer per possible average RLC throughput per mobile ( ). For

each layer, area is covered if the average RLC throughput per mobile exceeds the user-defined thresholds. Each layer is


• Colour per ApplicationThroughput Per Mobile

Atoll displays the average application throughput per mobile ( ) provided on each pixel. The average

application throughput per mobile is calculated as follows:

Where,


is the application throughput of each HSDPA bearer user. For further information on the calculation of the

application throughput, see "Colour per Application Throughput" on page 280.

Coverage consists of several layers with a layer per possible average application throughput per mobile

( ). For each layer, area is covered if the average application throughput per mobile exceeds the user-

defined thresholds. Each layer is assigned a colour and displayed with intersections between layers.

Probability of Having a Certain RLC Peak Rate

This result can be obtained only if you have selected an HSDPA radio bearer in the Condition tab.

• Colour per Cell Edge Coverage Probability

Atoll shows areas where the selected HSDPA radio bearer is available with different cell edge coverage probabilities.

Coverage consists of several layers with a layer per cell edge coverage probability defined in the Display tab. For each

layer, area is covered if the selected HSDPA radio bearer is available. Each layer is assigned a colour and displayed with


6.5.2.6 HSUPA Prediction Study A dedicated HSUPA study is available with different calculation and display options. Atoll determines on each pixel the

best HSUPA bearer that can be obtained; it can consider either a single HSUPA bearer user or several ones on each pixel.

For further information on the HSUPA bearer selection, see "HSUPA Bearer Allocation Process" on page 240. By caclu-

lating this study with suitable display options, it is possible:

• To analyse the power required by the selected terminal,

• To analyse the required E-DPDCH quality,

• To analyse rates and throughputs.

Let us assume each pixel on the map corresponds to one or several users with HSUPA capable terminal, mobility and

HSUPA service. Each user may be using a specific carrier or all of them. Moreover, he does not create any interference.

Note that the HSUPA service area is limited by the pilot quality and the A-DPCH-EDPCCH quality.

T MAC DL

x

T MAC DL average

T RLC DL average

T RLC DL average

T RLC p – eak DL

x

x 1=

nHSDPA

¦nHSDPA

----------------------------------------------------=

nHSDPA

T RLC p – eak DL

x

T RLC DL average

T applicationDL

average

T applicationDL average

T applicationDL

x

x 1=

nHSDPA

¦nHSDPA

------------------------------------------------------=

nHSDPA

T applicationDL

x

T applicationDL average





Parameters used as input for the HSUPA prediction study are:

• The cell UL load factor,

• The cell UL reuse factor,

• The cell UL load factor due to HSUPA,

• The maximum cell UL load factor,

• The number of HSUPA users within the cell if the study is calculated for several users.

These parameters can be either simulation outputs, or user-defined cell inputs. In the last case, When no value is defined

in the Cells table, Atoll uses the following default values:

• Uplink load factor = 50%

• Uplink reuse factor = 1

• Uplink load factor due to HSUPA = 0%

• Maximum uplink load factor = 75%

• Number of HSUPA users = 1

6.5.2.6.2 Calculation Options

Atoll can calculate the HSUPA coverage prediction in one of two ways:

• HSUPA resources can be dedictated to a single user : On each pixel, the user is processed as if he is the only

user in the cell i.e he will use the entire remaining load after allocating capacity to all R99 users.

• HSUPA resources can be shared by HSUPA users defined or calculated per cell: Atoll considers several

HSUPA bearer users per pixel. After allocating capacity to all R99 users, the remaining load of the cell will be

shared equally between all the HSUPA bearer users. When the coverage prediction is not based on a simulation,

the number of HSUPA users is taken from the cell properties. The displayed results of the coverage prediction will

be an average result for one user.

6.5.2.6.3 Display OptionsThe following display options are available in the prediction property dialogue.

Colour per Required E-DPDCH Ec/Nt

Atoll displays on each pixel the E-DPDCH Ec/Nt required to obtain the selected HSUPA bearer. Coverage consists of

several layers with a layer per threshold. For each layer, area is covered if . Each layer is


Colour per Required Terminal Power

Atoll displays on each pixel the terminal power required to obtain the selected HSUPA bearer. The required terminal

power is calculated from the required E-DPDCH Ec/Nt. Coverage consists of several layers with a layer per threshold. For

each layer, area is covered if . Each layer is assigned a colour and displayed with intersectionsbetween layers.

Colour per MAC Rate

Atoll displays the MAC rate ( ) provided on each pixel. The MAC rate is calculated as follows:

Where,

is the transport block size (in kbits) for the selected HSUPA bearer; it is defined for each HSUPA bearer in the

HSUPA Radio Bearers table.

is the duration of one TTI for the selected HSUPA bearer; it is defined for each HSUPA bearer in the HSUPA Radio

Bearers table.

Coverage consists of several layers with a layer per possible MAC rate ( ). For each layer, area is covered if the

MAC rate exceeds the user-defined thresholds. Each layer is assigned a colour and displayed with intersections between

layers.

Colour per RLC Peak Rate

After selecting the HSUPA bearer, Atoll reads the corresponding RLC peak rate. This is the highest rate that the selected

HSUPA bearer can provide on each pixel.

Coverage consists of several layers with a layer per possible RLC peak rate ( ). For each layer, area is covered

if the RLC peak rate can be provided. Each layer is assigned a colour and displayed with intersections between layers.

Ec

Nt -------

© ¹§ ·

E DPDCH –

re q

Threshold t

P term

reqThreshold

t

R MAC UL

R MAC UL Sblock

UL

T TTI

----------------=

Sblock UL

T TTI

R MAC UL

R RLC p – eak UL



© Forsk 2010 AT282_TRG_E2 283


Colour per Minimum RLC Throughput

Atoll displays the minimum RLC throughput ( ) provided on each pixel. The minimum RLC throughput corre-

sponds to the RLC throughput obtained for a given BLER and the maximum number of retransmissions. It is calculated as

follows:

Where,

is the residual BLER for the selected uplink transmission format (HSUPA bearer with retransmis-

sions). It is read in the quality graph defined for the quartet “reception equipment-selected bearer-number of retransmis-

sions-mobility” (HSUPA Quality Graphs tab in the Reception equipment properties). This graph describes the variation of

BLER as a function of the measured quality (E-DPDCH Ec/Nt). Knowing the E-DPDCH Ec/Nt, Atoll finds the correspond-

ing BLER.

is the maximum number of retransmissions for the selected HSUPA bearer. This figure is read in the HSUPA Bearer

Selection table.

Coverage consists of several layers with a layer per possible minimum RLC throughput ( ). For each layer, area

is covered if the minimum RLC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and


Colour per Average RLC Throughput

When HARQ (Hybrid Automatic Repeat Request) is used, the required average number of retransmissions is smaller and

the RLC throughput is an average RLC throughput ( ). This is the RLC throughput obtained for a given BLER and

the average number of retransmissions. It is calculated as follows:

is the residual BLER for the selected uplink transmission format (HSUPA bearer with retransmis-

sions). It is read in the quality graph defined for the quartet “reception equipment-selected bearer-number of retransmis-

sions-mobility” (HSUPA Quality Graphs tab in the Reception equipment properties). This graph describes the variation of

BLER as a function of the measured quality (E-DPDCH Ec/Nt). Knowing the E-DPDCH Ec/Nt, Atoll finds the correspond-

ing BLER.

The average number of retransmissions ( ) is determined from early termination probabilities defined for the

selected HSUPA bearer (in the HSUPA Bearer Selection table). The Early Termination Probability graph shows the prob-

ability of early termination ( ) as a function of the number of retransmissions ( ). Atoll calculates the average number

of retransmissions ( ) as follows:

Coverage consists of several layers with a layer per possible average RLC throughput ( ). For each layer, area

is covered if the minimum RLC throughput exceeds the user-defined thresholds. Each layer is assigned a colour and


Colour per Application Throughput

Atoll displays the application throughput ( ) provided on each pixel. The application throughput represents the

net throughput after deduction of coding (redundancy, overhead, addressing, etc.). This one is calculated as follows:

Where:

and respectively represent the scaling factor between the application throughput and the minimum RLC

(Radio Link Control) throughput and the throughput offset. These two parameters model the header information and other

supplementary data that does not appear at the application level. They are defined in the service properties.

Coverage consists of several layers with a layer per possible application throughput ( ). For each layer, area is

covered if the application throughput exceeds the user-defined thresholds. Each layer is assigned a colour and displayed


T RLC Min – UL

T RLC Min – UL R RLC p – eak

UL1 B LE R HSUPA – u

N Rtx

----------------------------------------------------------------------------------=

BLER HSUPA

N Rtx

N Rtx

T RLC Min –

DL

T R LC Av – UL

T RLC Av – UL R RLC p – eak

UL1 B LE R HSUPA – u

N Rtx av

----------------------------------------------------------------------------------=

BLER HSUPA N Rtx

N Rtx av

p N Rtx

N Rtx av

N Rtx av

p N Rtx p N Rtx 1 – – N Rtx u

N Rtx 1=

N Rtx max

¦

p N Rt x max

-----------------------------------------------------------------------------------------------------=

T RLC Av – DL

T application

UL

T applicationUL

M b T RLC Min – UL

SF Rateu 'R – =

SF Rate 'R

T applicationUL




Colour per Average Application Throughput

Atoll displays the average application throughput ( ) provided on each pixel. It is calculated as follows:

Where:

and respectively represent the scaling factor between the average application throughput and the average

RLC (Radio Link Control) throughput and the throughput offset. These two parameters model the header information and

other supplementary data that does not appear at the application level. They are defined in the service properties.

Coverage consists of several layers with a layer per possible average application throughput ( ). For each

layer, area is covered if the average application throughput exceeds the user-defined thresholds. Each layer is assigned

a colour and displayed with intersections between layers.

6.6 Automatic Neighbour AllocationAtoll permits the automatic allocation of intra-technology neighbours in the current network. Two allocation algorithms are

available, one dedicated to intra-carrier neighbours and the other for inter-carrier neighbours.

The intra-technology neighbour allocation algorithms take into account all the cells of TBC transmitters. It means that all

the cells of TBC transmitters of your .atl document are potential neighbours.

The cells to be allocated will be called TBA cells. They must fulfil following conditions:


• They satisfy the filter criteria applied to the Transmitters folder,• They are located inside the focus zone,

• They belong to the folder on which allocation has been executed. This folder can be either the Transmitters folder

or a group of transmitters or a single transmitter.

Only TBA cells may be assigned neighbours.

In this section, the following are explained:

• "Neighbour Allocation for All Transmitters" on page 284.

• "Neighbour Allocation for a Group of Transmitters or One Transmitter" on page 288.

• "Importance Calculation" on page 288.

6.6.1 Neighbour Allocation for All Transmitters

We assume that we have a reference, cell A, and a candidate neighbour, cell B. When the automatic neighbour allocation

starts, Atoll checks the following conditions:

1. The distance between both cells must be less than the user-definable maximum inter-site distance. If the distance

between the reference cell and the candidate neighbour is greater than this value, then the candidate neighbour

is discarded.


Carriers: This option enables you to select the carrier(s) on which you want to run the allocation. You may choose one or

more carriers. Atoll will allocate neighbours to cells using the selected carriers.

Force co-site cells as neighbours: This option enables you to force cells located on the reference cell site in the candidate

neighbour list. This constraint can be weighted among the others and ranks the neighbours through the importance field

(see after).

Force adjacent cells as neighbours (only for intra-carrier neighbours): This option enables you to force cells geographically

adjacent to the reference cell in the candidate neighbour list.This constraint can be weighted among the others and ranks

the neighbours through the importance field (see after).


T applicat ion Av –

ULM b T RLC Av –

ULSF Rateu 'R – =

SF Rate 'R


Note:


Notes:

• Adjacence criterion: Let CellA be a candidate neighbour cell of CellB. CellA is considered

adjacent to CellB if there exists at least one pixel in the CellB Best Server coverage area

where CellA is Best Server (if several cells have the same best server value) or CellA is the

second best server that enters the Active Set (respecting the HO margin of the allocation).



© Forsk 2010 AT282_TRG_E2 285


Force neighbour symmetry: This option enables user to force the reciprocity of a neighbourhood link. Therefore, if the refer-

ence cell is a candidate neighbour of another cell, this one wi ll be considered as candidate neighbour of the reference cell.


force/forbid a cell to be candidate neighbour of the reference cell.Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current neighbours

and carries out a new neighbour allocation. If not selected, the existing neighbours are kept.

3. There must be an overlapping zone ( ) with a given cell edge coverage probabil ity:

• Intra-carrier neighbours: intra-carrier handover is a soft handover.

The reference cell A and the candidate cell B are located inside a continuous layer of cells with carrier c1 (c1 is the selected

carrier on which you run the allocation).

S A is the area where the cell A is the best serving cell. It means that the cell A is the first one in the active set.

- The pilot signal received from the cell A is greater than the minimum pilot signal level.

- The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0).

- The pilot quality from A is the best.

SB is the area where the cell B can enter the active set.

- The pilot signal received from the cell B is greater than the minimum pilot signal level.

- The pilot quality from B is greater than the pilot quality from A minus the Ec/I0 margin. The Ec/I0 margin has

the same meaning as the AS-threshold defined in the Cell properties. So, it should logically have the same

value.

• Inter-carrier neighbours: inter-frequency handover is a hard handover. It is needed in a multi-carrier W-CDMA net-

work:

• When this option is checked, adjacent cells are sorted and listed from the most adjacent to

the least, depending on the above criterion. Adjacence is relative to the number of pixels

satisfying the criterion.

Figure 6.14: Overlapping Zone for Intra-carrier Neighbours

S A SB




- To balance loading between carriers and layers (1st case),

- To make a coverage reason handover from micro cell frequency to macro cells (2nd case).

1st case: the reference cell A is located inside a continuous layer of cells with carrier c1 (c1 is the selected carrier on which

you run the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.

S A is the area where the cell A is not the best serving cell of its layer but can enter the active set.


- The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0).

- The pilot quality from A is not the highest one. It is strictly lower than the best pilot quality received and greater

than the best pilot quality minus the Ec/I0 margin.

SB is the area where the cell B is the best serving cell of its layer.


- The pilot quality from B exceeds a user-definable minimum value (minimum Ec/I0).

- The pilot quality from B is the highest one.

2nd case: the reference cell A is located on the border of a layer with carrier c1 (c1 is the selected carrier on which you run

the allocation) and the candidate cell B belongs to a layer of cells with carrier c2.

S A is the area where the pilot quality from the cell A starts significantly decreasing but the cell A is still the best serving cell

of its layer (since it is on the border).


- The pilot quality from A is the highest one

- The pilot quality from A is lower than a user-definable minimum value (minimum Ec/I0) plus the Ec/I0 margin.

SB is the area where the cell B is the best serving cell of its layer.


- The pilot quality from B exceeds a user-definable minimum value (minimum Ec/I0).

- The pilot quality from B is the highest one.

Figure 6.15: Overlapping Zone for Inter-carrier Neighbours - 1st Case

Note:

• Two ways enable you to determine the I 0 value:

1 - Global Value: A percentage of the cell maximum power is considered. If the % of

maximum power is too low, i.e. if , Atoll takes into account the pilot

power of the cell. Then, I 0 represents the sum of values calculated for each cell.

2 - Defined per Cell: Atoll takes into account the total downlink power defined per cell. I0

represents the sum of total transmitted powers.

% P max u P pi lot



© Forsk 2010 AT282_TRG_E2 287


Atoll calculates the percentage of covered area ( ) and compares this value to the % minimum covered

area. If this percentage is not exceeded, the candidate neighbour B is discarded.

4. The importance of neighbours.

For information on the importance calculation, see "Importance Calculation" on page 288.

Importance values are used by the allocation algorithm to rank the neighbours according to the allocation reason. Atoll

lists all neighbours and sorts them by importance value so as to eliminate some of them from the neighbour list if the maxi-

mum number of neighbours to be allocated to each transmitter is exceeded. If we consider the case for which there are

15 candidate neighbours and the maximum number of neighbours to be allocated to the reference cell is 8. Among these

15 candidate neighbours, only 8 (having the highest importance values) will be allocated to the reference cell. Note that

specific maximum numbers of neighbours (maximum number of intra-carrier neighbours, maximum number of inter-carrier

neighbours) can be defined at the cell level (property dialogue or cell table). If defined there, this value is taken into account

instead of the default one available in the Neighbour Allocation dialogue.



Therefore, a neighbour may be marked as exceptional pair, co-site, adjacent, coverage or symmetric. For neighbours

accepted for co-site, adjacency and coverage reasons, Atoll displays the percentage of area meeting the coverage condi-

tions and the corresponding surface area (km2), the percentage of area meeting the adjacency conditions and the corre-

sponding surface area (km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.

Figure 6.16: Overlapping Zone for Inter-carrier Neighbours - 2 nd Case

Notes:

• No simulation or prediction study is needed to perform an automatic neighbour allocation.

When starting an automatic neighbour allocation, Atoll automatically calculates the path

loss matrices if not found.

• Even if no specific terminal, mobility or service is selected in the automatic allocation, it is

interesting to know that the algorithm works such as finding the maximum number of

neighbours by selection the multi-service traffic data as follows:

Service: selection of the one with the lowest body loss.

Mobility: no impact on the allocation, no specific selection.

Terminal: selection of the one with the greatest (Gain - Loss) value, and, if equal, the one

with the lowest noise figure.

• The neighbour lists may be optionally used in the power control simulations to determine

the mobile's active set.

• A forbidden neighbour must not be listed as neighbour except if the neighbourhoodrelationship already exists and the Delete existing neighbours option is unchecked when




• The force neighbour symmetry option enables the users to consider the reciprocity of a

neighbourhood link. This reciprocity is allowed only if the neighbour list is not already full.

Thus, if the cell B is a neighbour of the cell A while the cell A is not a neighbour of the cell

B, two cases are possible:

1st case: There is space in the cell B neighbour list: the cell A will be added to the list. It will

be the last one.

S A SB

S A

---------------------- 100 u





© Forsk 2010 AT282_TRG_E2 289


Where

6.6.3.2 Importance of Inter-carrier Neighbours

As indicated in the table below, the neighbour importance depends on the neighbourhood cause; this value varies between

0 to 100%.

Except the case of forced neighbours (importance = 100%), priority assigned to each neighbourhood cause is determined

using the Importance Function (IF). The IF considers two factors for calculating the importance:

• The co-site factor (C) which is a Boolean

• The overlapping factor (O) meaning the percentage of overlapping

The IF is user-definable using the Min importance and Max importance fields.

The IF evaluates importance as follows:

Where

Yes Yes

Notes:



fields, neighbours will be ranked in the following order:

i. Co-site neighbours

ii. Adjacent neighbours

iii. Neighbours based on coverage overlapping

• If the ranges of the importance factors overlap, the neighbours may not be ranked

according to the neighbourhood cause.

• The ranking between neighbours from the same category depends on the factors (A) and

(O).

• The default value of Min(O) = 1% ensures that neighbours selected for symmetry will have

an importance greater than 0%. With a value of Min(O) = 0%, neighbours selected for

symmetry will have an importance field greater than 0% only if there is some coverage

overlapping.

Min C ' C Max O O 100% Max O – A +^ `+ 60% 40% 30% O 70% A +^ `+

' X Max X Min X – =


value

Existing neighbour If the Delete existing neighbours option is not selectedExisting

importance

Exceptional pair If the Force exceptional pairs option is selected 100 %

Co-site transmitter If the Force co-site cells as neighbours option is selected IF


coverage conditionsIf the % minimum covered area is exceeded IF

Symmetric neighbourhood

relationshipIf the Force neighbour symmetry option is selected IF

Factor Min importance Default value Max importance Default value

Overlapping factor (O) 1% 60%

Co-site factor (C) 60% 100%

Co-site Neighbourhood cause IFResulting IF using the default

values from the table above

No

Yes

Notes:





ii. Neighbours based on coverage overlapping



Min O Max O

Min C Max C

Min O ' O O + 1% 59% O +

Min C ' C O + 60% 40% O +





6.7 Primary Scrambling Code AllocationDownlink primary scrambling codes enable you to distinguish cells from one another (cell identification).

By default, there are 512 primary scrambling codes numbered (0...511).

The cells to which Atoll allocates scrambling codes are referred to as the TBA cells (cells to be allocated). TBA cells fulfil

following conditions:

- They are act ive,

- They satisfy the filter criteria applied to the Transmitters folder,

- They are located inside the focus zone,

- They belong to the folder on which allocation has been executed. This folder can be either the Transmitters

folder or a group of transmitters or a single transmitter.

6.7.1 Automatic Allocation Description

6.7.1.1 Options and Constraints

The scrambling code allocation algorithm can take into account following constraints and options:

1. Neighbourhood between cells,

You may consider:

• First order neighbours: The neighbours of TBA cells listed in the Intra-technology neighbours table,

• Second order neighbours: The neighbours of neighbours,

• Third order neighbours: The neighbour’s neighbour’s neighbours.

2. Cells fulfilling a criterion on Ec/I0 (option “Additional Overlapping Conditions”),

For a reference cell “A”, Atoll considers all the cells “B” that can enter the active set on the area where the reference cell

is the best server (area where (Ec/I0) A exceeds the minimum Ec/I0 and is the highest one and (Ec/I0)B is within a Ec/I0

margin of (Ec/I0) A).

3. Reuse distance,

• The ranking between neighbours from the same category depends on the factor (O).

• The default value of Min(O) = 1% ensures that neighbours selected for symmetry will have

an importance greater than 0%. With a value of Min(O) = 0%, neighbours selected for

symmetry will have an importance field greater than 0% only if there is some coverage

overlapping.

Note:


Notes:

• In the context of the primary scrambling code allocation, the term "neighbours" refers to

intra-carrier neighbours.

• Atoll can take into account inter-technology neighbour relations as constraints to allocatedifferent scrambling codes to the UMTS neighbours of a GSM transmitter. In order to

consider inter-technology neighbour relations in the scrambling code allocation, you must

make the Transmitters folder of the GSM .atl document accessible in the UMTS .atl

document. For information on making links between GSM and UMTS .atl documents, see

the User Manual .

• Atoll considers symmetry relationship between a cell, its first order neighbours, its second

order neighbours and its third order neighbours.

Note:

• Atoll considers either a percentage of the cell maximum powers or the total downlinkpower used by the cells in order to evaluate I0. In this case, I0 equals the sum of total

transmitted powers. When this parameter is not specified in the cell properties, Atoll uses

50% of the maximum power.

Notes:

• Reuse distance is a constraint on the allocation of scrambling codes. A code cannot be

reused at a cell that is not at least as far away as the reuse distance from the cell allocated

with the particular code.



© Forsk 2010 AT282_TRG_E2 291


4. Exceptional pairs,

5. Domains of scrambling codes,

6. The number of primary scrambling codes per cluster. In Atoll, we call "cluster", a group of scrambling codes as

defined in 3GPP specifications. 3GPP specifications define 64 clusters consisting of 8 scrambling codes (in this

case, clusters are numbererd from 0 to 63). However, you can define another value (e.g. if you set the number of

codes per cluster to 4, scrambling codes will be distributed in 128 clusters).

When the allocation is based on a Distributed strategy (Distributed per Cell or Distributed per Site), this parameter

can also be used to define the interval between the primary scrambling codes assigned to cells on a same site.

The defined interval is applied by adding the following lines in the Atoll.ini file:

For more information about setting options in the atoll.ini file, see the Administrator Manual .

7. The carrier on which the allocation is run: It can be a given carrier or all of them. In this case, either Atoll

independently plans scrambling codes for the different carriers, or it allocates the same primary scrambling code

to each carrier of a transmitter if the option "Allocate carriers identically" is selected.8. The possibility to use a maximum of codes from the defined domains (option "Use a Maximum of Codes"): Atoll

will try to spread the scrambling code spectrum the most.

9. The "Delete All Codes" option: When selecting this option, Atoll deletes all the current scrambling codes and

carries out a new scrambling code allocation. If not selected, the existing scrambling codes are kept.

In addition, it depends on the selected allocation strategy. Allocation strategies can be:

• Clustered allocation: The purpose of this strategy is to choose for a group of mutually constrained cells, scrambling

codes among a minimum number of clusters. In this case, Atoll will preferentially allocate all the codes within the

same cluster.

• Distributed per cell allocation: This strategy consists in using as many clusters as possible. Atoll will preferentially

allocate codes from different clusters.

• One cluster per site allocation: This strategy allocates one cluster to each site, then, one code from the cluster to

each cell of each site. When all the clusters have been allocated and there are still sites remaining to be allocated,

Atoll reuses the clusters as far as possible at another site.

• Distributed per site allocation: This strategy allocates a group of adjacent clusters to each site, then, one cluster

to each transmitter on the site according to its azimuth and finally, one code from the cluster to each cell of each

transmitter. The number of adjacent clusters per group depends on the number of transmitters per site you have

in your network; this information is required to start allocation based on this strategy. When all the groups of adja-

cent clusters have been allocated and there are still sites remaining to be allocated, Atoll reuses the groups of

adjacent clusters as far as possible at another site.

In the Results table, Atoll only displays scrambling codes allocated to TBA cells.

6.7.1.2 Allocation Process

For each TBA cell, Atoll lists all cells which have constraints with the cell. They are referred to as near cells. The near cells

of a TBA cell may be:

• Its neighbour cells: the neighbours listed in the Intra-technology neighbours table (options “Existing neighbours”

and "First Order"),

• The neighbours of its neighbours (options “Existing neighbours” and “Second Order”),

• The third order neighbours (options “Existing neighbours” and “Third Order”),

• The cells that fulfil Ec/I0 condition (option “Additional Overlapping Conditions”),

• The cells with distance from the TBA cell less than the reuse distance,

• The cells that make exceptional pairs with the TBA cell.

Additional constraints are considered when:

• The cell and its near cells are neighbours of a same GSM transmitter (only if the Transmitters folder of the GSM

.atl document is accessible in the UMTS .atl document),

• The neighbour cells cannot share the same cluster (for the "Distributed per site" allocation strategy only).

These constraints have a certain weight taken into account to determine the TBA cell priority during the allocation process

and the cost of the scrambling code plan. During the allocation, Atoll tries to assign different scrambling codes to the TBA

cell and its near cells. If it respects all the constraints, the cost of the scrambling code plan is 0. When a cell has too many

constraints and there are not anymore scrambling codes available, Atoll breaks the constraint with the lowest cost so as

to generate the scrambling code plan with the lowest cost. For information on the cost generated by each constraint, see

"Cell Priority" on page 293.

• Scrambling code reuse distance can be defined at cell level. If this value is not defined,

then Atoll will use the default reuse distance defined in the Scrambling Code Automatic

Allocation dialogue.

Note:

• When no domain is assigned to cells, Atoll considers the 512 primary scrambling codes

available.

[PSC]

ConstantStep = 1




6.7.1.2.1 Single Carrier Network

The allocation process depends on the selected strategy. Algorithm works as follows:

Strategies: Clustered and Distributed per Cell

Atoll processes TBA cells according to their priority. It allocates scrambling codes starting with the highest priority cell and

its near cells, and continuing with the lowest priority cells not allocated yet and their near cells. For information on calcu-

lating cell priority, see "Cell Priority" on page 293.

Strategy: One Cluster per Site

All sites which have constraints with the studied site are referred to as near sites.

Atoll assigns a cluster to each site, starting with the highest priority site and its near sites, and continuing with the lowest

priority sites not allocated yet and their near sites. When all the clusters have been allocated and there are still sites

remaining to be allocated, Atoll reuses the clusters at another site. When the Reuse Distance option is selected, the algo-

rithm reuses the clusters as soon as the reuse distance is exceeded. Otherwise, when the option is not selected, the algo-

rithm tries to assign reused clusters as spaced out as possible.

Then, Atoll allocates a primary scrambling code from the cluster to each cell located on the sites (codes belong to the

assigned clusters). It starts with the highest priority cell and its near cells and goes on with the lowest priority cells not

allocated yet and their near cells.

For information on calculating site priority, see "Site Priority" on page 295. For information on calculating cell priority, see


Strategy: Distributed per Site

All sites which have constraints with the studied site are referred to as near sites.

Atoll assigns a group of adjacent clusters to each site, starting with the highest priority site and its near sites, and contin-

uing with the lowest priority sites not allocated yet and their near sites. When all the groups of adjacent clusters have been

allocated and there are still sites remaining to be allocated, Atoll reuses the groups of adjacent clusters at another site.

When the Reuse Distance option is selected, the algorithm reuses the groups of adjacent clusters as soon as the reuse

distance is exceeded. Otherwise, when the option is not selected, the algorithm tries to assign reused groups of adjacent

clusters as spaced out as possible. Then, Atoll assigns each cluster of the group to each transmitter of the site according

to the transmitter azimuth and selected neighbourhood constraints (options "Neighbours in Other Clusters" and "Second-

ary Neighbours in Other Clusters"). Then, Atoll allocates a primary scrambling code to each cell located on the transmit-

ters (codes belong to the assigned clusters). It starts with the highest priority cell and its near cells and goes on with the

lowest priority cells not allocated yet and their near cells.

For information on calculating site priority, see "Site Priority" on page 295. For information on calculating cell priority, see


Determination of Groups of Adjacent Clusters

In order to determine the groups of adjacent clusters to be used, Atoll proceeds as follows: It defines theoretical groupsof adjacent clusters, independently of the defined domain, considering the 512 primary scrambling codes available and

the specified number of codes per cluster (if this one is set to 8, 64 clusters are supposed to be available). It starts the

division in group from the cluster 0 (hard coded) and takes into account the maximum number of transmitters per site user-

specified in order to determine the number of clusters in each group and then, the number of possible groups.

Let us assume that the number of codes per cluster is set to 8 and the maximum number of transmitters per site in the

network is 3. In this case, we have the following theoretical groups:

If no domain is assigned to cells, Atoll can use all these groups for the allocation. On the other hand, if a domain is used,

the tool compares adjacent clusters really available in the assigned domain to the theoretical groups and only keeps adja-

cent clusters mapping the theoretical groups.

Let us assume that we have a domain consisted of 12 clusters: clusters 1 to 8 and clusters 12 to 15.

Therefore, Atoll will be able to use the following groups of adjacent clusters:

• Group 2 with cluster 3, 4 and 5,

• Group 3 with cluster 6, 7 and 8,

• Group 6 with cluster 12, 13 and 14.

• The clusters 1, 2 and 15 will not be used.

If a domain does not contain any adjacent clusters, the user is warned through the 'Event Viewer'.

6.7.1.2.2 Multi-Carrier Network

In case you have a multi-carrier network and you run the scrambling code allocation on all the carriers, the allocation proc-

ess depends on the allocation strategy as detailed above and in addition, wether the option "Allocate Carriers Identically"

is selected or not.

Group 1 Group 2 Group 3 Group 4 ... Group 21

Cluster 0

Cluster 1

Cluster 2

Cluster 3

Cluster 4

Cluster 5

Cluster 6

Cluster 7

Cluster 8

Cluster 9

Cluster 10

Cluster 11

...

Cluster 61

Cluster 62

Cluster 63



© Forsk 2010 AT282_TRG_E2 293


When the option is not selected, algorithm works for each strategy, as explained above. On the other hand, when the

option is selected, allocation order changes. It is no longer based on the cell priority but depends on the transmitter priority.

All transmitters which have constraints with the studied transmitter will be referred to as near transmitters.

In case of a "Per cell" strategy (Clustered and Distributed per cell), Atoll starts scrambling code allocation with the highest

priority transmitter and its near transmitters and continues with the lowest priority transmitters not allocated yet and their

near transmitters. The same scrambling code is assigned to each cell of the transmitter.

In case of the "One cluster per site" strategy, Atoll assigns a cluster to each site and then, allocates a scrambling code to

each transmitter. It starts with the highest priority transmitter and its near transmitters and continues with the lowest priority

transmitters not allocated yet and their near transmitters. The same scrambling code is assigned to each cell of the trans-

mitter.

In case of the "Distributed per site" strategy, Atoll assigns a group of adjacent clusters to each site, then a cluster to eachtransmitter and finally, allocates a scrambling code to each transmitter. It starts with the highest priority transmitter and its

near transmitters and continues with the lowest priority transmitters not allocated yet and their near transmitters. The same

scrambling code is assigned to each cell of the transmitter.

For information on calculating transmitter priority, see "Transmitter Priority" on page 295.

6.7.1.3 Priority Determination

6.7.1.3.1 Cell Priority

Scrambling code allocation algorithm in Atoll allots priorities to cells before performing the actual allocation. Priorities

assigned to cells depend upon how much constrained each cell is and the cost defined for each constraint. A cell without

any constraint has a default cost, , equal to 0. The higher the cost on a cell, the higher the priority it has for the scrambling

code allocation process.

There are six criteria employed to determine the cell priority:

• Scrambling Code Domain Criterion

The cost due to the domain constraint, , depends on the number of scrambling codes available for the allocation.

The domain constraint is mandatory and cannot be broken.

When no domain is assigned to cells, 512 scrambling codes are available and we have:

When domains of scrambling codes are assigned to cells, each unavailable scrambling code generates a cost. The higher

the number of codes available in the domain, the less will be the cost due to this criterion. The cost is given as:

• Distance Criterion

The constraint level of any cell i depends on the number of cells ( j ) present within a radius of "reuse distance" from its

centre. The total cost due to the distance constraint is given as:

Each cell j within the reuse distance generates a cost given as:

Where

is a weight depending on the distance between i and j . This weight is inversely proportional to the inter-cell distance.

For a reuse distance of 2000m, the weight for an inter-cell distance of 1500m is 0.25, the weight for co-site cells is 1 and

the weight for two cells spaced out 2100m apart is 0.

is the cost of the distance constraint. This value can be defined in the Constraint Cost dialogue.

• Exceptional Pair Criterion

The constraint level of any cell i depends on the number of exceptional pairs ( j ) for that cell. The total cost due to excep-

tional pair constraint is given as:

Where

is the cost of the exceptional pair constraint. This value can be defined in the Constraint Cost dialogue.

• Neighbourhood Criterion

Note:

• When cells, transmitters or sites have the same priority, processing is based on an

alphanumeric order.

C

C i Do m

C i Do m 0 =

C i Do m 512 Number of scrambling codes in the domain – =

C i Dist C j Dist i

j

¦=

C j Dist i w d ij c udis cetan

=

w d ij

c dis cetan

C i EP c EP i j –

j

¦=

c EP




The constraint level of any cell i depends on the number of its neighbour cells j , the number of second order neighbours k

and the number of third order neighbours l .

Let’s consider the following neighbour schema:

The total cost due to the neighbour constraint is given as:

Each first order neighbour cell j generates a cost given as:

Where

is the importance of the neighbour cell j .

is the cost of the first order neighbour constraint. This value can be defined in the Constraint Cost dialogue.

Because two first order neighbours must not have the same scrambling code, Atoll considers the cost created by two first

order neighbours to be each other.

Each second order neighbour cell k generates a cost given as:

Where

is the cost of the second order neighbour constraint. This value can be defined in the Constraint Cost dialogue.

Because two second order neighbours must not have the same scrambling code, Atoll considers the cost created by two

second order neighbours to be each other.

Each third order neighbour cell l generates a cost given as:

Where

is the cost of the third order neighbour constraint. This value can be defined in the Constraint Cost dialogue.

Because two third order neighbours must not have the same scrambling code, Atoll considers the cost created by two

third order neighbours to be each other.

Figure 6.17: Neighbourhood Constraints

Note:

• Atoll considers the highest cost of both links when a neighbour relation is symmetric and the

importance value is different.

In this case, we have:

And

C i N C j N1 i C j j c – N1 i

j c¦+

j

¦© ¹¨ ¸§ ·

C k N2 i C k k c – N2 i

k c¦+

k

¦© ¹¨ ¸§ ·

C l N3 i C l l c – N3 i

l c¦+

l

¦© ¹¨ ¸§ ·

+ +=

C j N1 i I j c uN1

=

I j

c N1

C j j c – N1 i C j N1 i C j c N1 i +

2 -----------------------------------------------------------=

C k N2 i M ax C j N1 i C k N1 j u C j c N1 i C k N1 j c u ( , ) c uN2

=

c N2

C k k c – N2 i C k N2 i C k c N2 i +2

-------------------------------------------------------------=

C l N3 i Ma x C j N1 i C k N1 j C l N1 k uu C j c N1 i C k N1 j c u C l N1 k u

C j N1 i C k c N1 j u C l N1 k c u C j c N1 i C k c N1 j c u C l N1 k c u© ¹¨ ¸§ ·

c N3u=

c N3

C l l c – N3 i C l N3 i C l c N3 i +

2 -----------------------------------------------------------=

C j N1 i Max I i j – I j i – c uN1

=

C k N2 i Max C j N1 i C k N1 j u C j N1 k C i N1 j u( , ) c uN2

=



© Forsk 2010 AT282_TRG_E2 295


• GSM Neighbour Criterion

This criterion is considered when the co-planning mode is activated (i.e. the Transmitters folder of the GSM .atl document

is made accessible in the UMTS .atl document) and inter-technology neighbours have been allocated. If the cell i is neigh-

bour of a GSM transmitter, the cell constraint level depends on how many cells j are neighbours of the same GSM trans-

mitter. The total cost due to GSM neighbour constraint is given as:

Where

is the cost of the GSM neighbour constraint. This value can be defined in the Constraint Cost dialogue.

• Cluster Criterion

When the "Distributed per Site" allocation strategy is used, you can consider additional constraints on allocated clusters

(one cell, its first order neighbours and its second order neighbours must be assigned scrambling codes from different clus-

ters). In this case, the constraint level of any cell i depends on the number of first and second order neighbours, j and k .

The total cost due to the cluster constraint is given as:

Where

is the cost of the cluster constraint. This value can be defined in the Constraint Cost dialogue.

Therefore, the total cost due to constraints on any cell i is defined as:

With

6.7.1.3.2 Transmitter Priority

In case you have a multi-carrier network and you run scrambling code allocation on "all" the carriers with the option "allo-

cate carriers identically", algorithm in Atoll allots priorities to transmitters. Priorities assigned to transmitters depend on

how much constrained each transmitter is and the cost defined for each constraint. The higher the cost on a transmitter,

the higher the priority it has for the scrambling code allocation process.

Let us consider a transmitter Tx with two cells using carriers 0 and 1. The cost due to constraints on the transmitter is given

as:

With and

Here, the domain available for the transmitter is the intersection of domains assigned to cells of the transmitter. The

domain constraint is mandatory and cannot be broken.

6.7.1.3.3 Site Priority

In case of "Per Site" allocation strategies (One cluster per site and Distributed per site), algorithm in Atoll allots priorities

to sites. Priorities assigned to sites depend on how much constrained each site is and the cost defined for each constraint.

The higher the cost on a site, the higher the priority it has for the scrambling code allocation process.

Let us consider a site S with three transmitters; each of them has two cells using carriers 0 and 1. The cost due to

constraints on the site is given as:

With and

Here, the domain considered for the site is the intersection of domains available for transmitters of the site. The domain

constraint is mandatory and cannot be broken.

6.7.2 Allocation Examples

6.7.2.1 Allocation Strategies and Use a Maximum of Codes

In order to understand the differences between the different allocation strategies and the behaviour of algorithm when

using a maximum of codes or not, let us consider the following sample scenario:

C i N 2G c N 2G j Tx 2G –

j

¦=

c N 2G

C i Cluster C j N1 i c Cluster u

j

¦ C k N2 i c Cluster u

k

¦+=

c Cluster

C i C i Do m C i U +=

C i U C i Dist C i EP C i N C i N 2G C i Cluster + + + +=

C Tx C Tx Do m C +Tx

U =

C Tx U Ma x i Tx C i U = C Tx Do m 512 Number of scrambling codes in the domain – =

C S C S U C S Do m +=

C S U Ma x

Tx S C Tx U = C S Do m 512 Number of scrambling codes in the domain – =




Let Site0, Site1, Site2 and Site3 be four sites wi th 3 cells using carrier 0 whom scrambling codes have to be allocated out

of three clusters consisted of 8 primary scrambling codes. This implies that the domain of scrambling codes for the four

sites is from 0 to 23 (cluster 0 to cluster 2). The reuse distance is supposed to be less than the inter-site distance. Only

co-site neighbours exist.

The following section lists the results of each combination of options with explanation where necessary.

6.7.2.1.1 Strategy: Clustered

Since the restrictions of neighbourhood only apply to co-sites with the same importance and sites distances are greater

than reuse distances, every cell has the same priority. Then, scrambling code allocation to cells is performed in an alpha-

numeric order.

Figure 6.18: Primary Scrambling Codes Allocation

Without ‘Use a Maximum of Codes’ With ‘Use a Maximum of Codes’

Atoll starts allocating the codes from the start of cluster 0

at each site.

As it is possible to use a maximum of codes, Atoll starts

allocation at the start of a different cluster at each site.

When a cluster is reused, and there are non allocated

codes left in the cluster, Atoll first allocates those codes

before reusing the already used ones.



© Forsk 2010 AT282_TRG_E2 297


6.7.2.1.2 Strategy: Distributed


than reuse distances, every cell has the same priority. Then, scrambling code allocation to cells is performed in an alpha-

numeric order.

6.7.2.1.3 Strategy: ‘One Cluster per Site


than reuse distances, every site has the same priority. Then, cluster allocation to sites is performed in an alphanumeric

order.


Atoll allocates codes from different clusters to each cell of

the same site. Under given constraints of neighbourhood

and reuse distance, same codes can be allocated to each

site’s cells.

Atoll allocates codes from different clusters to each site’s

cells. As it is possible to use a maximum of codes, Atoll

allocates the codes so that there is least repetition of codes.


In this strategy, a cluster of codes is limited to be used at

just one site at a time unless all codes and clusters have

been allocated and there are still sites remaining to be allo-cated. In this case Atoll reuses the clusters as far as possi-

ble at another site.

When it is possible to use a maximum of codes, Atoll can

allocate different codes from a reused cluster at another

site.




6.7.2.1.4 Strategy: ‘Distributed per Site


than reuse distances, every site has the same priority. Then, the group of adjacent clusters allocation to sites is performed

in an alphanumeric order.

6.7.2.2 Allocate Carriers Identically

In order to understand the behaviour of algorithm when using the option "Allocate Carriers Identically" or not, let us

consider the following sample scenario:

Let Site0, Site1, Site2 and Site3 be four sites with 3 cells using carrier 0 and 3 cells using carrier 1. Scrambling codes have

to be allocated out of 3 clusters consisted of 8 primary scrambling codes. This implies that the domain of scrambling codes

for the five sites is from 0 to 23 (cluster 0 to cluster 2). The reuse distance is supposed to be less than the inter-site

distance. Only co-site neighbours exist. Allocation algorithm will be based on the "One Cluster per Site" strategy and the

option "Use a Maximum of Codes" is selected.

In both cases (with and without ’Allocate Carriers Identically’), every site has the same priority. Then, cluster allocation to

sites is performed in an alphanumeric order.


In this strategy, a group of adjacent clusters is limited to be

used at just one site at a time unless all codes and groups

of adjacent clusters have been allocated and there are still

sites remaining to be allocated. In this case (here only one

group of adjacent clusters (clusters 0, 1 and 2) is available),

Atoll reuses the group at another site.

When it is possible to use a maximum of codes, Atoll can

allocate different codes from a reused group of adjacent

cluster at another site.

Without ‘Allocate Carriers Identically’ With ‘Allocate Carriers Identically’

Atoll allocates one cluster at each site as detailed in the

previous section. Then, it allocates a code from the cluster

to each cell of the site so as to use a maximum of codes.

In this case, Atoll allocates one cluster at each site and

then, one code to each transmitter so as to use a maximum

of codes. Then, the same code is given to each cell of the

transmitter.



© Forsk 2010 AT282_TRG_E2 299


6.8 Automatic GSM-UMTS Neighbour Allocation

6.8.1 Overview

You can automatically calculate and allocate neighbours between GSM and UMTS networks. In Atoll, it is called inter-

technology neighbour allocation.

Inter-technology handover is used in two cases:

• When the UMTS coverage is not continuous. In this case, the UMTS coverage is extended by UMTS-GSM

handover into the GSM network,• And in order to balance traffic and service distribution between both networks.

Note that the automatic inter-technology neighbour allocation algorithm takes into account both cases.

In order to be able to use the inter-technology neighbour allocation algorithm, you must have:

• An .atl document containing the GSM network, GSM.atl, and another one describing the UMTS network,

UMTS.atl,

• An existing link on the Transmitters folder of GSM.atl into UMTS.atl.

The external neighbour allocation algorithm takes into account all the GSM TBC transmitters. It means that all the TBC

transmitters of GSM.atl are potential neighbours. The cells to be allocated will be called TBA cells which, being cells of

UMTS.atl, satisfy following conditions:


• They satisfy the filter criteria applied to Transmitters folder,

• They are located inside the focus zone,

• They belong to the folder for which allocation has been executed. This folder can be either the Transmitters folder

or a group of transmitters subfolder.Only UMTS TBA cells may be assigned neighbours.

6.8.2 Automatic Allocation Description

The allocation algorithm takes into account criteria listed below:

• The inter-transmitter distance,

• The maximum number of neighbours fixed,

• Allocation options,

• The selected allocation strategy,

Two allocation strategies are available: the first one is based on distance and the second one on coverage overlapping.

We assume we have a UMTS reference cell, A, and a GSM candidate neighbour, transmitter B.

6.8.2.1 Algorithm Based on Distance

When automatic allocation starts, Atoll checks following conditions:

1. The distance between the UMTS reference cell and the GSM neighbour must be less than the user-definable

maximum inter-site distance. If the distance between the UMTS reference cell and the GSM neighbour is greater

than this value, then the candidate neighbour is discarded.

Candidate neighbours are sorted in descending order with respect to distance.




Force co-site cells as neighbours: It enables you to automatically include GSM transmitters located on the same site as

the reference UMTS cell in the candidate neighbour list. This option is automatically selected.


force/forbid a GSM transmitter to be candidate neighbour of the reference UMTS cell.

Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current neighbours

and carries out a new neighbour allocation. If not selected, existing neighbours are kept.




mum number of neighbours to be allocated to each cell is exceeded. If we consider the case for which there are 15 candi-

date neighbours and the maximum number of neighbours to be allocated to the reference cell is 8. Among these 15

candidate neighbours, only 8 (having the highest importance values) will be allocated to the reference cell. Note that the

Note:

• Transmitter azimuths are taken into account to evaluate the inter-transmitter distance (for

further information on inter-transmitter distance calculation, please refer to "Calculation of

Inter-Transmitter Distance" on page 302)




maximum number of inter-technology neighbours can be defined at the cell level (property dialogue or cell table). If defined

there, this value is taken into account instead of the default one available in the Neighbour Allocation dialogue.

As indicated in the table below, the neighbour importance depends on the neighbourhood cause; this value varies between

0 to 100%.

Where is the distance between the UMTS reference cell and the GSM neighbour and is the maximum inter-site

distance.



Therefore, a neighbour may be marked as exceptional pair, co-site, or distance. For neighbours accepted for distance

reasons, Atoll displays the distance from the reference cell (m). Finally, if cells have previous allocations in the list, neigh-

bours are marked as existing.

6.8.2.2 Algorithm Based on Coverage Overlapping

When automatic allocation starts, Atoll checks following conditions:1. The distance between the UMTS reference cell and the GSM neighbour must be less than the user-definable

maximum inter-site distance. If the distance between the UMTS reference cell and the GSM neighbour is greater

than this value, then the candidate neighbour is discarded.




Force co-site cells as neighbours: It enables you to automatically include GSM transmitters located on the same site as

the reference UMTS cell in the candidate neighbour list. This option is automatically selected.


force/forbid a GSM transmitter to be candidate neighbour of the reference UMTS cell.

Delete existing neighbours: When selecting the Delete existing neighbours option, Atoll deletes all the current neighbours

and carries out a new neighbour allocation. If not selected, existing neighbours are kept.

3. There must be an overlapping zone ( ) with a given cell edge coverage probabili ty.

Four different cases may be considered for S A:

- 1st case: S A is the area where the cell A is the best serving cell of the UMTS network.

- The pilot signal received from A is greater than the minimum pilot signal level,

- The pilot quality from A exceeds a user-definable minimum value (minimum Ec/I0) and is the highest one.

In this case, the Ec/I0 margin must be equal to 0dB and the max Ec/I0 option disabled.

- 2nd case: S A represents the area where the pilot quality from the cell A strats decreasing but the cell A is still

the best serving cell of the UMTS network.

The Ec/I0 margin must be equal to 0dB, the max Ec/I0 option selected and a maximum Ec/I0 user-defined.


- The pilot quality from A exceeds the minimum Ec/I0 but is lower than the maximum Ec/I0.

- The pilot quality from A is the highest one.

- 3rd case: S A represents the area where the cell A is not the best serving cell but can enter the active set.

Here, the Ec/I0 margin has to be different from 0dB and the max Ec/I0 option disabled.


- The pilot quality from A is within a margin from the best Ec/I0, where the best Ec/I0 exceeds the minimum

Ec/I0.

- 4th case: S A represents the area where:



value


importance


Co-site transmitter If the Force co-site cells as neighbours option is selected 100 %


distance conditionsIf the maximum distance is not exceeded 1

d

d max

------------ –

d d max

Note:

• Here, real inter-transmitter distance is considered.

S A SB



© Forsk 2010 AT282_TRG_E2 301


- The pilot quality from A is within a margin from the best Ec/I0 (where the best Ec/I0 exceeds the minimum

Ec/I0) and lower than the maximum Ec/I0.

In this case, the margin must be different from 0dB, the max Ec/I0 option selected and a maximum Ec/I0 user-

defined.

Two different cases may be considered for SB:

- 1st case: SB is the area where the cell B is the best serving cell of the GSM network.

In this case, the margin must be set to 0dB.

- The signal level received from B on the BCCH TRX type exceeds the user-defined minimum threshold and

is the highest one.

- 2nd case: The margin is different from 0dB and SB is the area where:

- The signal level received from B on the BCCH TRX type exceeds the user-defined minimum threshold and

is within a margin from the best BCCH signal level.

Atoll calculates the percentage of covered area ( ) and compares this value to the % minimum covered

area. If this percentage is not exceeded, the candidate neighbour B is discarded.

Candidate neighbours fulfilling coverage conditions are sorted in descending order with respect to % of covered area.




mum number of neighbours to be allocated to each cell is exceeded. If we consider the case for which there are 15 candi-

date neighbours and the maximum number of neighbours to be allocated to the reference cell is 8. Among these 15

candidate neighbours, only 8 (having the highest importance values) will be allocated to the reference cell. Note that the

maximum number of inter-technology neighbours can be defined at the cell level (property dialogue or cell table). If defined

there, this value is taken into account instead of the default one available in the Neighbour Allocation dialogue.

As indicated in the table below, the neighbour importance depends on the cause; this value varies between 0 to 100%.

Except the case of forced neighbours (importance = 100%), priority assigned to each neighbourhood cause is determined

using the Importance Function (IF). The IF considers two factors for calculating the importance:

• The co-site factor (C) which is a Boolean

• The overlapping factor (O) meaning the percentage of overlapping

The IF is user-definable using the Min importance and Max importance fields.

The IF evaluates importance as follows:

Where

Guidelines for the automatic allocationWhen the automatic allocation is based on coverage overlapping, we recommend you to perform two successive auto-

matic allocations:

- A first allocation in order to find handovers due to non-continuous UMTS coverage. In this case, you have to select the

max Ec/I 0 option and define a high enough value.

- A second allocation in order to complete the previous list with handovers motivated for reasons of traffic and service

distribution. Here, the max Ec/I 0 option must be disabled.

Neighbourhood reason WhenImportance

value


importance


Co-site transmitter If the Force co-site cells as neighbours option is selected IF


coverage conditionsIf the % minimum covered area is exceeded IF

Factor Min importance Default value Max importance Default value

Overlapping factor (O) 1% 60%

Co-site factor (C) 60% 100%

Co-site neighbourhood reason IFResulting IF using the default

values from the table above

No

Yes

S A SB

S A

---------------------- 100 u

Min O Max O

Min C Max C

Min O ' O O + 1% 59% O +

Min C ' C O + 60% 40% O +







Therefore, a neighbour may be marked as exceptional pair, co-site or coverage. For neighbours accepted for co-site and

coverage reasons, Atoll displays the percentage of area meeting the coverage conditions and the corresponding surface

area (km2). Finally, if cells have previous allocations in the list, neighbours are marked as existing.

6.8.2.3 Appendices

6.8.2.3.1 Delete Existing Neighbours Option

As explained above, Atoll keeps the existing inter-technology neighbours when the Delete existing neighbours option is

not checked. We assume that we have an existing allocation of inter-technology neighbours.

A new TBA cell i is created in UMTS.atl. Therefore, if you start a new allocation without selecting the Delete existing neigh-

bours option, Atoll determines the neighbour list of the cell i .

If you change some allocation criteria (e.g. increase the maximum number of neighbours or create a new GSM TBC trans-

mitter) and start a new allocation without selecting the Delete existing neighbours option, it examines the neighbour list of

TBA cells and checks allocation criteria if there is space in their neighbour lists. A new GSM TBC transmitter can enter theTBA cell neighbour list if allocation criteria are satisfied. It wil l be the first one in the neighbour list.

6.8.2.3.2 Calculation of Inter-Transmitter Distance

When allocation algorithm is based on distance, Atoll takes into account the real distance ( in m) and azimuths of anten-

nas in order to calculate the effective inter-transmitter distance ( in m).

where x = 0.5% so that the maximum D variation does not exceed 1%.

The formula above implies that two cells facing each other will have a smaller effective distance than the real physical

distance. It is this effective distance that will be taken into account rather than the real distance.

Notes:





ii. Neighbours based on coverage overlapping



• The ranking between neighbours from the same category depends on the factor (O).

Notes:

• No prediction study is needed to perform an automatic neighbour allocation. When starting

an automatic neighbour allocation, Atoll automatically calculates the path loss matrices if

not found.

• A forbidden neighbour must not be listed as neighbour except if the neighbourhood

relationship already exists and the Delete existing neighbours option is unchecked when




• In the Results, Atoll displays only the cells for which it finds new neighbours. Therefore, if a

TBA cell has already reached its maximum number of neighbours before starting the new

allocation, it will not appear in the Results table.

Figure 6.19: Inter-Transmitter Distance Computation

Note:

• This formula is not used when allocation algorithm is based on coverage overlapping. In

this case, real inter-transmitter distance is considered.

D

d

d D 1 x Ecosu x Dcosu – + u=



Chapter 7CDMA2000 1xRTT 1xEV-DO Networks

This chapter provides descriptions of all the algorithms for calculations, analyses, automatic allocations, simulations and prediction studies available

in CDMA2000 projects.


