121334268 transition from gsm to umts

GSM to UMTS TransitionRF Engineering GuidelineEG: GSMUTR

401-380-373 Issue 1.1 July 2000

Lucent Technologies - Proprietary This document contains proprietary information of Lucent Technologies and is not to be disclosed or used except in accordance with applicable agreements Copyright 2000 Lucent Technologies Unpublished and Not for Publication All Rights Reserved

Copyright 2000 by Lucent Technologies. All Rights Reserved. This material is protected by the copyright laws of the United States and other countries. It may not be reproduced, distributed, or altered in any fashion by any entity (either internal or external to Lucent Technologies), except in accordance with applicable agreements, contracts, or licensing, without the express written consent of the Customer Training and Information Products organisation and the business management owner of the material. For permission to reproduce or distribute, please contact: The Manager, RF Systems & Capacity Engineering Group 01793 883275 (domestic) (44) 1793 883275 (international)

Notice Every effort was made to ensure that the information in this information product was complete and accurate at the time of printing. However, information is subject to change.

Contents

1. ABOUT THIS DOCUMENT1.1. 1.2. 1.3. 1.4. Purpose Contents Scope Audience

11 2 3 3

2. INTRODUCTION TO THE UMTS AIR INTERFACE2.1. Background

55 5 5 7 8 8 8 11 13 13 14 15 17

Frequency allocation Standards UMTS summary 2.2. Band plan

Satellite allocation Terrestrial allocation 2.3. 2.4. UTRAN air interface attributes Channel mapping on the air interface

Access stratum Logical channels Transport channels Physical channels

Issue 1.1 - July 2000

Lucent Technologies Proprietary See Notice on first page

iii

GSM to UMTS Transition RF Engineering Guideline

2.5.

Channel spreading, coding and modulation

23 23 27 29 29 29 31 34 35 36 36 36 37 44 44 45 46

Uplink Downlink Synchronisation codes 2.6. Physical channel frame structure

Uplink time slot fields Downlink time slot fields 2.7. Speech coding

Transcoder Free Operation GSM Full Rate codec GSM Half Rate codec GSM Enhanced Full Rate codec Adaptive Multi-Rate codec types 2.8. Codec speech quality

Fixed rate codecs Disadvantages Adaptive Multi Rate codecs

3. MOBILE HANDOVER3.1. 3.2. 3.3. 3.4. Handover types Cell sets Preparation for UTRAN to UTRAN handover Preparation for UTRAN to GSM handover

4949 51 52 53 54 58 58 58 59

Silence Duration parameters 3.5. Handover execution

UTRAN soft handover UTRAN to GSM handover 3.6. GSM to UTRAN handover

iv



GSM to UMTS Transition - RF Engineering Guideline

4. SUBSCRIBER SERVICES4.1. 4.2. 4.3. Coding and interleaving for subscriber services Services multiplexing Rate matching

6162 62 63 63 63 64 64 64 64 64 65 65 66

Uplink Downlink 4.4. Control channel coding and interleaving

Dedicated Control Channel Downlink Common Control Channels 4.5. Channel mapping examples-1

8kbs bearer - speech 144kbs bearer - data 384kbs-1 bearer - data 480kbs bearer - data-1 -1

5. LUCENT EQUIPMENT5.1. Node-B (BTS)

6768 69 70 70 75 77 78 78 78 80 84 87 87 88 89

Distributed Milli-cell Microcell (Ultra-small cell) Milli-cell BTS traffic capacity BTS further reference 5.2. BTS Antennas

No existing network Existing single band network Existing dual band network Dual band and tri band GSM/UTRAN diplexers Broadband power divider Broadband indoor antennas Antenna feeder Masthead amplifier



v


Repeater Active/smart/adaptive antennas General antenna comments Further reference antennas and BTS accessories 5.3. Radio Network Controller

90 91 92 92 93 96 99 99 99 104 107 107 110 111 111 112 112 113 116 117 117 122 122 122 122 123 123 123

RNC characteristics RNC further reference 5.4. Radio Resource Control software

Radio resource allocation Radio Resource Allocation functions Radio Access Bearer parameters Physical channel related parameters Reverse outer loop power control Further reference power control system Power control parameters Radio Resource Control software further reference 5.5. Handover

Measurement reporting Measurement messages Measurement performance Soft (and softer) handover algorithm Hard handover algorithm UTRAN GSM handover algorithm UTRAN GSM GPRS handover algorithm Handover control software further reference 5.6. Lucent equipment capacity

Hardware (Lucent Network Release 1.0) Software (Lucent Release 0.1)

6. RF NETWORK COVERAGE AND CAPACITY DESIGN6.1. Frequency planning

125126 126 127

Frequency planning criteria Example UTRAN band assignment United Kingdom

vi




6.2.

Code assignment parameters

129 129 129 130 130 131 131 131 133 145 148 149 150 152 153 159 159 160 160 161 173 174 174 175 175 175 176 177 177 178 178

Primary Synchronisation Channel Code Secondary Synchronisation Channel Code Scrambling code Code assignment summary 6.3. Air interface link power budget

Background Effect of coding scheme on link power budget Link power budget elements Example link power budgets Margins for fading and building attenuation 6.4. Estimating coverage and traffic capacity

Fixed cell loading Adaptive cell loading Estimating base station numbers Land-use classification UMTS Dimensioning tool further reference 6.5. Airpro coverage and traffic distribution prediction software

Introduction Principal features Airpro default values 6.6. BTS and antenna settings optimisation software

Introduction Applications Cellular radio standards Optimising strategy RF model Results Comparison trials 6.7. 6.8. Inter-system boundary Further reference



vii


7. RF NETWORK PRACTICAL IMPLEMENTATION7.1. 7.2. Use of existing sites EMC at existing GSM sites

179179 180 180 181 183 183 184 185 190 191 193 198 198 199 200 200 200 201 202 203 203 204 205 206 207 207 208 208 209 210 212 214

Transmitter isolation Receiver performance degradation Inter-system isolation criteria Estimating inter-system isolation Antenna coupling Separate UTRAN and GSM antenna systems Single UTRAN and GSM antenna system UTRAN BTS spurious emissions Worked example of co-siting GSM and UTRAN 7.3. Use of repeaters

Introduction Design Uniform range extension Cascaded range extension Repeater gain and composite noise factor Donor cell shrinkage Median repeater link budget calculation Median repeater link budget adjustment Repeater donor cell antennas Summary 7.4. Use of microcells

Embedded and non-embedded microcells Microcell problems Co-channel macrocells and microcells Macrocells and microcells on different channels Dual layer UTRAN 7.5. Masthead amplifiers

Without masthead amplifier With masthead amplifier Summary

viii




7.6.

Practical antenna considerations

214 214 215 215 217 218 219 219 220 222 222

Intermodulation products Front-to-back ratio Variable electrical down-tilt 7.7. Transmit diversity

UTRAN implementation Effect of UTRAN implementation 7.8. 7.9. 7.10. Coverage areas for different services Survey test equipment Site selection and design optimisation

RF network engineering requirements

8. DEPLOYMENT WORKED EXAMPLE8.1. Scenario

225225 226 226 229 230 230 232 233 237 241 241 244 247 260 260 261

Coverage area Existing base station sites 8.2. 8.3. Background Estimating coverage and capacity

Coverage model Traffic model System model Analysis results 8.4. Coverage prediction with Airpro (CE 5)

Existing 900MHz coverage Existing 1800MHz coverage Predicted UTRAN coverage Results Conclusion 8.5. EMC when using a common antenna



ix


900MHz receiver - in-band power 1800MHz receiver - in-band power UTRAN receiver - in-band power GSM 900MHz receiver out-of-band power GSM 1800MHz receiver out-of-band power UTRAN receiver out-of-band power Conclusion

261 262 262 263 264 265 266

APPENDIX A SILENCE DURATION PARAMETERS ACRONYMS

267 269

x




About this Document

11. About this Document1.1. PurposeThis document describes the principal Radio Frequency (RF) engineering implications when cellular radio networks make the transition from the GSM air interface standard to the Universal Mobile Telecommunications System (UMTS) standard. In particular it discusses the following areas: Air interface Handover Subscriber services Lucent equipment RF network coverage and capacity RF network practical implementation Future developments

Owing to the wide range of environments in which UMTS systems will be deployed, and the differing operational and commercial priorities of network operators, the choice of implementation technique will be territory and customer specific. As a result, this document can only provide an outline guide to the main issues involved in the network design and deployment



1


process. It is not a substitute for the detailed analysis and design that will be required on a site by site basis.

1.2.

ContentsChapter 2 - Air Interface

This chapter provides a brief introduction to the main characteristics of the radio link, including the function of transport channels and their mapping to the physical channels transmitted over the air. It describes the concept of channel codes, which are used to implement Code Division Multiple Access (CDMA) to the shared RF channel (frequency), and identifies the main codes applied to the air interface. It also describes spreading factor (and associated de-spreading gain) and bearer data rate concepts. As this document deals with the physical layer of the link, protocol aspects are not covered. Chapter 3 - Handover

This chapter describes the main UMTS handover types, with particular emphasis on inter system handover with a neighbouring or overlaid GSM network. Chapter 4 - Subscriber Services

This chapter describes the basic subscriber services, from the viewpoint of their associated bearer data rates and spreading factors. Later chapters use these concepts when looking at the effect of the anticipated mix of services on RF network capacity and design. Chapter 5 - Lucent Equipment

This chapter describes the network elements that form the base station system. It describes the main attributes of Lucents equipment, and equipment capacity and interdependencies as they relate to RF network capacity and design. Chapter 6 - RF Network Coverage and Capacity Design

This chapter describes the network design process from the RF perspective, including radio link power budget calculation, estimated coverage techniques, and traffic modelling. It also describes Lucent software tools and techniques for network design. Chapter 7 - RF Network Practical Implementation

This chapter discusses some of the practical issues involved in upgrading an existing network from GSM to UMTS.

2




Areas considered include inter system interference, equipment that can be shared between systems, and the extent to which a UMTS service can be provided from an existing infrastructure of GSM base station sites. Chapter 8 - Deployment Worked Example

This chapter describes a worked example of UMTS deployment. Where possible, the data used is based on actual traffic projections and infrastructure configurations provided by existing network operators. Appendix A - Silence Duration Parameters

Appendix A reproduces an extract from ETSI UMTS xx.15 v1.0.0 1999-02 regarding the Definition and Setting of Silence Duration Parameters.

1.3.

Scope

This document covers the radio engineering implications of the transition from the use of the GSM air interface to that of UMTS, for the provision of cellular radio services. It is primarily intended to cover circumstances where existing GSM network operators wish to upgrade their network, in whole or part, to provide a UMTS service. In view of the range of this subject, many topics can only be covered in outline. However, sufficient information is included to identify the major issues involved and to identify areas for further planning and investigation. This document describes the features and facilities available in Lucent Network Release 1.0 software.

1.4.

Audience

This document concentrates on the World Radio Conference (WRC) Region 1 (Europe, MiddleEast and Africa) as this is where UMTS networks will be first established. However, many of the subjects covered will be relevant to other WRC regions. It is intended for use by the following groups: Engineers undertaking first-time design of GSM to UMTS upgrades Technicians who want an appreciation of UMTS design considerations Sales or support staff Managers or administrative staff who want a high-level understanding of UMTS design processes



3


A working knowledge of the following topics will be useful: GSM air interface (such as power budget and the basic structure of control and traffic channels) CDMA concepts

4




Introduction to the UMTS Air Interface

2

2. Introduction to the UMTS Air InterfaceThis chapter describes the main features of the UMTS radio link. It concentrates on the system aspects that have most impact on the design and implementation of a customer network, rather than on the design of individual network elements.

2.1.

Background

Frequency allocationThe World Radio Conference (WRC) has allocated a frequency band around 2GHz for use by Public Land Mobile Networks (PLMNs) to support third generation public mobile phone and data services. These are expected gradually to replace the existing second generation networks (largely based on GSM and IS95). The International Telecommunications Union Radio (ITU-R) has overall responsibility for defining the third generation system, known as the International Mobile Telecommunications 2000 (IMT-2000). The IMT-2000 system will be an integrated system that allows terminals to access both satellite and terrestrial based stations. Bands 1885MHz to 2025MHz and 2110MHz to 2200MHz are allocated to the terrestrial component. Bands 1980MHz to 2021MHz and 2170MHz to 2200MHz are allocated to the satellite component.

StandardsThe ITU-R has produced high-level documents covering the performance, service type, and inter-working requirements for IMT-2000.



5


Various international standards bodies such as the European Telecommunications Standards Institute (ETSI) are responsible for the detailed technical specifications of the equipment required to provide an IMT-2000 compatible service. A number of different standards are likely to emerge; but they are expected to have sufficient inter-working capability to allow an integrated IMT-2000 service for subscribers. IMT-2000 networks will support five interface standards: IMT-DS Frequency Division Duplex (FDD) version of UMTS (discussed in this document) IMT-MC IMT-TC IMT-SC IMT-FT US CDMA 2000 standard Time Division Duplex (TDD) version of UMTS GSM EDGE (IS-136) standard DECT standardrd

The four Technical Specification Groups (TSGs) of the ETSI-supported 3 Generation Partnership Project (3GPP) have approved the detailed specification parts of their submission to the ITU-R for the IMT-2000 radio interface standard. This is a terrestrial radio interface specification known as the Universal Terrestrial Radio Access Network (UTRAN). The UTRAN is based on a Wide-band Code Division Multiple Access (WCDMA) air interface. The ITU-R intend to approve the detailed specification of the IMT-2000 radio interface in May 2000, based on submissions received from the international standards bodies.

6




UMTS summaryA UMTS network can consist of one or more access networks, using different radio access systems, linked to the same core network. Together they form a single UMTS network. The term UMTS covers all the network elements in both the access network and the core network. The UTRAN specified by ETSI is one such UMTS access network. It supports wide area terrestrial mobile telecommunications services, using the Frequency Division Duplex (FDD) IMT-2000 bands. This document concentrates primarily on the ETSI FDD UTRAN. The UTRAN comprises the network elements that correspond to the Base Station Subsystem (BSS) in a GSM network: Base Transceiver Station (BTS) Base Station Controller (BSC)

In the UTRAN, the equivalent of the GSM BTS is referred to as either the BTS or the Node B. The equivalent of the GSM BSC is referred to as the Radio Network Controller (RNC). The UTRAN BTS and RNC together form the Radio Network System (RNS). Owing to differences in the radio standards currently used, and consequently those used to provide a IMT-2000 service, this document concentrates on WRC Region 1, where the UTRAN RNS will be used as the UMTS access network.



7


2.2.

Band plan

The WRC of 1992 allocated 230MHz of the 2GHz spectrum to IMT-2000 services with the intention of providing a uniform band plan for all three WRC regions. The allocated spectrum consists of two blocks: 140MHz for TDD and FDD uplinks from 1885 to 2025MHz 90MHz for FDD downlinks from 2110 to 2200MHz

Satellite allocationWithin each block, two 30MHz sub-blocks with a 190MHz duplex separation are allocated to satellite operation: 1980MHz to 2010MHz for uplinks 2170MHz to 2200MHz for downlinks

Terrestrial allocationFDD systems 1920 to 1980MHz Uplink Mobile Transmit (60MHz band, duplex separation 190MHz) 2110 to 2170MHz Downlink Base Transmit (60MHz band, duplex spacing 190MHz)

TDD systems 1885 to 1920MHz (35MHz band) 2010 to 2025MHz (15MHz band)

8




The IMT-2000 band plan and that of the existing major PLMN systems is illustrated in the following figure, according to their primary geographic area of use.

Figure 1 IMT-2000 band plan compared to existing PLMN systems

In many areas, part of the IMT-2000 band is already in use by other PLMN systems. The lower TDD IMT-2000 band will have to coexist with the Digital European/Enhanced Cordless Telephone/Telecommunications (DECT) and Personal Handy-phone System (PHS) systems. DECT DECT is widely used in the EU, Africa, Asia, and Caribbean to provide cordless telephony, wireless Private Branch Exchanges (PBX), and Wireless Local Loop (WLL) services. DECT uses dynamic channel allocation, and except where used for WLL applications, is usually deployed on a self-provision basis. This means there is limited scope for frequency coordination with a wide area IMT-2000 network. The output power of most DECT systems is limited to 250mW peak, 10mW average. For indoor systems, interference to outdoor IMT-2000 systems is further reduced by building attenuation.



9


PHS PHS was designed in Japan. In recent years subscriber numbers have declined sharply, and several networks have closed. PHS can be considered as functionally similar to DECT. It is rarely used outside SE Asia, Australia and New Zealand, owing to problems arising from its partial use of DECT frequencies. Similar problems to those found with DECT are likely to be experienced when coexistence with TDD IMT-2000 is attempted. PCS The Personal Communication System (PCS) 1900 is currently being deployed as a second generation cellular system in the Americas. Some areas are adopting a CDMA air interface based on the IS95 standard (J-STD-008), and others the GSM air interface. Whichever air interface is used, there is a significant overlap with the IMT-2000 TDD and FDD uplink band. In view of this, the American Federal Communications Commission (FCC) has allocated a different band for third generation cellular systems. The FCC has adopted licensing and service rules governing operation for 30MHz of the 700MHz spectrum auctioned in the spring of 2000. It has established two license bands, one of 20MHz (two paired 10MHz bands) and one of 10MHz (two paired 5MHz bands) that can be used for advanced wireless services, including third generation broadband wireless access. Spectrum is made up in part from 746-764MHz and 776-794MHz (TV channels 60-62 and 64-66). The FCC will auction the licenses in six Economic Area Groupings across America and will allow interested parties to bid for both license bands in one area. The remaining 6MHz will be used as guard band comprising 4MHz (two paired 2MHz bands) and 2MHz (two paired 1MHz bands). FCC plans to invite comments on technical and operational issues regarding these frequencies. In many parts of America, third generation cellular services may not be able to use these frequencies until 2007 owing to the use of television channels 60 to 69. GSM 1800 / DCS 1800 There is no direct overlap between the GSM 1800 (also known as DCS 1800) downlink band, and the IMT-2000 TDD and FDD uplink bands. However, in view of the possibility of co-siting base station equipment, and sharing antenna systems, particular care is required to ensure adequate isolation in order to avoid interference. This topic is covered in Chapter 7 RF Network Practical Implementation.

10




2.3.

UTRAN air interface attributes

This document concentrates on the ETSI UTRAN system. There are two versions of UTRAN, one that uses TDD mode and one that uses FDD mode. This document deals mainly with the terrestrial FDD version of UTRAN. This is the standard that forms the basis for the wide area rd deployment of 3 generation PLMNs.



11


The principal air interface attributes of the FDD and the TDD UTRANs are:

Feature Uplink frequency /MHz Downlink frequency /MHz Channel bandwidth /MHz Carrier raster /MHz Duplex separation /MHz Frequency stability /ppm Chip rate /Mcs-1 Spreading factor BSS separation codes Modulation Modulation filter Power control Power control steps /dB Minimum output power /dBm Power control dynamic range /dB Power control sample rate /kHz Channel coding & interleaving for services tolerating BER > 10-6 Channel coding & interleaving for services requiring BER < 10-6 Modulation symbol rate /M symbol s-1 Radio super-frame length /ms Radio frame length /ms Radio slot per frame Channel allocation

Terrestrial FDD UTRAN 1920-1980 2110-2170 5 0.2 130 (min) but variable 0.05 base, 0.1 mobile 3.84 4 to 256 Gold code 10ms, 38400 chips, length 241 -1 QPSK Root raised cosine roll-off factor 0.22 Fast closed loop & slow quality loop 0.25 to 1.5 -50 80 uplink, 30 downlink 1.6 Convolutional, rate 1/2 or 1/3 Turbo coding 0.016 to 1.024 720 10 15 of 666.7us 2,560 chips Network controlled

TDD UTRAN 1885-1920 & 2010-2025 1885-1920 & 2010-2025 5 0.2 N/A 0.05 base, 0.1 mobile 3.84 1 to 16 Scrambling code of length 16 chips QPSK Root raised cosine roll-off factor 0.22 Open loop & slow closed loop 0.25 to 1.5 -50 80 uplink, 30 downlink

Convolutional, rate 1/2 or 1/3 Turbo coding 0.256 to 4.096 240 10 15 of 666.7us 2,560 chips Dynamic

12




Feature Handover control

Terrestrial FDD UTRAN Mobile assisted measurement of signal level & timing. GSM measurements supported. 1 macro, 2 micro, 3 pico +21 [4] +33 [1], +27 [2], +24 [3], +10 [5], 0 [6]

TDD UTRAN Probing for ODMA

Base output power class Mobile output power /dBm [class number]

+21 [4] +33 [1], +27 [2], +24 [3], +10 [5], 0 [6]

Table 1 Air interface attributes

2.4.

Channel mapping on the air interface

This section summarises the mapping of logical channels and transport channels to the physical channels transmitted over the air interface.

Access stratumThe access stratum on the air interface is divided into three layers: Layer 1 Layer 2 Layer 3

Layer 1 Layer 1 is the physical layer. Signalling and traffic data is borne on the air interface by physical channels. The physical channels are defined by code set and frequency in FDD mode and by code, timeslot, and frequency in TDD mode. Layer 2 Layer 2 is divided into two sub-layers: Medium Access Control (MAC) layer (lower layer). The MAC layer is responsible for the random access procedures, physical link control, error protection, ciphering, multiplexing, and channel mapping to the physical layer (Layer 1) Radio Link Control (RLC) layer (upper layer).



13


The RLC layer is responsible for logical link control, and acknowledgement /unacknowledgement of data transfer Layer 3 Layer 3 is the Radio Resource Control (RRC) layer. The RRC layer is responsible for coordination and control of bearers, monitoring processes, power control, measurement reporting, paging, and broadcast control functions. In order to define a process for each different type of information, sets of logical channels are mapped onto transport channels, and ultimately physical channels are defined. Logical channels are defined between the RLC and the MAC. Transport channels are defined between the MAC and the physical layer (Layer 1).

Logical channelsThe following logical channels are used to transfer signalling information: Broadcast Control Channel (BCCH) -downlink The BCCH is a downlink broadcast channel which carries system information. There are two types: BCCH-Constant (BCCH-C) and BCCH-Variable (BCCH-V), the data on which may be constantly updated. Paging Control Channel (PCCH) - downlink The PCCH is a downlink channel which carries paging messages. It is used when the network does not know the location cell of the mobile or the mobile is in sleep mode. Common Control Channel (CCCH) The CCCH is a bi-directional channel which carries data when the mobile has no RRC connection to the network. Dedicated Control Channel (DCCH) The DCCH is a bi-directional channel which carries point-to-point dedicated control data between the network and a mobile. It is used when a dedicated connection has been established through RRC connection set up procedures. ODMA Common Control Channel (OCCCH) The OCCCH is a bi-directional channel which carries control data directly between mobiles. It is used when the mobile has no RRC connection with the network. ODMA Dedicated Control Channel (ODCCH)

14




The ODCCH is a bi-directional point-to-point channel which carries dedicated control data directly between mobiles. It is used when a dedicated connection has been established through the RRC connection set-up procedures. Shared Channel Control Channels (SHCCH) TDD mode only The SHCCH is used in TDD mode only. It is a bi-directional channel which carries control data for the uplink and downlink shared channels. Synchronisation Control Channel (SCCH) TDD mode only The SCCH is used in TDD mode only. It is a downlink channel which carries the location and structure of the BCCH. The following logical channels are used to transfer user data: Dedicated Traffic Channel (DTCH) The DTCH is a bi-directional dedicated point-to-point channel which carries user data between the network and the mobile. ODMA Dedicated Traffic Channel (ODTCH) The ODTCH is a dedicated point-to-point channel which carries user data directly between mobiles, used as a relay link. Common Traffic Channel (CTCH) The CTCH is a uni-directional point-to-multipoint channel which carries user data for a specified group of mobiles.

Transport channelsThe information is transferred from the MAC layer and mapped onto physical channels via a set of transport channels. There are two types of transport channel: Common transport channels Dedicated transport channels

Common transport channels Broadcast Channel (BCH) downlink The BCH is transmitted from the base station to all mobiles in the cell coverage area and broadcasts system configuration information.



15


Paging Channel (PCH) downlink The PCH is also broadcast from the base station to all mobiles in the cell coverage area. It pages mobiles when they are in idle mode and only their Location Area, not their specific cell, is known.

Forward Access Channel (FACH) downlink The FACH is transmitted from the base station to mobiles and carries relatively small amounts of control data and very short packets of user data, without the use of power control.

Random Access Channel (RACH) uplink The RACH is a contention-based uplink channel used for initial access, non-real time dedicated control or very short packets of traffic data.

ONMA Random Access Channel (ORACH) The ORACH performs a similar function to the RACH when a relay link is used.

Common Packet Channel (CPCH) FDD mode only The CPCH is a contention-based channel carrying bursty traffic data in shared mode using fast power control.

Downlink Shared Channel (DSCH) The DSCH is a downlink channel shared between several mobiles for carrying control or traffic data.

DSCH Control Channel The DSCH Control Channel is a downlink channel used in conjunction with the DSCH for signalling of DSCH resource allocation.

Broadcast Channel (BCH) The BCH is a downlink broadcast channel carrying system information for the whole cell.

Synchronisation Channel (SCH) TDD mode only The SCH is a downlink TDD mode channel carrying synchronisation data for the whole cell. Note: This SCH has no connection with the physical channel SCH used in FDD mode.

Uplink Shared Channel (USCH) TDD mode only The USCH is an uplink TDD mode channel shared by several mobiles for carrying control or traffic data.

16




Dedicated transport channels Dedicated Channel (DCH) duplex channel pair For the duration of a call, one DCH is transmitted from the mobile to the base station and one DCH is transmitted from the base station to the mobile. They form a duplex circuit that can be used to carry a number of different types of user data (or logical channels). Fast Uplink Signalling Channel (FAUSCH) The FAUSCH is an uplink channel used to allocate dedicated channels in conjunction with the FACH. ODMA Dedicated Channel (ODCH) The ODCH is dedicated to one mobile when used for relay links. With the exception of the FAUSCH, each transport channel has an associated transport format. This is defined by a combination of encoding, interleaving, bit rate, and mapping onto physical channels. Some transport channels may use variable formats.

Physical channelsThe transport channels are mapped to the physical channels transmitted over the air interface. This mapping process is flexible and for some logical channels there are several options depending on their function and the type of information being transferred. Different transport and physical channels are used in the uplink and downlink directions. Depending on the type and data rate of the transport channel, coding and multiplexing/ demultiplexing may be applied prior to the data being carried by the physical channel. The channels carrying broadcast information are directly mapped. That is, BCCH to BCH and PCCH to PCH. Mapping for the other control and traffic channels is more flexible. For example, the downlink DCCH can be mapped onto either the FACH or the DSCH depending on the information requirements. In the uplink, the DCCH may take information from the CPCH, FAUSCH, RACH, USCH or DCH. The logical channel DTCH has access to a similar range of transport channels. The CCCH can only use the RACH and FACH for bi-directional communication. Downlink mapping of transport to physical channels Three common control channels are used in the downlink: BCH, PCH and FACH. Each have a specific coding process and are mapped onto a physical channel that is specific to common control. For DCH coding, multiplexing and mapping to physical channels is similar to that for the uplink.



17


If a mobile is in soft handover (establishes a connection with a new cell while still communicating with its current serving cell) the same data may be received by the mobile from multiple base stations, each using different physical channels. Prior to DCH transmission in the physical layer, information is added at Layer 1. During a soft handover the physical layer information bits received from the target base stations will be different from those received from the serving base stations. Also, the composition of the CCTrCH may not be the same for all base stations involved in the soft handover. Transmit Power Control (TPC) bits and Transport Format Combination Indicator (TFCI) bits are added to the DCH (as Layer 1 information) before it is transmitted over the physical channel. During soft handover, the TPC bits received from the neighbour base stations will differ from those received from the serving base station. The TFCI bits are identical.

BCH PCH

Code Code

OVSF Code 1 Prim. CCPCH OVSF Code 2 Sec. CCPCH OVSF Code 3 DPCH

FACH DCH(comprising DCCH, DTCH)

Code (and mux/divide if CCTrCH exceeds capacity of one Physical Channel)

TPC and TFCI added prior to transmission for UE power control

DCH(comprising DCCH, DTCH)


OVSF Code 4 DPCH

TPC and TFCI added prior to transmission for UE power control Base Station

DCH(comprising DCCH, DTCH)


OVSF Code n DPCH

TPC and TFCI added prior to transmission for UE power control

Figure 2 Mapping of downlink transport and physical channels

18




Uplink mapping of transport to physical channels Two common control channnels are used in the uplink: RACH and FAUSCH. Each one has specific coding mechanisms and is mapped onto a physical channel. The DCH channels (DCCH and DTCH) are coded and multiplexed to form a Coded Composite Transport Channel (CCTrCH). Depending on its data rate the CCTrCH is then mapped onto one physical channel or demultiplexed onto several physical channels. The physical channel used for the CCTrCH carries data from the DCH transport channel only. The DCH data is not multiplexed with data from other transport channels. Transmit Power Control (TPC) bits and Transport Format Combination Indicator (TFCI) bits are added to the DCH (as Layer 1 information) before it is transmitted over the physical channel.RACHOVSF Code 1

CodePRACH OVSF Code 2

DCH

Code & Mux.DCH DCH(comprising DCCH and DTCH)

Divide If CCTrCH exceeds capacity of one Physical Channel

DPCH OVSF Code 3 DPCH Base Station OVSF Code n

CodeDPCH

Transmit Power Control (TPC) Transmit Format Combination Indicators (TFCI)added prior to transmission for BTS power control

Figure 3 Mapping of uplink transport and physical channels Physical channel types The physical channels are defined by their basic resource characteristics, in terms of code and frequency plans, modulation and transmission. A number of different physical channel types are used in both the uplink and downlink. Downlink physical channels Downlink physical channels are grouped into four types: Synchronisation Channel (SCH) The SCH transmits the synchronisation codes used by the mobile to synchronise to a base station. It comprises a primary and a secondary channel, which are transmitted simultaneously.



19


Common Pilot Channel (CPICH) The CPICH provides the phase reference for the downlink common channels, and is implemented as either a Primary or Secondary Pilot Channel.

Common Control Physical Channel (CCPCH) The CCPCH is the bearer for the Broadcast Channel (BCH), the Paging Channel (PCH), or the Forward Associated Control Channel (FACCH). If the CCPCH carries the BCH it is the Primary CCPCH, if it carries either the PCH or FACCH it is the Secondary CCPCH.

Dedicated Physical Channel (DPCH) The DPCH carries the downlink Dedicated Channel (DCH) transport channel together with Layer 1 data comprising the Channel Associated Pilot, Transmit Power Control (TPC) bits, and Transmit Format Combination Indicator (TFCI) bits. The DPCH can be considered to be formed from the Dedicated Physical Data Channel (DPDCH) carrying the Dedicated Channel (DCH) and the Dedicated Physical Control Channel (DPCCH) carrying the Layer 1 data, time multiplexed together to form the DPCH

The downlink physical channel arrangement is shown in the following diagram:

20




Primary SCH(Sync. code, short Golay code, already chipped. BTS specific)

Synchronisation Channel (SCH)

(Layer 1 Data - chip rate, framing, group of Golay codes)

Secondary SCH

Primary CPICH

Common Pilot Channel (CPICH)Secondary CPICH

Primary CCPCH Base Station(BCH at 32 kb/s)

Common Control Physical Channel (CCPCH)

(Paging (PCH) and Forward Access (FACH) channels)

Secondary CCPCH

DTCH DCCH

(Dedicated Physical Data Channel (DPDCH)) (DCH Transport Channel)

(Dedicated Physical Control Channel (DPCCH)) (Channel Associated Pilot, TPC, TFCI)

Dedicated Physical Channel (DPCH)

Figure 4 Downlink physical channels



21


Uplink physical channels The uplink physical channels are: Physical Random Access Channel (PRACH) The PRACH carries the Random Access Channel (RACH) and the FAUSCH. Dedicated Physical Channel (DPCH) The DPCH carries the uplink Dedicated Channel (DCH) transport channel, together with Layer 1 data comprising the Channel Associated Pilot, Transmit Power Control (TPC) bits, Transmit Format Combination Indicator (TFCI), and Feedback Information (FBI) bits. The DPCH can be considered to be formed from the Dedicated Physical Data Channel (DPDCH) carrying the Dedicated Channel (DCH) and the Dedicated Physical Control Channel (DPCCH) carrying the Layer 1 data. These are fed separately to the I & Q ports of the mobiles QPSK modulator to form the DPCH. The uplink physical channel arrangement is shown in the following diagram:Random Access Channel (RACH)

Physical Random Access Channel (PRACH)

DTCH DCCH

(Dedicated Physical Data Channel (DPDCH)) (DCH Transport Channel)

Base Station

(Dedicated Physical Control Channel (DPCCH))

Dedicated Physical Channel (DPCH)

(Channel Associated Pilot, TPC, TFCI)

Figure 5 Uplink physical channels

22




2.5.

Channel spreading, coding and modulation

This section describes channel spreading, coding and modulation concepts in the uplink and downlink.

UplinkSpreading Dedicated Physical Data and Control Channels Dual-channel Binary Phase Shift Keying (BPSK) modulation is used to spread the Dedicated Physical Data Channels (DPDCH) and Dedicated Physical Control Channels (DPCCH). The uplink DPDCH and DPCCH are mapped to the I & Q branches of the modulator respectively. The I & Q branches are spread to the chip rate with two Orthogonal Variable Spreading Factor (OVSF) codes and then scrambled using a mobile specific scrambling code cscramb. This process is illustrated in the following diagram:

Orthogonal Variable Spreading Factor (OVSF) Codes

cos (t)

DPDCHOVSF Codes

I I+jQ

Cscramb

Real p(t)

+Qp(t) Imag

sin (t)

+

DPCCH

Spreading & Modulation for Uplink DPDCH/DPCCH

Figure 6 Uplink spreading and modulation When more than one code is used for transmission, additional uplink DPDCHs may be transmitted on either the I or Q branches. For each branch, each additional DPDCH is assigned its own channel code. Uplink DPDCH channels on different branches may share a common channel code.



23


Spreading Physical Random Access Channels The message component of the Physical Random Access Channel (PRACH) is spread and modulated in a similar manner to that used for the uplink dedicated physical channels. The uplink DPDCH is replaced with the data part and the DPCCH is replaced with the control part. The scrambling code used for the message component is chosen based on the base stationspecific preamble code in use. Channel codes Orthogonal Variable Spreading Factor (OVSF) codes are used as channel codes, which ensure that a number of mobiles can share the same RF channel (frequency) without causing unacceptable interference. These codes allow Code Division Multiple Access (CDMA) to the shared RF channel (frequency). The OVSF codes are illustrated in the following code tree:

Figure 7 OVSF code tree for channel codes The code tree defines the code length used to provide the specified spreading factor. The higher user data rate services use shorter codes and hence lower spreading factors (and associated de-spreading gain). A given mobile cannot use all channel codes simultaneously. A channel code can only be used by a mobile if no other code on the path from the specific code to the root of the code tree, or in the sub-tree below the specific code, is used by that mobile. Thus the number of available channel codes is not fixed, but depends on the data rate and associated spreading factor of each physical channel used.

24




For each call, the mobile is allocated at least one uplink channel code, for an uplink DPCCH. Usually, at least one further uplink channel code is allocated for an uplink DPDCH. Additional uplink channel codes may be allocated if the mobile needs more DPDCHs. All channel codes used for the DPDCH must be orthogonal to the channel code used for the DPCCH. As each mobile using the same RF channel uses a unique uplink scrambling code, no coordination of the allocation of uplink channel codes to mobiles is needed. They are allocated in a predefined order that exploits the design of the scrambling codes used by the mobile transmitter. The mobile and the network may negotiate the number and length (spreading factor) of the channel codes needed for the call, and the network allocates the necessary codes. Scrambling codes To allow identification during inter-cell handover, each mobile is assigned a unique code which is not repeated at other cells. The uplink uses either short or long scrambling codes, depending on the capabilities of the base station receiver. In both cases complex scrambling is used, in which each code allocation consists of a pair of I & Q codes. Short scrambling codes are used in cells that use a sophisticated receiver with a multiple user detector and interference canceller. With short codes, the cross correlation between different physical channels and users does not vary with time as it does with long codes. Consequently the cross correlation matrices used in the advanced receiver have to be updated less frequently, reducing the complexity of the receiver design. The base station informs the mobile of its code allocation using the Access Grant message (although it is possible to change the code allocation during a call). Owing to their better interference averaging properties, long codes are used if the base station does not support multiple user detection. Each long code maps to an indicated short code, and the Access Grant message informs the mobile whether it is to use the indicated short code or the corresponding long code. Random access codes The base station broadcasts a cell specific spreading code for the preamble part of the Random Access Message. Additional codes may be broadcast if the traffic loading is high. These preamble spreading codes must be co-ordinated between cells to avoid interference. A realvalued 256 chip orthogonal Gold code is used, and all 256 codes may be used by the system. A preamble signature code is used that carries one of 16 different orthogonal complex signatures of length 16 based on a set of orthogonal Gold codes of length 16. The base station broadcasts the signatures that may be used in a cell.



25


The preamble signature specifies one of the 16 nodes in the code tree that correspond to channel codes of length 16. The sub-tree below the specified node is used to spread the message part of the Random Access Message. In addition to spreading, the message part is also scrambled with a 10ms complex code. This code is cell specific, and is associated with the spreading code used for the preamble part. Although the scrambling code is the same for each access slot, the scrambling codes do not collide in different access slots, as the slots are time shifted by 1.25ms. Modulation Quadrature Phase Shift Keying (QPSK) is used with a chip rate of 3.84M chips/s. To reduce the linearity requirements on the power amplifier and hence improve its efficiency, 0 the scrambling codes are designed so that N-1 out of N consecutive chips produce +/- 90 rotations of the I&Q multiplexed data and control channels. The remaining 1 out of N chips produces 0,+/- 900, or 1800 rotation. In addition to scrambling codes design, compatible uplink channel codes must be chosen. This limits the phase transitions of the baseband signal, prior to input to a pulse-shaping filter which ensures the resulting modulated signal is constrained within the RF spectrum channel mask (that is, adjacent channel interference is limited).

26




DownlinkSpreading Dedicated Physical Channel and Common Control Physical Channel QPSK data modulation is used to spread Dedicated Physical Channels (DPCH) and Common Control Physical Channels (CCPCH) in the downlink. Pairs of baseband data bits are taken in serial form and applied in parallel to the I & Q branches of the spreading and modulation chain. The I & Q branches are then spread to the chip rate with the same channel code cch (that is, real spreading) and then scrambled using the same cell specific scrambling code cscramb. This is illustrated in the following diagram:

Orthogonal Variable Spreading Factor (OVSF) Code (c ch)

cos (t)

Dedicated Physical Channel (DPCH) and Common Control Physical Channel (CCPCH)

I I+jQSerial to Parallel Conversion OVSF Code (c ch)

Cscramb

Real p(t)

+Qp(t) Imag

sin (t)

+

Spreading & Modulation for Downlink DPCH & CCPCHs

Figure 8 Downlink spreading and modulation Each physical channel uses a different channel code, but the same scrambling code. In addition to the DPCH and CCPCH the Synchronisation Channel (SCH) is also multiplexed onto the downlink transmission. The SCH is transmitted intermittently, one code word per slot, and multiplexed onto the downlink after the DPCH and CCPCH have been scrambled. The SCH is therefore not orthogonal to the other downlink channels. This is illustrated in the following diagram:



27


Figure 9 Downlink multiplexing of synchronisation channel Channel codes The OVSF codes used in the uplink are also used as downlink channel codes, to ensure orthogonality between the channels operating at different data rates and spreading factors. The code assignment restrictions that apply to the uplink on a per-mobile basis also apply on the downlink on a per-cell basis. The channel code used for the Primary CCPCH is predefined for all cells within the network. The BCH broadcasts the channel code used for the Secondary CCPCH. The channel codes used for the downlink Dedicated Physical Channels are allocated by the network. The base station tells the mobile which channel codes to receive by, in the downlink Access Grant message response to the uplink Random Access request. This channel code set may be changed during a call (for example, after a handover or change of service type). The change of channel code set is negotiated over the Dedicated Control Channel (DCCH).

28




Scrambling codes A total of 512 scrambling codes are available. They are divided into 32 sub-groups of 16 codes to aid rapid cell searching. A scrambling code is assigned to a cell at initial deployment, and the mobile is advised of the code during its cell search process. As a solution to cell congestion, the facility to assign more than one scrambling code to a cell may be adopted in the future.

Synchronisation codesSynchronisation is required for the mobile to decode the downlink data when first receiving a base station, and in the uplink direction for the base station to decode the mobiles random access requests. Primary and secondary codes are orthogonal Golay codes of length 256. They are formed by combining pairs of m-sequence codes and transmitted at the system chip rate. Modulation As on the uplink, Quadrature Phase Shift Keying (QPSK) with a chip rate of 3.84M chips/s is used on the downlink, with similar pulse shaping and filtering.

2.6.

Physical channel frame structure

A 10ms frame structure is used for all information transmitted on both uplink and downlink physical channels. The 10ms frame structure is divided into 15 time slots, each of 666.7s duration. 72 of these 10ms frames form one super-frame of 720ms duration. The data rate carried by each 10ms time slot depends on the type of transport channel carried and the type of service provided.

Uplink time slot fieldsEach uplink 666.7s radio time slot may contain a number of fields, the data capacity of which varies according to the format that is used. When the time slot carries the DPDCH, it has one data field per time slot (Ndata), which contains the number of data bits shown below:



29


Slot Format

Channel Bit -1 Rate /kbs 15 30 60 120 240 480 960

Channel Symbol Rate -1 /k symbols s 15 30 60 120 240 480 960

Spreading Factor 256 128 64 32 16 8 4

Bits per Frame 150 300 600 1200 2400 4800 9600

Bits per Slot

Ndata Number of Data Bits 10 20 40 80 160 320 640

0 1 2 3 4 5 6

10 20 40 80 160 320 640

Table 2 Formats for DPDCH data fields in uplink radio time slot When the time slot carries the DPCCH, it has four data fields per time slot Pilot (Npilot ) Transport Format Combination Indication (TFCI) (NTFCI ) Feedback Information (FBI) (NFBI) Transmit Power Control (TPC) (NTPC)

The bit assignment according to slot format is shown below:Slot Format Channel Bit Rate -1 /kbs Channel Symbol Rate /k symbol -1 ss 15 15 15 15 15 15 Spreading Factor Bits per Frame Bits per Slot Npilot Number of Pilot Data Bits 6 8 5 7 6 5 NTPC Number of TPC Data Bits 2 2 2 2 2 1 NTFCI Number of TFCI Data Bits 2 0 2 0 0 2 NFBI Number of FBI Data Bits 0 0 1 1 2 2

0 1 2 3 4 5

15 15 15 15 15 15

256 256 256 256 256 256

150 150 150 150 150 150

10 10 10 10 10 10

Table 3 Formats for DPCCH data fields in uplink radio time slot

30




Downlink time slot fieldsThe downlink Secondary Common Control Physical Channel (CCPCH) has three data fields per time slot: Transport Format Combination Indication (TFCI) (NTFCI ) Data (Ndata) Pilot (Npilot )Slot Format Channel Bit Rate -1 /kbs Channel Symbol Rate /k symbol -1 s 15 15 30 30 60 120 240 480 960 Spreading Factor Bits per Frame Bits per Slot Ndata Number of Data Bits 12 10 32 30 64 144 296 616 1256 Npilot Number of Pilot Data Bits 8 8 8 8 8 8 16 16 16 NTFCI Number of TFCI Data Bits 0 2 0 2 8 8 8 8 8

0 1 2 3 4 5 6 7 8

30 30 60 60 120 240 480 960 1920

256 256 128 128 64 32 16 8 4

300 300 600 600 1200 2400 4800 9600 19200

20 20 40 40 80 160 320 640 1280

Table 4 Secondary CCPCH data fields in downlink radio time slot with pilotSlot Format Channel Bit Rate /kbs-1 Channel Symbol Rate /k symbol -1 s 15 15 30 30 60 120 240 480 960 Spreading Factor Bits per Frame Bits per Slot Ndata Number of Data Bits 20 18 40 38 72 152 312 632 1272 Npilot Number of Pilot Data Bits 0 0 0 0 0 0 0 0 0 NTFCI Number of TFCI Data Bits 0 2 0 2 8 8 8 8 8

0 1 2 3 4 5 6 7 8

30 30 60 60 120 240 480 960 1920

256 256 128 128 64 32 16 8 4

300 300 600 600 1200 2400 4800 9600 19200

20 20 40 40 80 160 320 640 1280

Table 5 Secondary CCPCH data fields in downlink radio time slot without pilot The downlink time slots carry both the Dedicated Physical Control Channel (DPCCH) and Dedicated Physical Channel (DDPCH) in an interleaved manner, and have five data fields per time slot:



31


Transport Format Combination Indication (TFCI) (NTFCI ) DPCCH

Data Field 1 (Ndata1) DPDCH

Transmit Power Control (NTPC ) DPCCH

Data Field 2 (Ndata2) DPDCH

Pilot (Npilot ) DPCCH

32




Slot Format

Ch. Bit Rate -1 kbs

Ch. Symbol Rate -1 ksyms

SF

Bits per Frame DPD CH 60 30 240 210 210 180 150 120 510 480 450 420 900 2100 4320 9120 18720

Bits per Frame DPDC CH 90 120 60 90 90 120 150 180 90 120 150 180 300 300 480 480 480

Bits per Frame Total

Bits per Slot

DPD CH Bits per Slot Ndata1

DPD CH Bits per Slot Ndata2 2 2 14 14 12 12 8 8 28 28 24 24 56 120 240 496 1008

DPDC CH Bits per Slot NTFCI 0 2 0 2 0 2 0 2 0 2 0 2 8 8 8 8 8

DPDC CH Bits per Slot NTPCI 2 2 2 2 2 2 2 2 2 2 2 2 4 4 8 8 8

DPDC CH Bits per Slot NpliotI 4 4 2 2 4 4 8 8 4 4 8 8 8 8 16 16 16

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

15 15 30 30 30 30 30 30 60 60 60 60 120 240 480 960 1920

7.5 7.5 15 15 15 15 15 15 30 30 30 30 60 120 240 480 960

512 512 256 256 256 256 256 256 128 128 128 128 64 32 16 8 4

150 150 300 300 300 300 300 300 600 600 600 600 1200 2400 4800 9600 19200

10 10 20 20 20 20 20 20 40 40 40 40 80 160 320 640 1280

2 0 2 0 2 0 2 0 6 4 6 4 4 20 48 112 240

Table 6 Downlink DPDCH and DPCCH data fields



33


2.7.

Speech coding

Prior to transmission over a digital radio system, the analogue audio signal must be digitised. This function can be performed by a coder/decoder (codec). There are two types of codec: Waveform codecs Voice codecs (also known as vocoders)

Waveform codecs replicate the analogue waveform as faithfully as possible. Voice codecs extract the essential intelligibility information from speech, and reproduce it in a comprehensible manner, without regard to the accuracy with which the original analogue signal is reproduced. Where bandwidth efficiency is not essential, such as in the landline telephone network, waveform codecs are used (for example, ITU-T G.711 1972). Where bandwidth efficiency is essential, such as in mobile radio systems, voice codecs are normally used. As accuracy of the original analogue signal reproduction is not a measure of vocoder performance, their performance is based on subjective tests of perceived voice quality. One commonly used measure is the Mean Opinion Score (MOS). This is based on a jury marking the intelligibility and speaker recognition of pre-defined phrases spoken through the vocoder system. The MOS value of 4.3 for the land-line standard ITU-T G.711 for Pulse Code Modulation (PCM) 64 bit is normally used as a base reference level. In the case of residential subscribers and small businesses requiring only a few lines, G.711 circuits are not terminated at the customer premises, but are presented by an analogue local loop. The quality degradation introduced by the local loop varies greatly, depending on its length and transmission means. For example, several miles of analogue catenary with loading coils (typically used in rural locations) can introduce significant degradation. It has been reported that a typical G.711 circuit presented by analogue local loop has a MOS of 4.0. A wide variety of vocoder techniques are used depending on the application involved; but generally the faster the resultant data stream, the better the voice quality and resulting MOS value. When comparing MOS values, the comparison must be made when the vocoders are operating at the same Bit Error Rate (BER).

34




Currently the GSM and UMTS standards define six codecs supported by the UTRAN air interface: GSM Full Rate GSM Half Rate GSM Enhanced Full Rate GSM Adaptive Multi-Rate Full Rate Half Rate

UMTS Adaptive Multi-Rate

The definition of the common codec list in 3GPP for GSM and UMTS follows the specifications given in ITU Q.765.5.

Transcoder Free OperationThe UMTS Technical Specifications outline the 3GPP internal codec list for both GSM and UMTS codecs to be used by the Bearer Independent Call Control protocol to set up or modify a call in Transcoder Free Operation (TrFO). TrFO allows the transport of speech signals in the coded domain from one mobile to another through the radio access network and core network, and possibly through an additional transit network. This enables high speech quality, low transmission costs and high flexibility. Codec type selection and resource allocation is negotiated out-of-band before and after call setup. Possible Codec (re-)configuration, Rate Control and Discontinuous Transmission (DTX) signaling may be performed after call setup by additional in-band signaling, or by a combination of in-band and out-of-band signaling. Up to release 99, GSM does not support TrFO and specifies Tandem Free Operation (TFO) instead. TFO offers similar advantages to TrFO but is based on pure in-band signaling after call setup. The UMTS Technical Specifications allow interaction between TrFO and TFO. They also provide a GSM evolutionary path toward TFO.



35


GSM Full Rate codecThe Regular Pulse Excited Long Term Prediction (RPE-LTP) vocoder was selected for GSM in 1989. This generates an output bit rate of 13kb/s, and has a maximum MOS of 3.7. This data rate is termed the Full Rate for a GSM system. The RPE-LTP vocoder is formed from the following sections: LTP filter, Linear Prediction Coder (LPC), and de-emphasis filter. (The RPE-LTP is also known as the RPE-LPC.) DTX can be independently enabled in the uplink and/or downlink, as defined by the network on a cell basis. It cannot be negotiated at call setup or during the call. The DTX scheme uses one Silence Information Descriptor (SID) frame to mark the end of a speech burst and to start comfort noise generation. Identical SID frames for comfort noise updates are sent in speech pauses about every 480 ms, aligned with the cells TDMA frame structure. TFO allows the reception of GSM-FR DTX information for the downlink direction in all cases.

GSM Half Rate codecThe GSM Half Rate codec type supports one fixed Codec Mode with 5.6kbs . DTX may be used as in GSM Full Rate. Owing to poor speech quality, this codec is not widely used in GSM networks.-1

GSM Enhanced Full Rate codecThe GSM Enhanced Full Rate codec type supports one fixed Codec Mode with 12.2kbs . Again, DTX may be enabled in the uplink and/or the downlink independently, as defined by the network on a cell basis and cannot be negotiated at call setup or during the call. The DTX scheme uses one SID frame to mark the end of a speech burst and to start comfort noise generation. It is important to note that the parameters for the start of comfort noise generation are calculated at the transmitter side from the previous eight speech frames. A DTX hangover period therefore needs to be applied at the transmitter side before sending the first SID frame. SID frames with incremental information for comfort noise updates are sent in speech pauses approximately every 480 ms, aligned with the TDMA frame structure of the cell. The defined TFO allows reception of GSM EFR DTX information for the downlink direction in all cases.-1

36




Adaptive Multi-Rate codec typesAdaptive Multi Rate (AMR) is a new mobile technology that introduces a speech and channel -1 -1 codec able to support both GSM full rate (22.8kbs gross bit rate) and half rate (11.4kbs gross bit rate) channel modes. For each channel mode a number of different Codec Mode bit rates can be employed. The Lucent GSM AMR feature is realised using a speech codec located in the Speech Transcoding Frame (STF-2000) and the mobile, and a channel codec located in the BTS-2000 and the mobile. AMR differs from existing GSM speech codecs, in that it can adapt its data rate (speech coding) and error protection level (channel coding) in accordance with the prevailing radio channel conditions. By selecting the most appropriate channel mode (AMR Half Rate or AMR Full Rate) and the Codec Mode (combination of speech and channel bit rates), AMR is able to offer a balance between speech quality and network capacity. The sampling rate is 8 000 samples/s leading to a bit rate for the encoded bit stream of 4.75, -1 5.15, 5.90, 6.70, 7.40, 7.95, 10.2 or 12.2kbs . The coding scheme for the multi-rate coding modes is the Algebraic Code Excited Linear Prediction Coder (ACELP). The multi-rate ACELP coder is referred to as MR-ACELP. The AMR Codec algorithm is applied in GSM and UMTS in three different codec types: Full Rate AMR Half Rate AMR UMTS AMR



37


The AMR operating modes are shown below.

Channel type

Source coding bit rate 12.2 kbit/s

TCH/FS/AMR (TCH/AFS)

10.2 kbit/s (GSM EFR) 7.95 kbit/s 7.40 kbit/s (IS136 EFR) 6.70 kbit/s 5.90 kbit/s 5.15 kbit/s 4.75 kbit/s 7.95 kbit/s

TCH/HS/AMR (TCH/AHS)

7.40 kbit/s (IS136 EFR) 6.70 kbit/s 5.90 kbit/s 5.15 kbit/s 4.75 kbit/s

Table 7 AMR Codec Modes At call set-up the network selects a suitable Codec Mode set (containing a maximum of 4 Codec Mode bit rates) appropriate for the call. The uplink and downlink must use the same Codec Mode set. However, during the call the respective links may use different Codec Mode bit rates from within the chosen Codec Mode set. During a call, Codec Mode adaptation is controlled by the BTS. Quality measurements made by the BTS (uplink) and MS (downlink) are compared against pre-defined threshold/hysteresis values from which the BTS is able to decide whether a Codec Mode change is required. The BTS informs the MS and TRAU of any Codec Mode changes using in-band signalling. In theory Codec Mode adaptation can take place every speech frame. However, due to propagation delays and necessary filtering in the codec adaptation functions, a lower rate is recommended.

38




MSSPE CHE

BTSCHD

TRAUSPD

Uplink Speech Data Codec Mode Indication (uplink) Sug. Mode Command (downlink)

Codec Adaptation

Codec Adaptation

Codec Mode Command (uplink) Codec Mode Indication (downlink) Downlink Speech DataCHD CHE SPE

SPD

Figure 10 Codec mode signalling for GSM AMR In the uplink direction, the mobile indicates the Codec Mode employed on the current and next speech frames and suggests a Codec Mode to be used for the downlink. In the downlink direction the BTS uses the uplink quality measurements to generate an uplink Codec Mode command instructing the mobile to use a new Codec Mode at the earliest opportunity (earliest speech frame) as well as to indicate the Codec Mode for the downlink. This is based on either the mobile suggested mode or a BTS selected mode. The channel mode adaptations are controlled by the BSC using the Radio Link Control (RLC) algorithms. The BSC is also responsible for the initial Codec Mode set selection. Channel mode changes occur far less frequently, no more than a few times per minute. Channel mode changes can take place in either of the following circumstances: At call set-up and after handover During a call - dynamic channel mode adaptation (AMR handover)

Both links must use the same channel mode.



39


AMR benefits AMR provides improved speech quality in both FR and HR modes. Recent characterisation tests on the AMR codec conducted by ETSI have shown AMR Full Rate (AFS) to be superior to GSM EFR, particularly under bad interference conditions. AMR Half Rate (AHS) was equivalent or better than GSM FR under good radio channel conditions down to 16dB CIR. AMR parameters The AMR has additional parameters, which are optional at the originating side but mandatory for the terminating side: Active Codec Set (ACS). Eight bits In FR AMR and HR AMR up to four modes may be selected by setting the corresponding bits to 1 In HR AMR only four out of the lower six modes can be selected In UMTS AMR all eight modes may be selected If the ACS is not specified at the originating side, all modes are supported there. The terminating side may then select freely If ACS is not provided, SCS and MACS (see below) also cannot be provided

Supported Codec Set (SCS). Eight bits In FR AMR and UMTS AMR up to eight modes may be selected by setting the corresponding bits to 1 In HR AMR only the lower six modes may be selected If the SCS is not specified at the originating side, all modes are supported there. The terminating side may then select freely If SCS is not provided, MACS (see below) also cannot be provided

Maximal number of Codec Modes (MACS). Three bits In FR AMR and HR AMR, one to four Codec Modes are allowed within the ACS In UMTS AMR, one to eight Codec Modes are allowed within the ACS If MACS is not specified at the originating side, then the maximum number of modes is supported there. The terminating side may then select freely

Initial Codec Mode. Three bits. One of the Codec Modes within the ACS is indicated as the starting mode If the ICM is not specified at the originating side, the terminating side may select freely

40




-

The Length Indicator field (LI) is set to 3, 4, 5 or 6 at the originating side, depending on how many parameters are specified. The terminating side returns the selected codec with a full set of parameters. Hence LI is always set to 6 by the terminating side. If any node in the path from the originating side to the terminating side does not support the parameter set offered by the originating side, it may restrict it. If necessary the missing, optional parameter octets are inserted then

GSM AMR codec types The GSM AMR codec types comprise eight Full Rate and six Half Rate different Codec Modes -1 (for example, 12.2 4.75 kbs ). The active Codec Mode is selected from the Active Codec Set (ACS) by the network (Codec Mode Command) with assistance by the mobile station (Codec Mode Request). This Codec Mode Adaptation, also termed Rate Control, can be performed every 40 ms by going one Codec Mode up or down within the ACS. The Codec Modes in uplink and downlink at one radio leg may be different. In Tandem Free Operation both radio legs (A and B) are considered by the Distributed Rate Control algorithm for the optimal selection of the active Codec Mode in each direction (uplink A and then downlink B, and uplink B and then downlink A respectively). The worst of both radio legs determines the highest allowed Codec Mode, respectively the maximum allowed rate. Besides this Maximum Rate Control the active Codec Mode may sometimes be frozen to a fixed mode by either of the two radio legs to allow a smooth handover procedure (Exact Rate Control). All rate control commands are transmitted in-band: on the radio interface, the BTS-TRAU interface and the TRAU-TRAU interface. The Active Codec Set is configured at call setup or reconfigured during the call. It consists of from one to four Codec Modes (MACS) at a given time, selected from the Supported Codec Set. The maximum number of MACS and the Supported Codec Set may be constrained by the network to consider resources and radio conditions. The same Active Codec Sets are available in both the uplink and downlink. Different sets may be selected in the uplink and downlink. At start up of Tandem Free Operation, both Active Codec Sets are taken into account to determine the common Active Codec Set. In a later phase the Supported Codec Sets and MACSs of both radio legs may be taken into account to find the optimum Common Active Codec Set. All configuration data and update protocols are transmitted in-band. DTX scheme The DTX scheme of the AMR Codec Type marks the end of a speech burst with a specific SID_FIRST frame. SID_FIRST does not contain comfort noise parameters. It starts the comfort noise generation with parameters calculated at receiver side from the most recently received seven speech frames. Therefore, a DTX hangover period needs to be applied at transmitter side before sending the SID_FIRST.



41


Absolutely coded SID_UPDATE frames follow approximately every eighth frame (160 ms) in speech pauses. SID_UPDATE frames are sent independently of the cells TDMA frame structure and are related only to the source signal. Typically, in the uplink direction, an ONSET frame precedes a new speech burst. DTX on or off is defined by the network on a cell basis. The defined Tandem Free Operation allows the reception of GSM-AMR DTX information for the downlink direction in all cases. Note: The DTX scheme of the Enhanced Full Rate codec type is not compatible with the DTX -1 scheme of the AMR codec type in Codec Mode 12.2kbs , although the speech modes of the two codec types are bit identical. UMTS AMR codec types The UMTS AMR codec type comprises eight different Codec Modes (for example, 12.2 -1 4.75kbs ). The active Codec Mode is selected from the Active Codec Set by the network. This is known as Codec Mode adaptation. Codec Mode adaptation can be performed every 20 ms by going to any arbitrary Codec Mode within the ACS. The Codec Modes in uplink and downlink at one radio leg may be different. In TFO or TrFO operation, both radio legs (A and B) are considered for the optimal selection of the active Codec Mode in each direction (uplink A and then downlink B) by a Distributed Rate Control algorithm. The worst value of the two radio legs determines the highest allowed Codec Mode (maximum allowed rate). This selection process is known as maximum rate control. The active Codec Mode may sometimes be frozen to a fixed mode by any of the two radio legs to allow a smooth handover procedure (exact rate control). All rate control commands are transmitted in-band on the IU and A interface and out-of-band on the radio interface. UMTS AMR Active Codec Set functionality is the same as for GSM AMR codec types. At call setup the originating side sends the AMR parameter set in the Codec List. The terminating side then selects a suitable Active Codec Set from the given information and sends it back. If the terminating side does not support TrFO, a transcoder is allocated at a suitable position in the path, as close as possible to the terminating side. This transcoder can use inband signaling to install TFO after call setup. Then, at TFO start up, both Active Codec Sets are taken into account to determine the common Active Codec Set. In a later phase the Supported Codec Sets and MACSs of both radio legs may be taken into account to find the optimum Common Active Codec Set. All configuration data and update protocols are transmitted in-band on the TFO interface, but (possibly) out of band within the UMTS network. For more information on TFO refer to GSM Recommendation 08.62 & Technical Specification 28.062.

42




Source Controlled Rate scheme The Source Controlled Rate (SCR) scheme of the default AMR codec type marks the end of a speech burst with a specific SID_FIRST frame. SID_FIRST does not contain Comfort Noise parameters. It starts comfort noise generation with parameters calculated at receiver side from the latest received seven speech frames. A DTX hangover period must therefore be applied at transmitter side before sending the SID_FIRST. Absolutely coded SID_UPDATE frames follow approximately every eighth frame (160 ms) in speech pauses. SID_UPDATE frames are sent independently of the cells timing structure and are related only to the source signal. Typically, an ONSET frame does not exist in UMTS networks but may be received in TFO from the distant partner. It marks the beginning of a speech burst. SCR on or off is defined by the network on a cell or call basis. TFO and TrFO always allow the reception of AMR SCR information for the downlink. The SCR scheme of UMTS AMR codec types is fully compatible with the DTX scheme of GSM FR AMR and HR AMR. For compatibility with other systems, the UMTS AMR Codec application may optionally support various other DTX and rate control schemes (such as GSM-EFR, TDMA-EFR, TDMA-US1, PDC-EFR). It is not currently possible to establish TFO between UMTS and these other systems, but this is likely to be supported in a later release. Further reference Details of these codec types and their related procedures such as DTX and rate control are described in the respective standards documents.



43


2.8.

Codec speech quality

This section summarises the perceived speech quality of the various speech codecs.

Fixed rate codecs Full rate (13kbs-1) Regular Pulse Excited Long Term Prediction (RPE-LTP) Adopted 1989, with a maximum MOS of 3.7-1

Enhanced full rate (13kbs ) Advanced Code Linear Predictive (A-CELP) Adopted 1995, with a maximum MOS of 4.0-1

Half rate (5.6kbs ) Vector Sum Excited Linear Predictive (V-SELP) Adopted 1994, with a maximum MOS of 3.5

-

Figure 11 GSM Fixed Rate Codecs MOS against CIR

44




Advantages Enhanced full rate vocoders provide subscribers with improved perceived speech quality, close to that of a land-line link. Enhanced full rate can increase the perceived speech quality MOS from 3.7 to 4.0, (compared to 4.3 for land-line) Half rate vocoders provide increased network capacity (for example, an increase of up to 40%, when 60% of mobiles are capable of half rate operation) New base stations are not required Additional back-haul link capacity is not required

Disadvantages Subscribers may not consider half rate speech quality as acceptable, particularly in mobile to mobile calls Additional network capacity provided by the use of Half-Rate Codecs depends on the proportion of subscribers with half rate mobiles Enhanced full rate improved speech quality is only available to subscribers with new, enhanced full rate mobiles Enhanced full rate and half rate speech coding will be superseded by Adaptive Multi Rate speech coding from 2001 (this will require new mobiles and network infrastructure upgrades)



45


Adaptive Multi Rate codecsAdaptive Multi Rate devices provide the benefits of high speech quality from the EFR Codec combined with the bandwidth efficiency of using HR channels. The following graph shows measurements taken during ETSI characterisation tests of the AMR codec for AMR clean speech in AFS and AHS modes, compared against GSM EFR, FR and HR.

5

W irelin e Q u ality R eg io n IT U -T G .7 11 P C M - 64 & A n alo g u e L o cal L o o p

4 M ean O pin ion S c or e

3AF S EFR AH S FR HR

2

1 C IR (d B )No Err ors 19 16 13 10 7 4

Figure 12 Speech quality (MOS) against CIR for GSM speech codecs Advantages Greater robustness to channel errors in FR mode under poor radio channel conditions Increased network capacity through codec operation in AMR Half Rate mode and allowing a lower CIR compared to existing GSM codecs Significant capacity gains (up to 90%) can be achieved through AMR mode Extra capacity is freed up for expected increase in data traffic Extended coverage deeper into buildings AMR operation can be tailored to meet different operator needs

46




AMR can be used in the following modes: Full Rate mode for maximum robustness to channel errors Half Rate mode to maximise network capacity Mixed Full/Half Rate mode. This allows operators to balance the trade-off between speech quality and network capacity in accordance with prevailing radio channel and traffic load conditions



47


This page is intentionally left blank

48




Mobile Handover

33. Mobile HandoverThis chapter outlines the three main types of UMTS mobile handover, with particular emphasis on inter system handover between a UTRAN and a GSM network.

3.1.

Handover types

Three types of mobile handover are supported: Soft / softer handover A soft handover takes place when the mobile establishes a connection with a new cell, while it is still in communication with its current serving cell. This can only take place if both cells use the same RF channel (frequency). A softer handover is a special type of soft handover that takes place if both the cells involved are controlled from the same BTS (or Node B) network element. The main difference is in the implementation within the UTRAN. Uplink maximal ratio combining instead of selection combining can be used for the softer handover, as the former is performed at the BTS, while the latter takes place at the Radio Network Controller (RNC) network element. UTRAN to UTRAN hard handover A UTRAN to UTRAN hard handover may take place under the following conditions: Handover between cells operating on different RF channels (frequencies). For example when cells are allocated different numbers of RF channels, belong to different networks, or to different cell layers



49


Handover between cells operating in different duplex modes (FDD or TDD). For example, when entering a building with a local high speed TDD network Handover between cells when operating in packet mode, when soft handover is not needed Handover between different radio transceivers at the same cell, operating on different RF channels

UTRAN to GSM handover A UTRAN to GSM handover occurs when a mobile is handed over between a UTRAN and a GSM traffic channel, where an isolated area of UTRAN coverage exists within a GSM network. This type of handover is to be expected in the early stages of UTRAN deployment, when UTRAN coverage may not be contiguous over network operators areas. Additionally, to promote competition some countries are encouraging the issue of UTRAN licences to companies that do not operate an existing GSM network. Under these circumstances, existing GSM licenses have been amended to require support for roaming by UTRAN subscribers

To support inter frequency hard handover, (for example, between UTRANs or between a UTRAN and a GSM network), the mobile must conduct a cell search on a different frequency from that used for the serving cell, without interrupting the data flow associated with the call in progress. This can be achieved by either of the following methods: Using a dual receiver Slotted downlink transmission

50




3.2.

Cell sets

The mobile maintains three cell sets: Active Set This lists all UTRAN cells that are currently assigning a downlink Dedicated Physical Channel (DPCH) to the mobile. It includes any cell with which the mobile is in soft or softer handover. Handover Monitoring Set This lists all the cells (UTRAN and GSM) that the mobile has been tasked to monitor. The list corresponds to the BCCH Allocation (BA) list sent on the Slow Associated Control Channel (SACCH) in the GSM network. A separate set can be defined to monitor for cell selection, corresponding to the BCCH Allocation (BA) list sent on the Broadcast Control channel (BCCH) in the GSM network. Handover Target Set This lists the cells received by the mobile at a level sufficient for the associated DPCHs to be decoded, but which are not currently in the Active Set. These cells may be on different frequencies from the current serving cell, and part of a UTRAN or GSM network.



51


3.3.

Preparation for UTRAN to UTRAN handover

The Handover Monitoring Set is sent to the mobile via the BCH of the serving cell or via specific signalling on the Dedicated Channel (DCH). The Handover Monitoring Set contains the following data for each monitored cell: Downlink scrambling code Cell ID number

The network knows the mapping of cell scrambling codes to synchronisation codes (groups indicated by the secondary synchronisation channel). During the neighbour cell measurement process the mobile uses the primary and secondary synchronisation channels to synchronise to the cells, together with knowledge of the scrambling codes in use. From both the serving and neighbour cells, the mobile measures the downlink: Received signal level Relative timing between serving and neighbour cells, based on the phase difference of the scrambling codes Bit Error Rate (BER) / Block Error Rate (BLER)

52




3.4.

Preparation for UTRAN to GSM handover

The UTRAN super-frame structure is 720ms in length and comprises 72 frames of 10ms each. The GSM super-frame structure is 6.12s in length and comprises 26 frames of 120ms each. As 720ms is a multiple of 120ms, it is possible to use compatible timing for inter system measurements. A dual-standard (UTRAN/GSM) mobile, when operating in UTRAN mode, can use the GSM Frequency Correction Channel (FCH) and GSM Synchronisation Channel (SCH) data to maintain the timing between a UTRAN carrier and a GSM carrier. The effect is as if there were two asynchronous GSM carriers. A dual-standard mobile can make GSM network measurements during the idle periods in the downlink transmission that are created using downlink slotted mode. The use of slotted mode is controlled by the UTRAN, which informs the mobile which frame is slotted. Slotted frames are achieved by compressing the data that would normally be sent in a 10ms frame, by either code puncturing or reducing the spreading factor, so that it can be sent in only 5ms. This creates a 5ms silent period (the Silence Duration) during which the mobile can make GSM

121334268 transition from gsm to umts

Documents

GSM to UMTS

Mobile EvolutionsFrom GSM to UMTS

UMTS/GSM ANTENNA - Sectron

Mobile Communications Systems GSM/UMTS

Migracion de Gsm a Umts

GSM and UMTS - University of Hong Kongkcleung/courses/csis7304/Summer.2006/... · GSM and UMTS Market GSM Overview Services Sub-systems Components UMTS/IMT-2000 ... Architecture of

Transition from GSM to UMTS

Kpi for umts and gsm

Atoll Gsm Umts Lte Acp

Presentaz. gsm umts

Overview gsm,gprs& umts

Novero GSM/UMTS MMC бустер

LTE UMTS GSM Network Optimization

GSM and UMTS RF Colocation

Gsm gprs-umts

Gsm Umts (Ojo)

GSM AND UMTS MOBILITY SIMULATOR

Instalación GSM/UMTS Estación Base

Antenna GSM/UMTS

GSM, UMTS Migration to LTE

GSM to UMTS Transition

GSM/UMTS ANTENNA - Sectron

UMTS Frente a GSM

gsm & umts

PowerPoint Presentationsouthafrica.ni.com/sites/default/files/1 Keynote - Nancy.pdf · UMTS 2100 UMTS 850 UMTS 900 UMTS 1900 GPS GSM 850 GSM 900 GSM 1800 GSM 1900 Bluetooth 4.1 (LE)

GSM-UMTS-OUTAN - Anleitung zur Installation der GSM‑UMTS ... · UMTS-Antenne-IXX103410|98-4043010|Version1.0 DE GSM-UMTS-OUTAN Anleitung zur Installation der GSM‑UMTS‑OUT Antenne

Gsm Umts Lte

3gpp Kpi Umts Gsm

GSM, GPRS, EDGE & UMTS

Gsm and Umts Networks

Alttc Gsm Umts Migration

Gsm Umts Lte Inter-worki

GSM, GPRS and UMTS Overview

Nastar V600R008 (GSM&UMTS)

Gsm Gprs Umts