_____________________________________________________________________

Title:
Submission of Proposed Radio Transmission Technologies
Source:
SMG2
_____________________________________________________________________
Attachment 2 of Circular Letter 8/LCCE/47 contains the Cover Sheet for Submission of
Proposed Radio Transmission Technologies which has to be completed and submitted by
proponents together with all the relevant material on the proposed RTT. This would enable ITU
to maintain an updated catalogue of all submitted RTTs.

Cover Sheet for Submission of Proposed Radio Transmission Technologies

The information listed below will be used for cataloguing radio transmission technologies for IMT-2000 by the ITU
and will be posted electronically.
This cover sheet (and additional information, if applicable) should be attached when an evaluation group submits a
proposal on radio transmission technologies for IMT-2000.
1. Proponent
a) Name of proponent: __________________ ETSI/SMG/SMG2 _________________________
b) Proponent category:
ITU-R membership: Yes __ No ___
Regional/National standards body: Yes __ (Name:__ ETSI ________) No ___
Industry group: Yes ___ (Name:________________) No __
Other: (Name:________________) No __
c) Contact point
Name:
Niels P. S. Andersen
Organization:
Tele Danmark A/S NMR
Address:
Spotorno Alle 12
DK-2630 Taastrup
Denmark
Tel:
+45 43 586378
Fax:
+45 43 710382
Email:
npa@tdk.dk
2. Proposal identification
a) Name of the proposed RTTs (list all the names) (if the proponent submits multiple proposals): _ UTRA (UMTS
Terrestrial Radio Access) _
b) Status of proposal:
Revision ___ (former proposed RTTs name:_____________)
New proposal __
3. Proposed RTT(s) service environment (check as many as appropriate)
Indoor __ Outdoor to indoor pedestrian __
Vehicular __ Satellite ___
4. Attachments
Technology template for each test environment __
Requirements and objectives template __
IPR statement __ (outline of the current situation)
Other (any additional inputs which the proponent may consider relevant to the evaluation) __
5. Has the proposal already been submitted to an evaluation group registered with ITU?
Yes __ (Name of evaluation group: ___ ETSI-SMG2 _______, Date of submission:_ 29/1/1998 _)
No ___
6. Other information
a) Name of person submitting form:
K. H. Rosenbrock (ETSI Director-General)
1. Date:

The ETSI UMTS Terrestrial Radio Access (UTRA) ITU-R RTT

Candidate Submission
1.

INTRODUCTION

This document contains the ETSI UMTS terrestrial radio access (UTRA) RTT candidate submission. The UTRA
network is currently being developed in ETSI SMG2 and this document reflects the status as of May/June 1998.
Thus, any modification to this RTT can be made as a result of that process.
The document is divided into a main part containing a description of the radio access system and two Annexes.
Annex A contains the answers to the RTT template. Annex B provides the answers to the fulfilment of requirements
template while Annex C shows the capacity and coverage analysis for the evaluated test cases.
The main part of this document has a definition and abbreviation section as shown in Section 2. Section 3 describes
the general architecture of the radio access network. Section 4 defines the Layer 2 and 3 of the radio protocol, i.e.
from the radio resource management sub-layer to the MAC sub-layer as defined in ITU-R recommendation M.1035.
Finally the Physical layer is described in Sections 5 and 6 for the FDD mode and TDD mode respectively.
Interoperability is discussed in Section 7.

2.
2.1

DEFINITIONS, ABBREVIATIONS AND SYMBOLS

Definitions

Active Set
Set of radio links simultaneously involved in a specific communication service between an MS and a UTRAN.
Cell
Geographical area served from one UTRAN Access Point. A cell is defined by a cell identity broadcast from the
UTRAN Access Point.
Coded Composite Transport Channel (CCTrCH)
A data stream resulting from encoding and multiplexing of one or several transport channels.
Iu
The interconnection point (interface) between the RNS and the Core Network. It is also considered as a reference
point.
Iub
Interface between the RNC and the Node B.
Iur
Interface between two RNSs.
Logical Channel
A logical channel is a radio bearer, or part of it, dedicated for exclusive use of a specific communication process.
Different types of logical channel are defined according to the type of information transferred on the radio
interface.
Node B
A logical node responsible for radio transmission / reception in one or more cells to/from the UE. Terminates the Iub
interface towards the RNC.
Physical Channel
In FDD mode, a physical channel is defined by code, frequency and, in the uplink, relative phase (I/Q). In TDD
mode, code, frequency, and time-slot define a physical channel.
Physical channel data stream
In the uplink, a data stream that is transmitted on one physical channel.
In the downlink, a data stream that is transmitted on one physical channel in each cell of the active set.

Radio access bearer

The service that the access stratum provides to the non-access stratum for transfer of user data between MS and CN.
Radio Access Network Application Part
Radio Network Signalling over the Iu.
Radio Network Subsystem Application Part
Radio Network Signalling over the Iur.
Radio frame
A radio frame is a numbered time interval of 10 ms duration used for data transmission on the radio physical
channel. A radio frame is divided into 16 slots of 0.625 ms duration. The unit of data that is mapped to a radio frame
(10 ms time interval) may also be referred to as radio frame.
Radio link
A set of (radio) physical channels that link an MS to a UTRAN access point.
Radio link addition
A [soft handover] procedure whereby a branch through a new [sector of a cell] is added in case some of the already
existing branches were using [sectors] of the same cell.
Radio link removal
A [soft handover] procedure whereby a branch through a new [sector of a cell] is removed in case some of the
remaining existing branches use [sectors of] that cell.
Radio Network Controller
This equipment in the RNS is in charge of controlling the use and the integrity of the radio resources.
Radio Network Subsystem
Either a full network or only the access part of a UMTS network offering the allocation and the release of specific
radio resources to establish means of connection in between an UE and the UTRAN.
A Radio Network Subsystem is responsible for the resources and transmission/reception in a set of cells.
Serving RNS
A role an RNS can take with respect to a specific connection between an UE and UTRAN. There is one Serving
RNS for each UE that has a connection to UTRAN. The Serving RNS is in charge of the radio connection between a
UE and the UTRAN. The Serving RNS terminates the Iu for this UE.
Drift RNS
The role an RNS can take with respect to a specific connection between an UE and UTRAN. An RNS that supports
the Serving RNS with radio resources when the connection between the UTRAN and the UE need to use cell(s)
controlled by this RNS is referred to as Drift RNS
RRC connection
A point-to-point bi-directional connection between RRC peer entities on the UE and the UTRAN sides, respectively.
An UE has either zero or one RRC connection.
Signalling connection
An assured-mode link between the user equipment and the core network to transfer higher layer information between
peer entities in the non-access stratum.
Signalling link
Provides an assured-mode link layer to transfer the MS-UTRAN signalling messages as well as MS - Core Network
signalling messages (using the signalling connection).
Transport channel
The channels that are offered by the physical layer to Layer 2 for data transport between peer L1 entities are denoted
as Transport Channels.
Different types of transport channels are defined by how and with which characteristics data is transferred on the
physical layer, e.g. whether using dedicated or common physical channels are employed.
Transport Format
A combination of encoding, interleaving, bit rate and mapping onto physical channels.
Transport Format Indicator (TFI)
A label for a specific Transport Format within a Transport Format Set.

Transport Format Set

A set of Transport Formats. For example, a variable rate DCH has a Transport Format Set (one Transport Format for
each rate), whereas a fixed rate DCH has a single Transport Format.
UTRAN access point
The UTRAN-side end point of a radio link. A UTRAN access point is a cell.
User Equipment
A Mobile Equipment with one or several UMTS Subscriber Identity Module(s).

2.2

Abbreviations

For the purposes of this specification the following abbreviations apply.

ARQ
AAL
ATM
BCCH
BER
BLER
BS
BSS
BPSK
CA
CAA
CBR
CCC
CCCH
CCPCH
CCTrCH
CD
CDA
CDMA
CN
CTDMA
CRC
DCA
DCH
DCCH
DC-SAP
DL
DPCH
DPCCH
DPDCH
DRNS
DRX
DTX
DS-CDMA
FACH
FDD
FDMA
FEC
FER
HCS
HO
GMSK
GSM
ITU
JD
kbps

Automatic Repeat Request

Application Adaptation Layer
Asynchronous Transfer Mode
Broadcast Control Channel
Bit Error Ratio
Block Error Ratio
Base Station
Base Station System
Binary Phase Shift Keying
Capacity Allocation
Capacity Allocation Acknowledgement
Constant Bit Rate
ControlCall Control
Common Control Channel
Common Control Physical Channel
Coded Composite Transport Channel
Capacity Deallocation
Capacity Deallocation Acknowledgement
Code Division Multiple Access
Core Network
Code Time Division Multiple Access
Cyclic Redundancy Check
Dynamic Channel Allocation
Dedicated Channel
Dedicated Control Channel
Dedicated Connection Service Access Point
Downlink
Dedicated Physical Channel
Dedicated Physical Control Channel
Dedicated Physical Data Channel
Drift RNS
Discontinuous Reception
Discontinuous Transmission
Direct-Sequence Code Division Multiple Access
Forward Access Channel
Frequency Division Duplex
Frequency Division Multiple Access
Forward Error Correction
Frame Error Ratio
Hierarchical Cellular Structures
Handover
Gaussian Minimum Shift Keying
Global System for Mobile Communication
International Telecommunication Union
Joint Detection
kilo-bits per second

L1
L2
L3
LAC
LLC
MA
MAC
MAHO
Mcps
ME
MM
MO
MOHO
MS
MT
NRT
ODMA
OVSF
PC
PCH
PDU
PHY
PhyCH
QoS
QPSK
PG
PRACH
PUF
RACH
RANAP
RF
RLC
RLCP
RNC
RNS
RNSAP
RR
RRC
RRM
RT
RU
RX
SAP
SCH
SDCCH
SDU
SF
SIR
SMS
SP
SRNS
TCH
TDD
TDMA
TFI
TPC
TX
UUE
UL
UMTS

Layer 1 (physical layer)

Layer 2 (data link layer)
Layer 3 (network layer)
Link Access Control
Logical Link Layer
Multiple Access
Medium Access Control
Mobile Assisted Handover
Mega Chip Per Second
Mobile Equipment
Mobility Management
Mobile Originated
Mobile Originated Handover
Mobile Station
Mobile Terminated
Non-Real Time
Opportunity Driven Multiple Access
Orthogonal Variable Spreading Factor (codes)
Power Control
Paging Channel
Protocol Data Unit
Physical layer
Physical Channel
Quality of Service
Quaternary Phase Shift Keying
Processing Gain
Physical Random Access Channel
Power Up Function
Random Access Channel
Radio Access Network Application Part
Radio Frequency
Radio Link Control
Radio Link Control Protocol
Radio Network Controller
Radio Network Subsystem
Radio Network Subsystem Application Part
Radio Resource
Radio Resource Control
Radio Resource Management
Real Time
Resource Unit
Receive
Service Access Point
Synchronisation Channel
Stand-alone Dedicated Control Channel
Service Data Unit
Spreading Factor
Signal-to-Interference Ratio
Short message Service
Switching Point
Serving RNS
Traffic Channel
Time Division Duplex
Time Division Multiple Access
Transport Format Indicator
Transmit Power Control
Transmit
UserUser Equipment
Uplink
Universal Mobile Telecommunications System

USIM
UTRA
UTRAN
VA
VBR

3.
3.1

UMTS Subscriber Identity Module

UMTS Terrestrial Radio Access
UMTS Terrestrial Radio Access Network
Voice Activity
Variable Bit Rate

THE RADIO ACCESS NETWORK ARCHITECTURE

General Architecture

Figure 1 shows the assumed UMTS architecture as outlined in ETSI/SMG. The focus in this section is on the radio
interface of the access stratum. This figure shows a that there will be an access stratum part containing basically all
the radio specific parts providing certain services to the non-access stratum through service access points (SAP).

Non-Access Stratum

Access Stratum
MS

Radio
(Uu)

UTRAN

Iu

Core Network

Figure 1. Assumed UMTS Architecture

Figure 2 shows a simplified UMTS architecture with the external reference points and interfaces to the UTRAN.
(The terminal can be named both as Mobile Station (MS), User Equipment (UE) or Mobile Equipment (ME).)

CN
Iu
UTRAN
Uu
UE

U TR AN
CN
UE

U M T S T errestrial R ad io A ccess N etw ork

C ore N etw ork
U ser E q u ip em et

Figure 2. UMTS Architecture

3.2

Basic Principles

Some basic principles agreed are:

Logical separation of signalling and data transport networks
Macro diversity is fully handled in the UTRAN
UTRAN and CN functions are fully separated from transport function, i.e. the fact that some UTRAN or CN
function resides in the same equipment, as some transport functions does not make the transport functions part of
the UTRAN or the CN.

3.2.1

Mobility Handling

It is generally agreed to contain radio access specific procedures within UTRAN. This means that all cell level
mobility should be handled within UTRAN. Also the cell structure of the radio network should not necessarily be
known outside the UTRAN.
When a dedicated connection exists to the UE, the UTRAN shall handle the radio interface mobility of the UE. This
includes procedures such as soft handover.
When a dedicated connection does not exist to the UE, no UE information in UTRAN is needed. In this case, the
mobility is handled directly between UE and CN outside access stratum (e.g. by means of registration procedures).
When paging the UE, the CN indicates a 'geographical area' that is translated within UTRAN to the actual cells that
shall be paged. A 'geographical area' shall be identified in a cell-structure independent way. One possibility is the use
of 'Location Area identities'.
During the lifetime of the dedicated connection, the registrations to the CN are suppressed by the UE. When a
dedicated connection is released, the UE performs a new registration to the CN, if needed.
Thus the UTRAN does not contain any permanent 'location registers' for the UE, but only temporary contexts for the
duration of the dedicated connection. This context may typically contain location information (e.g. current cell(s) of
the UE) and information about allocated radio resources and related connection references.

3.3
3.3.1

UTRAN logical architecture

UTRAN Architecture

The UTRAN consists of a set of Radio Network Subsystems connected to the Core Network through the Iu and
interconnected together through the Iur as shown in Figure 3.

C o re N etw o rk

Iu

Iu

RNS

RNS
Iu r

C e lls

Figure 3. UTRAN Architecture

Each RNS is responsible for the resources of its set of cells.
For each connection between User Equipment and the UTRAN, one RNS is the Serving RNS. When required, Drift
RNSs support the Serving RNS by providing radio resources as shown in Figure 4. The role of an RNS (Serving or
Drift) is on a per connection basis between a UE and the UTRAN.

C o r e N e tw o r k

Iu
DRNS

SRN S
Iu r

C e lls

UE

Figure 4. Serving and Drift RNS

3.3.2

RNS Architecture

A RNS consists of a Radio Network Controller and one or more abstract entities currently called Node B as shown in
Figure 5. Node B is connected to the RNC through the Iub interface.

RNC

Iub

Node B

Iub

Node B

Cells
Figure 5. RNS Architecture
The RNC is responsible for the Handover decisions that require signalling to the UE.
The RNC comprises a combining/splitting function to support macro diversity between different Node B.
The functions and internal structure of Node B is for further studies.
However, a Node B can comprise an optional combining/splitting function to support macro diversity inside a Node
B.

3.4
3.4.1

Function descriptions
List of functions

Functions related to overall system access control

System information broadcasting
Functions related to radio channel ciphering
Radio channel ciphering
Radio channel deciphering
Functions related to handover
Radio environment survey
Handover decision
Macro-diversity control
Handover Control
Handover execution
Handover completion
SRNS Relocation
Inter-System handover
Functions related to radio resource management and control
Radio bearer connection set-up and release (Radio Bearer Control)
Reservation and release of physical radio channels
Allocation and de-allocation of physical radio channels
Packet data transfer over radio function
RF power control
RF power setting
Radio channel coding
Radio channel decoding
Channel coding control
Initial (random) access detection and handling

3.4.2

Functions description

3.4.2.1
Functions related to overall system access control
System access is the means by which a UMTS user is connected to the UMTS in order to use UMTS services and/or
facilities. User system access may be initiated from either the mobile side, e.g. a mobile originated call, or the
network side, e.g. a mobile terminated call.
3.4.2.1.1

System information broadcasting

This function provides the mobile station with the information that is needed to camp on a cell and to set up a
connection in idle mode and to perform handover or route packets in communication mode. The tasks may include:
access rights
frequency bands used
configuration of transport channels, PCH, FACH and RACH channel structure of the cell etc
network and cell identities
information for location registration purposes
UE idle mode cell selection and cell re-selection criteria
UE transmission power control information
UE access and admission control information
Because of its close relation to the basic radio transmission and the radio channel structure, the basic control and
synchronisation of this function should be located in UTRAN.
3.4.2.2
3.4.2.2.1

Functions related to radio channel ciphering

Radio channel ciphering

This function is a pure computation function whereby the radio transmitted data can be protected against an nonauthorised third party. Ciphering may be based on the usage of a session-dependent key, derived through signalling
and/or session dependent information.
This function is located in the UE and in the UTRAN.
3.4.2.2.2

Radio channel deciphering

This function is a pure computation function that is used to restore the original information from the ciphered
information. The deciphering function is the complement function of the ciphering function, based on the same
ciphering key.
This function is located in the UE and in the UTRAN.
3.4.2.3
3.4.2.3.1

Functions related to handover

Radio environment survey

This function performs measurements on radio channels (current and surrounding cells) and translates these
measurements into radio channel quality estimates. Measurements may include:
1. received signal strengths (current and surrounding cells),
2. estimated bit error ratios, (current and surrounding cells),
3. estimation of propagation environments (e.g. high-speed, low-speed, satellite, etc.),
4. transmission range (e.g. through timing information),
5. Doppler shift,
6. synchronisation status,
7. Received interference level.
In order for these measurements and the subsequent analysis to be meaningful, some association between the
measurements and the channels to which they relate should be made in the analysis. Such association may include
the use of identifiers for the network, the base station, the cell (base station sector) and/or the radio channel.
This function is located in the UE and in the UTRAN.
3.4.2.3.2

Handover decision

This function consists of gathering estimates of the quality of the radio channels (including estimates from
surrounding cells) from the measuring entities and to assess the overall quality of service of the call. The overall
quality of service is compared with requested limits and with estimates from surrounding cells. Depending on the
outcome of this comparison, the macro-diversity control function or the handover control function may be activated.
This function may also include functionality to assess traffic loading distribution among radio cells and to decide on
handing over traffic between cells for traffic reasons.

The location of this function is depending on the handover principle chosen.

if network only initiated handover, this function is located in the UTRAN;
if mobile only initiated handover, this function is located in the UE;
if both the mobile and the network can initiate handover, this function will be located in both the UTRAN and
the UE.
3.4.2.3.3

Macro-diversity control

Upon request of the Handover Decision function, this function control the duplication/ replication of information
streams to receive/ transmit the same information through multiple physical channels (possibly in different cells)
from/ towards a single mobile terminal.
This function also controls the combining of information streams generated by a single source (diversity link), but
conveyed via several parallel physical channels (diversity sub-links). Macro diversity control should interact with
channel coding control in order to reduce the bit error ratio when combining the different information streams. This
function controls macro-diversity execution which is located at the two endpoints of the connection element on
which macro-diversity is applied (diversity link), that is at the access point and also at the mobile termination.
In some cases, depending on physical network configuration, there may be several entities which combine the
different information streams, e.g. one entity combines information streams on radio signal basis, another combines
information streams on wire-line signal basis.
This function is typically located in the UTRAN. However, depending on the physical network architecture, some bit
stream combining function within the CN may have to be included in the control.
3.4.2.3.4

Handover Control

In the case of switched handover, this function is responsible for the overall control of the handover execution
process. It initiates the handover execution process in the entities required and receives indications regarding the
results.
Due to the close relationship with the radio access and the Handover Decision function, this function should be
located in the UTRAN.
3.4.2.3.5

Handover execution

This function is in control of the actual handing over of the communication path. It comprises two sub-processes:
handover resource reservation and handover path switching. The handover resource reservation process will
reserve and activate the new radio and wire-line resources that are required for the handover. When the new
resources are successfully reserved and activated, the handover path switching process will perform the final
switching from the old to the new resources, including any intermediate path combination required, e.g. handover
branch addition and handover branch deletion in the soft handover case.
This function is located in the UTRAN for UTRAN internal path switching and in the CN for CN path switching.
3.4.2.3.6

Handover completion

This function will free up any resources that are no longer needed. A re-routing of the call may also be triggered in
order to optimise the new connection.
This function is located both in the UTRAN and in the CN.
3.4.2.3.7

SRNS Relocation

The SRNS Relocation function co-ordinates the activities when the SRNS role is to be taken over by another RNS.
SRNS relocation implies that the Iu interface connection point is moved to the new RNS.
This function is located in the UTRAN and the CN.
3.4.2.3.8

Inter-System handover

The Inter-system handover function enables handover to and from e.g. GSM BSS.
This function is located in the UTRAN, the UE and the CN.
3.4.2.4
Functions related to radio resource management and control
Radio resource management is concerned with the allocation and maintenance of radio communication resources.
UMTS radio resources must be shared between circuit mode (voice and data) services and other modes of service
(e.g. packet data transfer mode and connectionless services).
3.4.2.4.1

Radio bearer connection set-up and release (Radio Bearer Control)

This function is responsible for the control of connection element set-up and release in the radio access sub network.
The purpose of this function is
1. to participate in the processing of the end-to-end connection set-up and release,
2. and to manage and maintain the element of the end-to-end connection, which is located in the radio access
sub network.

10

In the former case, this function will be activated by request from other functional entities at call set-up/release. In
the latter case, i.e. when the end-to-end connection has already been established, this function may also be invoked to
cater for in-call service modification or at handover execution. This function interacts with the reservation and
release of physical (radio) channels function.
This function is located both in the UE and in the UTRAN.
3.4.2.4.2

Reservation and release of physical radio channels

This function consists of translating the connection element set-up or release requests into physical radio channel
requests, reserving or releasing the corresponding physical radio channels and acknowledging this reservation/
release to the requesting entity.
This function may also perform physical channel reservation and release in the case of a handover. Moreover, the
amount of radio resource required may change during a call, due to service requests from the user or macro-diversity
requests. Therefore, this function must also be capable of dynamically assigning physical channels during a call.
Note: This function may or may not be identical to the function reservation and release of physical radio channels.
The distinction between the two functions is required e.g. to take into account sharing a physical radio
channel by multiple users in a packet data transfer mode.
This function is located in the UTRAN.
3.4.2.4.3

Allocation and de-allocation of physical radio channels

This function is responsible, once physical radio channels have been reserved, for actual physical radio channel
usage, allocating or de-allocating the corresponding physical radio channels for data transfer. Acknowledging this
allocation/de-allocation to the requesting entity is for further study.
Note: This function may or may not be identical to the function reservation and release of physical radio channels.
The distinction between the two functions is required e.g. to take into account sharing a physical radio
channel by multiple users in a packet data transfer mode.
This function is located in the UTRAN.
3.4.2.4.4

Packet data transfer over radio function

This function provides packet data transfer capability across the UMTS radio interface. This function includes
procedures which:
1. provide packet access control over radio channels,
2. provide packet multiplexing over common physical radio channels,
3. provide packet discrimination within the mobile terminal,
4. provide error detection and correction,
5. provide flow control procedures.
This function is located in both the UE and in the UTRAN.
3.4.2.4.5

RF power control

In order to minimise the level of interference (and thereby maximise the re-use of radio spectrum), it is important
that the radio transmission power is not higher than what is required for the requested service quality. Based on
assessments of radio channel quality, this function controls the level of the transmitted power from the mobile station
as well as the base station.
This function is located in both the UE and in the UTRAN.
3.4.2.4.6

RF power setting

This function adjusts the output power of a radio transmitter according to control information from the RF power
control function. The function forms an inherent part of any power control scheme, whether closed or open loop.
This function is located in both the UE and in the UTRAN.
3.4.2.4.7

Radio channel coding

This function introduces redundancy into the source data flow, increasing its rate by adding information calculated
from the source data, in order to allow the detection or correction of signal errors introduced by the transmission
medium. The channel coding algorithm(s) used and the amount of redundancy introduced may be different for the
different types of transport channels and different types of data.
This function is located in both the UE and in the UTRAN.

11

3.4.2.4.8

Radio channel decoding

This function tries to reconstruct the source information using the redundancy added by the channel coding function
to detect or correct possible errors in the received data flow. The channel decoding function may also employ a
priori error likelihood information generated by the demodulation function to increase the efficiency of the decoding
operation. The channel decoding function is the complement function to the channel coding function.
This function is located in both the UE and in the UTRAN.
3.4.2.4.9

Channel coding control

This function generates control information required by the channel coding/ decoding execution functions. This may
include channel coding scheme, code rate, etc.
This function is located in both the UE and in the UTRAN.
3.4.2.4.10

Initial (random) access detection and handling

This function will have the ability to detect an initial access attempt from a mobile station and will respond
appropriately. The handling of the initial access may include procedures for a possible resolution of colliding
attempts, etc. The successful result will be the request for allocation of appropriate resources for the requesting
mobile station.
This function is located in the UTRAN.

3.5

Description of UTRAN interfaces

3.5.1

Iu interface, assumptions

3.5.1.1

Streamlining functions

3.5.1.1.1

Access Network Triggered Streamlining

One Access Network triggered function needed over the Iu interface is the function for SRNS Relocation. SRNS
Relocation needs support from the Core Network to be executed.
C o r e N e tw o r k

C o r e N e tw o r k

Iu
DRNS

Iu

SRN S

SRN S

RNS

Iu r

C e lls

UE

UE

B e f o r e S R N S R e lo c a tio n

A f te r S R N S R e lo c a tio n

Figure 6. Serving RNS Relocation

3.5.2

Iu interface protocol

The Radio Network signalling over Iu consists of the Radio Access Network Application Part (RANAP). The
RANAP consists of mechanisms to handle all procedures between the CN and UTRAN. It is also capable of
conveying messages transparently between the CN and the UE without interpretation or processing by the UTRAN.
Over the Iu interface the RANAP protocol is, e.g. used for:
Facilitate a set of general UTRAN procedures from the Core Network such as paging -notification as defined
by the general SAP.
Separate each User Equipment (UE) on the protocol level for mobile specific signalling management as
defined by the dedicated SAP.
Transfer of transparent non-access signalling as defined in the dedicated SAP.
Request of various types of UTRAN Radio Access Bearers through the dedicated SAP.
Perform the streamlining function.
The Access Stratum provides the Radio Access Bearers

12

Various transmission possibilities exist to convey the bearers over the Iu to the Core Network. It is therefore
proposed to separate the Data Transport Resource and traffic handling from the RANAP (Figure 7). This resource
and traffic handling is controlled by the Transport Signalling. A Signalling Bearer carries the Transport Signalling
over the Iu interface.
Radio
Network
layer

Iu Data
Streams

RANAP

Transport
Signalling

Transport
layer
Signalling
Bearer

Data
Transport

Figure 7. Separation of RANAP and transport over Iu

The RANAP is terminated in the SRNS.

3.5.3

Description of UTRAN internal interfaces

3.5.3.1
Iur Interface
The Iur interface connects a SRNS and a DRNS.
This interface should be open.
The information exchanged across the Iur is categorised as below:
One or more Iur Data stream which comprises
Radio frames
Simple, commonly agreed Quality estimate
Synchronisation information
Signalling
Addition of Cells in the DRNS which may lead or not to the addition of an new Iur Data stream
Removal of Cells in the DRNS
Modify Radio bearer characteristics
Note: This list of procedures is not the full list over Iur interface.
From a logical stand point, the Iur interface is a point to point interface between the SRNS and all the DRNS, i.e.
there is no deeper hierarchy of RNSs than the SRNS and DRNS. However, this point to point logical interface
should be feasible even in the absence of a physical direct connection between the two RNSs.
3.5.3.1.1

Functional split over Iur Interface

Note: This is only an initial list.

3.5.3.1.1.1
Macro-diversity Combining/Splitting
DRNS may perform macro-diversity combining/splitting of data streams communicated via its cells. SRNS performs
macro-diversity combining/splitting of Iur data streams received from/sent to DRNS(s), and data streams
communicated via its own cells.
The internal DRNS handling of the macro-diversity combining/splitting of radio frames is controlled by the DRNS.
3.5.3.1.1.2

Control of Macro-diversity Combining/Splitting Topology

When requesting the addition of a new cell for a UE-UTRAN connection, the SRNS can explicitly request to the
DRNS a new Iur data stream, in which case the macro-diversity combining and splitting function within the DRNS is
not used for that cell. Otherwise, the DRNS takes the decision whether macro-diversity combining and splitting
function is used inside the DRNS for that cell i.e. whether a new Iur data stream shall be added or not.

13

3.5.3.1.1.3

Handling of DRNS Hardware Resources

Allocation and control of DRNS hardware resources, used for Iur data streams and radio interface
transmission/reception in DRNS, is performed by DRNS.
3.5.3.1.1.4

Allocation of Downlink Channelisation Codes

Allocation of downlink channelisation codes of cells belonging to DRNS is performed in DRNS.

Note that this does not imply that the signalling of the code allocation to the UE must be done from the DRNS.
3.5.3.1.2

Iur Interface protocol

The signalling information across Iur interface as identified in section 0 is called Radio Network Subsystem
Application Part (RNSAP).
The RNSAP is terminated in the SRNS and in the DRNS.
As already stated in Section 0 a clear separation shall exist between the Radio Network Layer and the Transport
Layer. It is therefore proposed to separate the Data Transport resource and traffic handling from the RNSAP (Figure
8). This resource and traffic handling is controlled by the Transport Signalling. A Signalling Bearer carries the
Transport Signalling over the Iur interface.
Radio
Network
layer

Iur Data
Streams

RNSAP

Transport
Signalling

Transport
layer
Signalling
Bearer

Data
Transport

Figure 8. Separation of RNSAP and transport over Iur

3.5.3.2
Iub Interface
Note: This description is applicable if the Iub interface will be standardised as an open interface
The Iub interface connects a RNC and a Node B.
The information transferred over the Iub reference point can be categorised as follows:
1. Radio application related signalling
The Iub interface allows RNC and Node B to negotiate about radio resources, for example to add and delete
cells controlled by the Node B to support communication of the dedicated connection between UE and SRNS.
2. Radio frame data blocks
The Iub interface provides means for transport of uplink and downlink radio frame data blocks between RNC
and Node B. This transport can use pre-defined transmission links or switched connections.
3. Quality estimations of uplink radio frames and synchronisation data
The macro-diversity combining function of the RNC uses Node B quality estimations of the uplink radio frame
data blocks. There is also a need for accurate time synchronisation between the soft handover branches.
The information in category 3 is tightly coupled to the radio frame data blocks in category 2. Therefore, category 2
and 3 information is multiplexed on the same underlying transport mechanism (e.g. switched connection), and is
together referred to as an Iub data stream.
The Iub data stream shall follow the same specification as the Iur data stream.
Over the Iub interface between the RNC and one Node B, one or more Iub data streams are established, each
corresponding to one or more cells belonging to the Node B.

14

3.5.3.2.1

Functional split over Iub

Note: This is only an initial list.

3.5.3.2.1.1
Macro-diversity Combining of Radio Frame Data Blocks
Node B may perform macro-diversity combining/splitting of data streams communicated via its cells. RNC performs
macro-diversity combining/splitting of Iub data streams received from/sent to several Node B(s).
3.5.3.2.1.2

Control of Macro-diversity Combining/Splitting Topology

When requesting the addition of a new cell for a UE to UTRAN connection, the RNC can explicitly request to the
Node B a new Iub data stream, in which case the macro-diversity combining and splitting function within the Node B
is not used for that cell. Otherwise, the Node B takes the decision whether macro-diversity combining and splitting
function is used inside the Node B for that cell i.e. whether a new Iub data stream shall be added or not.
The Node B controls the internal Node B handling of the macro-diversity combining/splitting.
3.5.3.2.1.3

Soft Handover Decision

To support mobility of the UE to UTRAN connection between cells, UTRAN uses measurement reports from the
MS and detectors at the cells. [The mechanisms for this are FFS.]
The RNC takes the decision to add or delete cells from the connection.
3.5.3.2.1.4

Handling of Node B Hardware Resources

Mapping of Node B logical resources onto Node B hardware resources, used for Iub data streams and radio interface
transmission/reception, is performed by Node B.
3.5.3.2.1.5

Allocation of Downlink Channelisation Codes

Allocation of downlink channelisation codes of cells belonging to Node B is performed in Node B.

Note that this does not imply that the signalling of the code allocation to the UE must be done from Node B.

3.6

UTRAN internal bearers

For all open interfaces, one mandatory set of protocols must be specified. However, a clear separation between the
Radio Network functions and the Transport functions should allow this Transport layer to be exchanged to another
one with minimum impact on the Radio Network functions.

3.6.1

User Data Bearers

ATM and AAL type 2 (ITU-T recommendations I363.2 and I.366.1) is used as the standard transport layer for Soft
Handover data stream across the Iur interface.

3.6.2

Signalling Bearers

Note: These requirements are initial requirements. Other requirements may be added later on.
3.6.2.1
Signalling Bearer Requirements for Iu Interface
Over the Iu interface the RANAP protocol requires:
A connectionless transport of RANAP messages to facilitate e.g. paging.
A connection oriented transport of RANAP messages e.g. to facilitate messages belonging to a specific User
equipment (UE) during a call.
A reliable connection to make the RANAP simpler.
Support of signalling inactivity testing of a specific UE connection.
3.6.2.2
Signalling Bearer Requirements for Iur Interface
There exist at least two major types of soft handover over the Iur interface:
1. The case when a new physical transmission (Iur data stream) is set up over the Iur interface to provide an
additional cell.
2. The case when existing transmission (Iur data stream) is used over the Iur interface when an additional cell is
added in the DRNS. In this case the DRNS must be able to identify the UE in order to perform the adding of
the cell. Consequently a UE context must exist in the DRNS.
Over the Iur interface the RNSAP protocol requires:
A connection oriented transport of RNSAP messages, i.e. one signalling bearer connection for each DRNS for
a particular UE.
A reliable connection to make the RNSAP simpler.
Support of signalling inactivity testing of a specific UE connection.

15

4.

RADIO INTERFACE ARCHITECTURE

4.1

Radio interface protocol architecture

4.1.1

Overall protocol structure

The radio interface is layered into three protocol layers:

the physical layer (L1),

the data link layer (L2),

network layer (L3).

Layer 2 is split into two sub-layers, Link Access Control (LAC) and Medium Access Control (MAC).
Layer 3 and LAC are divided into Control (C-) and User (U-) planes.
In the C-plane, Layer 3 is partitioned into sub-layers where the lowest sub-layer, denoted as Radio Resource Control
(RRC), interfaces with layer 2. The higher layer signalling such as Mobility Management (MM) and Call Control
(CC) are assumed to belong to the non-access stratum, and therefore not in the scope of SMG2. On the general level,
the protocol architecture is similar to the current ITU-R protocol architecture, ITU-R M.1035.
Figure 9 shows the radio interface protocol architecture. Each block in Figure 9 represents an instance of the
respective protocol. In the U-plane, the shaded LAC protocol may belong to the non-access stratum. Service Access
Points (SAP) for peer-to-peer communication are marked with circles at the interface between sub-layers. The SAP
to the physical layer provides the transport channels.

C-plane signalling

U-plane information

L3
RRC

L2/LAC
LAC

LAC

LAC

RLCP

LAC

RLCP

RLCP

L2/MAC
MAC

L1

PHY

Figure 9. Radio Interface protocol architecture (Service Access Points marked by circles)

16

4.1.2

Layer 1 Services and functions

4.1.2.1
L1 Services
The physical layer offers information transfer services to MAC and higher layers. The physical layer transport
services are described by how and with what characteristics data are transferred over the radio interface. An adequate
term for this is Transport Channel1.
4.1.2.1.1

Transport channels

A general classification of transport channels is into two groups:

common channels (where there is a need for in-band identification of the MSs when particular MSs are
addressed) and
dedicated channels (where the MSs are identified by the physical channel, i.e. code and frequency)
Common transport channel types are:
1. Random Access Channel(s) (RACH) characterised by:
existence in uplink only,
collision risk,
open loop power control,
limited data field, and
requirement for in-band identification of the MSs.
2. Forward Access Channel(s) (FACH) characterised by:
existence in downlink only,
possibility to use beam-forming,
possibility to use slow power control,
lack of fast power control and
requirement for in-band identification of MSs.
3. Broadcast Control Channel (BCCH) characterised by:
existence in downlink only,
low fixed bit rate and
requirement to be broadcast in the entire coverage area of the cell.
4. Paging Channel (PCH) characterised by:
existence in downlink only,
possibility for sleep mode procedures and
requirement to be broadcast in the entire coverage area of the cell.
The only type of dedicated transport channel is the:
1. Dedicated Channel (DCH) characterised by:
possibility to use beam-forming,
possibility to change rate fast (each 10ms),
fast power control and
inherent addressing of MSs.

This should be clearly separated from the classification of what is transported, which relates to the concept of
logical channels. Thus DCH is used to denote that the physical layer offers the same type of service for both control
and traffic.

17

To each transport channel, there is an associated Transport Format (for transport channels with a fixed or slow
changing rate) or an associated Transport Format Set (for transport channels with fast changing rate). A Transport
Format is defined as a combination of encoding, interleaving, bit rate and mapping onto physical channels. A
Transport Format Set is a set of Transport Formats. E.g., a variable rate DCH has a Transport Format Set (one
Transport Format for each rate), whereas a fixed rate DCH has a single Transport Format.
4.1.2.1.2

Model of physical layer of the MS

Figure 10 shows a model of the MSs physical layer in the uplink.

The model shows that one or several DCHs can be processed and multiplexed together by the same coding and
multiplexing unit. The detailed functions of the coding and multiplexing unit are yet to be defined. The single output
data stream from the coding and multiplexing unit is denoted Coded Composite Transport Channel (CCTrCH).
The data stream of the CCTrCH is fed to a data splitter unit that splits the CCTrCHs data stream onto one or several
Physical Channel Data Streams.
The current configuration of the coding and multiplexing unit (transport format) is either signalled to, or optionally
blindly detected by, the network for each 10 ms frame. If the configuration is signalled, the Transport Format
Indicator (TFI) bits represent it. Note that the TFI signalling only consists of pointing out the current transport
formats within the already configured transport format sets. In the uplink there is only one TFI representing the
current transport formats on all DCHs simultaneously. The physical channel data stream carrying the TFI is mapped
onto the physical channel carrying the power control bits and the pilot.
The random access transport channel (RACH) is the only common type transport channel in the uplink. RACHs are
always mapped one-to-one onto physical channels, i.e. there is no physical layer multiplexing of RACH. The MAC
layer handles Service multiplexing.
DCH DCH DCH

DCH

Coding and
multiplexing

Coding

RACH
Coding

Coded Composite
Transport Channel
(CCTrCH)

Transport
Format
Indicator
(TFI)

Splitter
Physical Channel
Data Streams
Phy CH

Phy CH

Phy CH

TPC

Phy CH

Phy CH

Figure 10. Model of the MSs physical layer uplink

Figure 11 shows the model of the MSs physical layer for the downlink.
For the DCHs the model is quite similar as the uplink model. Differences are mainly due to the soft and softer
handover. Further, the pilot, TPC bits and TFIs are time multiplexed onto the same physical channel(s) as the DCHs.
The mapping between DCHs and physical channel data streams works in the same way as for the uplink. Note
however, that the number of DCHs, the coding and multiplexing etc. may be different in uplink and downlink.
Further, the definition of physical channel data is somewhat different from the uplink.
Note that it is logically one and the same physical data stream in the active set of cells, even though physically there
is one stream for each cell. The same processing and multiplexing is done in each cell. The only difference between
the cells is the actual codes, and these codes of course correspond to the same spreading factor.
The physical channels carrying the same physical data stream are combined in the MS receiver, excluding the pilot,
and in some cases the TPC bits and transport format indicators (TFIs). TPC bits received on certain physical
channels may be combined, e.g. physical channels from cells belonging to the same site (softer handover), provided
that UTRAN has informed the MS that the TPC information on these channels is identical. The TFIs may also be
combined provided that all physical data streams are identical in the associated cells. Figure 11 shows the case where
one of the physical data streams is only transmitted in one of the cells, while two other physical data streams are

18

transmitted in three cells, i.e. there are two different active sets for the MS. This would be the situation if e.g. a
certain type of service should not employ soft handover whereas other simultaneous services should. Since the
number of DCHs and thereby the combinations of transport format sets now will be different between different cells,
the TFIs will also differ. In this example the TFIs transmitted from Cell 2 and Cell 3 will be exactly identical and
may therefore be combined by the MS. However, the TFI from Cell 1 will be different. If different active sets
between physical data streams are allowed, UTRAN must inform the MS of what TFIs are identical. Note that
physical channel data streams that are related to the same CCTrCH are always transmitted in the same set of cells.
There are three types of common transport channels in the downlink, namely BCCH, FACH and PCH. Downlink
common transport channels are mapped one-to-one onto separate physical channels. The MAC layer handles Service
multiplexing.
FACH

BCCH

PCH

DCH DCH DCH

DCH

Decoding and
demultiplexing

Decoding

Decoding Decoding Decoding

Coded Composite
Transport Channel
(CCTrCH)
MUX
Physical Channel
Data Streams
Phy CH

Phy CH

Phy CH

Cell 1
Cell 2
Cell 3

Phy CH
Phy CH
Phy CH

Phy CH
Phy CH
Phy CH

Phy CH

TPC stream 1, TFI 1

TPC stream 2, TFI 2
TPC stream 3, TFI 2

Figure 11. Model of the MSs physical layer downlink

4.1.2.2
L1 Functions
The physical layer performs the following main functions:

FEC encoding/decoding of transport channels

Measurements

Macro diversity distribution/combining and soft handover execution

Multiplexing/de-multiplexing of transport channels and of coded composite transport channels

Mapping of coded composite transport channels on physical channels

Modulation and spreading/demodulation and de-spreading of physical channels

Frequency and time (chip, bit, slot, frame) synchronisation

Closed-loop power control

Power weighting and combining of physical channels

RF processing

4.1.3
4.1.3.1
4.1.3.1.1

Layer 2 Services and Functions

MAC sub-layer
MAC services

The main responsibility of MAC is to handle the access to the physical layer, i.e. the mapping or/and multiplexing of
user information and control signalling to transport channels.
The MAC layer provides the following services to the LAC [RLCP] sub-layer:

Establishment and release of MAC connections

Peer-to-peer transportation of LAC [RLCP] PDUs

19

4.1.3.1.2

MAC functions

The functions of MAC include:

Multiplexing/de-multiplexing of higher layer PDUs into/from transport blocks delivered to/from the
physical layer on transport channels. MAC should support service multiplexing at least for common transport
channels, since the physical layer does not support multiplexing of these channels. Multiplexing at MAC level
should also be supported onto DCHs for the case where the physical layer cannot offer sufficiently many DCHs
or transport formats for each of these

Selection of transport format within the transport format set. During communication MAC selects the
appropriate transport format within an assigned transport format set for each active transport channel depending
on source rate and radio resource limitations. The selection can be done on a 10ms frame basis or slower.
Depending on the selected transport format one or more PDUs from higher layer may be mapped onto a
transport block, consisting of one or more 10ms frames. The substantially slower process of setting up or
modifying the transport channels, and thereby the transport format set assignments, are handled by the RRC
protocol.

Priority handling. In the mapping of data onto transport channels, and in the selection of transport formats,
MAC may prioritise data differently. For instance, MAC may block PDUs of a certain higher layer instance, or
select transport formats corresponding to low rates for those PDUs, when there are PDUs from a higher layer
instance of higher priority.

Identification of MSs on common transport channels. When a particular MS is addressed on a common

downlink channel, or when an MS is using the RACH, there is a need for in-band identification of the MS. Since
the MAC layer handles the access to, and multiplexing onto, the transport channels, the identification
functionality is naturally also placed in MAC.

Contention resolution on RACH. The unambiguous separation of different MSs using the contention based
RACH channel is also naturally handled by MAC.

Dynamic scheduling. A scheduling function may be applicable for packet data on common as well as on
dedicated channels. The scheduling function is basically a rapidly operating (10ms basis or slightly slower)
resource allocation function, closely connected to the transport format selection and thereby a MAC function.

Note: above list of MAC functions may not be complete

4.1.3.1.3

Open issues

Main open issues are:

whether RLCP is part of MAC or a separate sub-layer
whether ciphering should be done by MAC or not.

4.1.3.2
4.1.3.2.1

RLCP
RLCP services

The RLCP layer provides LAC with either an assured/non-assured mode service (adds overhead) or a transparent
service (does not add overhead). The assured/non-assured mode service uses, in case of assured mode,
retransmission techniques that are optimised for the physical layer.

Assured mode operation. In the assured mode operation a reliable link, using ARQ, is maintained between the
peer protocol entities using RLCP service. Variable bit rates are supported.

Unassured mode operation. In the unassured mode operation, a link is maintained between the peer protocol
entities using RLCP service. No ARQ is performed. Variable bit rates are supported.

Transparent mode operation. The data stream will pass the RLCP without that the RLCP will append any
overhead to the data stream.

20

4.1.3.2.2

RLCP functions

The following functions are proposed:

Segmentation and assembly of LAC PDUs,

Automatic Repeat Request (ARQ). Either a Selective Repeat or a Go Back N ARQ is proposed.

Concerning the segmentation function it is proposed that LAC PDUs are transformed into reasonably small fixed
size RLCP PDUs, the size of which is determined by:

The smallest possible bit rate,

The frequency with which the rate may change.

4.1.3.2.3

Example of segmentation in RLCP

Assume that an MS is able to transmit with the following rates: {16 kbps, 32 kbps, 64 kbps}. The rates correspond to
the transmission rates at the RLCP level. The period in which the rate is not allowed to change is 10 ms. Thus,
following the rule stated above, the RLCP PDU is 160 bits.
In Figure 12 an illustration is given of how RLCP PDUs are transmitted. First, two RLCP PDUs are transmitted in a
10 ms frame. The rate of the channel is then 32 kbps. After 10 ms the rate is changed to 16 kbps. Now only one
RLCP PDU is transmitted during a 10 ms frame.
RLCP PDU

RLCP PDU

160 bits

RLCP PDU

160 bits

RLCP PDU

160 bits

10 ms

160 bits

10 ms

32 kbit/s

16 kbit/s

10 ms
16 kbit/s

Change of rate
Figure 12. Transmission of RLCP PDUs.

4.1.3.2.4

Open issues

Main open issues are:

whether RLCP retransmissions are needed if a LAC exists in U-plane,
whether ciphering should be done by RLCP, and
whether RLCP is part of the MAC or a separate protocol sub-layer.
4.1.3.3
4.1.3.3.1

LAC sub-layer
LAC services

The LAC sub-layer provides the following services to layer 3:

Establishment and release of LAC connections

Assured peer-to-peer transportation of L3 PDUs,

Unassured peer-to-peer transportation of L3 PDUs,

Transparent transportation of L3 PDUs (no protocol overhead).

21

4.1.3.3.2

LAC functions

The LAC provides data link layer functions to higher layers. The LAC is physical layer independent but it should be
designed for the characteristics of the radio environment.
The functions of LAC include:
Automatic Repeat Request (ARQ),
Flow control,
In-sequence delivery of LAC SDUs to higher layers,
Segmentation and assembly of higher layer PDUs.

4.1.3.3.3

Open issues

Main open issues are:

whether LAC in U-plane is needed assuming RLCP exists,
whether LAC in U-plane belongs to the access stratum or to the non-access stratum,
whether ciphering should be done by LAC in C-plane or not.

4.1.4
4.1.4.1
4.1.4.1.1

Layer 3 - RRC Services and Functions

RRC services
General Control

The General Control provides an information broadcast service. This service broadcasts information to all UEs in a
certain geographical area. The basic requirements from such service are:

It should be possible to broadcast non-access stratum information in a certain geographical area.

The information is transferred on an unassured mode link. Unassured mode means that the delivery of the
broadcast information can not be guaranteed (typically no retransmission scheme is used). It seems reasonable to
use an unassured mode link since the information is broadcast to a lot of UEs and since broadcast information
often is repeated periodically.

It should be possible to do repeated transmissions of the broadcast information (how it is repeated is controlled
by the non-access stratum).

The point where the UE received the broadcast information should be included, when the access stratum
delivers broadcast information to the non-access stratum.

4.1.4.1.2

Notification

The Notification provides paging and notification broadcast services. The paging service sends information to a
specific UE(s). The information is broadcast in a certain geographical area but addressed to a specific UE(s). The
basic requirements from such service are:

It should be possible to broadcast paging information to a number of UEs in a certain geographical area.

The information is transferred on an unassured mode link. It is assumed that the protocol entities in non-access
stratum handle any kind of retransmission of paging information.

The notification broadcast service broadcasts information to all UEs in a certain geographical. The basic
requirements from this service are typically the same as for the information broadcast service of the General Control
SAP:

It should be possible to broadcast notification information in a certain geographical area.

The information is transferred on an unassured mode link.

22

4.1.4.1.3

Dedicated Control

The Dedicated Control provides services for establishment/release of a connection and transfer of messages using
this connection. It should also be possible to transfer a message during the establishment phase. The basic
requirements from the establishment/release services are:

It should be possible to establish connections (both point and group connections).

It should be possible to transfer an initial message during the connection establishment phase. This message
transfer has the same requirements as the information transfer service.

It should be possible to release connections.

The information transfer service sends a message using the earlier established connection. It is possible to specify the
quality of service requirements for each message. A finite number of quality of service classes will be specified, but
currently no class has been specified. In order to get an idea of the basic requirements, the CC and MM protocols in
GSM are used as a reference. A GSM based core network is chosen since it is one main option for UMTS.
Considering the existing GSM specification of CC and MM the basic requirements from the information transfer
service are (these are services provided by RR and the data link layer in GSM):

Assured mode link for transfer of messages

This assured mode link guarantees that the CC and MM messages are transferred to the corresponding side.
Assured mode means that the delivery of the paging information can be guaranteed (some kind of retransmission
scheme is used). A connection between two DC SAPs using an assured mode link has already been introduced
and is called signalling connection. This link should also guarantee that no messages are lost or duplicated
during handover.

Preserved message order

The order of the transferred messages is preserved.

Priority handling
If SMS messages should be transported through the control plane it should be possible to give higher priority to
signalling messages.

The CC and MM protocols also expect other services, which can not be supported by the current primitives of the
Dedicated Control SAP, e.g. indication of radio link failure.

4.1.4.2
RRC functions
The Radio Resource Control (RRC) layer handles the control plane signalling of Layer 3 between the MSs and
URAN.
An initial proposal (not a complete list) for functions of RRC include:
Establishment, reconfiguration and release of a RRC connection between the MS and UTRAN,
Establishment, reconfiguration and release of Radio Access Bearers,
Assignment and release of radio resources to signalling radio bearer and radio access bearers within the RRC
connection,
Terminal mobility functions for the RRC connection, including handover and other mobility functions necessary
for packet data, e.g. cell/paging area update procedures,
MS measurement reporting and control of the reporting,
Outer loop power control,
Broadcast of system information,
Paging/notification.

23

5.

LAYER 1 DESCRIPTION (FDD MODE)

5.1
5.1.1

Transport channels and physical channels (FDD)

Transport channels

Transport channels are the services offered by Layer 1 to the higher layers.
5.1.1.1
5.1.1.1.1

Dedicated transport channel

DCH - Dedicated Channel

The Dedicated Channel (DCH) is a downlink or uplink transport channel that is used to carry user or control
information between the network and a mobile station. The DCH thus corresponds to the three channels Dedicated
Traffic Channel (DTCH), Stand-alone Dedicated Control Channel (SDCCH), and Associated Control Channel
(ACCH) defined within ITU-R M.1035. The DCH is transmitted over the entire cell or over only a part of the cell
using lobe-forming antennas.
5.1.1.2
5.1.1.2.1

Common transport channels

BCCH - Broadcast Control Channel

The Broadcast Control Channel (BCCH) is a downlink transport channel that is used to broadcast system- and cellspecific information. The BCCH is always transmitted over the entire cell.
5.1.1.2.2

FACH - Forward Access Channel

The Forward Access Channel (FACH) is a downlink transport channel that is used to carry control information to a
mobile station when the system knows the location cell of the mobile station. The FACH may also carry short user
packets. The FACH is transmitted over the entire cell or over only a part of the cell using lobe-forming antennas.
5.1.1.2.3

PCH - Paging Channel

The Paging Channel (PCH) is a downlink transport channel that is used to carry control information to a mobile
station when the system does not know the location cell of the mobile station. The PCH is always transmitted over
the entire cell.
5.1.1.2.4

RACH - Random Access Channel

The Random Access Channel (RACH) is an uplink transport channel that is used to carry control information from a
mobile station. The RACH may also carry short user packets. The RACH is always received from the entire cell.

5.1.2

Physical channels

5.1.2.1
The physical resource
The basic physical resource is the code/frequency plane. In addition, on the uplink, different information streams
may be transmitted on the I and Q branch. Consequently, a physical channel corresponds to a specific carrier
frequency, code, and, on the uplink, relative phase (0 or /2).
5.1.2.2
5.1.2.2.1

Uplink physical channels

Dedicated uplink physical channels

There are two types of uplink dedicated physical channels, the uplink Dedicated Physical Data Channel (uplink
DPDCH) and the uplink Dedicated Physical Control Channel (uplink DPCCH).
The uplink DPDCH is used to carry dedicated data generated at Layer 2 and above, i.e. the dedicated transport
channel (DCH). There may be zero, one, or several uplink DPDCHs on each Layer 1 connection.
The uplink DPCCH is used to carry control information generated at Layer 1. The Layer 1 control information
consists of known pilot bits to support channel estimation for coherent detection, transmit power-control (TPC)
commands, and an optional transport-format indicator (TFI). The transport-format indicator informs the receiver
about the instantaneous parameters of the different transport channels multiplexed on the uplink DPDCH, see further
Section 3. There is one and only one uplink DPCCH on each Layer 1 connection.
Frame structure
Figure 13 shows the frame structure of the uplink dedicated physical channels. Each frame of length 10 ms is split
into 16 slots, each of length Tslot = 0.625 ms, corresponding to one power-control period. A super frame corresponds
to 72 consecutive frames, i.e. the super-frame length is 720 ms.

24

Data
Ndata bits

DPDCH

TPC
NTPC bits

Pilot
Npilot bits

DPCCH

TFI
NTFI bits

0.625 ms, 10*2k bits (k=0..6)

Slot #1

Slot #2

Slot #i

Slot #16

Tf = 10 ms

Frame #1 Frame #2

Frame #i

Frame #72

Tsuper = 720 ms
Figure 13. Frame structure for uplink DPDCH/DPCCH
The parameter k in Figure 13 determines the number of bits per uplink DPDCH/DPCCH slot. It is related to the
spreading factor SF of the physical channel as SF = 256/2k. The spreading factor may thus range from 256 down to
4. Note that an uplink DPDCH and uplink DPCCH on the same Layer 1 connection generally are of different rates,
i.e. have different spreading factors and different values of k.
The exact number of bits of the different uplink DPCCH fields in Figure 13 (Npilot, NTPC, and NTFI) is yet to be
determined.

5.1.2.2.2
5.1.2.2.2.1

Common uplink physical channels

Physical Random Access Channel

The Physical Random Access Channel (PRACH) is used to carry the RACH. It is based on a Slotted ALOHA
approach, i.e. a mobile station can start the transmission of the PRACH at a number of well-defined time-offsets,
relative to the frame boundary of the received BCCH of the current cell. The different time offsets are denoted
access slots and are spaced 1.25 ms apart as illustrated in Figure 14. Information on what access slots are available
in the current cell is broadcast on the BCCH.

1.25 ms

Access slot #1

Random-access burst

Access slot #2

Random-access burst

Access slot #i

Random-access burst
Offset of access slot #i

Access slot #8

Random-access burst

Frame boundary
Figure 14. Access slots.
The structure of the random access burst of Figure 14 is shown in Figure 15. The random access burst consists of
two parts, a preamble part of length 1 ms and a message part of length 10 ms. Between the preamble part and the

25

message part there is an idle time period of length 0.25 ms (preliminary value). The idle time period allows for
detection of the preamble part and subsequent on-line processing of the message part.
Random-access burst
Preamble part
1 ms

Message part
0.25 ms
10 ms
Figure 15. Structure of the Random Access burst.

Preamble part
The preamble part of the random-access burst consists of a signature of length 16 complex symbols (1j). Each
preamble symbol is spread with a 256 chip real Orthogonal Gold code. There are a total of 16 different signatures,
based on the Orthogonal Gold code set of length 16 (see Section 5.3.1.2.3.1
Preamble spreading code for more
details).

Message part
The message part of the random-access burst has the same structure as the uplink dedicated physical channel. It
consists of a data part, corresponding to the uplink DPDCH, and a Layer 1 control part, corresponding to the uplink
DPCCH, see

Data part

Pilot symbols

Rate information

10 ms

Figure 16. The data and control parts are transmitted in parallel. The data part carries the random access request or
small user packets. The spreading factor of the data part is limited to SF{256, 128, 64, 32} corresponding to
channel bit rates of 16, 32, 64, and 128 kbps respectively. The control part carries pilot bits and rate information,
using a spreading factor of 256. The rate information indicates which channelisation code (or rather the spreading
factor of the channelisation code) is used on the data part, see further Section 5.3.1.2.3 Random access codes.

Data part

Pilot symbols

Rate information

10 ms

Figure 16. The message part of the random access burst.

26

Figure 17 shows the structure of the data part of the Random-Access burst. It consists of the following fields (the
values in brackets are preliminary values):
Mobile station identification (MS ID) [16 bits]. The MS ID is chosen at random by the mobile station at the time
of each Random-Access attempt.
Required Service [3 bits]. This field informs the base station what type of service is required (short packet
transmission, dedicated-channel set-up, etc.)
An optional user packet
A CRC to detect errors in the data part of the Random-Access burst [8 bits].
MS ID Req. Ser.

Optional user packet

CRC

Figure 17. Structure of Random-Access burst data part.

5.1.2.3
5.1.2.3.1

Downlink physical channels

Dedicated physical channels

There is only one type of downlink dedicated physical channel, the Downlink Dedicated Physical Channel (downlink
DPCH).
Within one downlink DPCH, dedicated data generated at Layer 2 and above, i.e. the dedicated transport channel
(DCH), is transmitted in time-multiplex with control information generated at Layer 1 (known pilot bits, TPC
commands, and an optional TFI). The downlink DPCH can thus be seen as a time multiplex of a downlink DPDCH
and a downlink DPCCH, compare Section 5.1.2.2.1 Dedicated uplink physical channels.
Frame structure
Figure 18 shows the frame structure of the downlink DPCH. Each frame of length 10 ms is split into 16 slots, each of
length Tslot = 0.625 ms, corresponding to one power-control period. A super frame corresponds to 72 consecutive
frames, i.e. the super-frame length is 720 ms.
Figure 18. Frame structure for downlink DPCH.
DPCCH
TFI
Pilot
TPC
Npilot bits NTPC bits NTFI bits

DPDCH
Data
Ndata bits

0.625 ms, 20*2k bits (k=0..6)

Slot #1

Slot #2

Slot #i

Slot #16

Tf = 10 ms

Frame #1

Frame #2

Frame #i

Frame #72

Tsuper = 720 ms
The parameter k in Figure 18 determines the total number of bits per downlink DPCH slot. It is related to the
spreading factor SF of the physical channel as SF = 256/2k. The spreading factor may thus range from 256 down to
4.
The exact number of bits of the different downlink DPCH fields in Figure 18 (Npilot, NTPC, NTFI, and Ndata) is yet to be
determined.
Note that connection-dedicated pilot bits are transmitted also for the downlink in order to support the use of
downlink adaptive antennas.
When the total bit rate to be transmitted on one downlink connection exceeds the maximum bit rate for a downlink
physical channel, multi-code transmission is employed, i.e. several parallel downlink DPCHs are transmitted for one
connection using the same spreading factor. In this case, the Layer 1 control information is put on only the first

27

downlink DPCH. The additional downlink DPCHs belonging to the connection do not transmit any data during the
corresponding time period, see Figure 19.
Multiple codes may also transmitted in order to transmit different transport channels on different codes (code
multiplex). In that case, the different parallel codes may have different spreading factors and the Layer 1 control
information is transmitted on each code independently.
DPCCH

DPDCH

Transmission
Power

Physical Channel 1

Transmission
Power

Physical Channel 2

Transmission
Power

Physical Channel L

One slot (0.625 ms)

Figure 19. Downlink slot format in case of multi-code transmission.
5.1.2.3.2
5.1.2.3.2.1

Common physical channels

Primary Common Control Physical Channel (CCPCH)

The Primary CCPCH is a fixed rate (32 kbps, SF=256) downlink physical channels used to carry the BCCH.
Figure 20 shows the frame structure of the Primary CCPCH. The frame structure differs from the downlink DPCH in
that no TPC commands or TFI is transmitted. The only Layer 1 control information is the common pilot bits needed
for coherent detection.

28

Data
12 bits

Pilot
8 bits
0.625 ms, 20 bits

Slot #1

Slot #2

Slot #i

Slot #16

Tf = 10 ms

Frame #1

Frame #2

Frame #i

Frame #72

Tsuper = 720 ms
Figure 20. Frame structure for Primary Common Control Physical Channel.
5.1.2.3.2.2

Secondary Common Control Physical Channel

The secondary CCPCH is used to carry the FACH and PCH. It is of constant rate. However, in contrast to the
Primary CCPCH, the rate may be different for different secondary CCPCH within one cell and between cells, in
order to be able to allocate different amount of FACH and PCH capacity to a cell. The rate and spreading factor of
each secondary CCPCH is broadcast on the BCCH. The set of possible rates is the same as for the downlink DPCH,
see Section 5.1.2.3.1
Dedicated physical channels.
The frame structure of the Secondary CCPCH is shown in Figure 21.
x

Data
Ndata bits

Pilot
Npilot bits

0.625 ms, 20*2k bits (k=0..6)

Slot #1

Slot #2

Slot #i

Slot #16

Tf = 10 ms

Frame #1

Frame #2

Frame #i

Frame #72

Tsuper = 720 ms
Figure 21. Frame structure for Secondary Common Control Physical Channel.
The FACH and PCH are mapped to separate Secondary CCPCHs. The main difference between a CCPCH and a
downlink dedicated physical channel is that a CCPCH is not power controlled. The main difference between the
Primary and Secondary CCPCH is that the Primary CCPCH has a fixed predefined rate while the Secondary CCPCH
has a constant rate that may be different for different cells, depending on the capacity needed for FACH and PCH.
Furthermore, a Primary CCPCH is continuously transmitted over the entire cell while a Secondary CCPCH is only
transmitted when there is data available and may be transmitted in a narrow lobe in the same way as a dedicated
physical channel (only valid for a Secondary CCPCH carrying the FACH).
5.1.2.3.2.3
Synchronisation Channel
The Synchronisation Channel (SCH) is a downlink signal used for cell search. The SCH consists of two sub
channels, the Primary and Secondary SCH. Figure 22 illustrates the structure of the SCH:

29
Tslot = 2560 chips

chips
Primary SCH

cp

i,1
Secondary SCH cs

256 chips

cp

cp

csi,2

csi,16

Tframe = 16*Tslot
cp : Primary Synchronization Code
csi,k: One of 17 possible Secondary Synchronization Codes

(csi,1, csi,2, ..., csi,16) encode cell specific long scrambling code group i
Figure 22. Structure of Synchronisation Channel (SCH).
The Primary SCH consists of an unmodulated orthogonal Gold code of length 256 chips, the Primary
Synchronisation Code, transmitted once every slot. The Primary Synchronisation Code is the same for every base
station in the system and is transmitted time-aligned with the BCCH slot boundary as illustrated in Figure 22.
The Secondary SCH consists of repeatedly transmitting a length 16 sequence of unmodulated Orthogonal Gold
codes of length 256 chips, the Secondary Synchronisation Codes, transmitted in parallel with the Primary
Synchronisation channel. Each Secondary Synchronisation code is chosen from a set of 17 different Orthogonal Gold
codes of length 256. This sequence on the Secondary SCH indicates which of the 32 different code groups (see
Section 5.3.2.2.2 Scrambling code) the base station downlink scrambling code belongs. 32 sequences are used to
encode the 32 different code groups each containing 16 scrambling codes. The 32 sequences are constructed such
that their cyclic-shifts are unique, i.e., a non-zero cyclic shift less than 16 of any of the 32 sequences is not equivalent
to some cyclic shift of any other of the 32 sequences. Also, a non-zero cyclic shift less than 16 of any of the
sequences is not equivalent to itself with any other cyclic shift less than 16. This property is used to uniquely
determine both the long code group and the frame timing in the second step of acquisition (see Section 5.5.2.1
Initial cell search). The following sequences are used to encode the 32 different code groups each
containing 16 scrambling codes (note that ci indicates the ith Secondary Short code of the 17 Orthogonal Gold
codes).
(c1 c1 c2 c11 c6 c3 c15 c7 c8 c8 c7 c15 c3 c6 c11 c2 )
(c1 c2 c9 c3 c10 c11 c13 c13 c11 c10 c3 c9 c2 c1 c16 c16 )
(c1 c3 c16 c12 c14 c2 c11 c2 c14 c12 c16 c3 c1 c13 c4 c13 )
(c1 c4 c6 c4 c1 c10 c9 c8 c17 c14 c12 c14 c17 c8 c9 c10 )
(c1 c5 c13 c13 c5 c1 c7 c14 c3 c16 c8 c8 c16 c3 c14 c7 )
(c1 c6 c3 c5 c9 c9 c5 c3 c6 c1 c4 c2 c15 c15 c2 c4 )
(c1 c7 c10 c14 c13 c17 c3 c9 c9 c3 c17 c13 c14 c10 c7 c1 )
(c1 c8 c17 c6 c17 c8 c1 c15 c12 c5 c13 c7 c13 c5 c12 c15 )
(c1 c9 c7 c15 c4 c16 c16 c4 c15 c7 c9 c1 c12 c17 c17 c12 )
(c1 c10 c14 c7 c8 c7 c14 c10 c1 c9 c5 c12 c11 c12 c5 c9 )
(c1 c11 c4 c16 c12 c15 c12 c16 c4 c11 c1 c6 c10 c7 c10 c6 )
(c1 c12 c11 c8 c16 c6 c10 c5 c7 c13 c14 c17 c9 c2 c15 c3 )
(c1 c13 c1 c17 c3 c14 c8 c11 c10 c15 c10 c11 c8 c14 c3 c17 )
(c1 c14 c8 c9 c7 c5 c6 c17 c13 c17 c6 c5 c7 c9 c8 c14 )
(c1 c15 c15 c1 c11 c13 c4 c6 c16 c2 c2 c16 c6 c4 c13 c11 )
(c1 c16 c5 c10 c15 c4 c2 c12 c2 c4 c15 c10 c5 c16 c1 c8 )
(c1 c17 c12 c2 c2 c12 c17 c1 c5 c6 c11 c4 c4 c11 c6 c5 )
(c2 c8 c11 c15 c14 c1 c4 c10 c10 c4 c1 c14 c15 c11 c8 c2 )
(c2 c9 c1 c7 c1 c9 c2 c16 c13 c6 c14 c8 c14 c6 c13 c16 )
(c2 c10 c8 c16 c5 c17 c17 c5 c16 c8 c10 c2 c13 c1 c1 c13 )
(c2 c11 c15 c8 c9 c8 c15 c11 c2 c10 c6 c13 c12 c13 c6 c10 )
(c2 c12 c5 c17 c13 c16 c13 c17 c5 c12 c2 c7 c11 c8 c11 c7 )
(c2 c13 c12 c9 c17 c7 c11 c6 c8 c14 c15 c1 c10 c3 c16 c4 )
(c2 c14 c2 c1 c4 c15 c9 c12 c11 c16 c11 c12 c9 c15 c4 c1 )
(c2 c15 c9 c10 c8 c6 c7 c1 c14 c1 c7 c6 c8 c10 c9 c15 )
(c2 c16 c16 c2 c12 c14 c5 c7 c17 c3 c3 c17 c7 c5 c14 c12 )
(c2 c17 c6 c11 c16 c5 c3 c13 c3 c5 c16 c11 c6 c17 c2 c9 )
(c2 c1 c13 c3 c3 c13 c1 c2 c6 c7 c12 c5 c5 c12 c7 c6 )
(c2 c2 c3 c12 c7 c4 c16 c8 c9 c9 c8 c16 c4 c7 c12 c3 )
(c2 c3 c10 c4 c11 c12 c14 c14 c12 c11 c4 c10 c3 c2 c17 c17 )
(c2 c4 c17 c13 c15 c3 c12 c3 c15 c13 c17 c4 c2 c14 c5 c14 )
(c2 c5 c7 c5 c2 c11 c10 c9 c1 c15 c13 c15 c1 c9 c10 c11 )

30

The multiplexing of the SCH with the other downlink physical channels (DPCH and CCPCH) is illustrated in Figure
23. The figure illustrates that the SCH is only transmitted intermittently (one codeword per slot) and also that the
SCH is multiplexed after long code scrambling of the DPCH and CCPCH. Consequently, the SCH is nonorthogonal to the other downlink physical channels.
Lower position during
256 chips per slot
0
1
SCH

cp
0

1
cs

DPCH
& CCPCH

cch,1

To IQ modulator

cscramb
cch,N

Figure 23. Multiplexing of SCH.

The use of the SCH for cell search is described in detail in Section 5.5.2

5.1.3

Cell search.

Mapping of Transport Channels to Physical Channels

Figure 24 summarises the mapping of transport channels to physical channels.

Transport Channels
Physical Channels
BCCH
Primary Common Control Physical Channel (Primary CCPCH)
FACH
Secondary Common Control Physical Channel (Secondary CCPCH)
PCH
RACH
Physical Random Access Channel (PRACH)
DCH
Dedicated Physical Data Channel (DPDCH)
Synchronisation Channel (SCH)
Figure 24. Transport-channel to physical-channel mapping.
Mapping Method of PCH to Common Control Physical Channel
The mapping method is shown in Figure 25.
The PCH is divided into several groups in one super-frame, and layer 3 information is transmitted in each group.
Each group of PCH shall have information amount worth of 4 slots, and consists of a total of 6 information parts: 2
Paging Indication (PI) parts - for indicating whether there are MS-terminated calls or not, and 4 Mobile User
Identifier (MUI) parts - for indicating Identity of the paged mobile user.
In each group, PI parts are transmitted ahead of MUI parts.
In all groups, 6 information parts are allocated with a certain pattern in the range of 24 slots. By shifting each pattern
by 4 slots, multiple 288 groups of PCH can be allocated on one Secondary Common Control Physical Channel.

31

PCH

Superframe (720 ms)

Frame (10 ms)

PCH
PCH

PCH

PCH

PCH

Slot (0.625 ms)

MUI MUI MUI

2
3
4
PI1 PI2

MUI1

PCH information
for group #1

MUI MUI MUI

2
3
4
PI1 PI2

MUI1

MUI MUI MUI

2
3
4
PI1 PI2

PCH information
for group #2

MUI1

PCH information
for group #3

Figure 25. PCH mapping method.

5.2
5.2.1

Multiplexing, channel coding and interleaving (FDD)

Transport-channel coding/multiplexing

Figure 26 illustrates the overall concept of transport-channel coding and multiplexing. The following steps can be
identified:
Channel coding, including optional transport-channel multiplexing
Static rate matching
Inter-frame interleaving
Transport-channel multiplexing
Dynamic rate matching
Intra-frame interleaving
The different steps are described in detail below

32

TC

Channel coding +
optional TC multiplex

Static rate matching

Inner interleaving
(inter-frame)

Transport-channel
multiplexing

Dynamic rate matching

(uplink only)

Inner interleaving
(intra-frame)

TC

TC

TC

Coding +
interleaving

Coding +
interleaving

Rate
matching

Rate
matching

Interleaving
(optional)

Interleaving
(optional)

Multiplex

Rate
matching

Interleaving

Figure 26. Coding and multiplexing of transport channels.

Note that although the coding, static rate matching, and inter-frame interleaving is done in parallel chains for
different transport channels, some co-ordination in the parameter setting may be needed when adding, removing, or
modifying transport channels (indicated by the dashed box in Figure 26).
The output after the inner interleaving is typically mapped to one DPDCH. Only for the very highest bit rates the
output is split onto several DPDCHs, i.e. multi-code transmission.
Primarily, transport channels are coded and multiplexed as described above, i.e. into one data stream mapped on one
or several physical channels. However, an alternative way of multiplexing services is to use code multiplexing,
which corresponds to having several parallel multiplexing chains as in Figure 26, resulting in several data stream,
each mapped to one or several physical channels.
5.2.1.1
Channel coding
Channel coding is done on a per-transport-channel basis, i.e. before transport-channel multiplexing.
The following options are available for the transport-channel specific coding, see also Figure 27:
Convolutional coding
Outer Reed-Solomon coding + Outer interleaving + Convolutional coding
Turbo coding
Service-specific coding, e.g. unequal error protection for some types of speech codecs.

33

Convolutional
coding

Reed-Solomon
coding

Outer
interleaving

Convolutiona
coding

Turbo
coding

Service-specific
coding

Figure 27. Channel coding in UTRA/FDD.

5.2.1.1.1

Convolutional coding

Convolutional coding is typically applied for services that require a BER in the order of 10-3. Convolutional coding
is also, in concatenation with RS coding + outer interleaving, applied to services that require a BER in the order of
10-6, see also Section 5.2.1.1.2
Outer Reed-Solomon coding and outer interleaving.
Table 1 lists the possible parameters for the convolutional coding.
Table 1. Generator polynomials for the convolutional codes.
Rate
1/3

Constraint
length
9
9

Generator
polynomial 1
557
561

Generator
polynomial 2
663
753

Generator
polynomial 3
711
N/A

Typically, rate-1/3 convolutional coding is applied to dedicated transport channels (DCHs) in normal (non-slotted)
mode while rate convolutional coding is applied to DCHs in slotted mode, see Section 5.5.4.2.1.1
Slotted
mode.
5.2.1.1.2

Outer Reed-Solomon coding and outer interleaving

Reed-Solomon coding + outer interleaving, is, in concatenation with inner convolutional coding, typically applied to
transport channels that require a BER in the order of 10-6.
The RS-coding is of approximate rate 4/5 using the 256-ary alphabet.
The outer interleaving is symbol-based block interleaver with interleaver width equal to the block length of the RS
code. The interleaver span is variable in the range 20 ms to 150 ms.
5.2.1.1.3

Turbo coding

The use of Turbo coding for high data rate (above 32 kbps), high quality services, is currently being investigated
within ETSI. Turbo codes of rate 1/3 and (for the highest data rates), have been proposed to replace the
concatenation of convolutional and Reed-Solomon codes. ETSI is awaiting further results of simulations illustrating
the performance of Turbo Codes.

34

The block diagram for the basic Turbo Encoder is shown in Figure 28.
Info
bits
Parity
Bits

Constituent
Encoder #1

Puncture
Parity
Bits

Constituent
Encoder #2

Interleaver

Figure 28. Block diagram of a Turbo code encoder.

If Turbo codes are shown to give an improved FEC for high quality services, compared with the existing proposal,
then the basic FEC coding for the UTRA/FDD will be as shown in Figure 29.

BER = 10

-3

BER = 10

-6

Convolutional
coding

Channel
interleaving

Turbo
coding

Channel
interleaving

Service-specific coding

Figure 29. FEC coding for UTRA/FDD when turbo codes are used.
5.2.1.1.4

Service specific coding

The service-specific-coding option allows for additional flexibility of the UTRA Layer 1 by allowing for additional
coding schemes, in addition to the standard coding schemes listed above. One example is the use of unequal-errorprotection coding schemes for certain speech-codecs.
5.2.1.2
Inner inter-frame interleaving
Inner inter-frame bit interleaving is carried out on a per-transport-channel basis on those transport-channels that can
allow for and require interleaving over more than one radio frame (10 ms). The span of the inner inter-frame
interleaving can vary in the range 20 ms to 150 ms.
5.2.1.3
Rate matching
Two types of rate matching is carried out:
Static rate matching carried out on a slow basis, typically every time a transport channel is added or removed
from the connection.
Dynamic rate matching carried out on a frame-by-frame (10 ms) basis
5.2.1.3.1

Static rate matching

Static rate matching is used for two different reasons:

to adjust the coded transport channel bit rate to a level where minimum transmission quality requirements of each
transport channel is fulfilled with the smallest differences in channel bit energy
to adjust the coded transport channel bit rate so that the maximum total bit rate after transport channel
multiplexing is matched to the channel bit rate of the uplink and downlink dedicated physical channel
The static rate matching is based on code puncturing and unequal repetition.
Note that, although static rate matching is carried out prior to transport-channel multiplexing, the rate matching must
be co-ordinated between the different transport channels.

35

5.2.1.3.2

Dynamic rate matching

Dynamic rate matching is carried out after the multiplexing of the parallel coded transport channels and is used to
match the total instantaneous rate of the multiplexed transport channels to the channel bit rate of the uplink DPDCH.
Dynamic rate matching uses unequal repetition and is only applied to the uplink. On the downlink, discontinuous
transmission (DTX) is used when the total instantaneous rate of the multiplexed transport channels does not match
the channel bit rate.
5.2.1.3.3

Rate matching algorithm

Lets denote:

S N = {N1 , N 2 ,..., N L } = ordered set (in ascending order from left to right) of allowed number of bits per block

NC = number of bits per matching block

S0 = d1 , d 2 ,..., d NC = set of NC data bits

P = maximum amount of puncturing allowed (tentatively 0.2, for further study)
The rate-matching rule is as follows:
find Ni and Ni+1 so that Ni NC < Ni+1
j=0

N
z = i +1
NC
if ( z > 1 & N C N i )
repeat every bit from set Sj z times

NC = NC z
N Ni
< P)
if ( C
NC
x = NC
y = NC-Ni

S j = d1 , d 2 ,..., d NC

do while y > 1

x
x
z = , k =
z
y
x = x-k
y = y-k
puncture every zth bit from set Sj
form new set Sj+1 from not punctured bits of set Sj
j = j+1
end do
if y == 1
puncture last bit from set Sj
else
x = NC
y = Ni+1-NC

S j = d1 , d 2 ,..., d NC

do while y > 1

x
x
z = , k =
z
y
x = x-k
y = y-k

36

repeat every zth bit from set Sj

form new set Sj+1 from not repeated bits of set Sj
j = j+1
end do
if y == 1
repeat first bit from set Sj
5.2.1.4
Transport-channel multiplexing
The coded transport channels are serially multiplexed within one radio frame. The output after the multiplexer
(before the inner interleaving) will thus be according to Figure 30.
10 ms (one radio frame)
TC-1

TC-2

TC-M

Figure 30. Transport channel multiplexing.

As an option, transport channels may be multiplexed within the channel-coding unit, typically after outer RS coding
but before outer interleaving.
5.2.1.5
Inner intra-frame interleaving
Inner intra-frame interleaving over one radio frame (10 ms) is applied to the multiplexed set of transport channels.

5.2.2

Automatic Repeat Request (ARQ)

The details of the UTRA ARQ schemes are not yet specified. Therefore, the impact on Layer 1, e.g. if soft
combining of retransmitted packets is to take place, is not yet fully specified.

5.2.3

Coding for layer 1 control

5.2.3.1
Transport-format-indicator coding
The TFI bits are encoded using bi-orthogonal (32, 6) block code. The coding procedure is as shown in Figure 31.
TFI word (1-6 bits)

Biorthogonal
(32, 6) coding

Mapping to
physical channel

Figure 31. Channel coding of TFI bits.

The length of the TFI code word is 32 bits. Thus there are 2 bits of (encoded) TFI in every slot of the radio frame.
The code words of the bi-orthogonal block code are from the level 6 of the code three of OVSF codes defined in
chapter 4.3.2.1. The code words, C6(i), i = 0, ..., 31, form an orthogonal set,
of 32 functions. By taking the binary complements of the code words of

SC6 = {C6 (0), C6 (1),..., C6 (31)} ,

SC6 , another set,

SC6 = C6 (0), C6 (1),..., C6 (31) is formed. These two sets are mutually bi-orthogonal yielding total of 64
different code words.
Mapping of the TFI bits to the code words is done as shown in the Figure 32.

37

RI bits

32 bits

MSB

000000
000001
000010
000011

LSB

C6(0)
C6(0)
C6(1)
C6(1)

64 functions

111110
111111

C6(31)
C6(31)
Figure 32. Mapping of TFI bits to bi-orthogonal code words.

Bits of the TFI code words are time multiplexed to DPCCH as shown in the Figure 33. Within a slot the more
significant bit is transmitted before the less significant bit.
MSB

TFI code word

LSB

31 30 29 28

Slot 1

1 0

Slot 2

Slot 16

Radio frame 10 ms

Figure 33. Time multiplexing of the bits of a TFI code word to radio frame.

5.3

Spreading and modulation (FDD)

5.3.1

Uplink spreading and modulation

5.3.1.1
Spreading
Uplink Dedicated Physical Channels (uplink DPDCH/DPCCH)
Figure 34 illustrates the spreading and modulation for the case of a single uplink DPDCH. Data modulation is dualchannel QPSK, where the uplink DPDCH and DPCCH are mapped to the I and Q branch respectively. The I and Q
branch are then spread to the chip rate with two different channelisation codes cD/cC and subsequently complex
scrambled by a mobile-station specific complex scrambling code cscramb
Channelization
codes (OVSF)
cD

cos(t)
I

DPDCH

cscramb

p(t)

I+jQ

cC
DPCCH

Real

sin(t)
Imag

p(t)

Figure 34. Spreading/modulation for uplink DPDCH/DPCCH.

38

For multi-code transmission, each additional uplink DPDCH may be transmitted on either the I or the Q branch. For
each branch, each additional uplink DPDCH should be assigned its own channelisation code. Uplink DPDCHs on
different branches may share a common channelisation code.
PRACH
The spreading and modulation of the message part of the Random-Access burst is basically the same as for the
uplink dedicated physical channels, see Figure 34, where the uplink DPDCH and uplink DPCCH are replaced by the
data part and the control part respectively. The scrambling code for the message part is chosen based on the basestation-specific preamble code, the randomly chosen preamble sequence, and the randomly chosen access slot
(random-access time-offset), see Section 5.1.2.2.2.1 Physical Random Access Channel. This guarantees that two
simultaneous Random-Access attempts that use different preamble codes and/or different preamble sequences will
not collide during the data part of the Random-Access bursts.
5.3.1.2
Code generation and allocation
5.3.1.2.1

Channelisation codes

The channelisation codes of Figure 34 are the same type of OVSF codes as for the downlink, see Figure 37. For the
uplink, the restrictions on the allocation of channelisation codes given in Section 5.3.2.2.1
Channelisation
codes are only valid within one mobile station.
Each connection is allocated at least one uplink channelisation code, to be used for the uplink DPCCH. In most
cases, at least one additional uplink channelisation code is allocated for a uplink DPDCH. Further uplink
channelisation codes may be allocated if more than one uplink DPDCH are required.
As different mobile stations use different uplink scrambling codes, the uplink channelisation codes may be allocated
with no co-ordination between different connections. The uplink channelisation codes are therefore always allocated
in a pre-defined order. The mobile-station and network only need to agree on the number and length (spreading
factor) of the uplink channelisation codes. The exact codes to be used are then implicitly given.
5.3.1.2.2

Scrambling codes

Either short or long scrambling codes should be used on uplink.

5.3.1.2.2.1
Short scrambling code
The short scrambling code is a complex code cscramb = cI+jcQ, where cI and cQ are two different codes from the
extended Very Large Kasami set of length 256.
The network decides the uplink short scrambling code. The mobile station is informed about what short scrambling
code to use in the downlink Access Grant message that is the base-station response to an uplink Random Access
Request.
The short scrambling code may, in rare cases, be changed during the duration of a connection.
5.3.1.2.2.2
Long scrambling code
The long uplink scrambling code is typically used in cells without multi-user detection in the base station. The
mobile station is informed if a long scrambling code should be used in the Access Grant Message following a
Random-Access request and in the handover message.
What long scrambling code to use is directly given by the short scrambling code. No explicit allocation of the long
scrambling code is thus needed.
The scrambling code sequences are constructed as the position wise modulo 2 sum of 40960 chip segments of two
binary m-sequences generated by means of two generator polynomials of degree 41. Let x, and y be the two msequences respectively. The x sequence is constructed using the primitive (over GF(2)) polynomial 1+X3+X41 . The y
sequence is constructed using the polynomial 1+ X20+X41 . The resulting sequences thus constitute segments of a set
of Gold sequences.
The scrambling code for the quadrature component is a 1024-chip shifted version of the in-phase scrambling code.
The uplink scrambling code word has a period of one radio frame of 10 ms.
Let n40 n0 be the binary representation of the scrambling code number n (decimal) with n0 being the least
significant bit. The x sequence depends on the chosen scrambling code number n and is denoted xn, in the sequel.
Furthermore, let xn(i) and y(i) denote the i:th symbol of the sequence xn and y, respectively
The m-sequences xn and y are constructed as:
Initial conditions:
xn(0)=n0 , xn(1)= n1 , =xn(39)= n39 , xn(40)= n40
y(0)=y(1)= =y(39)= y(40)=1
Recursive definition of subsequent symbols:
xn(i+41) =xn(i+3) + xn(i) modulo 2, i=0,, 241-43,
y(i+41) = y(i+20)+y(i) modulo 2, i=0,, 241-43.
The definition of the n:th scrambling code word for the in phase and quadrature components follows as (the left most
index correspond to the chip scrambled first in each radio frame):

39
CIlong,n = < xn(0)+y(0), xn(1)+y(1), ,xn(40959)+y(40959) >,
CQlong,n = < xn(1024)+y(1024), xn(1025)+y(1025), , xn(41983) + y(41983) >,
again all sums being modulo 2 additions.
Now, the complex long scrambling code Clong,n is defined by:
Clong,n = (CIlong,n + jCQlong,n ) =
= < ((xn(0)+y(0)) + j(xn(1024)+y(1024))), ,
((xn(40959)+y(40959)) + j(xn(41983) + y(41983))) >
The code generator must be able to generate the sequence shifted arbitrarily from the initial state.
5.3.1.2.3

Random access codes

Preamble spreading code

5.3.1.2.3.1

The spreading code for the preamble part is cell specific and is broadcast by the base station. More than one
preamble code can be used in a base station if the traffic load is high. The preamble codes must be code planned,
since two neighbouring cells should not use the same preamble code.
The code used is a real-valued 256 chip Orthogonal Gold code. All 256 codes are used in the system. The preamble
codes are generated in the same way as the codes used for the downlink synchronisation channel and are defined in
Section 5.3.2.2.3 Synchronisation codes.
5.3.1.2.3.2

Preamble signature

The preamble part carries one of 16 different orthogonal complex signatures of length 16, <P0, P1, ..., P15>. The
signatures are based on a set of Orthogonal Gold codes of length 16 and are specified in Table 2. The base station
broadcasts which signatures are allowed to be used in a cell.
Table 2. Preamble signatures. A = 1+j.
Preamble symbols
P0

PA

P2

P3

P4

P5

P7

P8

P9

P10

P11

P12

P13

P14

P15

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

10

-A

-A

-A

-A

-A

-A

-A

11

-A

-A

-A

-A

-A

12

-A

-A

-A

-A

-A

13

-A

-A

-A

-A

-A

-A

-A

-A

-A

14

-A

-A

-A

-A

-A

15

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

-A

16

-A

-A

-A

-A

-A

-A

-A

-A

-A

Signature

5.3.1.2.3.3

P6

Channelisation codes for the message part

The signature in the preamble specifies one of the 16 nodes in the code-tree that corresponds to channelisation codes
of length 16, as shown in Figure 35. The sub-tree below the specified node is used for spreading of the message part.
The control (Q-branch) is spread with the channelisation code of spreading factor 256 in the lowest branch of the
sub-tree. The data part (I-branch) can use any of the channelisation codes from spreading factor 32 to 256 in the
upper-most branch of the sub-tree. However, the system may restrict the set of codes (spreading factors) actually

40

allowed in the cell, through the use of a BCCH message.

SF = 32
SF = 16

SF = 128
SF= 64

SF = 256
Data

Signature 1

Control

.
.
.
Data

Signature 16

Control

Figure 35. Channelisation codes for the random access message part.
Since the control part is always spread with a known channelisation code of length 256, it can be detected by the
base station. The rate information field of the control part informs the base station about the spreading factor used on
the data part. With knowledge of the sub-tree (obtained from the preamble signature) and the spreading factor
(obtained from the rate information), the base station knows which channelisation code is used for the data part.
This structure allows for simultaneous detection of multiple random access messages arriving in the same access slot,
as long as different signatures are used.
5.3.1.2.3.4

Scrambling code for the message part

In addition to spreading, the message part is also subject to scrambling with a 10 ms complex code. The scrambling
code is cell-specific and has a one-to-one correspondence to the the spreading code used for the preamble part. Note
that although the scrambling code is the same for every access slot, there is no scrambling-code collision problems
between different access slots due to the 1.25 ms time shifts between the access slots.
The scrambling codes used are from the same set of codes as is used for the other dedicated uplink channels. The
first 256 codes are used for the random access channel. The generation of these codes is explained in Section
5.3.2.2.2
Scrambling code.
5.3.1.3
5.3.1.3.1

Modulation
Modulating chip rate

The modulating chip rate is 4.096 Mcps. This basic chip rate can be extended to 8.192 or 16.384 Mcps.
5.3.1.3.2

Pulse shaping

The pulse-shaping filters are root-raised cosine (RRC) with roll-off =0.22 in the frequency domain.
5.3.1.3.3

Modulation

QPSK modulation is used.

5.3.2

Downlink spreading and modulation

5.3.2.1
Spreading
Figure 36 illustrates the spreading and modulation for the downlink DPCH. Data modulation is QPSK where each
pair of two bits are serial-to-parallel converted and mapped to the I and Q branch respectively. The I and Q branch
are then spread to the chip rate with the same channelisation code cch (real spreading) and subsequently scrambled by
the same cell specific scrambling code cscramb (real scrambling).

41

cos(t)
I

DPDCH/DPCCH

SP

p(t)
cch

sin(t)

cscramb

p(t)

cch: channelization code

cscramb: scrambling code
p(t): pulse-shaping filter (root raised cosine, roll-off 0.22)

Figure 36. Spreading/modulation for downlink dedicated physical channels.

For multi-code transmission, each additional downlink DPCH should also be spread/modulated according to Figure
36. Each additional downlink DPCH should be assigned its own channelisation code.
5.3.2.2
Code generation and allocation
5.3.2.2.1

Channelisation codes

The channelisation codes of Figure 36 are Orthogonal Variable Spreading Factor (OVSF) codes that preserve the
orthogonality between downlink channels of different rates and spreading factors. The OVSF codes can be defined
using the code tree of Figure 37.
c4,1 = (1,1,1,1)
c2,1 = (1,1)
c4,2 = (1,1,-1,-1)
c1,1 = (1)
c4,3 = (1,-1,1,-1)
c2,2 = (1,-1)
c4,4 = (1,-1,-1,1)
SF = 1

SF = 2

SF = 4

Figure 37. Code-tree for generation of Orthogonal Variable Spreading Factor (OVSF) codes.
Each level in the code tree defines channelisation codes of length SF, corresponding to a spreading factor of SF in
Figure 36. All codes within the code tree cannot be used simultaneously within one cell. A code can be used in a cell
if and only if no other code on the path from the specific code to the root of the tree or in the sub-tree below the
specific code is used in the same cell. This means that the number of available channelisation codes is not fixed but
depends on the rate and spreading factor of each physical channel.
The channelisation code for the BCCH is a predefined code which is the same for all cells within the system.
The channelisation code(s) used for the Secondary Common Control Physical Channel is broadcast on the BCCH.
The channelisation codes for the downlink dedicated physical channels are decided by the network. The mobile
station is informed about what downlink channelisation codes to receive in the downlink Access Grant message that
is the base-station response to an uplink Random Access request. The set of channelisation codes may be changed
during the duration of a connection, typically as a result of a change of service or an inter-cell handover. A change of
downlink channelisation codes is negotiated over a DCH.
5.3.2.2.2

Scrambling code

The total number of available scrambling codes is 512, divided into 32 code groups with 16 codes in each group.
The grouping of the downlink codes is done in order to facilitate a fast cell search, see Section 5.5.2
Cell
search. The downlink scrambling code is assigned to the cell (sector) at the initial deployment. The mobile station
learns about the downlink scrambling code during the cell search process, see Section 5.5.2
Cell search.
The scrambling code sequences are constructed as the position wise modulo 2 sum of 40960 chip segments of two
binary m-sequences generated by means of two generator polynomials of degree 18. Let x, and y be the two
sequences respectively. The x sequence is constructed using the primitive (over GF(2)) polynomial 1+X7+X18 . The y
sequence is constructed using the polynomial 1+X5+X7+ X10+X18 . The resulting sequences thus constitute segments
of a set of Gold sequences.
The scrambling codes are repeated for every 10 ms radio frame.

42

Let n17 n0 be the binary representation of the scrambling code number n (decimal) with n0 being the least
significant bit. The x sequence depends on the chosen scrambling code number n and is denoted xn, in the sequel.
Furthermore, let xn(i) and y(i) denote the i:th symbol of the sequence xn and y, respectively
The m-sequences xn and y are constructed as:
Initial conditions:
xn(0)=n0 , xn(1)= n1 , =xn(16)= n16 , xn(17)= n17
y(0)=y(1)= =y(16)= y(17)=1
Recursive definition of subsequent symbols:
xn(i+18) =xn(i+7) + xn(i) modulo 2, i=0,,218-20,
y(i+18) = y(i+10)+y(i+7)+y(i+5)+y(i) modulo 2, i=0,, 218-20.
All sums of symbols are taken modulo 2.
The definition of the n:th scrambling code word follows as (the left most index correspond to the chip scrambled first
in each radio frame):
Cscramb,n = < xn(0)+y(0), xn(1)+y(1), ,xn(40959)+y(40959) >,
again all symbol sums being modulo 2 additions.
The index n runs from 0 to 511 giving 512 distinct 40960 chip segments of a corresponding Gold code sequence.
The leftmost chip in Cscramb,n corresponds to the first chip in a 10 ms radio frame and the rightmost to the last.
The sign of the I- and Q-branch component is changed if and only if the corresponding chip in Cscramb,n equals 1.
The code generator must be able to generate the sequence shifted arbitrarily from the initial state.
5.3.2.2.3

Synchronisation codes

The Primary and Secondary code words, Cp and {C1,,C17} respectively, consist of pair wise mutually orthogonal
Gold codes of length 256. The Primary SCH is furthermore chosen to have good aperiodic auto correlation
properties. The code sequences are constructed with the help of two binary m-sequences of length 255, x, and y,
respectively. The x sequence is constructed using the polynomial 1+X2+X3+X4+X8 . The y sequence is constructed
using the polynomial 1+X3+X5+X6+X8 .
Before we define the Primary and Secondary code words, we define the set of orthogonal Gold codes.
Let n7 n0 be the binary representation of the scrambling code number n (decimal) with n0 being the least
significant bit. The x sequence depends on the chosen code number n and is denoted xn in the sequel. Furthermore,
let xn(i) and y(i) denote the i:th symbol of the sequence xn and y, respectively
The m-sequences xn and y are constructed as:
Initial conditions:
xn(0)=n0 , xn(1)= n1 , =xn(6)= n6 , xn(7)= n7
y(0)=y(1)= =y(6)= y(7)=1
Recursive definition of subsequent symbols:
xn(i+8) =xn(i+4) + xn(i+3) + xn(i+2) + xn(i) modulo 2, i=0,, 246,
y(i+8) = y(i+6)+ y(i+5)+ y(i+3)+y(i) modulo 2, i=0,, 246.
The definition of the n:th SCH code word follows (the left most index correspond to the chip transmitted first in each
slot):
CSCH,n = < 0, xn(0)+y(0), xn(1)+y(1), ,xn(254)+y(254) >,
All sums of symbols are taken modulo 2.
Note that the code words always start with a constant 0. symbol.
Before modulation and transmission these binary code words are converted to real valued sequences by the
transformation 0 -> +1, 1 -> -1.
The Primary and Secondary code words are defined in terms of CSCH,n and the definition of Cp and {C1,,C17} now
follows as:
Cp= CSCH, 0
and
Ci = CSCH, i , i=1,,17
5.3.2.3
Modulation
1.1.1.1.1

Modulating chip rate

The modulating chip rate is 4.096 Mcps. This basic chip rate can be extended to 8.192 or 16.384 Mcps.
5.3.2.3.2

Pulse shaping

The pulse-shaping filters are root raised cosine (RRC) with roll-off =0.22 in the frequency domain.
5.3.2.3.3

Modulation

QPSK modulation is used.

43

5.4
5.4.1

Radio transmission and reception (FDD)

General

The information presented in this section is based on a chip rate of 4.096 Mcps. Appropriate adjustments should be
made for higher chip rate options.

5.4.2

Frequency bands and channel arrangement

5.4.2.1
Proposed frequency bands for operation
UTRA/FDD is designed to operate in the following paired band:
Table 3. Proposed frequency band for UTRA/FDD
2110 2170 MHz
Mobile station receive
Base station transmit

1920 1980 MHz

Mobile station transmit
Base station receive
Deployment in other frequency bands is not precluded.

5.4.2.2
Carrier spacing
The nominal channel spacing is 5 MHz, but this can be adjusted to optimise performance in particular deployment
scenarios. The channel raster is 200 kHz, which means that the carrier frequency must be a multiple of 200 kHz.
5.4.2.3
TX RX frequency separation
The minimum transmit to receive separation is 130 MHz when operating in the paired band defined in Table 3. If
used in other frequency bands like the American PCS band the minimum separation would be 80 MHz.
5.4.2.4
Variable duplex distance
UTRA/FDD should support a variable duplex distance, i.e. Dduplexer = Fdown Fup is not necessary a constant but is, in
general, allowed to vary within certain limits. The specific limits for the duplex distance applicable for different
frequency bands and terminal classes are yet to be determined.

5.4.3

Service classes

5.4.3.1
Terminal service classes
A number of different service classes will be used to define the data rate and code allocation for a UTRA/FDD
terminal. Possible types of service class profiles are 144 kbps, 384 kbps and 2048 kbps.

5.4.4

Transmitter characteristics

The output power is given in terms of power level at the antenna connector of the equipment. For equipment with
integral antenna only, a reference antenna with a gain of 0 dBi is assumed.
5.4.4.1
Mobile station output power
The mobile station output power profile would be used to define a range of terminal output powers for use in
different system scenarios. The power class would be based on the mobile stations peak power for example 30 dBm.
For mobile station using directive antennas for transmission, a class dependent limit will be placed on the maximum
EIRP (Equivalent Isotropic Radiated Power).
5.4.4.2
Base station output power
The base station output power profile would be used to cater for different system scenarios. The power class would
be based on the peak power specified for the base stations.
5.4.4.3
Output power dynamics
The transmitter uses fast closed-loop Carrier/Interference based power control and slow quality-based power control
on both the uplink and downlink.
Table 4. Output power dynamics for UL and DL
Uplink (UL)

Downlink (DL)

Power control steps

Variable 0.25-1.5 dB

Variable 0.25-1.5 dB

Minimum transmit power

-50 dBm

Power control cycles per second

1.6 kHz

1.6 kHz

Power control dynamic

80 dB

30 dB

] dBm

44

5.4.4.4

Output RF spectrum emissions

5.4.4.4.1

Out of band emissions

The assumed spectrum mask has been derived from simulations on a real wide band amplifier as shown in Figure 38
below. These emission levels will be dependent on the power class and code allocation of the mobile and base
station.
Transmitter masks

Power density [dB/Hz] relative the carrier

10

20

30

40
MS mask
50
BS mask
60

70
10

0
2
frequency [MHz]

10

Figure 38. Assumed spectrum masks.

5.4.4.4.2

Spurious emissions

The limits for spurious emissions at frequencies greater than 250% of the necessary bandwidth would be based on
the applicable tables from ITU-R Recommendation SM.329. Further guidance would be taken from the ERC
recommendation that is currently under progress.
5.4.4.5
Adjacent channel protection (ACP)
Adjacent channel protection (ACP) is the ratio of the transmitted power and the power measured after a receiver
filter in the adjacent channel.
The ACP envisaged for 5 MHz channel spacing is in the order of 35 dB to 40 dB. The possibility is being considered
of dynamically relaxing the ACP requirements for mobile stations under conditions when this would not lead to
significant interference (with respect to other systems or UMTS operators). This would be carried out under network
control, primarily to facilitate reduction in MS power consumption.
5.4.4.6

Occupied bandwidth

The channel bandwidth is less than 5 MHz based on a chip rate of 4.096 Mcps.
5.4.4.7

Frequency stability

The frequency stability for the mobile and base station is indicated in Table 5.
Table 5. Mobile and base station frequency stability.

5.4.5

Mobile station

Base station

3 PPM (unlocked), 0.1 PPM (locked)

0.05 PPM

Receiver characteristics

A Rake receiver or any other suitable receiver structure using coherent reception in both channel impulse response
estimation, and code tracking procedures is assumed.
5.4.5.1
Diversity characteristics
Three forms of diversity are available in UTRA / FDD:

45

Table 6. Diversity characteristics for UTRA/FDD.

Time diversity

Channel coding and interleaving in both up link and down link.

Multi-path
diversity

Rake receiver or other suitable receiver structure with maximum combining.

Additional processing elements can increase the delay-spread performance due
to increased capture of signal energy.
Antenna diversity with maximum ratio combing in the base station and
optionally in the mobile stations. Possibility for downlink transmit diversity in
the base station.

Space diversity

5.4.5.2
Reference sensitivity level
The reference sensitivity for the following services; 8 kbps, 144 kbps, 384 kbps and 2048 kbps are specified in the
link budget template for a number of test environments and multi-path channel classes.
5.4.5.3
BER noise floor level
The BER noise floor level for voice services is significantly less than 10-3 BER. The BER noise floor level for data
services is significantly less than 10-6 BER.
5.4.5.4
Maximum tolerable delay spread
To maintain the voice and data service quality requirements the UTRA/FDD concept allows for a time dispersion
spread suitable for the various propagation models specified in UMTS 30.03 (which contains the models defined in
ITU-R recommendation M.1225).
5.4.5.5
Maximum tolerable Doppler spread
The maximum tolerable Doppler spread is 1000 Hz, which at a 2 GHz carrier frequency corresponds to a maximum
velocity of about 500 km/hr. Parameters determining system performance are not necessarily optimised for this value
of Doppler spread.

5.5
5.5.1
5.5.1.1
5.5.1.1.1

Physical layer procedures (FDD)

Power control
Uplink power control
Closed loop power control

The uplink closed loop power control adjusts the mobile station transmit power in order to keep the received uplink
Signal-to-Interference Ratio (SIR) at a given SIR target.
The base station should estimate the received uplink DPCCH power after RAKE combining of the connection to be
power controled. Simultaneously, the base station should estimate the total uplink received interference in the current
frequency band and generate a SIR estimate SIRest. The base station then generates TPC commands according to the
following rule:
SIRest > SIRtarget,UL TPC command = down
SIRest < SIRtarget,UL TPC command = up
Upon the reception of a TPC command, the mobile station should adjust the transmit power of both the uplink
DPCCH and the uplink DPDCH in the given direction with a step of TPC dB. The step size TPC is a parameter that
may differ between different cells, in the region 0.25 1.5 dB.
In case of receiver diversity (e.g., space diversity) or softer handover at the base station, the TPC command should
be generated after diversity combining.
In case of soft handover, the following procedure is considered:
in the base stations a quality measurement is performed on the received signals; in case the quality measurement
indicates a value below a given threshold, an increase command is sent to the mobile, otherwise a decrease
command is transmitted; all the base stations in the active set send power control commands to the mobile;
the mobile compares the commands received from different base stations and increases its power only if all the
commands indicate an increase value (this means that all the receivers are below the threshold); in case one
command indicates a decrease step (that is, at least one receiver is operating in good conditions), the mobile
reduces its power; in case more than one decrease commands are received by the mobile, the mobile station
should adjust the power with the largest step in the down direction ordered by the TPC commands received
from each base station in the active set.

46

the quality threshold for the base stations in the active set should be adjusted by the outer loop power control (to
be implemented in the network node were soft handover combining is performed).

5.5.1.1.2

Outer loop (SIR target adjustment)

The outer loop adjusts the SIR target used by the closed-loop power control. The SIR target is independently
adjusted for each connection based on the estimated quality of the connection. In addition, the power offset between
the uplink DPDCH and uplink DPCCH may be adjusted. How the quality estimate is derived and how it affects the
SIR target is decided by the radio-resource management, i.e. it is not a physical-layer issue.
5.5.1.1.3

Open-loop power control

Open-loop power control is used to adjust the transmit power of the physical Random-Access channel. Before the
transmission of a Random-Access burst, the mobile station should measure the received power of the downlink
Primary CCPCH over a sufficiently long time to remove effects of the non-reciprocal multi-path fading. From the
power estimate and knowledge of the Primary CCPCH transmit power (broadcast on the BCCH) the downlink pathloss including shadow fading can be found. From this path loss estimate and knowledge of the uplink interference
level and the required received SIR, the transmit power of the physical Random-Access channel can be determined.
The uplink interference level as well as the required received SIR are broadcast on the BCCH.
5.5.1.2
Downlink power control
5.5.1.2.1

Closed loop power control

The downlink closed loop power control adjusts the base station transmit power in order to keep the received
downlink SIR at a given SIR target.
The mobile station should estimate the received downlink DPCH power after RAKE combining of the connection to
be power controled. Simultaneously, the mobile station should estimate the total downlink received interference in
the current frequency band. The mobile station then generates TPC commands according to the following rule:
SIRest > SIRtarget,DL TPC command = down
SIRest < SIRtarget,DL TPC command = up
Upon the reception of a TPC command, the base station should adjust the transmit power in the given direction with
a step of TPC dB. The step size TPC is a parameter that may differ between different cells, in the range 0.25 1.5
dB.
In case of receiver diversity (e.g., space diversity) at the mobile station, the TPC command should be generated after
diversity combining.
5.5.1.2.2

Outer loop (SIR target adjustment)

The outer loop adjusts the SIR target used by the closed-loop power control. The SIR target is independently
adjusted for each connection based on the estimated quality of the connection. In addition, the power offset between
the downlink DPDCH and DPCCH may be adjusted. How the quality estimate is derived and how it affects the SIR
target is decided by the radio-resource management, i.e. it is not a physical-layer issue.

5.5.2

Cell search

5.5.2.1
Initial cell search
During the initial cell search, the mobile station searches for the base station to which it has the lowest path loss. It
then determines the downlink scrambling code and frame synchronisation of that base station. The initial cell search
uses the synchronisation channel (SCH), shown in Figure 39 below (repeated from Section 5.1.2.3.2.3
Synchronisation Channel).
Tslot = 2560 chips

chips
Primary SCH

cp

i,1
Secondary SCH cs

256 chips

cp

cp

csi,2

csi,16

Tframe = 16*Tslot
cp : Primary Synchronization Code
csi,k: One of 17 possible Secondary Synchronization Codes

(csi,1, csi,2, ..., csi,16) encode cell specific long scrambling code group i
Figure 39. Structure of synchronisation channel (SCH).
This initial cell search is carried out in three steps:

47

Step 1: Slot synchronisation

During the first step of the initial cell search procedure the mobile station uses the primary SCH to acquire slot
synchronisation to the strongest base station. This is done with a single matched filter (or any similar device)
matched to the primary synchronisation code cp which is common to all base stations. The output of the matched
filter will have peaks for each ray of each base station within range of the mobile station, see Figure 40. Detecting
the position of the strongest peak gives the timing of the strongest base station modulo the slot length. For better
reliability, the matched-filter output should be non-coherently accumulated over a number of slots.
Matched
filter (cp)

Slot-wise
accumulation

Find
maximum

Timing modulo Tslot

Tslot

Two rays from BSi

One ray from BSj

Figure 40. Matched-filter search for primary synchronisation code to slot synchronisation
(timing modulo the slot length).
Step 2: Frame synchronisation and code-group identification
During the second step of the initial cell search procedure, the mobile station uses the secondary SCH to find frame
synchronisation and identify the code group of the base station found in the first step. This is done by correlating the
received signal at the positions of the Secondary Synchronisation Code with all possible (16) Secondary
Synchronisation Codes. Note that the position of the Secondary Synchronisation Code is known after the first step.
The outputs of all the 17 correlators for 16 consecutive secondary SCH locations are used to form the decision
variables. The decision variables are obtained by non-coherently summing the correlator outputs corresponding to
each 16 length sequence out of the 32 possible sequences and its 16 cyclic shifts giving a total of 512 decision
variables. Note that the cyclic shifts of the sequences are unique (see Section 5.1.2.3.2.2 Secondary Common
Control Physical Channel). Thus, by identifying the sequence/shift pair that gives the maximum correlation value,
the code group as well as the frame synchronisation is determined.
Step 3: Scrambling-code identification
During the third and last step of the initial cell-search procedure, the mobile station determines the exact scrambling
code used by the found base station. The scrambling code is identified through symbol-by-symbol correlation over
the Primary CCPCH with all scrambling codes within the code group identified in the second step. Note that, from
step 2, the frame boundary and consequently the start of the scrambling code is known. Correlation must be carried
out symbol-wise, due to the unknow data of the primary CCPCH. Also, in order to reduce the probability of
wrong/false acquisition, due to combat background noise/interference, averaging the correlator outputs over a
sequence of symbols (diversity) might be required before using the outputs to determine the exact scrambling code.
After the scrambling code has been identified, the Primary CCPCH can be detected, super-frame synchronisation can
be acquired and the system- and cell specific BCCH information can be read.
5.5.2.2
Idle mode cell search
When in idle mode, the mobile station continuously searches for new base stations on the current and other carrier
frequencies. The cell search is done in basically the same way as the initial cell search. The main difference
compared to the initial cell search is that an idle mobile station has received a priority list from the network. This
priority list describes in which order the downlink scrambling codes should be searched for and does thus
significantly reduce the time and effort needed for the scrambling-code search (step 3). Also the complexity in the
second step may be reduced if the priority list only includes scrambling codes belonging to a subset of the total set of
code groups. The priority list is continuously updated to reflect the changing neighbourhood of a moving mobile
station.
5.5.2.3
Active mode cell search
When in active mode, the mobile station continuously searches for new base stations on the current carrier
frequency. This cell search is carried out in basically the same way as the idle mode cell search. The mobile station
may also search for new base stations on other carrier frequencies using the slotted mode, see Section 5.5.4.2.1.1
Slotted mode.

5.5.3

Random access

The procedure of a random access request is:

48

1. The mobile station acquires synchronisation to a base station

2. The mobile station reads the BCCH to get information about:
2.1

The preamble spreading code(s) /message scrambling code(s) used in the cell

2.2

The available signatures

2.3

The available access slots

2.4

The available spreading factors for the message part

2.5

The interference level at the base station

2.6

The primary CCPCH transmit power level

3. The mobile station selects a preamble spreading code/message scrambling code

4. The mobile station selects a spreading factor for the message part.
5. The mobile station estimates the downlink path loss (by using information about the transmited and
received power level of the primary CCPCH), and determines the required uplink transmit power (by
using information about the interference level at the base station).
6.

The mobile station randomly selects an access slot and signature from the available access
slots and signatures.

7.
8.

The mobile station transmits its random access burst.

The mobile station waits for an acknowledgement from the base station. If no
acknowledgement is received within a predefined time-out period, the mobile station starts again from
step 5.

A typical implementation of the base-station random-access receiver for a given preamble code and preamble
sequence is illustrated in Figure 41. The received signal is fed to a matched filter, matched to the preamble code. The
output of the matched filter is then correlated with the preamble sequence. The output of the preamble correlator will
have peaks corresponding to the timing of any received Random-Access burst using the specific preamble code and
preamble sequence. The estimated timing can then be used in a ordinary RAKE combiner for the reception of the
data part of the Random-Access burst.
Preamble correlator
Matched
filter

Peak
detector
Preamble
sequence

Timing
estimator

Ts
RAKE

Figure 41. Base-station Random-Access receiver.

Upon reception of the Random-Access burst, the base station responds with an Access Grant message on the FACH.
In case the Random Access request is for a dedicated channel (circuit-switched or packet) and the request is granted,
the Access Grant message includes a pointer to the dedicated physical channel(s) to use. As soon as the mobile
station has moved to the dedicated channel, closed-loop power control is activated.

5.5.4

Handover

5.5.4.1
Soft handover

Intra-frequency handover

When in active mode, the mobile station continuously searches for new base stations on the current carrier
frequency. During the search, the mobile station monitors the received signal level from neighbouring base stations,
compares them to a set of thresholds, and reports them accordingly back to the base station. Based on this
information the network orders the mobile station to add or remove base station links from its active set. The active
set is defined as the set of base station from which the same user information is sent, simultaneously demodulated
and coherently combined, i.e. the set of base stations involved in the soft handover.
From the cell-search procedure, the mobile station knows the frame offset of the Primary CCPCH of potential softhandover candidates relative to that of the source base station(s) (the base stations currently within the active set).
When a soft handover is to take place, this offset together with the frame offset between the downlink DPCH and the
Primary CCPCH of the source base station, is used to calculate the required frame offset between the downlink

49

DPCH and the Primary CCPCH of the destination base station (the base station to be added to the active set). This
offset is chosen so that the frame offset between the downlink DPCH of the source and destination base stations at
the mobile-station receiver is minimised. Note that the offset between the downlink DPCH and Primary CCPCH can
only be adjusted in steps of one downlink DPCH symbol in order to preserve downlink orthogonality.
Softer handover
Softer handover is the special case of a soft handover between sectors/cells belonging to the same base station site.
Conceptually, a softer handover is initiated and executed in the same way as an ordinary soft handover. The main
differences are on the implementation level within the network. For softer handover, it is e.g. more feasible to do
uplink maximum-ratio combining instead of selection combining as the combining is done on the BTS level rather
than on the BSC level.
5.5.4.2
Inter-frequency handover
In UTRA/FDD the vast majority of handovers are within one carrier frequency, i.e. intra-frequency handover. Interfrequency handover may typically occur in the following situations:
Handover between cells to which different number of carriers have been allocated, e.g. due to different capacity
requirements (hot-spot scenarios).
Handover between cells of different overlapping orthogonal cell layers using different carrier frequencies
Handover between different operators/systems using different carrier frequencies including handover to GSM.
A key requirement for the support of seamless inter-frequency handover is the possibility for the mobile station to
carry out cell search on a carrier frequency different from the current one, without affecting the ordinary data flow.
UTRA/FDD supports inter-frequency cell search in two different ways, a dual-receiver approach and a slotteddownlink-transmission approach.
5.5.4.2.1

Measurements

5.5.4.2.1.1

Slotted mode

With slotted downlink transmission, it is possible for a single-receiver mobile station to carry out measurements on
other frequencies without affecting the ordinary data flow. The principle of slotted downlink transmission is
illustrated in Figure 42.
When in slotted mode, the information normally transmitted during a 10 ms frame is compressed in time. This can be
achieved by:
code puncturing, for lower compression factors,
changing the FEC rate, for higher compression factors.
Note that the idle slot is created without any loss of data as the number of information bits per frame is kept constant,
while the processing gain is reduced by increasing the coding rate. As illustrated in Figure 42, the instantaneous
transmit power is increased in the slotted frame in order to keep the quality (BER, FER, etc.) unaffected by the
reduced processing gain.

One frame
(10 ms)

Idle period available for

inter-frequency measurements

Figure 42. Downlink slotted transmission.

Although Figure 42 shows slotted transmission with a mid-frame idle-period, there are in general three types of
possible slotted transmission mechanisms, as illustrated in Figure 43.
End-of-frame idle period

Mid-frame idle period

Start-of-frame idle period

One frame (10 ms)

One frame (10 ms)

One frame (10 ms)

50

Figure 43. Possible idle period positions.

The default position is the mid-frame idle period. The start-of-frame and end-of-frame idle are supported in order to
be able to create an even longer double-frame idle period, as illustrated in Figure 44.
Double-frame idle period

One frame (10 ms)

Figure 44. Double-frame idle period.

When in slotted mode, slotted frames can occur periodically, as illustrated in Figure 42, or requested on demand.
The rate of and type of slotted frames is variable and depends on the environment and the measurement
requirements.
For UTRA-to-GSM inter-frequency handover considerations, see section 7.2 UTRA - GSM handover.
For services that allows for a larger delay, e.g. data services with interleaving over several frames, multiple frames
can be compressed together in order to create a short measurement slot. As an example, for a 2 Mbps service, with
interleaving of 5 frames (50 ms), a 5 ms idle slot can be created by puncturing only 10% of 5 frames, as illustrated in
Figure 45.

Compressed transmission during

one interleaver span

Figure 45. Multi-frame compressed mode for long-delay services.

5.5.4.2.1.2

Dual Receiver Approach

Mobile terminals equipped with receiver antenna diversity can switch one diversity branch periodically to another
frequency for measurement purposes. This results in a slight loss in diversity capability. Another option is to have a
separate receiver, dedicated for inter-frequency measurement purposes.
5.5.4.3
Paging control
5.5.4.3.1

BS operation

The MSs shall be grouped by a specified method, and paged by each group. At the BS, the corresponding group
number is designated, together with the terminating information that includes the MS ID number that had a
terminating call. The BS shall transmit the terminating information with the MUI part of PCH of the designated
group number.
For the PCH of the group which does not have terminating information:
The BS shall transmit the two PI parts (PI1 and PI2) in the PCH as all 0.
The MUI part shall not be transmitted.
For the PCH of the group which have terminating information:
The BS shall transmit the two PI parts (PI1 and PI2) in the PCH as all 1.
The MUI part shall be transmitted within the same PCH.
5.5.4.3.2

MS operation

The MS shall normally receive only the PI1. The (soft decision) majority decision process of PI1 shall be performed.
- Result of the process equal to 1 with high reliability:
The MUI part of the same PCH shall be received.
- Result of the process equal to 0 with high reliability:
Reception shall be kept OFF until the end of the current superframe.
- Result of the process with low reliability:
PI2 within the same PCH shall be received.
The majority decision process of PI2 shall be performed.
- Result of the process equal to 1 with high reliability:
The MUI part of the same PCH shall be received.
- Result of the process equal to 0 with high reliability:
Reception shall be kept OFF until the end of the current superframe.
- Result of the process with low reliability
The MUI shall be received.

51

When the MUI part is received, the existence of terminating calls for the MS shall be judged based on the
terminating information included in the MUI part.

5.6
5.6.1

Additional features and options (FDD)

Adaptive antennas

Adaptive antennas are recognised as a way to enhance capacity and coverage of the system. Solutions employing
adaptive antennas are already supported in the UTRA/FDD concept through the use of connection-dedicated pilot
bits on both uplink and downlink.

5.6.2

Multi-user detection

UTRA/FDD is designed to work without requiring joint detection of multiple user signals. However, the potential
capacity gains of such receivers in a UTRA/FDD system have been recognised and taken into account in the design
of the concept. In the uplink the possibility to use only short codes facilitates more advanced receiver structures with
reasonable complexity.

5.6.3

Downlink transmit diversity

Transmitter diversity in the downlink provides a means to significantly improve capacity and coverage of
UTRA/FDD, without the requirement for a second receiver chain in the mobile station that receiver diversity would
entail. However, a typical transmit diversity technique, such as delay transmit diversity, has two main drawbacks:
self-interference at locations with good SINR; and the requirement for additional Rake fingers in the mobile receiver.
In order to overcome these drawbacks, diversity schemes have been proposed for UTRA/FDD, that maintain the
orthogonality between diverse downlink transmit antennas, whilst offering significant advantages in the downlink
performance. Simulation results for the proposed techniques have shown a gain of up to 7 dB (compared with the
non-diversity case) for slow speed mobiles in a single path fading environment. In the proposed schemes, the
orthogonality between antennas, is maintained using either code, or time division.
5.6.3.1
Orthogonal Transmit Diversity
Orthogonal Transmit Diversity (OTD) utilises code division transmission diversity. The implementation of OTD is
as follows. Coded bits are split into two data streams and transmitted via two separate antennas. Different
orthogonal channelisation codes are used per antenna for spreading. This maintains the orthogonality between the
two output streams, and hence self-interference is eliminated in flat fading. Note that by splitting the coded data into
two separate data streams, the effective number of channelisation codes per user is the same as the case without
OTD.
The above structure is highly flexible, it may be easily extended to more antennas (4, 8, etc.)
OTD may be an optional feature that can be turned on only if needed. In addition, it is possible to support a mixture
of mobiles with and without OTD capability.
The additional required processing at the mobile station is small. Figure 46 illustrates Rake finger processing with
OTD. It is important to note that the Pilot signal is also split and transmitted on both antennas which allows coherent
detection of the signals received from both antenna. The data is processed using a Rake finger with parallel
processing capability. Both transmitted signal streams are received simultaneously at the same delay (for a given
multipath ray), hence no additional buffering and skewing of data is necessary. This significantly reduces the
hardware complexity/cost associated with OTD implementation.
Pilot integration

Data integration
c1
Scrambling code

Data
combiner

Pilot integration

Data integration
c2

Figure 46. Rake finger processing with OTD.

In the base station transmitter, the base-band processing (i.e. data splitting and separate spreaders) required for OTD
already exists with multi-code transmission in the downlink. From the OTD viewpoint, it is advantageous to employ

52

multi-code transmission for all data rates, and it is also recommended to match the number of codes assigned to the
user with the number of transmit antennas.
5.6.3.2
Time division transmit diversity
Two schemes have been put forward utilising time division transmission diversity for downlink UTRA/FDD mode
operation. The basic Base Station Transmitter block diagram for Time Transmission Diversity is shown in Figure 47.
In time division transmission diversity the signal is switched between antennas in one of two ways. Either, the signal
is switched according to a pattern decided by the base station, or it is switched depending on signalling received
from the mobile station.
Ant 0
RF

FIR

Pilot
TPC
M
U
X

TFI

Spreading

Switch
Ant 1

Data

Encoder

Interleaver
RF

FIR

Figure 47. Base station transmitter block diagram for time division transmission diversity.
5.6.3.2.1

Time Switched Transmission Diversity

Time switched transmission diversity (TSTD) is implemented using the block diagram exactly as shown in Figure
47. TSTD does not assume any change to the UTRA/FDD physical layer channel structure other than switching at
the filter input. There is no change to the channel coding, rate matching, interleaving and spreading within the
UTRA/FDD physical layer description.
TSTD is used for the transmission of downlink Dedicated Physical Channels (DPCHs). All other downlink channels,
i.e. the Common Control Physical Channels (CCPCHs) and the Synchronisation Channel (SCH), are transmitted
from a single antenna, without diversity. TSTD is implemented by transmitting consecutive slots of the downlink
DPCHs through two separate antennas. After scrambling, the spread time slots are switched consecutively to each
antenna (i.e. the baseband signal is switched before modulation is applied, between transmitter antennas, at a rate of
once every 0.625 ms).
The BCCH informs all mobile stations of the corresponding base stations capability for TSTD. The DPDCH and the
DPCCH in the same slot for a given mobile station, are then transmitted from one of the antennas. The next slot of
the DPCH is transmitted from the other antenna. The DPCHs of other users operating in TSTD mode, may have
different switching patterns in order to reduce the peak transmit power and peak to average power ratio in each
power amplifier.
The spread time slots are transmitted to each antenna sequentially as shown in Figure 48.
1 Frame (10 ms)
1.25 ms
0.625 ms
Non-Transmission
Diversity

TS
#0

Tx Antenna
#1

TS
#0

Tx Antenna
#2

TS
#1

TS
#2

TS
#3

TS
#2

TS
#1

TS
#4

TS
#5

TS
#4

TS
#3

TS
#6

TS
#7

TS
#6

TS
#5

Pilot

TS
#9

TS
#8

TS
#7

TPC TFI

TS
#8

TS
#10

TS
#11

TS
#10

TS
#9

Channel coded data

TS
#12

TS
#13

TS
#12

TS
#11

TS
#14

TS
#15

TS
#14

TS
#13

TS
#15

53

Figure 48. Switching pattern for Dedicated Physical Channels in TSTD.

5.6.3.2.2

Selection Transmit Diversity

Selection Transmit Diversity (STD) with fast closed loop control may be used to provide transmit diversity. For
STD, the structure of the Base Station Transmitter is as shown in Figure 49. The implementation of STD is as
follows. In the case of no soft handover, the base station antenna is dynamically selected, based on a fast transmit
antenna selection (AS) control signal, transmitted by the mobile station (similar to fast PC loop). The value of the AS
bit is determined, based on measurements on the antenna specific Primary CCPCH channel. The control loop speed
is 400 Hz (note: the exact AS control loop speed is for further study). In order to guarantee that the mobile station is
decoding the right downlink signal, the pilot symbols of the antennas are selected to be orthogonal with each other.
Ant 0
AS
FIR

RF

Pilot Select
TPC
M
U
X

RI

Spreading

Switch
Ant 1

Data

Encoder

Interleaver
FIR

RF

Figure 49. Selective Transmit Diversity: Base station transmitter block diagram.

5.6.4

Locationing function support

The wideband nature of the UTRA/FDD facilitates the high resolution in position location as the resolution
achievable is directly proportional to the channel symbol rate, in this case chip rate. The duration of one chip
corresponds to approximately 73 meters in propagation distance and if the delay estimation operates on the accuracy
of samples/chip then the achievable maximum accuracy is approximately 18 meters with the 4.096 Mcps chip rate.
Naturally there are then other inaccuracies that will cause degradation to the positioning but 18 meters can be
considered as kind of lower bound on the positioning performance. With higher sampling rate or chip rate the bound
is then naturally even lower.
With the UTRA/FDD concept the position location has been discussed in several ETSI/SMG2 input documents..
One example solution to use is the proposed power up function (PUF) which in the need for a MS to be heard by
several base stations will increase the transmission power over short interval. Other aspects of the position
mechanism are how the issue of actual measurement is done and whether that is based on loop around time or on
Time Difference Of Arrival (TDOA) or other measures.

6.

LAYER 1 DESCRIPTION (TDD MODE)

6.1

Logical channels and physical channels (TDD)

6.1.1

Logical channels

This chapter describes the logical channels that are required for data transfer. Logical channels are unidirectional and
can be divided into two categories:

Traffic channels and

Control channels
6.1.1.1
Traffic Channels
A traffic channel (TCH) is used for transferring user data and/ or layer 3 signalling data.

54

6.1.1.2

Control channels

Control channels carry layer 3 and MAC signalling data and they are also needed for the initial synchronisation of
the mobile station to the fixed radio network. Control channels can be further divided into:

Dedicated control channels and

Common control channels.

6.1.1.2.1

Dedicated Control Channels

Dedicated Control Channels (DCCH) are point to point control channels that carry connection oriented messages.
1.

Associated Control Channel (ACCH) is a point-to-point channel in the uplink or downlink direction. The
allocation of an ACCH is linked to the allocation of a TCH. The ACCH is used for RLC/MAC layer message
transfer, e.g. capacity and link control messages or power control messages.

2.

Stand-alone DCCH (SDCCH) is a point-to-point channel in the uplink or downlink direction. The allocation of
an SDCCH is not linked to the allocation of a TCH, MAC may or may not allocate SDCCH capacity to an MS
dependent upon circumstances. An SDCCH is used for the transfer of layer 3 and RLC/MAC layer messages.

6.1.1.2.2

Common Control Channels

Common Control Channels (CCCH) are point-to-multipoint or point-to-point control channels that carry
connectionless or connection oriented messages.
BCCH Broadcast Control Channel is a point-to-multipoint control channel in the downlink direction. The BCCH is
used for the broadcast of layer 3 and MAC information that describes a cell.
PCH

Paging Channel is a point-to-multipoint control channel in the downlink direction. The PCH is used for the
broadcast of layer 3 paging messages.

RACH Random Access Channel is a contention access uplink channel that can be used by MS to signal a number
of messages, e.g. capacity request when sending access request messages.
FACH the Forward Access Channel is a point-to-point or point-to-multipoint channel in the downlink direction that
is used for the transfer of MAC related signalling messages (e.g. resource allocations) or for the transfering
of small data packets.
SCH

Synchronisation Channel is a point-to-multipoint downlink channel that is used by mobile stations to

acquire frequency, chip and slot synchronisation and to obtain the current frame number in the TDMA
multi-frame structure.

6.1.2

Physical channels

6.1.2.1
Frame structure
In the following sections, an overview about the frame, time slot and code structure is outlined.
6.1.2.1.1

Time slots

The TDMA frame has a duration of 10 ms and is subdivided into 16 time slots (TS) of 625 s duration each. A time
slot corresponds to 2560 chips. The physical content of the time slots are the bursts of corresponding length as
described in section 6.1.2.2
Burst Types.
6.1.2.1.2

TDD Frame

Each 10 ms frame consists of 16 time slots, each allocated to either the uplink or the downlink (Figure 50). With
such a flexibility , the TDD mode can be adapted to different environments and deployement scenarios. In any
configuration at least one time slot has to be allocated for the downlink and at least one time slot has to be allocated
for the uplink.

55

frequency

10 ms
4.096
Mchip/s
625 s

time

Figure 50. The TDD frame structure

Examples for multiple and single switching point configurations as well as for symmetric and asymmetric UL/DL
allocations are given in Figure 51.

Multiple-switching-point configuration (symmetric DL/UL allocation):

10 ms

Multiple-switching-point configuration (asymmetric DL/UL allocation):

10 ms

Single-switching-point configuration (symmetric DL/UL allocation):

10 ms

Single-switching-point configuration (asymmetric DL/UL allocation):

10 ms

Figure 51. TDD frame structure examples

6.1.2.1.3

Spreading codes

Two options are being considered for the bursts that can be sent as described below. Both options allow a high
degree of bit rate granularity and flexibility, thus allowing the implementation of the whole service range from low to
high bit rates.
6.1.2.1.4

Multi-code transmission with fixed spreading

Within each time slot of length 625 s, an additional separation of user signals by spreading codes is used. This
means, that within one time slot of length 625 s, more than one burst of corresponding length as described in
Section 0 can be transmitted. These multiple bursts within the same time slot can be allocated to different users as
well as partly or all to a single user. For the multiple bursts within the same time slot, different spreading codes are
used to allow the distinction of the multiple bursts.
The bursts as described in Section 0 are designed in such a way, that up to 8 bursts can be transmitted within one
time slot, if the bursts are allocated to different users in the uplink. In the downlink or if several bursts in the time
slot are allocated to one single user in the uplink, even more than 8 bursts (e.g. 9 or 10) can be transmitted within
one time slot.

56

6.1.2.1.5

Single code transmission with variable spreading

Within each time slot of 625 s,

a mobile always uses single code transmission by adapting the spreading factor as a function of the data rate. This
limits the peak-to-average ratio of the modulated signal and consequently the stress imposed to the power
amplifier resulting in an improved terminal autonomy. Several mobiles can be received in the same time slot by
the base station, they are separated by their codes and the individual decoding can take profit of the joint
detection.
a base station should broadcast a single burst per mobile again by adapting the spreading as a function of the data
rate. High rate data transmissions requiring more than one timeslot per mobile can be supported by terminals
having the processing power for joint detection on a single slot : the required throughput occupies in a general
way an integer number of slots plus a fraction of an extra slot. Single burst transmission should occur in the
integer number of slots, while the extra slot can be occupied by a burst for the considered mobile plus extra
bursts for other mobiles, joint detection is only needed for this last time slot in the considered mobile.
6.1.2.2
Burst Types
As explained in the section 6.1.2.1.3
Spreading codes, two options are being considered for the spreading.
The bursts presented here are those corresponding to the option introduced in section 6.1.2.1.3.1.
6.1.2.2.1

Traffic bursts

Two types of traffic bursts are defined: The traffic burst 1 and the traffic burst 2. Both consist of two data symbol
fields, a midamble and a guard period. The traffic bursts 1 has a longer midamble of 512 chips than the traffic burst
2 with a midamble of 256 chips. Sample sets of midambles are given in sections 6.1.2.3.1
Sample Midamble
Code Set for Burst Type 1 and 6.1.2.3.2
Sample Midamble Code Set for Burst Type 2.
Because of the longer midamble, the traffic burst 1 is suited for the uplink, where up to 8 different channel impulse
responses have to be estimated. The traffic burst 2 can be used for the downlink and, if the bursts within a time slot
are allocated to less than four users, also for the uplink.
Thus the traffic burst 1 can be used for
uplink, independent of the number of active users in one time slot
downlink, independent of the number of active users in one time slot
The traffic burst 2 can be used for
uplink, if the bursts within a time slot are allocated to less than four users
downlink, independent of the number of active users in one time slot
The data fields of the traffic burst 1 are 61 symbols, i.e. 976 chips long, whereas the data fields length of the traffic
burst 2 is 69 symbols, i.e.1104 chips. The guard period for the traffic burst 1and 2 is 96 chip periods long.
The traffic bursts 1 and 2 are shown in Figure 52 and Figure 53. The contents of the traffic burst fields are described
in Table 7and Table 8.
Table 7. The contents of the traffic burst 1 fields
Chip number

Length of field

Length of field

Length of field

(CN)

in chips

in symbols

in s

0-975

976

61

238.3

Data symbols

976-1487

512

125.0

Midamble

1488-2463

976

61

238.3

Data symbols

2464-2559

96

23.4

Guard period

Data symbols
61 (976 chips)

Midamble
512 chips

Data symbols
61 (976 chips)

Contents of field

GP
96
CP

625 s
Figure 52. Burst structure of the traffic burst 1. GP denotes the guard period and CP the chip periods.

57

Table 8. The contents of the traffic burst 2 fields

Chip number

Length of field

Length of field

Length of field

(CN)

in chips

in symbols

in s

0-1103

1104

69

269.55

Data symbols

1104-1359

256

62.5

Midamble

1360-2463

1104

69

269.55

Data symbols

2464-2559

96

23.4

Guard period

Data symbols
69 (1104 chips)

Midamble
256 chips

Data symbols
69 (1104 chips)

Contents of field

GP
96
CP

625 s
Figure 53. Burst structure of the traffic burst 2. GP denotes the guard period and CP the chip periods.

The two different traffic bursts defined here are well-suited for the different applications mentioned above. It may be
possible to further optimize the traffic burst structure for specific applications, for instance for unlicensed operation.
6.1.2.2.2

Bursts for control channels

The beacon burst for the beacon time slot (downlink) has additional requirements to the midamble and the guard
period, respectively. The RACH performance can be improved by uplink power control. This will ensure that the
beacon measurement data used for the uplink power control algorithm are the most relevant ones.
Hence the beacon burst and thus the beacon time slot is located in time slot 0, whereas the access bursts and thus the
RACH is located for example in time slot 1. The time slots 2 and all following slots up to the switching point are
uplink time slots, the time slots after the switching point are downlink time slots. The location of the beacon channel,
RACH, DL and UL time slots are depicted in Figure 54.
DL

UL

Beacon
CH
TS 0

RACH

DL

..........

..........

TS 1
Switching Point

Figure 54. TDD frame structure with BCCH and RACH

6.1.2.2.2.1

Beacon burst

A mobile should be able to listen to the beacon time slot of his serving BTS and of the BTSs of adjacent cells. The
beacon burst will be sent with high power, e.g. with maximum power.
The beacon burst needs a longer midamble. The longer midamble is necessary due to synchronisation offsets
between BTSs, the synchronisation offset between MS and BTS due to propagation time, the delay spread and the
chip impulse filter. The total sum of time offsets is about 20-25 s.
To detect 8 cells simultaneously the beacon burst midamble has a length of about 200 s.
The time slot 0 is reserved for beacon bursts only. Further time slots can be allocated for beacon bursts, if it should
be necessary.

58

6.1.2.2.2.2

Access bursts

The mobiles send the access bursts randomly in the RACH channel. This leads to time-divided collision groups. The
usage of up to 8 orthogonal codes per time slot increases the amount of collision groups and RACH throughput,
respectively.
A further improvement is achieved by using two distinct access bursts, which can both be transmitted within one
time slot without collision. Access burst 1 uses only the first half of a time slot, access burst 2 the second. Both
access bursts are depicted in Figure 55 and Figure 56, respectively. The contents of the access burst fields are listed
in Table 9 and Table 10.
Table 9. The contents of the access burst 1 fields
Chip Number

Length of field

Length of field

Length of field

(CN)

in chips

in symbols

in s

0-335

336

21

82.0

Data symbols

336-847

512

125.0

Midamble

848-1183

336

21

82.0

Data symbols

1184-1279

96

23.4

Guard period

1279-2559

1280

312.5

Extended guard period

Data

Midamble

Data

GP

Contents of field

Extended GP

312.5 s

312.5 s
625 s

Figure 55. Access burst 1, GP denotes the guard period

Table 10. The contents of the access burst 2 fields

Chip Number

Length of field

Length of field

Length of field

(CN)

in chips

in symbols

in s

0-1279

1280

312.5

Extended guard period

1280-1615

336

21

82.0

Data symbols

1616-2127

512

125.0

Midamble

2128-2463

336

21

82.0

Data symbols

2464-2559

96

23.4

Guard period

Extended GP

Data

312.5 s

Midamble

312.5 s
625 s

Figure 56. Access burst 2, GP denotes the guard period

Contents of field

Data

GP

59

6.1.2.3
Training sequences for spread bursts
As explained in the section 6.1.2.1.3
Spreading codes, two options are being considered for the spreading.
The training sequences presented here are those corresponding to the option introduced in section 6.1.2.1.3.1.

Section 6.1.2.2.1 Traffic bursts contains a description of the spread speech/data bursts. These traffic bursts contain
Lm midamble chips, which are also termed midamble elements. The Lm elements
midamble codes

(k)

mi ; i=1,...,Lm; k=1,...,K; of the

m ( k ) ; k=1,...,K; of the K users are taken from the complex set

V m = {1, j, - 1, - j}.
The elements

(0-1)

( )
( )
mi k of the complex midamble codes m k fulfil the relation

mi( k ) {1, 1}; i = 1,..., Lm ; k = 1,..., K .

m (i k ) = ( j) i mi( k )
Hence, the elements

(k)

mi

of the complex midamble codes

(0-2)

m ( k ) of the K users are alternating real and imaginary.

With W being the number of taps of the impulse response of the mobile radio channels, the Lm binary elements
( )
mi( k ) ; i = 1,..., Lm ; k = 1,..., K ; of (6-2) for the complex midambles m k ; k=1,...,K; of the K users are

generated according to Steiners method from a single periodic basic code

m = m1 , m2 ,..., m Lm +( K 1) W
The elements

mi = mi P

mi {1, 1}; i = 1,...,( Lm + ( K 1)W ).

mi ; i = 1,...,( Lm + ( K 1)W ), of (6-3) fulfil the relation

i = ( P + 1),..., ( Lm + ( K 1)W ).
for the subset

The P elements

(0-4)

mi ; i = 1,..., P, of one period of m according to (6-3) are contained in the vector

m P = (m1 , m2 ,..., mP ) .
T

With

(0-3)

(0-5)

m according to (6-3) the Lm binary elements mi( k ) ; i = 1,...,L m ; k = 1,..., K ; of (8-2) for the midambles

of the K users are generated based on Steiners formula

mi( k ) = mi +( K k )W

i = 1,..., Lm ; k = 1,..., K .

(0-6)

In the following the term a midamble code set or a midamble code family denotes K specific midamble codes

m ( k ) ; k=1,...,K. Different midamble code sets m ( k ) ; k=1,...,K; are in the following specified based on different
periods m P according (6-5).
In adjacent cells of the cellular mobile radio system, different midamble codes sets
to guarantee a proper channel estimation.
As mentioned above a single midamble code set

m ( k ) ; k=1,...,K; should be used

m ( k ) ; k=1,...,K; consisting of K midamble codes is based on a

m P according to (6-5).
In the following several periods m P according Error! Reference source not found. which should be used to generate
single period

m ( k ) ; k=1,...,K; will be listed in tables in a hexadecimal representation. As shown in

Table 11 always 4 binary elements mi are mapped on a single hexadecimal digit.
Table 11. Mapping of 4 binary elements mi on a single hexadecimal digits
different midamble code sets

4 binary elements
-1 -1 -1 -1
-1 -1 -1 1
-1 -1 1 -1
-1 -1 1 1

mi

mapped on hexadecimal digit

0
1
2
3

60

-1 1 -1 -1
-1 1 -1 1
-1 1 1 -1
-1 1 1 1
1 -1 -1 -1
1 -1 -1 1
1 -1 1 -1
1 -1 1 1
1 1 -1 -1
1 1 -1 1
1 1 1 1
1 1 1 1

4
5
6
7
8
9
A
B
C
D
E
F

The mean degradations [2, equation (38)] which serve as a quality information of the periods
5) and hence of the specified midamble code sets
6.1.2.3.1

; k=1,...,K; will be also given.

Traffic bursts) the midamble has a length of Lm=512 , which is

m P according (6-5) for case of burst type 1.

Degradation in dB

m P of length P=456

C482462CA7846266060D21688BA00B72E1EC84A3D5B7194C8DA39E21A3CE12BF512
C8AAB6A7079F73C0D3E4F40AC555A4BCC453F1DFE3F6C82
56F3ACE0A65B96FC326A30B91665BD4380907C2B08DEC98C16A0B0339AEA855C3D
8BDD016E4C3E0F3DA5DF5C0891C851BA30A6C19ABE6C3ED4
1D566C76440333CBF3CA2A405386068E19A2D6A53560CC50138B3A15BF7D9683F95F
66FF096431363E09A514D61099DD3EAD52903BF4A27D14
9A0A349E49389CC184F7A3420D3FBE06B3A40BEE933D8E04E61FAA4A5214D918A1
ADD5BE25D833579FBCF17B422300D0CA1B419393F9722AA8
B760E5694E49169C225A2FBCDACCCA8847F8486A6A351EB7D045BA2271B2A4CB90
0404C0D2BBA00F80F963861BD7DCE748F0F10AE6B785D0F0
ECE93B83CE32E395405F7C889751970E84AFD632500B91E17C4E7846FE68D3C841013
5D3114D3281211214D1F5F1996A6B656259F11728AA52
DE1B6F6219A0AD1A3EB5EEA02173D704C3340AAE7310B93A21BCF979BC7B6C081
7003AA300B1704BCE62524EC48C505977A1570F6C6BA1A2D8
6.1.2.3.2

m P according to (6-

Sample Midamble Code Set for Burst Type 1

In the case of burst type 1 (see section 6.1.2.2.1

corresponding to:
K=8; W=57; P=456
Table 12. Sample Periods
Periods

(k )

0.649471
0.695320
0.705751
0.706513
0.707417
0.708587
0.711320

Sample Midamble Code Set for Burst Type 2

In the case of burst type 2 (see section 6.1.2.2.1

corresponding to:
K=3; W=64; P=192

Traffic bursts) the midamble has a length of Lm=256 , which is

61

Table 13. Sample Periods

m P according (6-5) for case of burst type 2.

Periods of length P=192

D4A124FE4D11BC14C258546A18C5DE0E3AA3F0617245DBFE
48D76A687E21D22321C5201977F620D7A4CB5945F5693A1C
9EEF5E79606DCAAB046769524691E09E816DC688ABC12030
D2369A2B704878F55B58A300C853A2F62233E6207E39F944
A26C7D9697B002714E9285D2AFC3AF1E233FC8C6C7486080
8A615F5D7EE05668415E626482E90B11C95305E4707015B5
5CC2D7409922FA463D2D14377EBCF0CC0E888426B06F0A82
A68238D5BD37B2B4C48B466B9815087898409AFCB804FA0B
6.1.2.3.3

Degradation in dB
0.615566
0.638404
0.663436
0.677739
0.686287
0.686660
0.688977
0.692613

Midamble Transmit Power

In the case of the downlink, 2K data blocks are transmitted in a burst simultaneously. Also in the uplink, if K greater
than one CDMA code are assigned to a single user, 2K data blocks are transmitted in a burst simultaneously by this
user. This is the so called multi-code uplink situation. In the downlink and the multi-code uplink, the mean power
used to transmit the midambles on the one hand and the 2K (or 2K) data blocks on the other hand shall be equal.
This shall be achieved by multiplying the midamble codes
attenuation or an amplification.

6.2
6.2.1

(k )

, k=1,...,K, with a proper real factor to achieve an

Multiplexing, channel coding and interleaving (TDD)

General

This section describes the services multiplexing, channel coding/interleaving and rate matching.
In the UTRA-TDD mode, the total number of basic physical channels (a certain time slot one spreading code on a
certain carrier frequency) per frame is given by the maximum number of time slots which is 16 and the maximum
number of CDMA codes per time slot. This maximum number of codes is 8 in case the different codes within one
time slot are allocated to different users in the uplink and is higher than 8 (e.g. 9 or 10) in the downlink or if several
codes are allocated to one single user in the uplink.
The service classes given in the following represent only a selection of all possibilities which are conceivable.
Two types of traffic bursts are used. They are described in section 6.1.2
Physical channels.

6.2.2

Multiplexing

In a same connection, multiple services could be treated with separate channel coding/interleaving and mapping to
different basic physical channels(slot/code), see Figure 57. In this way QoS can be separately and independently
controlled.
Coding/
interleaving

Parallel
services

Slot #1, Code #1

Coding/
interleaving

Slot #1, Code #2

Coding/
interleaving

Slot #M, Code #N

Figure 57. Service multiplexing (a)

A second alternative is time multiplexing at different points of the channel coding scheme, as shown in Figure 58.

62

Time
Mux

Outer
coding/interl.

Time
Mux

Inner
coding/interl.

Slot/Code #1
Time
Mux

Parallel
services
Time
Mux

Outer
coding/interl.

Time
Mux

Slot/Code #2

Slot/Code #N

Inner
coding/interl.

Figure 58. Service multiplexing (b)

After service multiplexing and channel coding, the multi-service data stream is mapped to one or, if the total rate
exceeds the upper limit for single-code transmission, several resource units.

6.2.3

Channel coding and interleaving

In Real Time (RT) services a FEC coding is used, instead Non Real Time (NRT) services could be well managed
with a proper combination of FEC and ARQ.
For the RT services two levels of QoS (10-3 , 10-6 ) have been considered as examples in Figure 59.
Only convolutional coding is used in case of BER=10-3, while a concatenated code scheme (Reed-Solomon, outer
interleaving and convolutional coding) or Turbo codes could be used to achieve BER=10-6.
BER=10-3

Inner coding
(conv.)

Outer coding
(RS)

BER=10-6

Outer
interleaving

Inner
interleaving

Inner coding
(conv.)

or

Inner
interleaving

Turbo Coding

Figure 59. FEC coding

6.2.3.1
Inner coding/interleaving
The convolutional coding rates change according to the rates of different services. The convolutional coding rates
from 1/4 to 1 have been chosen such that the complete system will be able to use as much as possible the same
decoding structure.
After convolutional coding, interleaving is used. For LDD services, inter-frame interleaving over two 10 ms frames
is applied. For LCD services an interleaving over 300 ms is applied.
6.2.3.2
Outer coding/interleaving
The outer RS coding, on GF(28) has different rate for different services. An outer interleaver to break the error burst
at the output of the Viterbi decoder is needed in addition to an inner interleaver for breaking the error bursts due to
fading.

6.2.4

Rate matching

To map the services on the air interface either puncturing or unequal repetition is used after channel coding.
This rate matching is performed considering both bursts:
burst 1 (long midamble) used in uplink;
burst 2 (short midamble) used in downlink as well as for uplink transmission in the case of multi-code
transmission.

6.2.5

Mapping of logical channels to physical channels

This section describes the way in which logical channels are mapped onto physical resources. A description of the
multi-frame structure is given in Section 6.2.6
Multi-frame structure

63

In the sequel, we use the terms physical channel and resource unit; a physical channel is defined as the association of
one time slot and one frequency. A resource unit is that part of a physical channel that is associated with one
spreading code. A physical channel therefore comprises of up to m resource units where m is the maximum number
of available codes in one time slot.
6.2.5.1
Traffic Channels
A traffic channel is mapped onto one or more sets of slots and codes within a frame. An interleaving period is
associated with each allocation. The frame is subdivided into slots that are available for uplink and downlink
information transfer. Each set of slots and codes over an interleaving period maps to a data unit and a data unit can
correspond to one or more FEC code blocks and one or more RLC protocol data units dependent from the service
being supported. The mapping is illustrated by the following diagram:
RLC PDU

RLC PDU

Coded bits

Coded bits

Frame 1

Frame 2

Frame 3

Figure 60. Mapping of PDU onto the physical bearer

For NRT packet data services an allocation is made only for a relatively short period of time. In general, for RT
services an allocation is made for a certain time period and a release procedure is necessary to release the resource.
For the efficient use of resources the slot/ code set allocated to a radio bearer may be changed from time to time and
the resources allocated to a VBR service may increase or decrease along with the changes in the data rate. Traffic
channels are power controlled, cf. section 6.5.3
Power Control.
6.2.5.2
Control Channels
The Broadcast Control Channel BCCH is transmitted in a predefined slot (e.g. slot 0) called beacon time slot
within particular frames in the multi-frame. All BS transmit their BCCH in the same beacon slot each using one of a
set of 8 systemwide defined beacon codes in the beacon time slot. Each transmission is made on a reference beacon
power level which is in the most cases higher than the allowed cell peak power level. Additional resource units
located anywhere in terms of slots /(non beacon) codes may be used to carry additional (secondary) BCCH
information. In frames where the BCCH is not transmitted then the beacon slot may be used for the transmission of
PCH, SCH and FACH signals. Each BS will use only the codes allocated to it for the beacon slot. It is also possible
to allocate to a particular BS more than one beacon code and to use these additional beacon codes for non BCCH
use, e. g. PCH or FACH. However, this requires a co-ordination in the cluster to avoid the overwriting of beacons of
the neighbouring cells.
A mobile station is therefore able to receive BCCH transmissions from up to seven neighbour base stations in
parallel together with its serving cell. Furthermore the reception of the BCCH bursts (together with those of PCH and
other beacon slot transmissions) will provide the signal strength measurements required for handover detection and
open loop power control.
The Paging Channel PCH can be mapped onto any combination of time slots and codes so that capacity can be
matched to requirements. The location of the PCH is indicated on the BCCH. The PCH has to allow an efficient
DRX. It is always transmitted at the maximum allowed power compatible with its location (beacon power or cell
peak power). Use of the beacon slots, in frames where the BCCH or the SCH is not transmitted, is the preferred
location. In this case each BS will use only the beacon code(s) allocated to it for beacon slot transmission.
The Random Access Channel RACH has an interleaving period of one frame and each transmission occupies only
one burst. To accommodate RACH a single uplink slot is subdivided into two sub-slots each capable of supporting
independent transmissions of one burst. The same slot may be used for RACH by more than one cell. Multiple
transmissions using different codes may be received in parallel. If needed more than one slot may be administrated

64

for the RACH. The location of slots allocated to RACH is indicated on the BCCH. The RACH uses open loop power
control.
The Forward Access Channel can be mapped onto any combination of downlink resource units including the
beacon slot. The location of the FACH is indicated on the BCCH and both, capacity and location can be changed, if
required. All mobile stations receive the FACH every frame. FACH may or may not be power controlled.
The Synchronisation Channel SCH enables a mobile station to synchronise on frequency and on chip and slot
level to and identify the position of the current frame within the multi-frame. A special synchronisation pattern is
used for this purpose. The synchronisation burst is transmitted on the beacon slot following a fixed scheme within
the beacon multi-frame. A mobile station will search for the synchronisation pattern on all beacon codes and after ist
detecting the code the MS can deduce the current frame number.
Each BS transmits once per multi-frame the SCH burst in the beacon slot using its allocated beacon code(s). The
allocation of all eight beacon codes, numbered p = 0, 1, .... 7, to all cells follows a defined reuse pattern so that all
codes are allocated. The SCH on the beacon code number p is transmitted in the frame number 3 * p in the multiframe, so that no two BS simultaneously transmit the SCH in the same frame and no SCH transmission is made in the
frames allocated for BCCH transmission. Thus a MS receiving the 8 BS of a cluster gets one SCH burst every 3rd
frames (30 ms). A MS receiving only one BS gets one SCH burst every 24th frame. To increase this rate, it is
possible to allocate another beacon code to the BS SCH.
An example for the mapping of the beacon slot on different frames is given in the following diagram:
24 frame multi frame
Frame 0

Frame 1

Frame 2

Frame 3

Frame 4

SCH - BS 0

BCCH - BS 0

BCCH - BS 0

PCH - BS 0

BCCH - BS 1

PCH - BS 1

SCH - BS 1

PCH - BS 1

FACH - BS 2

PCH - BS 2

PCH - BS 2

PCH - BS 2

BCCH - BS 3

PCH - BS 3

PCH - BS 3

PCH - BS 3

FACH - BS 4

PCH - BS 4

PCH - BS 4

PCH - BS 4

PCH - BS 5

PCH - BS 5

6
PCH - BS 7
Slot 0

FACH - BS 7

BCCH - BS 5

PCH - BS 5

PCH - BS 6
BCCH - BS 7

FACH - BS 6
PCH - BS 7

Slot 0

Slot 0

FACH - BS

PCH - BS 6

Slot 0

Figure 61. Example of mapping of the common control channels in the beacon slot
The Associated Control Channel ACCH is provided, if required, by allocating capacity within the TCH i.e. by inband signalling. If no TCH is in usage, though an ACCH is required, then the ACCH is provided through the
allocation of a resource unit every nth frame, where n is matched to service needs. An ACCH is not provided for all
services.
The Stand-alone Dedicated Control Channel SDCCH is provided by allocating one or more resource units (slotcode pairs) for one or more frames. In general, an SDCCH will be allocated only for the time required for the
transmission of one message.

6.2.6

Multi-frame structure

A strong requirement for the multi-frame structure comes from the realisation of low cost dual mode FDD-TDD
terminals and from the GSM compatibility of the UTRA proposal. In this respect the super-frame and multi-frame
structure for FDD and TDD mode have to be compatible and harmonised with GSM.
Thus in the proposed structure a multi-frame is composed by 24 frames each of length 10 ms. So the multi-frame
period is 240 ms (twice the GSM TCH-F multi-frame).
All frames in the traffic channel multi-frames are used to carry both user data and dedicated signalling because:
The use of a signalling frame like SACCH frame in GSM is avoided by the use of in-band dedicated signalling
or allocation of a SDCCH.
The most flexible method to distribute different user data blocks such as in-band signalling is under study.
There is no need for an idle slot to read BCCHs of adjacent cells as in GSM

65

Adjacent cells in the TDD network are frame-synchronised and share the common beacon slot. Thus a mobile
can read simultaneously BCCHs from all adjacent cells (provided that they are transmitted on different codes).
The bursty nature of TD-CDMA transmission and reception allows the MS in idle time slots to make
measurements on GSM and FDD networks. This is valid also for high bit rate users (BCCH and RACH slots
could also be used to this purpose)
The multi-frame length is therefore given by the common channel with the lowest bit rate in the present case the
SCH, if its multi-frame structure is compatible with the GSM TCH-F multi-frame. This leads to a multi-frame length
of 240 ms. Three TDD multi-frames match exactly into a FDD multi-frame ensuring the compatibility of both
components.

6.2.7

Examples mapping of services

In the examples shown in this section, Layer 2 signalling is indicated with symbol X. These bits are protected by a
previous block coding and added to data before convolutional encoding. This solution allows the possibility to have
variable signalling throughput.
Different values of X cause different puncturing levels.
6.2.7.1
Speech service
The 8 kbps bearer has been considered as speech service, in which a 160 data bits frame is defined as the elementary
block generated by speech encoder each 20 ms.
Figure 62 shows the channel coding rules to map this service on the two possible bursts.
20 ms

160 bits

CRC (8)

X bits

TAIL (8)

Convolutional (R=1/3)
528+3X bits
(3X+40)

Puncturing

488 bits
QPSK

(3X-24)

552 bits
1 block

244 symb

QPSK

1 block

276 symb
1 basic physical
channel

1 basic physical
channel

Figure 62. Channel Coding for 8 kbps speech service

In the case of fast signalling a whole speech block of 160 bits could be replaced by a signalling block.
The 240 (276) symbols obtained are mapped on 2 frames. The interleaving depth is 20 ms.
6.2.7.2

66

Data service
For data services that have to be protected to achieve P(E)=10-6 a concatenated coding or turbo coding could be
applied In the following examples a Reed-Solomon code, an outer interleaving and a convolutional code are used.
6.2.7.2.1

Data service 144 kbps LCD

A 144 kbps data service is provided coding 1440 bits each 10 ms frame with a Reed Solomon (180/225), then X
signalling bits and 8 tail bits are added.
The data block is convolutional coded with R=2/3 and mapped into 10 basic physical channels in the case of burst 1
or into 9 basic physical channels in the case of burst 2. Puncturing is used to match the data rate after channel coding
to the physical channels.
10 ms

1440 bits

RS code Rate 180/225

1800 bits

TAIL (8)

Convolutional (R=2/3)
2700+3/2X+12 bits
(272+3/2X)

(228+3/2X)
Puncturing

244 bits

276 bits
10 blocks

QPSK

9 blocks

QPSK

122 symb

138 symb
10 basic physical
channels

9 basic physical
channels

Figure 63. Channel Coding for 144 kbps data service

6.2.7.2.2

67

Data service 384 kbps LCD

A 384 kbps data service is provided coding 3840 bits each 10 ms frame with a Reed Solomon (200/245), then X
signalling bits and 3 block of 8 tail bits are added.
The data block is convolutional coded with R=2/3 and mapped into 26 basic physical channels in the case of burst 1
or into 24 basic physical channels in the case of burst 2. Puncturing is used to match the data rate after channel
coding to the physical channels.

10 ms

3840 bits

RS code Rate 200/245

4704 bits

TAIL (24)

Convolutional (R=2/3)
7056+3/2X+36 bits
(748+3/2X)

Puncturing

244 bits

(468+3/2X)

276 bits
26 blocks

QPSK

24 blocks

QPSK

122 symb

138 symb
26 basic physical
channels

24 basic physical
channels

Figure 64. Channel Coding for 384 kbps data service

68

6.2.7.2.3

Data service 512 kbps LCD

A 512 kbps data service is provided coding 5120 bits each 10 ms frame with a Reed Solomon (180/225), then X
signalling bits and 4 block of 8 tail bits are added.
The data block is convolutional coded with R=2/3 and mapped into 36 basic physical channels in the case of burst 1
or into 32 basic physical channels in the case of burst 2. Puncturing is used to match the data rate after channel
coding to the physical channels.

10 ms

5120 bits

RS code Rate 180/225

6400 bits

TAIL (32)

Convolutional (R=2/3)
9600+3/2X+48 bits
(864-3/2X)

(816+3/2X)
Puncturing

244 bits

276 bits
36 blocks

QPSK

32 blocks

QPSK

122 symb

138 symb
36 basic physical
channels

32 basic physical
channels

Figure 65. Channel Coding for 512 kbps data service.

6.2.7.2.4

Data service 2048 kbps LCD

A 2048 kbps data service is provided coding 20480 bits each 10 ms frame with a Reed Solomon (200/210), then X
signalling bits and 16 block of 8 tail bits are added.
The data block is convolutional coded with R=2/3 and mapped into 117 resource units.

69

10 ms

20480 bits

RS code Rate 200/210

21504 bits

Tail (16*8)

Convolutional (R=2/3)
32256+3/2X+192 bits
(3900+3/2X)

Puncturing

244 bits

(3744+3/2X)

276 bits
117 blocks

104 blocks

QPSK

QPSK

122 symb

138 symb
117 basic physical
channels

104 basic physical

channels

Figure 66. Channel Coding for 2048 kbps data service

6.3
6.3.1

Spreading and modulation (TDD)

General

In this chapter, there has been made a separation between the data modulation and the spreading modulation. The
data modulation is defined in Section 6.3.2 Data modulation and the spreading modulation in Section 6.3.3
Spreading modulation.
Table 14. Basic modulation parameters
Chip rate
Carrier spacing
Data modulation
Chip modulation
Spreading characteristics

4.096 Mchip/s
5.0 MHz
QPSK
root raised cosine
roll-off = 0.22
Orthogonal
Q chips/symbol

with Q = 2p, 1<= p <= 4

6.3.2

(0-7)

Data modulation

In this section, symbol rates and durations are defined (section 0) and the mapping of bits onto signal point
constellation is shown (section 6.3.2.2
Mapping of bits onto signal point constellation). Furthermore the pulse
shape for the transmission of the chips is determined (section 6.3.2.3 Pulse shape filtering).

70

6.3.2.1
Symbol rate
The symbol rate and duration are indicated below.
1
= 0.24414 s
Ts = Q. Tc, where Tc = chiprate
In particular for the fixed spreading option, Q=16 and Ts = 3.90625 s. This symbol rate comprise all slots,
including guard periods, midamble sequences, beacon and RACH slots.
6.3.2.2
Mapping of bits onto signal point constellation
In TDD a certain number K of CDMA codes can be assigned to either a single user or to different users who are
simultaneously transmitting bursts in the same time slot and in the same frequency band. The maximum possible
number of CDMA codes, which is smaller or equal to 16, depends on the actual interference situation and the service
requirements. In Section 6.1.2.2.1 Traffic bursts examples of bodies of such spread bursts for Q=16 associated with
a particular user are shown. Each user burst has 2 data carrying parts termed data blocks

d ( k ,i ) = ( d 1( k ,i ) , d (2k ,i ) ,..., d (Nk ,i ) ) T

Data block

i = 1, 2; k = 1,...,K.

(0-8)

d( k ,1) is transmitted before the midamble and data block d ( k ,2 ) after the midamble. Each of the N data
( , )

d nk i ; i=1, 2; k=1,...,K; n=1,...,N; of (6-7) has the symbol duration TS as already given.

symbols

The data modulation is QPSK, thus the data symbols

bl(,kn,i ) {0,1}

( , )

d nk i are generated from 2 interleaved and encoded data bits

l = 1,2; n = 1,..., N; k = 1,..., K; i = 1, 2

(0-9)

using the equation

Re{d (nk ,i ) } =
Im{d

( k ,i )
n

1
(2b1(,kn,i ) 1)
2
1
(2b2( ,kn,i ) 1)
2

}=

(0-10)

n = 1,..., N; k = 1,..., K; i = 1, 2.

Equation (6-9) corresponds to a QPSK modulation of the interleaved and encoded data bits

bl(,kn ,i ) of (6-8).

6.3.2.3
Pulse shape filtering
The pulse shape filtering is applied to each chip at the transmitter. In this context the term chip represents a single
element

(k )
( )
c q with q=1,...,Q; k=1,...,K; of a CDMA code c k ; k=1,...,K see also Section 6.3.3.2 CDMA codes.

The impulse response of the above mentioned chip impulse filter Cr0(t) shall be a root raised cosine. The
corresponding raised cosine impulse C0(t) is defined as

sin
C0 (t ) =

t
TC

TC

t
TC

t2
1 4 2 2
TC
cos

(0-11)

The roll-off factor shall be = 0.22. TC is the sampling time, respective chip duration:

TC =

1
= 0.24414 s
Chiprate

The impulse response C0(t) according to (6-10) and the energy density spectrum C0(f) of C0(t) are depicted in the
figure below:

71

C0(t)

C0(f)
in dB

1.2

10

1
10
0.8
20
0.6

30

40

0.4

50
0.2
60
0
70

0.2

2T

3T
4T
5T
time t (T is the sampling period)

6T

7T

8T

80
1/T

1/2T

0
Frequency f

1/2T

1/T

Figure 67. Basic impulse C0(t) and the corresponding energy density spectrum C0(f) of C0(t)

6.3.3

Spreading modulation

6.3.3.1

Basic spreading parameters

( , )

d nk i of (6-7) is spread with a CDMA code c

Each data symbol

6.3.3.2
The elements

(k )

of length Q=2, 4, 8, 16

CDMA codes
(k)

c q ; q=1,...,Q; k=1,...,K; of the CDMA codes c

(k )

; k=1,...,K; shall be taken from the complex set

V c = {1, j, -1, - j}.

(0-12)

In equation (6-12) the letter j denotes the imaginary unit. The CDMA codes
codes

(k )

are generated from binary CDMA

a ( k ) with elements aq( k ) , q=1,...,Q, k=1,...,K, using the relation

aq( k ) {1, 1}; q = 1,...,Q; k = 1,...,K.

c (qk ) = ( j) q aq( k )
Hence, the elements

( )

c qk of the CDMA codes c

(k )

(0-13)

are alternating real and imaginary. The following table shows

(k )

binary CDMA codes which can used for a ; k=1,...,K, in equation (6-13). These 16 orthogonal binary CDMA
codes are generated based on Walsh-Hadamard codes for Q=16 followed by a multiplication with a Pseudo Random
(PN) sequence. Typically K is smaller than 16 and therefore in equation (6-13) less than 16 binary CDMA codes are
needed. The CDMA codes given in the below table are one example. Other sets of 16 CDMA codes can be
generated by multiplying the 16 orthogonal binary Walsh-Hadamard CDMA codes with other PN sequences. In this
way, different sets of binary CDMA codes can be used in different cells.
Table 15. 16 Binary CDMA Codes
Code 1
Code 2
Code 3
Code 4
Code 5
Code 6
Code 7
Code 8
Code 9
Code 10
Code 11
Code 12
Code 13
Code 14
Code 15
Code 16

(-1 -1 1 1 1 -1 1 -1 1 -1 1 -1 1 1 -1 -1)T
(-1 -1 1 1 1 -1 1 -1 -1 1 -1 1 -1 1 1 1)T
(-1 -1 1 1 -1 1 -1 1 -1 1 -1 1 1 1 -1 -1)T
(-1 -1 1 1 -1 1 -1 1 1 -1 1 -1 -1 1 1 1)T
(-1 -1 -1 -1 -1 1 1 -1 1 -1 -1 1 -1 1 -1 -1)T
(-1 -1 -1 -1 -1 1 1 -1 -1 1 1 -1 1 1 1 1)T
(-1 -1 -1 -1 1 -1 -1 1 -1 1 1 -1 -1 1 -1 -1)T
(-1 -1 -1 -1 1 -1 -1 1 1 -1 -1 1 1 1 1 1)T
(-1 1 -1 1 1 1 -1 -1 1 1 -1 -1 1 1 1 -1)T
(-1 1 -1 1 1 1 -1 -1 -1 -1 1 1 -1 1 -1 1)T
(-1 1 -1 1 -1 -1 1 1 -1 -1 1 1 1 1 1 -1)T
(-1 1 -1 1 -1 -1 1 1 1 1 -1 -1 -1 1 -1 1)T
(-1 1 1 -1 -1 -1 -1 -1 1 1 1 1 -1 1 1 -1)T
(-1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 -1 1)T
(-1 1 1 -1 1 1 1 1 -1 -1 -1 -1 -1 1 1 -1)T
(-1 1 1 -1 1 1 1 1 1 1 1 1 1 1 -1 1)T

72

6.3.3.3

Spread signal of data symbols and data blocks

With the root raised cosine chip impulse filter Cr0(t) the spread signal
symbol

( k ,i )
n

( k ,i )

dn

( ,)

ki
d n (t ) belonging to an arbitrary data

can be expressed as

(t T0 ) = d

( k ,i )
n

(k )
q

Cro (t (q 1)Tc ) = d

( k ,i )
n

q =1

( j)

a q( k ) Cro (t (q 1)Tc ).

In equation (6-14) T0 denotes an arbitrary time shift. The transmitted signal belonging to the data block
7) transmitted before the midamble is

( k ,1)
n

(t ) =

( k ,1)
n

n=1

(0-14)

q =1

(k)
q

Cro (t (q 1)Tc nTc )

d( k ,1) of (6-

(0-15)

q =1

d( k ,2 ) of (6-7) transmitted after the midamble

Q
N
( k ,2 )
( k ,2 )
d n (t ) = d n
c (qk ) Cr0 (t (q 1)TC nTC NQTc Lm Tc ).

and for the data block

n=1

(0-16)

q =1

6.4

Radio transmission and reception (TDD)

6.4.1

Frequency bands and channel arrangement

6.4.1.1
Proposed frequency bands for operation
UTRA/TDD is designed to operate in any frequency band that will accomodate at least one 4,096 Mcps carrier.
6.4.1.2
Carrier raster
The channel raster is 200 kHz.
6.4.1.3
Tx - Rx Frequency Separation
Tx and Rx are not separated in frequency.

6.4.2

Service Class

See relevant chapter for FDD mode

6.4.3

Transmitter characteristics

6.4.3.1
Output power
The mobile station and base station output power profiles would be used to define a range of output powers for the
use in different system scenarios. The power class would be based on the peak power, e. g. 30 dBm for the terminals.
6.4.3.2
Output power dynamics
The transmitter uses fast closed-loop carrier/interference based power control and slow quality based power control
on both the up- and downlink. The step size is variable and in the range 1.5 ...3 dB with 100-800 steps/s The power
control dynamic is 80 dB on the uplink and 30 dB on the downlink.
6.4.3.3
Output RF spectrum emissions, adjacent channel power, occupied bandwidth, frequency
stability
See relevant chapters for FDD mode

6.4.4

Receiver characteristics

The receiver is typically a Joint Detection Receiver. Except for this the relevant chapters for the receiver
characteristic of the FDD system apply also for the TDD system.

6.5
6.5.1

Physical layer procedures (TDD)

Synchronisation of the TDD base stations

For a well working TDD-system synchronisation between BS is required. Because synchronisation is needed only on
frame basis (10 ms), the complexity for HW, SW and signalling is not very high.

73

The synchronisation concept will mainly base on a synchronisation via the existing air interface. It is subdivided into
3 hierarchical layers:
sub area synchronisation (strongly recommended)
main area synchronisation (recommended)
world wide synchronisation (optional)
The synchronisation within a sub area is done either by one beacon base station or by a main area beacon. All the
beacon base stations of a main area are synchronised by one main area beacon. The two described synchronisation
levels are implemented using either terrestrial radio links or physical lines (indoor). Each main/sub area can work as
a synchronised network for itself. The highest level of synchronisation can be implemented by using satellite links,
direct radio links, etc.. The main advantage of this concept is that each sub area can work stand alone if a
synchronisation link is lost. This hierarchical concept can be used by all operators within the sync. coverage area.

main area

sub area

main area beacon

beacon base station
synchronized base station

Figure 68. Synchronisation scenario

6.5.1.1
Synchronisation procedures
BS synchronisation is done by beacon base stations (normal base stations with this additional capability) via the
system radio interface. The synchronisation information is transmitted on a special logical down link channel. It is
mapped on a physical synchronisation burst on a reserved TS every 10th to 1000th frame, which is equivalent to 0.1
to 10 second (capacity loss < 0.1 %). Since the coverage area of the synchronisation information has to be larger
than that of traffic channels of this BS the midamble of the synchronisation burst has to have a high correlation gain
and/or the constraints to the transmitting power have to be weaker (only DL direction).
The propagation delay is no problem for smaller networks. For larger networks the co-ordinates of the BS are well
known. The propagation delay can then be easily calculated by a planning tool.
6.5.1.1.1

Indoor Synchronisation

In indoor environments, the distribution of the synchronisation information can be provided by LAN (e.g. optical
fibres). The distribution of synchronisation information via power lines is for further study.
Another alternative is synchronisation via air interface. The problem of receiving a good synchronisation signal in
modern buildings is the high path loss for outdoor to indoor propagation. So one BS (BS1) close to a window in the
last floor of the building will be synchronised to the outstanding network. BS1 delivers the synchronisation

74

information to a BS in the centre of this floor. This centre BS synchronises all other BS of this floor and additionally
the closest BS in the floor below. The synchronisation of all other floors is done in the same way.
6.5.1.2
Synchronisation procedures of beacon base stations
6.5.1.2.1

Leased Lines

These lines (PCM 24,30) cannot guarantee a constant and known time delay between different BS due to the
repeaters delay along the link.
The problem of the variable delays can be solved by delay measurements with the frame synchronisation word
(FSW). From a central point (e.g. BSC in GSM) a special FSW is sent towards the BS, which is reflecting this FSW
instantaneously. So, the entire delay can be measured and the single delay is calculated by dividing this round-trip by
2. If the forward and backward directions are not equal, the calculated single delay is erroneous.
6.5.1.2.2

Direct Radio Link

The locations of beacon BS are situated in a way that they all have line of sight to the main area beacon. Due to this
fact and because of using directional arials for these links it is expected that no additional transmit power is needed.
6.5.1.3
Synchronisation of main area beacons
6.5.1.3.1

Direct Radio Link

In countries with flat areas (e.g. in Africa) the sync. between main area beacons can be realised by direct radio link
on country specific frequencies.
6.5.1.3.2
6.5.1.3.2.1

Satellite Links
Global Positioning System (GPS)

A cheap solution for synchronisation of the main area beacon is GPS. The main disadvantage of this system is that it
is a military product, for which no one has paid, but the US. Army. So it can not be guaranteed that is always
available.
6.5.1.4
Conclusions
The advantage of this concept is that there is almost no additional effort needed for the realisation of a synchronised
network. Furthermore a synchronised network can provide secondary services:
radio controlled watches, with time zones
Navigation systems
Location Services
This concept allows the cost reduction of the synchronisation system by sharing the beacons between different
operators.
Taking a look at the well working synchronised analogous C-Netz in Germany, Portugal and South Africa shows that
a high increase of capacity is possible. The C-Netz was planned in Germany for 100.000 subscriber, but expanded to
more than 1 million.

6.5.2

Channel Allocation

For the UTRA-TDD mode a physical channel is characterised by a combination of its carrier frequency, time slot,
and spreading code as explained in the chapter on the physical channel structure
Channel allocation covers both:
resource allocation to cells (slow DCA)
resource allocation to bearer services (fast DCA)
6.5.2.1
Resource allocation to cells (slow DCA)
Channel allocation to cells follows the rules below:
A reuse one cluster is used in the frequency domain. In terms of an interference-free DCA strategy a timeslot-tocell assignment is performed, resulting in a time slot clustering. A reuse one cluster in frequency domain does not
need frequency planning. If there is more than one carrier available for a single operator also other frequency
reuse patters >1 are possible.
Any specific time slot within the TDD frame is available either for uplink or downlink transmission. UL/DL
resources allocation is thus able to adapt itself to time varying asymmetric traffic.
In order to accommodate the traffic load in the various cells the assignment of the timeslots (both UL and DL) to
the cells is dynamically (on a coarse time scale) rearranged (slow DCA) taking into account that strongly
interfering cells use different timeslots. Thus resources allocated to adjacent cells may also overlap depending on
the interference situation.

75

Due to idle periods between successive received and transmitted bursts, mobiles can provide the network with
interference measurements in time slots different from the currently used one. The availability of such
information enables the operator to implement the DCA algorithm suited to the network.
For instance, the prioritised assignment of time slots based on interference measurements results in a clustering in the
time domain and in parallel takes into account the demands on locally different traffic loads within the network.
6.5.2.2
Resource allocation to bearer services (fast DCA)
Fast channel allocation refers to the allocation of one or multiple physical channels to any bearer service Resource
units (RUs) are acquired (and released) according to a cell-related preference list derived from the slow DCA
scheme.
The following principles hold for fast channel allocation:
1. The basic RU used for channel allocation is one code / time slot / (frequency).
2. Multi-rate services are achieved by pooling of resource units. This can be made both in the code domain (pooling
of multiple codes within one time slot = multi-code operation) and time domain (pooling of multiple time slots
within one frame = multi-slot operation). Additionally, any combination of both is possible.
3. Since the maximal number of codes per time slot in UL/DL depends on several physical circumstances like,
channel characteristics, environments, etc. (see description of physical layer) and whether additional techniques
to further enhance capacity are applied (for example smart antennas),. the DCA algorithm has to be independent
of this number. Additionally, time hopping can be used to average inter-cell interference in case of low-medium
bit rate users.
4. Channel allocation differentiates between RT and NRT bearer services:
RT services: Channels remain allocated for the whole duration the bearer service is established. The
allocated resources may change because of a channel reallocation procedure (e.g. VBR).
NRT services: Channels are allocated for the period of the transmission of a dedicated data packet only
UDD channel allocation is performed using best effort strategy, i.e. resources available for NRT
services are distributed to all admitted NRT services with pending transmission requests. The number of
channels allocated for any NRT service is variable and depends at least on the number of current
available resources and the number of NRT services attempting for packet transmission simultaneously.
Additionally, prioritisation of admitted NRT services is possible.
5. Channel reallocation procedures (intra-cell handover) can be triggered for many reasons:
To cope with varying interference conditions.
In case of high rate RT services (i.e. services requiring multiple resource units) a channel reshuffling
procedure is required to prevent a fragmentation of the allocated codes over to many timeslots. This is
achieved by freeing the least loaded timeslots (timeslots with minimum used codes) by performing a channel
reallocation procedure.
When using smart antennas, channel reallocation is useful to keep spatially separated the different users in the
same timeslot.

6.5.3

Power Control

Power control is applied for UTRA/TDD to limit the interference level within the system thus reducing the inter-cell
interference level and to reduce the power consumption in the MS.
As mandatory power control scheme, a slow C-level based power control scheme (similar to GSM) is used both for
up- and downlink. Power control is made individually for each resource unit (code) with the following
characteristics:
Table 16. PC characteristics
Dynamic range
Power control rate
Step size
Remarks

Uplink
80 dB
variable; 2-100 cycles / second
2 dB
A cycle rate of 100 means that
every frame the power level is
controlled

Downlink
30 dB
variable; 2-100 cycles / second
2 dB
within one timeslot the powers of all
active codes are balanced to be within
a range of 20 dB

All codes within one timeslot allocated to the same bearer service use the same transmission power.
For RT services, in UL and DL a closed loop power control is used
For NRT services, both open loop power control and closed loop power control are used according to the MS
state and the operators needs
The initial power value is based on the path-loss estimate to the serving BS
In case of one user with simultaneous RT and NRT bearer service, the closed loop power control is used both for
RT and NRT bearer service. However, depending on the current services different power levels are used.

76

Optional enhancements concerning power control for further study:

Introduction of quality based power control

6.5.4

Cell Search

Cell Search is the procedure activated by the MS to find out a suitable BS which it can synchronise to.
Depending on the MS state, the cell search procedure can be performed in one of the following ways:
Initial Mode Cell Search
Idle Mode Cell Search
Active Mode Cell Search
6.5.4.1
Initial Mode Cell Search
The Initial Mode Cell Search procedure is activated by the MS at the power on.
As soon as the MS has been powered on, it tries to find a suitable BS to synchronise to. With Suitable means a BS
broadcasting the identities of the system/network the MS has access rights to and whose reference power level is
detected with the lowest path loss.
From the selected BS, the MS shall derive all the TDD-TDMA timings (i.e. chip, slot, frame configuration, multiframe, super-frame synchronism), the frequency synchronisation, the beacon code and all the system information
which are required to access to the network services.
During the first step of the procedure, the MS scans, over an open time window of 10 ms, for the synchronisation
pattern that a BS transmits on the Synchronisation Channel (SCH). As all the BSs transmit on the SCH each using
one of the eight defined beacon codes, the MS shall search for the synchronisation pattern on all the defined beacon
codes.
From the detection of the auto-correlation pulse, frequency; chip; time slot; frame and multi-frame synchronism and
path loss measurement can be derived.
Taking into account that the SCH is always located on a predetermined time slot of the frame (i.e. on the beacon time
slot), the MS can speed up the locking process by jointly detecting and analysing all the eight beacon codes. Then
the power measurements from several (frame synchronised) base stations can be obtained at the same time.
As a result of this first step, the MS has registered the TDMA timings and the beacon code from the strongest base
station.
In a following step, the MS tries to detect from the Broadcast Control Channel (BCCH) of the locked BS all the
other information (e.g. the switching point synchronism; the system identities; the RACH position etc.) which allows
to access the network services.
For this second step, the closed time window (i.e. a time window centred around the midamble field of the time slot)
is used.
6.5.4.2
Idle Mode Cell Search
The Idle Mode Cell Search procedure is activated by the MS when it is already synchronised to a BS but has no
physical channel allocated, that is when the MS is in the Idle state.
This procedure is activated when the locked BS is detected under a predetermined power threshold, which could also
depend on the link quality.
In the Idle state, the MS has all the TDD-TDMA timings from the locked BS, but still monitors the radio
environment in order to find out a stronger BS.
In order to save power consumption, the MS can perform the radio monitoring periodically.
By receiving the SCH, BCCH of other BS in the beacon time slot (also other common logical channels, for example
the PCH, which all provide a reference power level), the MS learns the information for a possible new cell selection.
Furthermore, the MS monitors all other time slots for interference measurements which are utilised by the BS later
when the MS tries to get into the cell via the RACH mechanism.
When a stronger BS is detected, the MS can lock to this new cell after checking access rights and aligning its TDDTDMA timings (by the detection of the BCCH and SCH).
6.5.4.3
Active Mode Cell Search
The Active Mode Cell Search procedure is activated by the MS when it is already synchronised to a BS with which
at least one physical channel is allocated, that is when the MS is in the Active State.
In the Active state, the MS periodically scans for the radio environment in order to keep updated the list of the
strongest cells, in respect to the serving one, it can detects. The list, which contains the identity and the Received
Signal Strength Intensity (RSSI) of each detected cell, is periodically forward to the serving BS (or it can be
forwarded on demand) and can be used to perform inter-cell handovers of the allocated physical channels.
By receiving the SCH, BCCH of other BS in the beacon time slot (also other common logical channels, for example
the PCH, which all provide a reference power level), the MS learns about its radio environment. Furthermore, the
MS monitors all time slots that are not occupied by the MS for interference measurements being utilised by the BS.

77

6.5.5

Random Access

The MS that needs to access the network services or needs more capacity shall transmit, to the selected BS, a random
access burst on the Random Access Channel (RACH).
The RACH can be positioned in one or more time slots of the uplink part of the frame, as indicated by the Broadcast
Control Channel (BCCH). The random access burst can be accommodated either in the first half or in the second half
of the assigned time slot(s), so that the time slot capacity is doubled. A further improvement of the capacity and, as a
consequence, a further reduction of collisions is achieved by allowing up to eight orthogonal codes per random
access time slot.
The network can regulate the RACH use by allowing separate access groups of MS at a time.
Upon reception of a random access burst, the selected BS shall answer to the MS by sending an access grant message
on the Forward Access Channel (FACH).
This message shall indicate the physical channels/time slots within the cell which are assigned to the MS.

6.5.6

Handover

In general, Hand-Over (HO) is considered as the change of physical channels (both at the radio interface and within
the fixed part of the access network) allocated to a call while maintaining this call. Within the scope of this chapter,
HO procedures are restricted to the procedures executed at the radio interface for inter-cell HO, i.e. HO between
different cells. Intra-cell HO (HO within a cell) can be defined as channel reallocation and is therefore within the
scope of a channel assignment (or channel allocation) procedure.
6.5.6.1
6.5.6.1.1

General matter
Assumptions

1. Cells are assumed to be synchronised on a frame scale basis.

2. If parallel bearer services are established on a single mobile terminal, they are being served by the same cell.
Consequently, a HO is performed for all these bearer services simultaneously.
3. For TDD, due to the slotted use of the frame and the time division between TX and RX, only hard HO is
mandatory. However, the proposal does not prevent the introduction of soft HO. The support of soft HO is for
further study.
6.5.6.1.2

Discrimination between NRT and RT bearer services

For the HO procedure it is reasonable to discriminate between Real-Time (RT) and Non-Real-Time (NRT) bearer
services due to the different requirements on the HO procedure:
RT bearer services have stringent delay requirements but have more relaxed BER requirements. Since HO shall
be seamless, i.e. not noticeable for the user, the HO procedure should cause no extra delay. A seamless HO for
RT services does not imply the need of a loss-less HO.
NRT bearer services have very low (possibly unconstrained) delay requirements but stringent BER requirements.
This implies HO of NRT services shall be loss-less, i.e. no data loss at the expense of a possible delay increase.
6.5.6.2
RT service HO
The basic HO scheme is similar to GSM, i.e. the basic scheme is a mobile assisted, network evaluated and decided,
hard HO using backward signalling. Nevertheless, enhancements are introduced to consider UMTS specific
traffic/service requirements.
6.5.6.2.1

HO initiation criteria and related measurements

HO initiation is based on the following non-exhaustive criteria list:

1. Receive quality of serving cell and neighbour cells
2. Receive level of serving cell and neighbour cells
3. MS-BS distance
4. Power budget
5. MS mobility (estimation of MS speed and direction)
6. Traffic reasons (overload handling)
Measurements performed by MS:
1. Receive level on downlink BCCH channel of serving and neighbour cells
2. Receive level on traffic channels of serving cell
3. Receive quality on downlink traffic and signalling channels of serving cell
The measurements are pre-processed and transmitted periodically to the network. Additionally, the MS can transmit
measurements instantaneously on certain events, e.g. on a fast signal strength decay or if ordered by the network
requesting extra measurement reports.
Measurements performed by BS:
1. Receive quality of uplink traffic and signalling channels
2. Interference level on idle channels

78

3. Estimation of MS speed and direction

Further, the BS maintains up-to-date information about the load situation.
6.5.6.2.2

HO decision

In the basic scheme, HO decision is performed in the network only. HO decision is based on:
Radio related criteria, e.g. path loss, power budget
Traffic and service related criteria, e.g. load situation.
6.5.6.2.3

HO execution

As a basic scheme, hard synchronous backward HO is assumed. This means the HO signalling to the MS is
performed on the old serving cell. Due to the synchronised network a fast HO access to the new cell is performed
ensuring a seamless HO.
6.5.6.3
NRT service HO
NRT bearer services require a loss-less HO. Thus, break duration time is of minor importance whereas an ARQ
protocol on higher network layers guarantees a loss-less data transfer. Since data of NRT bearer services is
transferred in a packet oriented mode over the air interface the following approach is proposed. HO is performed inbetween the transmission of data packets. In general, a HO for packet data services resembles a cell re-selection
process rather than a traditional HO.
In case an MS have been allocated at the same time both RT and NRT services, by default the HO for the RT service
is prioritised over the NRT service, i.e. the NRT service will follow the RT service into the new cell. However other
HO strategies are also possible, e.g. by assigning priorities to specific bearer services.
In case an MS uses only NRT services, a forward mobile evaluated (MEHO) HO with background control from the
network utilising broadcast HO parameters is proposed. For enabling the network to control NRT HO immediately, a
mobile assisted, network evaluated HO as an option for further study is considered as well.
6.5.6.3.1

HO initiation criteria and related measurements

HO initiation is based on the following criteria:

1. Receive level of serving cell and neighbour cells
2. Receive quality on traffic and/or signalling channels of serving cell
3. MS-BS distance
4. Power budget
5. MS mobility (estimation of MS speed and direction)
6. Traffic reasons (overload handling)
Measurements performed by MS:
1. The MS measures the receive level of the serving cell and neighbour cells on the downlink BCCH channel.
2. Receive quality of downlink traffic and signalling channels of the serving cell can be measured in terms of
BER/FER or estimated by ARQ repeat counters.
3. MS sojourn time in a cell.
6.5.6.3.2

HO decision

For NRT services the HO procedure is in fact a cell re-selection procedure which can be performed between two
successive packet transmissions. The cell re-selection is governed by:
1. Path-loss criterion
2. Cell priority
3. Routing area
4. MS mobility.
6.5.6.3.3

HO execution

The cell re-selection procedure is a forward type of HO, i.e. signalling channels are established in the new cell before
the next data packet starts.
Optionally, the network has the possibility to command a MS to perform a cell re-selection to a specific cell, e.g. for
load regulation purposes. This command overwrites the MS decision.
6.5.6.4
Intra-frequency HO and Inter-frequency HO
Intra-frequency HO is a HO between cells using the same (single) radio frequency whereas inter-frequency HO is a
handover between cells using different radio frequencies.
Different radio carriers are used to cover hot-spot scenarios or to offer a Hierarchical Cell Structure, or to allow for a
co-ordinated multi-operator deployment scenario. In any case cell synchronisation on a frame scale basis is assumed.
The inter-frequency HO procedure follows the same way and the same rules as the intra-frequency HO procedure,
with the following main exceptions:

79
The MS should be capable of performing radio measurements and cell identification during idle time slots by
listening to carrier frequencies other than the active one.
The inter-frequency HO is always performed as a hard handover.

6.6
6.6.1

Additional features and options (TDD)

Joint detection

Joint detection of simultaneously active CDMA codes in the uplink as well as the downlink will already be
performed in the introductory phase of the UTRA TDD mode. Therefore, this subject is treated in other sections of
this system description.

6.6.2

Adaptive antennas

In the UTRA TDD-component, adaptive antennas are supported through the use of connection dedicated midamble
sequences in both uplink and downlink (they are optional in the downlink). Moreover, the reciprocity between the
uplink and the downlink channel facilitates an efficient implementation of smart antennas. Although the UTRA TDD
component does not require the use of smart antennas, the resulting signal-to-interference-plus-noise-ratio (SINR)
can significantly be improved by incorporating various smart antenna concepts at the base station on the uplink as
well as the downlink.
These SINR gains may be exploited
to increase the capacity,
e.g., by reducing the amount of interference suffered (BS receiver) and created (BS transmitter) in the
system
to increase the quality,
to decrease the delay spread,
to reduce the transmission powers,
to reduce the electromagnetic pollution and user health hazards,
to enhance spatial user location due to the estimation of the dominant directions of arrivals,
or a combination thereof. Three different smart antenna concepts, namely
diversity antennas,
sector antennas,
and adaptive antenna arrays,
can be incorporated into the UTRA TDD mode.

6.6.3

Downlink transmit diversity

Downlink transmit diversity is supported by the UTRA TDD mode.

6.6.4

Locationing function support

The proposed BCCH concept and the fact that the base stations in a local area are synchronised facilitate the
implementation of mobile positioning algorithms in the UTRA TDD mode. Using the midambles transmitted during
every beacon slot, each mobile station can estimate the channel impulse response vectors corresponding to several
(up to seven) adjacent base stations. Thereby, time delay or delay difference measurements to these base stations are
obtained in a very efficient fashion. They are required as an input for mobile positioning algorithms.
If, moreover, an antenna array is located at a BS, estimates of the corresponding dominant directions of arrival at this
BS can also be obtained. Including these estimated arrival angles in the mobile positioning scheme will enhance the
accuracy of the resulting mobile positioning estimates significantly.

6.6.5

ODMA

6.6.5.1
Relaying and ODMA
The UTRA TDD mode is a suitable platform for the support of relaying. Relaying is a widely used technique for
radio packet data transmission both in commercial and military systems but it has so far not been widely used in
cellular systems. Relaying has the potential among others
to improve coverage and/or maximum user bit rates by reduced effective path loss, optimum link adaptation
and link diversity and
to increase capacity by lowering transmission powers and associated inter-cell interference.
The UTRA TDD design is sufficiently flexible to support both simple relaying and advanced relaying protocols such
as Opportunity Driven Multiple Access (ODMA) with negligible increase to the MS complexity or cost.
ODMA supports packet data transfer between an origin and destination via a network of intermediate relay nodes
(dedicated fixed relays or relaying enabled mobiles). TDD operation enables each node to receive other nodes
transmissions and build a connectivity table neighbours at each node exploiting path loss and delay information to.

80

This table is subsequently used to route packets across a network in a dynamic manner without incurring a significant
routing overhead.
6.6.5.2
Radio-Resource Organisation and Synchronisation
ODMA relaying requires MS to MS transmission allowing information to be sent from one mobile to another
without passing via BS. Each MS can receive broadcast-signalling information over a large cell area. Reception of
the broadcast information will allow frequency, chip and slot/frame synchronisation and determine connectivity/path
loss to the BS. The BCCH will also indicate which physical channels are available for conventional use and which
channels are reserved for MS-MS transmission. The MS-MS communications may use a different unpaired
frequency channel to the one generating the BCCH. In fact it may be feasible for the broadcast cell to be FDD. The
BS common channels will also be used for initial authentication and mobile location. An additional advantage of
receiving a BCCH is that it avoids violating any RX before TX regulations which may apply to the mobile.
Figure 69 shows an example how conventional TDD and MS-MS can be incorporated in the same frame structure.
The MS-MS resources are sub-divided into Calling and Traffic Channels. The Calling Channel is RACH like i.e.
random access with collision risk and the Traffic Channels are used for MS-MS transfers after negotiation on the
Calling Channel(s). Traffic Channels are preferably full time slots seized exclusively for one MS-MS
communication. Multi-code transmission is used to achieve high throughput if necessary to avoid excessive delay
supporting the ODMA operation. Use of higher order modulation would further assist even higher rate transmission.

10 ms
MS-MS Resource

Normal TDD
0

9 10 11 12 13 14 15

frequency
625 s

MS Traffic Channels

4.096
Mchip/s

Switch Point
time

MS Calling Channel

Figure 69. Example TDD Frame Structure Incorporating MS-MS Resources

An ODMA enabled MS will behave such that in MS-MS reserved slots it will be capable of receiving unless it has
something to transmit and then it should be capable of transmitting in any of the MS-MS slots.
A MS can choose to just monitor the Calling Channel slot until it determines a need to also use Traffic Channels.
This may be triggered by detecting its own address in a message field or by a requirement to source/sink data. To
access a Calling Channel it is proposed that the MS transmits a type 1 or 2 traffic burst using a randomly chosen
code. With joint detection the MS can simultaneously receive multiple signals with up to 30dB power difference and
thus resolve collisions. It is assumed that MS-MS transmissions take place over micro-cellular range. Thus, the
channel estimation can cope with the slight asynchronism between MS synchronised to the central BS.
The access to the traffic channels is based on a dynamic channel selection scheme based on interference
measurements by the seizing MS. The seizure of channels by MS can be indicated in the calling channel burst
structure as described below under addressing.
6.6.5.3
Idle-Mode Procedure
Each mobile requires some knowledge about other mobiles that it may communicate with and their relative
connectivity. How it acquires this information could be implementation specific. For example an ODMA system
would generate probe (RACH-like) signals to determine its neighbours and find end to end addresses. Probing is a
mechanism used to indicate mobile activity in the ODMA network. When a mobile station is switched on for the first
time it has no information about its surroundings. In this case the mobile will camp on one of the MS-MS CCH (after
establishing synchronisation with the central BS) which are used by all mobile stations to receive and broadcast
probes. With no ODMA system information stored in memory, the MS will begin a probing session, where the
mobile initially camps on a CCH and periodically broadcasts a probe packet. The neighbour list will initially be
empty. If another MS receives the broadcast packet it will register the probing MS as a neighbour and sends an

81

addressed probe in response. The response probe is transmitted at random to avoid contention with other mobiles
and typically one response is sent for every n broadcast probes received from a particular MS. The probe-response
mechanism enables each MS to build a neighbour list that should contain at least 5 MS.
When using the probing approach initially there is no connectivity information and so the probe power must start low
and ramp up until the required number of neighbouring MSs are determined. Additionally link adaptation
mechanisms could be used for setting the local connectivity area to contain at least 5 other MS. Probe
acknowledgements will appear on the Calling Channel (RACH-like). The acknowledgement will contain information
to help refine the power control.
If the probing mechanism is allowed to occur at any time the MSs must RX continuously which may reduce battery
life. To avoid this, a low duty cycle probing window co-ordinated by BS broadcast information can be used, i.e. the
sleeping MSs wake up periodically to send and receive probes (e.g. every minute) and then go back to sleep. The
window could be of the order of 0.5 seconds long. The BS has the capability to send a wakeup page to all the MSs
via the BSs paging channel. A sleeping MS that is then paged awake will stay active whilst it can detect local
ODMA transmissions. If it has not participated in such communication for a timeout period it will fall asleep.
Similarly it may decide to sleep after a long period of activity.
An alternative approach if feasible would be for some central intelligence to determine where all the mobiles are
located, their relative connectivity and somehow pass this information in an efficient manner to the MS.
Other MSs monitoring the probe/acks will determine connectivity between the nodes and themselves and refine their
own knowledge for future communications.
6.6.5.4
Addressing
There are 2 types of addressing to be considered, Relay-Relay and End-End i.e. the former manages a particular
relay hop and the latter identifies the origin and destination of the relayed transmission - within the cell. Note that all
messages will have a BTS as one end of a transmission - and so a BTS should have a special generic address e.g. 0.
It is assumed that each mobile has some unique end-to-end address e.g. MSISDN. The MSISDN should not be used
to address MS-MS transmissions, as these fields are unencrypted (or use encryption common to the cell). When a
MS registers onto the network it may be given a temporary identity (like a TMSI) which can be used for relaying
purposes. For efficiency the size of this identity (or derived version for relaying) should be kept to a minimum.
The probe information mapped onto a traffic burst and transmitted on a Calling Channel contains in the first half of
the transmission a message independent header and enables a relay transmission to be identified. The header reveals
source and destination addressing, link quality and power control parameters and which resources (Traffic Channels)
to be used next. The second part of the burst is message type dependent consisting of message type, source and
destination and flow information as message number, creation time, time to die and time elapsed.
6.6.5.5
Call Set-up
When a MO wishes to start a call it makes a conventional RACH access to the BS. A conventional
authentication/call set-up will take place but during the negotiation of resource it will be decided to use ODMA
mode. Firstly the BS will send a broadcast wakeup page to the MS relays. The BS will then ask the originating MS to
send a message to it via ODMA relaying which it then acknowledges. The initial route for these messages will be
based on knowledge acquired from the background probing. Alternatively, the BS could indicate the route to be used
to the mobile. The transmissions will be monitored by relays not directly involved in the link. These relays then
determine connectivity routes between the MO and BTS and are available to make further transmissions more
optimum and reliable. Other mobiles will fall asleep using the page-awake rules. A similar procedure is used for MT
calls.

6.7
6.7.1

System scenarios
Uncoordinated operation

A system requirement for uncoordinated residential operation is that systems can be bought and installed
independently. The reference points for power control will be different for the different systems and their spatial
separation can be arbitrarily small. Also, time synchronisation is very difficult to obtain leading to the loss of
orthogonality in the code domain not only in the uplink but also on the downlink. For these reasons it is has been
considered a requirement that time orthogonality is achieved between residential systems operating in close
proximity.

82

The consequence is that contrary to public systems which are synchronised and seek to maximise the interleaving
gain and hence performance and capacity, residential systems need to occupy as few slots as possible. In this way,
the scope for interference avoidance increases and more systems can be accommodated.
The unsynchronised base stations, upon installation and in periodic intervals thereafter, measure interference on all
slots and transmit the common control slot in the optimal position with regard to the slots in the frame used by other
systems.
1

System 1 :DL 1slot

:UL 1slot
1

System1 :DL 2slots

:UL 1slots

System 1 :DL 4slots

:UL 1slot
1

System 1 :DL 2slots

:UL 2slots

1 frame=10ms
Figure 70. Frame structure and timing relation example for more than one system and several asymmetry patterns.
The numbers in the grey boxes differentiate the common control channels of the different systems.

7.
7.1

INTEROPERABILITY
UTRA/FDD - UTRA/TDD handover

For terminals with both FDD and TDD capability the handover between the UTRA modes can be used. Both modes
use the same 10 ms frame length and can perform measurements on each other. The UTRA FDD mode can use the
slotted mode or other measurement ways described in Section 5.5.4.2.1.1
Slotted
mode
to
perform
measurements on the UTRA TDD mode. The UTRA FDD mode must search first the downlink activity part(s) in the
10 ms frame. As the UTRA TDD cells within the area are frame synchronised, the downlink/uplink timing obtained
for a single TDD cell is also valid for other cells belonging to the same network in the same area.
For the UTRA TDD mode, measurement time can be obtained between the activity periods (between
uplink/downlink transmission) to facilitate sufficient measurement frequency from UTRA FDD cells.
In the FDD mode, the mobile is continuously transmitting and receiving information. In order to perform a handover
to the TDD mode, it should be able to make measurements on TDD carriers. However, the spectral separation
between FDD carriers and TDD carriers may not be sufficient in some cases to be able to implement a filter to
protect the TDD receiver making the measurements. Therefore, the mobile might need to interrupt FDD transmission
in order to perform measurements in the TDD band. This can be implemented through a slotted mode in the uplink
direction similar to the one defined for the downlink transmission.
For both modes it is expected that the UTRA base station is able to indicate the channel numbers used for the FDD
and TDD cells in the area as well as the base station spreading/scrambling codes used. This does not cover the
unlicensed TDD use where handovers are not likely to happen as the networks are not likely to be inter-connected.

7.2

UTRA - GSM handover

The handover between UTRA and GSM system offering world-wide coverage already today has been one of the

83

main design criteria taken into account in the UTRA frame timing definition. The GSM compatible multi-frame
structure, with the super-frame being multiple of 120 ms, allows similar timing for inter-system measurements as in
the GSM system itself. The compatibility in timing is important, that when operating in UTRA mode, a multi-mode
terminal is able to catch the desired information from the synchronisation bursts in the synchronisation frame on a
GSM carrier with the aid of the frequency correction burst. This way the relative timing between a GSM and UTRA
carriers is maintained similar to the timing between two asynchronous GSM carriers.

7.2.1

UTRA/FDD to GSM handover

UTRA/FDD-GSM dual mode terminals can be implemented without simultaneous use of two receiver chains.
Although the frame length is different from GSM frame length, the GSM traffic channel and UTRA FDD channels
use similar 120 ms multi-frame structure. Similar timing can be naturally done with UTRA TDD mode as well.
A UTRA terminal can do the measurements either by requesting the measurement intervals in a form of slotted mode
where there are breaks in the downlink transmission or then it can perform the measurements independently with a
suitable measurement pattern. Independent measurements do not use slotted mode, but use dual receiver approach,
where the GSM receiver branch can operate independently of the UTRA FDD receiver branch.
For smooth inter-operation between the systems, information needs to be exchanged between the systems, in order to
allow UTRA base station to notify the terminal of the existing GSM frequencies in the area. Further more integrated
operation is needed for the actual handover where the current service is maintained, taking naturally into account the
lower data rate capabilities in GSM when compared to UMTS maximum data rates reaching all the way to 2 Mbits/s.
Measurements of GSM using slotted mode
6 ms idle periods (similar to that of GSM) can be created by using double-frame idle periods, as described in Section
5.5.4.2.1.1
Slotted mode. Therefore, it is possible to capture the GSM FCCH and SCH in the same way as in
GSM-to-GSM handover. The GSM Frequency Correction Channel (FCCH) and GSM Synchronisation Channel
(SCH) use one slot out of the eight GSM slots in the indicated frames with the FCCH frame with one time slot for
FCCH always preceding the SCH frame with one time slot for SCH. The principle is indicated in Figure 71.
GSM
FCCH & SCH
FCCH (Frequency Correction CH)

SCH (Synchronisation CH)

GSM TCH
UTRA FDD

GSM Idle Frame

6 ms UTRA FDD Idle Period

Figure 71. Example of GSM measurement timing relation between UTRA/FDD and GSM frame structures.
Alternatively, several shorter mid-frame idle periods (as described in Section 5.5.4.2.1.1 Slotted mode) with a certain
spacing and every GSM super-frame, can be used to capture the GSM FCCH and SCH. For instance, two 3 ms idle
periods every 120 ms, offset from each other by 30 ms, as illustrated in Figure 72.
GSM
FCCH & SCH
FCCH (Frequency Correction CH)

SCH (Synchronisation CH)

GSM TCH
UTRA FDD

GSM Idle Frame

3 ms UTRA FDD Idle Period

Figure 72. Another example of measurement timing relation between UTRA/FDD and GSM frame structures.

84

For the power measurements of GSM carriers, additional slotted frames will be used for single receiver FDD/GSM
mobiles. Requirements concerning the number of power measurements per slotted frame are for further study.

7.2.2

GSM to UTRA/FDD handover

The GSM system is likewise expected to be able to indicate also the UTRA FDD base station scrambling codes in
the area. This will make the cell identification simpler and after that the existing measurement practices in GSM,
between the slots or during idle slots, can be used for measuring the UTRA FDD mode when operating in GSM
mode.
As the UTRA FDD does not rely on any super-frame structure as with GSM to find out synchronisation, the terminal
operating in GSM mode is able to obtain UTRA FDD BS frame synchronisation once the UTRA FDD base station
scrambling code timing is acquired. The BS scrambling code has 10 ms period and is synchronised to UTRA FDD
common channels in the frame timing.

8.

85

ANNEX A -- ANSWERS TO TECHNOLOGY TEMPLATE

A1.1

Test environment support

A1.1.1

In what test environments will the SRTT operate?

Answer:
Indoor office (I), Outdoor to indoor and pedestrian (P), Vehicular (V), and Mixed-cell
pedestrian/vehicular (M)

A1.1.2

If the SRTT supports more than one test environment, what test environment does this
technology description template address?
Answer:
The template addresses all four test environments listed in A.1.1.

A1.1.3

Does the SRTT include any feature in support of FWA application? Provide detail about impact
of those features on the technical parameters provided in this template, stating whether the
technical parameters provided apply for mobile as well as for FWA applications.
Answer:
The proposal can be used for FWA applications. There are no differences in the radio
transmission technology parameters with respect to FWA than what is being used for the
cellular applications. The flexibility of the RTT allows for an optimisation of the transmission
and receiver chains according to the specific deployment scenario such as cellular or FWA.
Indeed, the RTT is designed to be future proof taking advantage of extended range technologies
such as adaptive antennas and antenna diversity in the downlink but also interference
cancellation techniques. It is also possible to include those kinds of techniques later, if
necessary, without requiring any frequency reconfiguration nor does it preclude the use of user
equipment not supporting those techniques like antenna diversity in the downlink.

A1.2

Technical parameters
Note: Parameters for both forward linear and reverse link should be described separately, if
necessary.

A1.2.1

What is the minimum frequency band required to deploy the system (MHz)?
Answer:
FDD mode: 25 MHz
TDD mode: 15 MHz
With these spectrum allocations, up to 2 Mbps user rate is possible. However, note that these
are the minimum spectrum requirements. Larger spectrum allocation is recommended for more
efficient operation. A larger spectrum allocation supporting two or more 5 MHz carriers would
e.g. allow for more efficent trunking or multiple cell layers.

A1.2.2

What is the duplex method: TDD or FDD?

Answer:
Both FDD and TDD modes are specified.

A1.2.2.1

What is the minimum up/down frequency separation for FDD?

Answer:
130 MHz for the UMTS/IMT-2000 band. A different minimum up/down frequency separation
may be applied for other frequency bands, e.g. for the American PCS band a separation of 80
MHz would apply...

86

A1.2.2.2

What is the requirement of transmit/receive isolation? Does the proposal require a duplexer in
either the mobile or base station.
Answer:
FDD mode: Duplexer needed in mobile station. Required transmit/receive isolation: 50 dB
(MS), 80 dB (BS). The required isolation in the BS can be achieved by separating the
transmitter and receiver antennas together with an appropriate receiver filter.
TDD mode: No duplexer needed.

A1.2.3

Does the RTT allow asymmetric transmission to use the available spectrum? Characterize.
Answer:
In both TDD and FDD modes asymmetric connections can be supported since it is possible to
set uplink and downlink bearer service characteristics independently.
In the FDD mode:
On an overall system level, it is possible, with the FDD mode to assign more carriers to the
downlink than uplink or vice versa.
In TDD mode:
The ratio of uplink to downlink capacity of a carrier can be adjusted by changing the ratio of
the number of uplink and downlink time slots within the frame.

A1.2.4

What is the RF channel spacing (kHz)? In addition, does the SRTT use interleaved frequency
allocation?
Note: Interleaved frequency allocation; allocating the 2nd adjacent channel instead of adjacent
channel at neighboring cluster cell is so called "interleaved frequency allocation". If a
proponent is going to employ this allocation type, proponent should be stated at A1.2.4 and fill
A1.2.15 of protection ratio for both of adjacent and 2nd adjacent channel.
Answer:
The RTT uses an RF channel raster of 200 kHz.
A fixed RF channel spacing is not defined. The carrier spacing can be flexibly chosen,
typically in the range
4.2-5.0 MHz (for 4.096 Mcps carrier) depending on the specific deployment scenario.
The SRTT does not use interleaved frequency allocation.

A1.2.5

What is the bandwidth per duplex RF channel (MHz) measured at the 3 dB down points? It is
given by (bandwidth per RF channel) x (1 for TDD and 2 for FDD). Please provide detail.
Answer:
FDD mode: 8.2 MHz, (16.4 MHzand 32.8 MHz for higher chip rates which are not yet
described)
TDD mode: 4.1 MHz

A1.2.5.1

Does the proposal offer multiple or variable RF channel bandwidth capability? If so, are
multiple bandwidths or variable bandwidths provided for the purposes of compensating the
transmission medium for impairments but intended to be feature transparent to the end user?
Answer:
The basic chip rate of the RTT is 4.096 Mcps corresponding to a channel bandwidth of
approximately 5 MHz. Additional chip rates 8.192 Mcps and 16.384 Mcps, corresponding to
bandwidths of approximately 10 MHz and 20 MHz respectively, are also specificied for the
FDD mode. These bandwidths are seen as future evolution of the RTT towards even higher
user rates (>2 Mbps). The different bandwidths are not used to compensate for transmission
medium impairments. The different bandwidths are transparent to the end user.

87

A1.2.6

What is the RF channel bit rate (kbps)?

NOTE 1 The maximum modulation rate of RF (after channel encoding, adding of in-band
control signalling and any overhead signalling) possible to transmit carrier over an RF
channel, i.e. independent of access technology and of modulation schemes.
Answer:
FDD mode UL: 16/32/64/128/256/512/1024 kbps (BPSK, chip rate = 4.096 Mcps, variable SF
= 4-256)
FDD mode DL: 32/64/128/256/512/1024/2048 kbps (QPSK, chip rate = 4.096 Mcps, variable
SF = 4-256)
TDD mode UL/DL: 512/1024/2048/4096 kbps (QPSK, chip rate = 4.096 Mcps, variable SF =
2-16)
Note 1: Multi-code transmission can be used to create composite RF channels with a combined
bit rate exceeding the maximum values given above

88

A1.2.7

Frame Structure : Describe the frame structure to give sufficient information such as;
- frame length
- the number of time slots per frame
- guard time or the number of guard bits
- user information bit rate for each time slot
- channel bit rate (after channel coding)
- channel symbol rate (after modulation)
- associated control channel (ACCH) bit rate
- power control bit rate.
Note 1: Channel coding may include FEC, CRC, ACCH, power control bits and guard bits.
Provide detail.
Note 2: Describe the frame structure for forward link and reverse link, respectively.
Note 3: Describe the frame structure for each user information rate
Answer:
Refer to system description
Frame length: 10 ms
Number of time slots per frame: 16 (time slot = power control period for FDD)
Guard time FDD mode: No guard time needed in FDD
Guard time TDD mode: 23.4 s
User information bit rate for each time slot: Variable
Channel bit rate (after channel coding and rate matching):
FDD mode:
UL: per IQ/branch (16/32/64/128/256/512/1024 kbps)
DL: 32/64/128/256/512/1024/2048 kbps
TDD mode: 512/1024/2048/4096 kbps
Channel symbol rate (after modulation):
FDD mode:

(16/32/64/128/256/512/1024 ksps

TDD mode:

256/512/1024/2048 ksps

Associated control channel bit rate: Variable. From an SRTT point-of-view, associated control
is not distinguished from traffic data.
Power-control bit rate: The power control command rate is 1.6 kHz for FDD and 100-800 Hz
for TDD
See the detailed description of the proposal for more information.
A1.2.8

Does the RTT use frequency hopping? If so characterize and explain particularly the impact
(e.g. improvements) on system performance.
Answer:
The RTT does not use frequency hopping.

A1.2.8.1

What is the hopping rate?

Answer: N/A

A1.2.8.2

What is the number of the hopping frequency sets?

Answer: N/A

A1.2.8.3

Are base stations synchronized or non-synchronized?

Answer: N/A

89

A1.2.9

Does the RTT use spreading scheme?

Answer:
Yes, the RTT uses Direct-Sequence spreading.

A1.2.9.1

What is the chip rate (Mchip/s): Rate at input to modulator.

Answer:
FDD: 4.096 Mcps (8.192 Mcps, 16.384 Mcps)
TDD: 4.096 Mcps

A1.2.9.2

What is the processing gain: 10 log (Chip rate / Information rate).

Answer:
The processing gain depends on the specific service. Assuming a span of the information rate
of 100 bps to 2.048 Mbps, the processing gain varies in the range 46-3 dB for a 4.096 Mcps
carrier.

A1.2.9.3

Explain the uplink and downlink code structures and provide the details about the types (e.g.
PN code, Walsh code) and purposes (e.g. spreading, identification, etc.) of the codes.
Answer:
FDD mode:
Channelisation codes (UL & DL): Orthogonal Variable Spreading Factor codes of length 2k
(for a 4.096 Mcps carrier)
Short scrambling codes (UL): Complex MS-specific code of length 256 chips (based on
extended Very-Large Kasami set).
Long scrambling code (UL): Complex MS-specific code of length 10 ms (40960 chips for a
4.096 Mcps carrier). Segment of different long Gold codes.
Scrambling code (DL): Real cell-specific code of length 10 ms (40960 chips for a 4.096 Mcps
carrier). Segment of different long Gold codes.
TDD mode:
Channelisation codes (UL & DL) : Orthogonal codes of length 2-16
Scrambling codes (UL & DL): Cell-specific PN codes of length 2-16

See the detailed description of the proposal for more information.

A1.2.10

Which access technology does the proposal use: TDMA, FDMA, CDMA , hybrid, or a new
technology?
In the case of CDMA which type of CDMA is used: Frequency Hopping (FH) or Direct
Sequence (DS) or hybrid? Characterize.
Answer:
FDD mode: Wide-band CDMA (Direct-Sequence).
TDD mode: Wide-band TDMA/CDMA (Direct-Sequence).

90

A1.2.11

What is the baseband modulation technique? If both the data modulation and spreading
modulation are required, please describe detail.
What is the peak to average power ratio after baseband filtering (dB)?
Answer:
FDD mode:
Data modulation: Dual-channel QPSK (UL), QPSK (DL)
Spreading modulation: QPSK (UL), BPSK (DL).

TDD mode:
Data modulation: QPSK (UL & DL)
Spreading modulation: QPSK (UL & DL)

Root raised cosine pulse shaping, roll-off factor 0.22 for FDD and TDD.
Resulting crest factor in the order of 5 dB for a single code case and up to 9 dB (TDD-mode)
for the multi-code transmission.
A1.2.12

What are the channel coding (error handling) rate and form for both the forward and reverse
links? e.g.
- Does the SRTT adopt FEC (Forward Error Correction) or other schemes?
Answer:
Default FEC:
Convolutional inner code (rate 1/3 or rate 1/2, constraint length K=9).
Optional outer Reed-Solomon code (rate TBD) for BER=10-6 circuit-switched services.
The use of Turbo codes for high-rate services is under consideration and will most likely
be adopted. However, the current evaluation results do not include Turbo codes.
Special FEC schemes, e.g. unequal error protection can be applied.

- Does the SRTT adopt unequal error protection? Please provide details.
Answer:
Unequal error protection can be applied (see above).

- Does the SRTT adopt soft decision decoding or hard decision decoding? Please provide
details.
Answer:
The decoding scheme of the SRTT is a receiver implementation issue and is not covered by
the RTT description. There is nothing in the RTT that prevents the use of either soft or hard
decision decoding.

- Does the SRTT adopt iterative decoding (e.g. turbo codes)? Please provide details.
Answer:
Turbo codes are under consideration (see above).
See the detailed description of the proposal for more information.

91

A1.2.13

What is the bit-interleaving scheme? Provide detailed description for both up link and down
link.
Answer:
Inner interleaving: Block interleaving with different interleaver spans (10 ms, 20 ms, 40 ms,
80 ms). Inter-frame interleaving (>10 ms) is applied on a transport-channel basis. Final intraframe interleaving (10 ms) is applied after transport-channel multiplexing.
Optional outer interleaving: Block interleaving with different interleaver spans (10 ms, 20
ms, 40 ms, 80 ms).

A1.2.14

Describe the taken approach for the receivers (MS and BS) to cope with multipath
propagation effects (e.g. via equalizer, RAKE receiver, etc.).
Answer:
FDD:
A RAKE receiver or any other suitable receiver structures can coherently combine multiple
paths and give diversity gains (the detailed receiver structure is implementation dependent).
Phase reference in the form of pilot symbols is available on both transmission directions.
TDD:
Typically, joint detection is used to coherently detect the data corresponding to different
CDMA codes and copes with multipath propagation effects at the MS as well as the BS (the
detailed receiver structure is implementation dependent). Phase reference in the form of a
pilot sequence is available in both transmission directions.

A1.2.14.1

Describe the robustness to intersymbol interference and the specific delay spread profiles that
are best or worst for the proposal.
Answer:
Within that range, the size of the delay spread does not, in itself, have any impact on the
performance. On the other hand, the shape of the delay spread may have an impact on the
performance.
FDD:
The interference due to time dispersion is suppressed by the processing gain. The exact
performance depends on the number of Rake fingers and the search window.
TDD:
Intersymbol interference is eliminated in the data detection process typically due to the
application of a Joint detection equalizer (the detailed detector structure is implantation
dependent)

A1.2.14.2

Can rapidly changing delay spread profiles be accommodated? Please describe.

Answer:
FDD:
Variations in the delay spread profiles in terms of amplitude and phase variations can be
tracked at least on a slot-by-slot (0.625 ms) basis. Additional variations in the delay spread
profile, such as the appearance/disappearance of rays can be tracked at least on a frame-byframe basis.
TDD:
All variations in the delay spread profile, including amplitude and phase variations as well as
the the appearance/disappearance of rays can be tracked at least on a slot-by-slot (0.625 ms)
basis.

92

A1.2.15

What is the Adjacent channel protection ratio?

In order to maintain robustness to adjacent channel interference, the SRTT should have some
receiver characteristics that can withstand higher power adjacent channel interference.
Specify the maximum allowed relative level of adjacent RF channel power in dBc. Please
provide detail how this figure is assumed.
Answer:
Preliminary studies has indicated that Adjacent Channel Protection (ACP) in the order of 30
to 35 dB will be sufficient between the operators in an uncoordinated operation. For co-sited
cells, the requirements are even less stringent.

A1.2.16

Power classes

A1.2.16.1

Mobile terminal emitted power: What is the radiated antenna power measured at the antenna?
For terrestrial component, please give (in dBm). For satellite component, the mobile terminal
emitted power should be given in EIRP (dBm).
Answer:
Not limited by the RTT.

A1.2.16.1.1

What is the maximum peak power transmitted while in active or busy state?
Answer:
Not limited by the RTT, typically 30 dBm and less.

A1.2.16.1.2

What is the time average power transmitted while in active or busy state? Provide detailed
explanation used to calculate this time average power.
Answer:
FDD mode:
Activity is 100 % if a mobile operates a dedicated channel. For packet transmission of the
common channels smaller TX active cycles possible.
TDD mode:
1 code (peak/average ratio 3.2 dB): Min. 14.8 dBm (1 timeslot), Max. 26.5 dBm (15
timeslots)8 codes (peak/average ratio 8.7dB): Min. 9.2 dBm (1 timeslot used), Max. 21 dBm
(15 timeslots used)
Calculation: Time average power =30dBm-peak/average ratio+10*log10(used
timeslots/frame/16).

A1.2.16.2

Base station transmit power per RF carrier for terrestrial component

A1.2.16.2.1

What is the maximum peak transmitted power per RF carrier radiated from antenna?
Answer:
Not limited by the RTT.

A1.2.16.2.2

What is the average transmitted power per RF carrier radiated from antenna?
Answer:
Not limited by the RTT.

93

A1.2.17

What is the maximum number of voice channels available per RF channel that can be
supported at one base station with 1 RF channel (TDD systems) or 1 duplex RF channel pair
(FDD systems), while still meeting G.726 performance requirements?
Answer:
FDD mode: There are a maximum of 256 orthogonal downlink channels available, some of
which must be allocated for downlink control channels. This leaves approximately 250
orthogonal channels for user traffic, such as voice. Normally, the cell capacity is interference
limited, i.e. the actual number of voice channels is lower than this number (exact number of
voice channels depends on operational conditions). Uplink is never limited by number of
orthogonal code channels, as the orthogonal code tree used is user specific in the uplink. In
some cases, e.g. for the case when adaptive antennas are used, the number of voice channels
per BS can be increased above 250 by applying multiple non-orthogonal code sets on the
downlink.
TDD mode: There are a maximum of 128 orthogonal downlink channels available, some of
which are allocated for downlink control channels. This leaves approximately 120 orthogonal
channels for user traffic. The reason why the maximum number of channels in TDD is only
50% of that in FDD is the UL/DL sharing of one 5 MHz carrier in TDD.

94

A1.2.18

Variable bit rate capabilities: Describe the ways the proposal is able to handle variable base
band transmission rates. For example, does the SRTT use:
-adaptive source and channel coding as a function of RF signal quality
Answer:
Source coding is not part of the RTT. Adaptive source coding as a function of RF quality is
possible. Adaptive channel coding as a function of RF signal quality is possible.
- Variable data rate as a function of user application?
Answer:
The user rate can vary on a 10 ms basis. See A1.2.18.1
- Variable voice/data channel utilization as a function of traffic mix requirements?
Answer:
The RTT allows for variable voice/data channel utilisation as a function of traffic mix
requirements.
Characterise how the bit rate modification is performed. In addition, what are the advantages
of your system proposal associated with variable bit rate capabilities?
Answer:
FDD: Different channel bit rates are possible by changing the spreading factor in factors of 2
from 256 down to 4. For the highest rates, multi-code transmission, i.e. transmission on
several parallel code channels, is used.
TDD: Different channel bit rates are possible by allocating a variable number of timeslots
and a variable number of codes to a connection and variable spreading codes.

On the uplink an arbitrary user bit rate after channel coding is matched to the closest possible
channel bit rate by code puncturing/repetition.
On the downlink an arbitrary user bit rate is matched to the chosen channel bit rate by
discontinuous transmission.
For variable-rate transmission, the rate can vary on a 10 ms basis. Explicit rate information,
to simplify decoding, may be transmitted on a physical control channel.
Multiple variable services can be multiplexed on one variable-rate physical channel or
multiplexed on different variable-rate physical channels.
The advantages with this approach are that the bit rate can be varied on a frame-by-frame
basis without any explicit resource allocation and negotiation. It also caters for the
independent quality control of each service on a multi-service connection.
A1.2.18.1

What are the user information bit rates in each variable bit rate mode?
Answer:
FDD mode
Variable user bit rates between 0 and 2.048 Mbit/s can be supported with 100 bit/s granularity,
with adjustments possible on a frame by frame basis. For a given connection, a sub-set of
these rates is chosen at call set-up. During the call, the rate can be varied between the rates
within the sub-set on a frame by frame basis. The sub-set of rates can also be changed during
a call, e.g. due to the addition or removal of services.
TDD mode
Variable user bit rates between 0 and 2.048 Mbit/s can be supported with a high degree of
flexibility by adjusting the number of codes and time slots used as well as by adjusting the
channel coding and burst types used.

95

A1.2.19

What kind of voice coding scheme or codec is assumed to be used in proposed RTT? If the
existing specific voice coding scheme or codec is to be used, give the name of it. If a special
voice coding scheme or codec (e.g. those not standardized in standardization bodies such as
ITU) is indispensable for the proposed RTT, provide detail, e.g. scheme, algorithm, coding
rates, coding delays and the number of stocastic code books.
Answer:
Different voice coding schemes can be supported since the supported RTT has a flexible
bearer capability supporting different bit rate allocation and voice coding frame length (e.g. 10
ms and 20 ms). Voice coding schemes envisaged to be used are the voice codecs used in the
GSM system, e.g. EFR and AMR coding schemes. The AMR is to be finalised end98.

A1.2.19.1

Does the proposal offer multplex voice coding rate capability? Provide detail.
Answer:
The RTT supports multiple voice coding rates through the chosen subset of the possible user
bit rates as indicated in A1.2.18.1.}

A1.2.20

Data services: Are there particular aspects of the proposed technologies which are applicable
for the provision of circuit-switched, packet-switched or other data services like asymmetric
data services? For each service class (A, B, C and D) a description of SRTT services should
be provided, at least in terms of bit rate, delay and BER/FER.
Note 1: See [draft new] Recommendation [FPLMTS.TMLG] for the definition of
- circuit transfer mode
- packet transfer mode
- connectionless service
and for the aid of understanding circuit switched and packet switched data services
Note 2: See ITU-T Recommendation I.362 for details about the service classes A, B, C and D
Answer:
All service classes can be supported with the proposed RTT.
The pooling of resource units bearer services at the radio interface with various data rates can
be achieved. Further, by variation of the spreading factor, power, coding rate and interleaving
depth various BER and delay requirements can be met.
For each service class dedicated bearer services at the radio interface are defined.
The bearer services at the radio interface are separated into low delay data (LDD), long
constrained delay (LCD) and unconstrained delay data (UDD) bearer services. The LDD
bearer is characterised by stringent delay (and stringent delay variation) requirements. In
contrary, the LCD bearer is characterised by less stringent delay (and delay variation)
requirements but more stringent BER requirements. Both LDD and LCD bearers can have a
constant or variable bit rate. Finally, the UDD bearer is characterised by unconstrained delay
requirements.
The following mapping may be used:
Class A: LDD
Class B: LDD-VBR
Class C: LCD
Class D: UDD

A1.2.20.1

For delay constrained, connection oriented. (Class A)

Answer:
The RTT provides user bit rates up to 2048 kbps. It can be set to any required value needed by
a particular service. For instance, the ISDN services N*64 kbps (up to 2048 kbps), where N is
an integer, can easily be supported. Various bit error ratios and/or frame error ratios and/or
delays (see other related items) are supported depending on what the service demands. The
RTT thus provides a flexible bearer concept.
For delay constrained, connection oriented (Class A).
The following non-comprehensive list gives some example supported LDD services:
8 kbit/s; Delay 20 ms; BER < 10-3
144 kbit/s; Delay 50 ms; BER < 10-6
384 kbit/s; Delay 50 ms; BER < 10-6

96

A1.2.20.2

For delay constrained, connection oriented, variable bit rate (Class B)

Answer:
Connection oriented variable bit rate service is supported since the bit rate can be controlled
both on the physical layer and on higher layers. Up to 2048 kbps is supported. See also
A1.2.18.1. The required QoS can be set according to what the service requires.
For delay constrained, connection oriented variable bit rate (Class B).
The following non-comprehensive list gives some example supported LDD-VBR services:
Peak data rate 64 kbit/s; Delay 50 ms; BER < 10-6; Granularity: 16 kbit/s
Peak data rate 144 kbit/s; Delay 50 ms; BER < 10-6; Granularity: 16 kbit/s
Peak data rate 384 kbit/s; Delay 50 ms; BER < 10-6; Granularity: 16 kbit/s
Peak data rate 2048 kbit/s; Delay 50 ms; BER < 10-6; Granularity: 32 kbit/s

A1.2.20.3

For delay unconstrained, connection oriented. (Class C)

Answer:
Class C services, e.g. best effort type, are supported. Depending on the service definitions
different QoS levels can be supported individually per user.
For delay unconstrained, connection oriented (Class C).
The following non-comprehensive list gives some example supported LCD services:
64 kbit/s; Delay 300 ms; BER < 10-6
144 kbit/s; Delay 300 ms; BER < 10-6
384 kbit/s; Delay 300 ms; BER < 10-6
2048 kbit/s; Delay 300 ms; BER < 10-6

A1.2.20.4

For delay unconstrained, connectionless. (Class D)

Answer:
The answer to A1.2.20.3 applies.
Connectionless or connection oriented packet services are considered to be a higher layer issue
since with connection less service it is assumed that no end-to-end connection exists, i.e. final
address is included in each packet. With connection oriented service it is meant that an end-toend connection is set up and then transferring the data without explicitly mentioning the final
address. Looking at the individual radio link both modes can of course be supported by the
RTT if the upper layers support both modes.
For delay unconstrained, connectionless (Class D).
The following non-comprehensive list gives some example supported UDD services:
64 kbit/s; Delay unconstrained; BER < 10-8
144 kbit/s; Delay unconstrained; BER < 10-8
384 kbit/s; Delay unconstrained; BER < 10-8
2048 kbit/s; Delay unconstrained; BER < 10-8

A1.2.21

Simultaneous voice/data services: Is the proposal capable of providing multiple user services
simultaneously with appropriate channel capacity assignment?
Note : The followings describe the different techniques that are inherent or improve to a great
extent the technology described above to be presented:
Description for both BS and MS are required in attributes from A2.22 through A1.2.23.2.
Answer:
Parallel services can be provided. The different services can have independent bit rate, biterror rate, delay, etc., and can have different transfer modes (packet/circuit-switched).

97

A1.2.22

Power control characteristics: Is power control scheme included in the proposal? Characterize
the impact (e.g. improvements) of supported power control schemes on system performance.
Answer:
3 types of power control are employed. One is the fast closed loop power control (FDD
mode only) which counters fading on a slot basis (0.625 ms): it is based on measurements on
SIR. The second one is the open loop power control, it is used only for the initial power
setting. The third one is the outer loop power control: it is based on BER and FER
measurements. It has the role to change the target C/I, when the situation of the mobile is
changing or for power control planning. It is done on a longer period basis. The use of fast
power control significantly improves the link-performance (BER as a function of Eb/N0)
especially in the case slow-moving mobile stations. For fast moving mobile stations (>100
km/h), there is less performance improvement due to fast power control.

A1.2.22.1

What is the power control step size in dB?

Answer:
FDD mode:
UL: Variable on a cell basis in the range 0.25-1.5 dB
DL: Variable on a cell basis in the range 0.25-1.5 dB
TDD mode:
The power control step size is variable, range 1.5 to 3 dB..

A1.2.22.2

What are the number of power control cycles per second?

Answer:
FDD mode: 1600
TDD mode: 100-800 depending on the exact UL/DL time slot allocation

A1.2.22.3

What is the power control dynamic range in dB?

Answer:
UL: 80 dB
DL: 30 dB

A1.2.22.4

What is the minimum transmit power level with power control?

Answer:
<-50 dBm for MS with highest power class (Maximum MS power <30 dBm, Power-control
dynamics > 80 dBm)

A1.2.22.5

What is the residual power variation after power control when RTT is operating? Please
provide details about the circumstances (e.g. in terms of system characteristics, environment,
deployment, MS-speed, etc.) under which this residual power variation appears and which
impact it has on the system performance.
Answer:
The residual power variation depends on the channel conditions, (Doppler spread and
frequency selectivity) and are difficult to specify in detail. The residual power variations are
fully included in the link-budget and capacity evaluations.

A1.2.23

Diversity combining in mobile station and base station: Are diversity combining schemes
incorporated in the design of the RTT?
Answer:
Yes

98

A1.2.23.1

Describe the diversity techniques applied in the mobile station and at the base station,
including micro diversity and macro diversity, characterizing the type of diversity used, for
example:
- time diversity

repetition, RAKE-receiver, etc.,

- space diversity

multiple sectors, multiple satellite, etc.,

- frequency diversity

FH, wideband transmission, etc.,

- code diversity

multiple PN codes, multiple FH code, etc.,

- other scheme.
Characterize the diversity combining algorithm, for example, switch diversity, maximal ratio
combining, equal gain combining. Additionally, provide supporting values for the number of
receivers (or demodulators) per cell per mobile user. State the dB of performance
improvement introduced by the use of diversity.
Answer:
Time diversity: Channel coding and interleaving in both uplink and downlink.
Multipath diversity: RAKE, joint detection or similar receiver structure with, typically,
maximum ratio combining is used in both BS and MS (implementation dependent).
Space diversity: Receive antenna diversity with, typically, maximum ratio combining can be
used in both uplink and downlink.
Transmit antenna diversity is under consideration for downlink.
Macro diversity: Soft (inter-site) handover with, typically, maximum ratio combining in
downlink, selection combining in uplink. Softer (inter-sector) handover with, typically.
Maximum ratio combining in both uplink and downlink.
Frequency diversity: Wideband carrier (equivalent to multi-path diversity).

For the mobile station: what is the minimum number of RF receivers (or demodulators) per
mobile unit and what is the minimum number of antennas per mobile unit required for the
purpose of diversity reception?
These numbers should be consistent to that assumed in the link budget template in Annex 2
and that assumed in the calculation of the capacity defined at A1.3.1.5.
Answer:
One RF receiver per mobile unit. One antenna per mobile unit.
A1.2.23.2

What is the degree of improvement expected in dB? Please also indicate the assumed
condition such as BER and FER.
Answer:
For receiver antenna diversity the diversity gain is 2.5 - 3.5 dB in required Eb/No for BER=10. If power control is disabled the gain is much higher for the low speed cases. On top of the
gain in reduced required Eb/No there is a gain in decreased transmitted power. This gain can be
up to 2.5 dB, depending on the environment.
3

Transmit diversity can also be employed, especially in the downlink. A gain similar to the gain
with receiver antenna diversity is expected.
All other diversity methods are inherent parts of the RTT concept and therefore it is difficult to
specify an explicit diversity gain figure in dB.

99

A1.2.24

Handover/Automatic Radio Link Transfer (ALT): Do the radio transmission technologies

support handover?
Characterize the type of handover strategy (or strategies) which may be supported, e.g. mobile
station assisted handover. Give explanations on potential advantages, e.g. possible choice of
handover algorithms. Provide evidence whenever possible.
Answer
The RTT supports automatic handover.
FDD: The handover scheme is based on a mobile assisted soft/softer handover mechanism and
hard handover.
The mobile station (MS) monitors the pilot signal levels received from neighbouring base
stations and reports to the network pilots crossing or above a given set of dynamic thresholds.
Based on this information the network orders the MS to add or remove pilots from its Active
Set. The Active Set is defined as the set of base station for which user signal is simultaneously
demodulated and coherently combined.
The same user information modulated by the appropriate base station code is sent from
multiple base stations. Coherent combining of the different signals from different sectored
antennas, from different base stations, or from the same antenna but on different multiple path
components is performed in the MS by the usage of Rake receivers.
Base stations with which the mobile station is in soft handover process the signal transmitted
by a mobile station. The received signal from different sectors of a base station (cell) can be
combined in the base station, and the received signal from different base stations (cells) can be
combined at the radio network controller. Soft handover results in increased coverage range on
the uplink. This soft handover mechanism results in truly seamless handover without any
disruption of service.
The spatial diversity obtained reduces the frame error rate in the handover regions and allows
for improved performance in difficult radio environment.
Furthermore, the RTT support various types of hard handover, e.g. inter-frequency handover.
The measurements to detect other frequency carriers available are made possible through the
use of measurement slots.
TDD: TDD provides two different handover mechanisms depending on the service type:
connection oriented or packet services.
For connection oriented services, the basic HO scheme is a mobile assisted, network evaluated
and decided hard handover using backward signalling. Appropriate measures are provided to
accelerate the HO procedure, e.g. in case of a corner effect. Furthermore the proposed RTT
does not prevent the introduction of soft handover. The support of soft handover is for further
study.
For packet services, the basic HO scheme is a mobile evaluated and decided hard handover
with background network control using forward signalling (cell reselection).
Potential advantages:
Seamless HO for connection oriented, loss-less HO for packet bearer services
Mobile assisted network evaluated and decided handover scheme is most appropriate for
RT services since it allows for both high flexibility in HO-algorithm design and
implementation, e.g. to meet operator specific requirements in various deployment
scenarios and system stability at high capacity.
Mobile evaluated and decided handover with network background control is most
appropriate for packet services since it allows for both resource savings on the air interface
by decentralised decision making and system integrity at high capacity by network coordinating measures.

100

A1.2.24.1

What is the break duration (sec) when a handover is executed? In this evaluation, a detailed
description of the impact of the handover on the service performance should also be given.
Explain how the estimate derived.
Answer:
FDD:
Soft handover: No break duration (make before break)
Hard handover: no loss for packet services due to ARQ
TDD:
For the basic scheme of hard HO the break duration on HO execution is the time interval
between suspension of transmission on the traffic and signalling channels of the serving cell
and the successful establishment of these on the new target cell.
This time is mainly dependent on the access procedure to the target cell. Since cells are
assumed to be synchronised on a frame basis a synchronous handover is executed, i.e. the MS
performs a HO access onto the traffic channels of the new cell with known synchronisation
resulting in a very short HO execution time.
Impact of HO on service performance:
RT services:
NRT services:

A1.2.24.2

since the break duration is very short, seamless HO is possible

ARQ mechanism ensures loss-less HO

For the proposed SRTT, can handover cope with rapid decrease in signal strength (e.g. street
corner effect)?
Give a detailed description of
- the way the handover detected, initiated and executed,
- how long each of this action lasts (minimum/maximum time in msec),
- the timeout periods for these actions.
Answer:
FDD: The MS continuously searches for signal from new and existing BS. It also maintains
two thresholds (e.g. pilot Ec/Io) based on current combined quality of the down link soft
handover legs to add newly detected BS or to drop existing BS from its soft handover active
set. The need to add or drop is sent in a message to the network, which determines whether or
not to execute the addition or deletion.
The time it takes to perform the above actions depends on the searcher and fixed
infrastructure. When compared to the initial cell access the procedure is much faster as only
the base stations indicated in the neighbour set need to be searched and thus the search time is
greatly reduced and thus dependent on the size of the base station set to be searched.
There is no time out period when soft or softer handover is performed.
TDD: The HO functionality can successfully cope with rapid field drop effects like e.g. the
street corner effect.
Special means are introduced to speed up the process during each phase of the HO:
1. Detection and initiation:
Fast measurement acquisition and neighbour cell identification due to synchronised
network
Signal strength trend analysis based on variable averaging window size and threshold
comparison as well as MS speed and MS moving direction estimates
2. Decision:
Network handles this HO type with highest priority
default (hot standby target cells) may be used
3. Execution:
For the TDD mode the network may be synchronised on frame basis. The handover execution
procedures can take this in account to execute the HO in a very short time.

101

A1.2.25

Characterize how does the proposed SRTT react to the system deployment in terms of the
evolution of coverage and capacity (e.g. necessity to add new cells and/or new carriers)
particularly in terms of frequency planning.
Answer:
No frequency planning needed due to frequency reuse 1.

A1.2.26

Sharing frequency band capabilities: To what degree is the proposal able to deal with spectrum
sharing among IMT-2000systems as well as with all other systems:
- spectrum sharing between operators,
- spectrum sharing between terrestrial and satellite IMT-2000 systems,
- spectrum sharing between IMT-2000 and non-IMT-2000 systems,
- other sharing schemes.
Answer:
For both FDD-mode and TDD-mode, sharing is always possible through frequency division.
Furthermore, for the TDD-mode, sharing the same frequency with another TDD system
including non-UMTS/IMT-2000 systems such as DECT and PHS is also possible due to the
TDMA component. Both fixed time division and interference avoidance in the time domain
using Dynamic Channel Allocation (DCA) can be used for this purpose. In addition, since
ODMA relays do not own dedicated radio resource but share it in an asynchronous fashion
with neighbouring nodes, they are tolerant to spectrum sharing.

A1.2.27

Dynamic channel allocation: Characterize the DCA schemes which may be supported and
characterize their impact on system performance (e.g. in terms of adaptability to varying
interference conditions, adaptability to varying traffic conditions, capability to avoid frequency
planning, impact on the reuse distance, etc.)
Answer:
FDD mode: DCA not generally needed for the FDD mode.
TDD mode: Slow DCA (allocation of time slots to cells) and fast DCA (allocation of a channel
to a certain call) can be distinguished. Additionally for the TDD mode the allocation of a slot
in a cell to uplink or download traffic is managed by the slow DCA, too.
The interference on different slots in the time frame may be different. Therefore, a DCA
algorithm that allocates the least interfered slots to ongoing calls with high QoS requirements
results in a considerable gain in quality and/or capacity. The capability to vary the ratio of
slots allocated in the uplink and in the downlink allows an optimal adaptation to the traffic
asymmetry. Since synchronised base stations are used, advanced combinations of fast and slow
DCA can be implemented in order to allocate the maximum amount of resources to the cell
with the momentarily highest amount of traffic.

A1.2.28

Mixed cell architecture: How well do the technologies accommodate mixed cell architectures
(pico, micro and macrocells)? Does the proposal provide pico, micro and macro cell user
service in a single licensed spectrum assignment, with handoff as required between them?
(terrestrial component only)
Note: Cell definitions are as follows:
pico - cell hex radius (r) < 100 m
micro - 100 m < (r) < 1000 m
macro - (r) > 1000 m
Answer:
Seamless handover is possible between cell layers.

102

A1.2.29

Describe any battery saver / intermittent reception capability

Answer:
In dedicated mode, i.e. during calls and in circuit oriented operation, the transmitter/receiver is
continuously on (FDD) and transmits/receives only in allocated timeslots (TDD), respectively.
Variable rate transmission is utilised whenever possible to reduce needed transmitted power.
Employing power control methods also reduces the transmitted power levels to a minimum.
With packet traffic, depending on the packet-access mode, the receiver and transmitter can be
used only periodically, i.e. being switched off until data is available for transmission or the
base station indicates the mobile station that is to be received. In the latter case, the polling is
done according to A1.2.29.1, which also explains the power saving during idle mode operation
of the terminal.

A1.2.29.1

Ability of the mobile station to conserve standby battery power: Please provide details about
how the proposal conserve standby battery power.
Answer:
In idle mode, the mobile station uses a sleep mode that permits that e.g. most of the circuits
can be turned off during the periods when the mobile station is not receiving. The mobile
station is only awakened for short periods to listen to e.g. the paging channel or the broadcast
channel.
See the detailed description of the proposal for more information.

A1.2.30

Signaling transmission scheme: If the proposed system will use radio transmission
technologies for signaling transmission different from those for user data transmission,
describe details of signaling transmission scheme over the radio interface between terminals
and base (satellite) stations.
Answer:
The signalling scheme for the RTT is basically the same as for user data. User data and
signalling are using the same L1 services.

A1.2.30.1

Describe the different signaling transfer schemes which may be supported, e.g. in connection
with a call, outside a call.
Does the SRTT support new techniques? Characterize.
Does the SRTT support signalling enhancements for the delivery of multimedia services?
Characterize.
Answer:
The RTT does not limit the use of any advanced techniques. The physical layer provides
means for transmission rate signalling which can be used also to indicate which services are
active and thus introduction of an associated control channel with service negotiation is
supported by the RTT.

A1.2.31

Does the SRTT support a Bandwidth on Demand (BOD) capability? Bandwidth on Demand
refers specifically to the ability of an end-user to request multi-bearer services. Typically this
is given as the capacity in the form of bits per second of throughput. Multi bearer services can
be implemented by using such technologies as multi carrier, multi time slot or multi codes. If
so, characterize these capabilities.
Note: BOD does not refer to the self-adaptive feature of the radio channel to cope with
changes in the transmission quality (see A1.2.5.1).
Answer:
Bandwidth on demand is supported in the range 0 bps to 2.048 Mbps user bit rate.
FDD mode: The bandwidth on demand possibility is implemented by one or more multi-codes
and variable spreading factor (4-256).
TDD mode: The bandwidth-on-demand possibility is implemented by multi-code (assigning
more than one code) and multi-slot (assigning more than one time slot) transmission.

103

A1.2.32

Does the SRTT support channel aggregation capability to achieve higher user bit rates?
Answer:
FDD mode: Channel aggregation to achieve higher rates is normally not needed for the FDD
mode, due to different bit rates of the physical channels (maximum 1024 kbps). Channel
aggregation (multi-code transmission) is supported and used for the highest user rates (up to 2
Mbps).
TDD mode: Channel aggregation (multi-code and multi-slot) is used, see A1.2.31.

A1.3

Expected Performances

A1.3.1

for terrestrial test environment only

A1.3.1.1

What is the achievable BER floor level (for voice)?

Note: BER floor level under BER measuring condition defined in Annex 2 using the data
rates indicated in section 1 of Annex 2.
Answer:
Significantly below BER = 10-3

A1.3.1.2

What is the achievable BER floor level (for data)?

Note: BER floor level under BER measuring condition defined in Annex 2 using the data rates
indicated in section 1 of Annex 2.
Answer:
Significantly below BER = 10-6 for circuit-switched data

A1.3.1.3

What is the maximum tolerable delay spread (in nsec) to maintain the voice and data service
quality requirements?
Note: The BER is an error floor level measured with the Doppler shift given in the BER
measuring conditions of ANNEX 2.
Answer:
Implementation dependent
Exact requirement is for further study but at least 50000 ns should be tolerated.

A1.3.1.4

What is the maximum tolerable doppler shift (in Hz) to maintain the voice and data service
quality requirements?
Note: The BER is an error floor level measured with the delay spread given in the BER
measuring conditions of ANNEX 2.
Answer:
Implementation dependent.
Exact requirement is for further study but at least 500 Hz should be tolerated.

A1.3.1.5

Capacity: The capacity of the radio transmission technology has to be evaluated assuming the
deployment models described in ANNEX 2 and technical parameters from A1.2.22 through
A1.2.23.2.

A1.3.1.5.1

What is the voice traffic capacity per cell (not per sector): Provide the total traffic that can be
supported by a single cell in Erlangs/MHz/cell in a total available assigned non-contiguous
bandwidth of 30 MHz (15 MHz forward/15 MHz reverse) for FDD mode or contiguous
bandwidth of 30 MHz for TDD mode. Provide capacities considering the model for the test
environment in ANNEX 2. The procedure to obtain this value in described in ANNEX 2.
The capacity supported by not a standalone cell but a single cell within contiguous service area
should be obtained here.
Answer:
See Simulation Results.

104

A1.3.1.5.2

What is the information capacity per cell (not per sector): Provide the total number of userchannel information bits which can be supported by a single cell in Mbps/MHz/cell in a total
available assigned non-contiguous bandwidth of 30 MHz (15 MHz forward / 15 MHz reverse)
for FDD mode or contiguous bandwidth of 30 MHz for TDD mode. Provide capacities
considering the model for the test environment in ANNEX 2. The procedure to obtain this
value in described in ANNEX 2. The capacity supported by not a standalone cell but a single
cell within contiguous service area should be obtained here.
Answer:
See Simulation Results. Furthermore, for the TDD mode, ODMA can increase capacity by
reducing effective path loss, optimum link adaptation and link diversity thus lowering
transmission power and associated inter-cell interference.

A1.3.1.6

Does the SRTT support sectorization? If yes, provide for each sectorization scheme and the
total number of user-channel information bits which can be supported by a single site in
Mbps/MHz (and the number of sectors) in a total available assigned non-contiguous
bandwidth of 30 MHz (15 MHz forward/15 MHz reverse) in FDD mode or contiguous
bandwidth of 30 MHz in TDD mode.
Answer:
The RTT supports use of the sectorisation. See also simulation results.

A1.3.1.7

Coverage efficiency: The coverage efficiency of the radio transmission technology has to be
evaluated assuming the deployment models described in ANNEX 2.

A1.3.1.7.1

2
What is the base site coverage efficiency in km /site for the lowest traffic loading in the voice
only deployment model? Lowest traffic loading means the lowest penetration case described in
ANNEX 2.
Answer:
See Link Budget Template.

A1.3.1.7.2

2
What is the base site coverage efficiency in km /site for the lowest traffic loading in the data
only deployment model? Lowest traffic loading means the lowest penetration case described in
ANNEX 2.
Answer:
See Link Budget Template. Furthermore, for the TDD mode, ODMA can increase coverage
efficiency by reducing effective path loss, optimum link adaptation and link diversity. This is
particularly useful for high rate data services.

A1.3.2

For satellite test environment only

A1.3.2.1-4

N/A

A1.3.3

Maximum user bit rate (for data): Specify the maximum user bit rate (kbps) available in the
deployment models described in ANNEX 2.
Answer:
Protocols are designed in such a way that user bit rate at least up to 2048 kbps.

A1.3.4

What is the maximum range in meters between a user terminal and a base station (prior to
hand-off, relay, etc.) under nominal traffic loading and link impairments as defined in Annex
2?
Answer:
See Link Budget Template.

A1.3.5

Describe the capability for the use of repeaters

Answer:
Repeaters can be used. In addition ODMA supports data transfer via a network of intermediate
relaying nodes (dedicated fixed relays or relaying enable mobiles).

105

A1.3.6

Antenna Systems : Fully describe the antenna systems that can be used and/or have to be used;
characterize their impacts on systems performance, (terrestrial only) e.g., does the RTT have
the capability for the use of :
- remote antennas: Describe whether and how remote antenna systems can be used to extend
coverage to low traffic density areas.
- distributed antennas: describe whether and how distributed antenna design are used, and in
which IMT-2000 test environments
- Smart antennas (e.g. switched beam, adaptive, etc.): describe how smart antennas can be
used, and in which IMT-2000 test environments
- other antenna systems
Answer:
Both FDD and TDD operating modes of UTRA are able to use all the standard types of Base
Station antennas. This includes those that provide omni-directional, sectored, fixed or variable
patterns.
Directive Antennas decrease the interference, leading to an increase in system capacity.
Both FDD and TDD mode support remote antenna systems, distributed antenna systems, and
smart antenna systems.
Signal-to-interference-plus-noise-ratio (SINR) can be improved significantly by incorporating
various smart antenna concepts on the uplink as well as the downlink. These SINR gains may
be exploited
to increase the capacity (mainly in urban areas),
e.g. by reducing the interference,
to increase the coverage (mainly in rural areas),
e.g., by increasing the cell size (range extension) or by improving the edge coverage,
to increase the link quality,
to decrease the delay spread,
to reduce the transmission powers, or a combination thereof.

A1.3.7

Delay (for voice)

Answer:
It depends on which speech codec is used. Different voice bearers can be supported. See
answer on A1.3.7.1 below.

A1.3.7.1

What is the radio transmission processing delay due to the overall process of channel coding,
bit interleaving, framing, etc., not including source coding? This is given as transmitter delay
from the input of the channel coder to the antenna plus the receiver delay from the antenna to
the output of the channel decoder. Provide this information for each service being provided. In
addition, a detailed description of how this parameter was calculated is required for both the
up-link and the down-link.
Answer:
Service specific delay (depends on interleaving/channel-coding setting). Minimum delay: 12
ms for 10 ms interleaving in FDD mode and 13 ms for interleaving over 2 frames in TDD
mode. Processing time of 2 ms included.

A1.3.7.2

What is the total estimated round trip delay in msec to include both the processing delay,
propagation delay (terrestrial only) and vocoder delay? Give the estimated delay associated
with each of the key attributes described in Figure 1 of Annex 3 that make up the total delay
provided.
Answer:
The round trip delay including source coding is implementation dependent. However, the
approximate delay is 13 ms excluding the speech codec delays as indicated by the answer on
A1.3.7.1. . In addition the speech coding delay should be added which is around 12 ms for
speech framing and processing assuming a 10 ms frame length speech codec. Other delays
such as delays on the interface between the BTS and the transcoder are not included.

106

A1.3.7.3

Does the proposed RTT need echo control?

Answer:
Echo control can be taken care of by mobile station design.

A1.3.8

What is the MOS level for the proposed codec for the relevant test environments given in
Annex 2? Specify its absolute MOS value and its relative value with respect to the MOS value
of ITU-T Recommendation G.711 (64 k PCM) and ITU-T Recommendation G.726 (32 k
ADPCM).
NOTE 1 - If a special voice coding algorithm is indispensable for the proposed RTT, the
proponent should declare detail with its performance of the codec such as MOS level. (See
A1.2.19)
Answer:
For the suggested AMR speech codec no exact values are available yet. For clean speech a
MOS value of 4.1 (+0.1 compared to G.726 and 0.1 compared to G.711) can be expected
depending on the source coding bit rate.

A1.3.9

Description on the ability to sustain quality under certain extreme conditions.

A1.3.9.1

System overload (terrestrial only) : Characterize system behavior and performance in such
conditions for each test services in Annex 2, including potential impact on adjacent cells.
Describe the effect on system performance in terms of blocking grade of service for the cases
that the load on a particular cell is 125%, 150%, 175%, and 200% of full load. Also describe
the effect of blocking on the immediate adjacent cells. Voice service is to be considered here.
Full load means a traffic loading which results in 1% call blocking with the BER of 10-3
maintained.
Answer:
FDD mode: Overload causes graceful degradation of system performance, e.g. by decreasing
the speech codec bit rate or increasing the BER.
TDD mode: Under overload conditions DCA can be used to increase the allocated resources to
the overloaded cell at the cost of the capacity of the neighbouring cells.

A1.3.9.2

Hardware failures: Characterize system behavior and performance in such conditions. Provide
detailed explanation on any calculation.
Answer:
Mostly implementation dependent. Radio bearer re-establishment is supported.

A1.3.9.3

Interference immunity: Characterize system immunity or protection mechanisms against

interference. What is the interference detection method? What is the interference avoidance
method?
Answer:
FDD mode: Interference is suppressed by the processing gain. Multi-user detection and/or
interference cancellation can be used but is not required.
TDD mode: Multi-user detection by means of joint detection is typically used as protection
against interference.

A1.3.10

Characterize the adaptability of the proposed SRTT to different and/or time varying conditions
(e.g. propagation, traffic, etc.) that are not considered in the above attributes of the section
A1.3.
Answer:
Adaptive transmit power is used. Adaptive channel coding can be used for TDD mode.

A1.4

Technology Design Constraints

A1.4.1

Frequency stability : Provide transmission frequency stability (not oscillator stability)

requirements of the carrier (include long term - 1 year frequency stability requirements in
ppm).

107

A1.4.1.1

For Base station transmission (terrestrial component only)

Answer:
0.05 ppm

A1.4.1.2

For Mobile station transmission

Answer:
3 ppm (unlocked), 0.1 ppm (locked)

A1.4.2

Out of band and spurious emissions: Specify the expected levels of base or satellite and mobile
transmitter emissions outside the operating channel, as a function of frequency offset f.
Answer:
The limits for spurious emissions at frequencies greater than +/-250% of the necessary
bandwidth would be based on the applicable tables from the ITU-R Recommendation SM.329.
further guidance would be taken from the CEPT ERC recommendations that are currently
under progress.

108

A1.4.3

Synchronisation requirements: Describe SRTTs timing requirements , e.g.

- Is BS-to-BS or satellite land earth station (LES)-to-LES synchronisation required?
Provide precise information, the type of synchronisation, i.e., synchronisation of carrier
frequency, bit clock, spreading code or frame, and their accuracy.
Answer:
FDD-mode: not required.
TDD mode:
Timeslot synchronisation is recommended to decrease the effort for neighbour cell interference
suppression. Frame synchronisation is recommended to speed up listening of neighbour cell
beacon information.
-Source stability of external reference frequency = 2N10-8 or higher.
-24 hours timing accuracy of base stations 1s.
- Is BS-to-network synchronisation required? (Terrestrial only).
Answer:
Yes. On a transmission level the base stations are synchronised to the transmission network.
The different base stations involved in soft handover (macro diversity) are synchronised per
active mobile station connection in soft handover. An accuracy of +/- 2 msec is acceptable for
the fixed transmission between base stations-to-network (radio network controller (RNC)) and
can be obtained through prioritised Iub (RNC-BS) and Iur (RNC-RNC) signalling
- State short-term frequency and timing accuracy of BS (or LES) transmit signal.
Answer:
A BS short-term frequency/timing accuracy of 0.05 ppm can be considered.
- State source of external system reference and the accuracy required, if used at BS (or
LES) (for example: derived from wire-line network, or GPS receiver).
Answer:
FDD mode: A non-dedicated Synchronisation Network is used for this BS clock reference
generation where the synchronisation network is superimposed on a PDH/SDH traffic
network. A dedicated synchronisation network (e.g. GPS based) is thus not needed as a source
for the BS synchronisation. Long-term stability is set by a PRC (Primary Reference Clock)
contained in or supplied to the core network (1x10-11 or better is expected).
TDD mode: Source stability of external reference frequency = 2N10-8 or higher.
- State free run accuracy of MS frequency and timing reference clock.
Answer:
3 ppm
- State base-to-base bit time alignment requirement over a 24 h period (s).
Answer:
FDD mode: The base-to-base bit time alignment over the radio interface is set by the
individual BS frequency/timing accuracy of 0.05 ppm as indicated above. There is no further
synchronisation needed between the different base stations from the radio transmission point
of view. The base stations involved in macro diversity/soft handover to a specific mobile
station are synchronised to that mobile station . That synchronisation is done individually per
mobile station when entering macro diversity mode. TDD mode: see BS-BS synchronisation
above.

A1.4.4

Timing jitter : For base (or LES) and mobile station give:
- the maximum jitter on the transmit signal,
- the maximum jitter tolerated on the received signal.
Timing jitter is defined as RMS value of the time variance normalized by symbol duration.
Answer:
TBD

109

A1.4.5

Frequency synthesizer : What is the required step size, switched speed and frequency range of
the frequency synthesizer of mobile stations?
Answer:
- Step size: 200 kHz
- Switched speed: 250 s
- Frequency range: 60 MHz

A1.4.6

Does the proposed system require capabilities of fixed networks not generally available today?
Answer:
No special requirements on transmission for the fixed network.

A1.4.6.1

Describe the special requirements on the fixed networks for the handover procedure. Provide
handover procedure to be employed in proposed SRTT in detail.
Answer:
N/A. Since the answer was no to the question above.

A1.4.7

Fixed network feature transparency

A1.4.7.1

Which service(s) of the standard set of ISDN bearer services can the proposed RTT pass to
users without fixed network modification.
Answer:
The RTT can provide bit rates up to 2048 kbps.

A1.4.8

Characterize any radio resource control capabilities that exist for the provision of roaming
between a private (e.g., closed user group) and a public IMT-2000operating environment.
Answer:
There are no additional radio resource control capabilities foreseen to support roaming
between private and public environments, as indicated above, than what already exists for
normal roaming support. To differentiate between different type of users/networks is seen as a
higher layer signalling issue which then will be used by the normal radio resource control
management functions to select the appropriate base stations.

A1.4.9

Describe the estimated fixed signaling overhead (e.g., broadcast control channel, power
control messaging). Express this information as a percentage of the spectrum which is used for
fixed signaling. Provide detailed explanation on your calculations.
Answer:
In downlink, system and cell specific information are broadcasted on the broadcast control
channel.
FDD mode: Reference (pilot) symbols for coherent detection, power control commands, and
rate information are provided in dedicated physical control channel (DPCCH). In uplink,
DPCCH uses fixed 16 kbps Q-branch channel, while DPDCH uses variable rate I-branch
channel. In downlink, DPCCH is time multiplexed with DPDCH, and its rate can be variable
depending on the DPDCH rate. The signalling overhead of DPCCH in dedicated physical
channel is ranging from 2.8% up to 25% in downlink, while 5.9% - 33% in uplink.
See the detailed description of the proposal for more information.
TDD mode:
This is mainly dependent on layer 2 and 3 protocol architecture and on the used layer2 and 3
algorithms. Layer 2 and 3 signalling overhead is currently being optimised.
See the detailed description of the proposal for more information.

110

A1.4.10

Characterize the linear and broadband transmitter requirements for base and mobile station.
(terrestrial only)
Answer:
Bandwidth:
MS and BTS: 60 MHz, depending on band allocation
Linearity:
Mobile Station:
For 2 equal power signals being separated by 200 kHz leading to an output level of 21 dBm
each the resulting intermodulation spectrum shall not exceed relative to peak spectrum:
-38 dB at 200 kHz to -90 dB at 800 kHz offset from higher/lower frequency signal
(linear decrease)
<=-95 dB at 1 MHz from higher/lower frequency signal and above
Base Station:
For 2 equal power signals being separated by 200 kHz leading to an output level of 6 dB
below nominal output level the resulting intermodulation spectrum shall not exceed relative to
peak spectrum:
-38 dB at 200 kHz to -90 dB at 800 kHz offset from higher/lower frequency signal
(linear decrease)
<=-95 dB at 1 MHz from higher/lower frequency signal and above

A1.4.11

Are linear receivers required? Characterize the linearity requirements for the receivers for base
and mobile station. (terrestrial only)
Answer:
Linear receivers are needed both for BS and MS. The 3rd order intercept point will be
specified between -10 dBm and -5 dBm.

A1.4.12

Specify the required dynamic range of receiver. (terrestrial only)

Answer:
80 dB for Automatic Gain Control.

111

A1.4.13

What are the signal processing estimates for both the handportable and the base station?
- MOPS (Mega Operation Per Second) value of parts processed by DSP
- gate counts excluding DSP
- ROM size requiements for DSP and gate counts in kByte
- RAM size requirements for DSP and gate counts in kByte
Note 1: At a minimum the evaluation should review the signal processing estimates (MOPS,
memory requirements, gate counts) required for demodulation, equalization, channel coding,
error correction, diversity processing (including RAKE receivers), adaptive antenna array
processing, modulation, A-D and D-A converters and multiplexing as well as some IF and
baseband filtering. For new technologies, there may be additional or alternative requirements
(such as FFTs etc.).
Note 2 : The signal processing estimates should be declared with the estimated condition such
as assumed services, user bit rate and etc.
Answer:
FDD mode:
8 kbps 2048 kbps: 5 86 (valid for both UL and DL)
Answer is given in million real multiplications with DSP and correlators included taken the
word length requirements relative to the DSP operation into account.
TDD mode:
It depends on the implementation, e.g. for a 110 kbps service around 15-20 MIPS is needed.
The convolutional encoding/decoding is not included in the figures as it is the same regardless
of the multiple access for the same data rate(s).

A1.4.14

Dropped calls: describe how the RTT handles dropped calls. Does the proposed RTT utilize a
transparent reconnect procedure that is, the same as that employed for handoff?
Answer:
The RTT supports the transparent reconnect procedure for handling dropped calls.

112

A1.4.15

Characterize the frequency planning requirements:

- Frequency reuse pattern: given the required C/I and the proposed technologies, specify the
frequency cell reuse pattern (e.g. 3-cell, 7-cell, etc.) and, for terrestrial systems, the
sectorization schemes assumed;
Answer:
1-cell reuse
Different sectorizations possible, e.g. 3 sectors/site, 6 sectors/site

- Characterize the frequency management between different cell layers;

Answer:
Frequency-separated cell layers. Additionally, reuse of frequencies of the macro layer possible
for TDD due to TDMA component.

- Does the SRTT use interleaved frequency allocation?

Answer:
No

- Are there any frequency channels with particular planning requirements?

Answer:
No
- all other relevant requirements.
NOTE 1 - The use of the second adjacent channel instead of the adjacent channel at a
neighbouring cluster cell is called "interleaved frequency planning". If a proponent is going to
employ an interleaved frequency plan, the proponent should state so in A1.2.4 and complete
A1.2.15 with the protection ratio for both the adjacent and second adjacent channel.
A1.4.16

Describe the capability of the proposed SRTT to facilitate the evolution of existing radio
transmission technologies used in mobile telecommunication systems migrate toward this
SRTT. Provide detail any impact and constraint on evolution.
Answer:
The detailed parameters of the RTT have been chosen with the easy implementation of dualmode UMTS/GSM.

A1.4.17

Are there any special requirements for base site implementation? Are there any features, which
simplify implementation of base sites? (terrestrial only)
Answer:
The base station configuration can be modular thus the number of user supported can be
increased modularly if desired, similar to introducing new TX/RX units to a GSM base station
with the difference being that RF hardware is not effected as only single RX/TX per base
station is required.

A1.5

Information required for terrestrial link budget template: Proponents should fulfill the link
budget template given in Table 1.3 of Annex 2 and answer the following questions.

A1.5.1

What is the base station noise figure (dB)?

Answer:
See Link Budget Template.

113

A1.5.2

What is the mobile station noise figure (dB)?

Answer:
See Link Budget Template.

A1.5.3

What is the base station antenna gain (dBi)?

Answer:
See Link Budget Template.

A1.5.4

What is the mobile station antenna gain (dBi)?

Answer:
See Link Budget Template.

A1.5.5

What is the cable, connector and combiner losses (dB)?

Answer:
See Link Budget Template.

A1.5.6

What are the number of traffic channels per RF carrier?

Answer:
Variable (depends on the rate of each traffic channel).

A1.5.7

What is the SRTT operating point (BER/FER) for the required Eb/N0 in the link budget
template?
Answer:
For speech BER = 10-3, for LCD BER = 10-6, for UDD BLER = 10%

A1.5.8

What is the ratio of intra-sector interference to sum of intra-sector interference and inter-sector
interference within a cell (dB)?
Answer:
Depends on the environment, in addition intra-cell interference cancellation, e.g. by means of
joint detection can be applied, especially for the TDD mode.

A1.5.9

What is the ratio of in-cell interference to total interference (dB)?

Answer:
Depends on the environment. In addition intra-cell interference cancellation, e.g. by means of
joint detection, can be applied and is typically used for the TDD mode.

A1.5.10

What is the occupied bandwidth (99%) (Hz)?

Answer:
Approximately 4.4 MHz

A1.5.11

What is the information rate (dBHz)?

Answer:
Service dependent.

A1.6

Satellite system configuration (applicable to satellite component only): Configuration details

in this subsection are not to be considered as variables. They are for information only.

A1.6.1-10

N/A

9.

114

ANNEX B: FULFILMENT OF REQUIREMENTS

Table 17. Technical Requirements and Objectives Relevant to the
Evaluation of Candidate Radio Transmission Technologies
IMT-2000 Item Description
Voice and data performance requirements
One-way end to end delay less than 40 ms**

Meets?*

Obj/Req

Source

Req

G.174, 7.5 Yes Y

No
Yes Y
Suppl.
No
F.720,
F.723,
G.114
Yes Y
G.174,
No
7.11 &
M.1079
7.3.1
Yes Y
G.174,
No
7.11 &
M.1079
7.3.1
Yes Y
M.1079
No
7.2.2
Yes Y
M.1034
No
10.8,
10.9

Req
For mobile videotelephone services, the IMT-2000 terrestrial component
should operate so that the maximum overall delay (as defined in ITU-T Rec.
F.720) should not exceed 400 ms, with the one way delay of the transmission
path not exceeding 150 ms
Speech quality should be maintained during 3% frame erasures over any 10 Req
second period. The speech quality criterion is a reduction of 0.5 mean opinion
score unit (5 point scale) relative to the error-free condition (G.726 at 32
kb/s)
DTMF signal reliable transport (for PSTN is typically less than one DTMF Req
errored signal in 104)

Voiceband data support including G3 facsimile

Req

Support packet switched data services as well as circuit switched data;

requirements for data performance given in ITU-T G.174

Req

Radio interfaces and subsystems, network related performance requirements

Network interworking with PSTN and ISDN in accordance with Q.1031 and Req
Q.1032
Meet spectral efficiency and radio channel performance requirements of
Req
M.1079
Provide phased approach with data rates up to 2 Mbit/s in phase 1
Obj
Maintain bearer channel bit-count integrity (e.g. synchronous data services
and many encryption techniques)
Support for different cell sizes, for example Mega cell Radius ~100-500 km
Macro cell Radius 35 km, Speed 500 km/h
Micro cell Radius 1 km, Speed 100 km/h
Pico cell Radius 50m, Speed 10 km/h
Application of IMT-2000 for fixed services and developing countries
Circuit noise - idle noise levels in 99% of the time about 100 pWp

Obj

Error performance - as specified in ITU-R F.697

Obj

Grade of service better than 1%

Obj

Obj

Obj

M.687-1
5.4
M.1034
12.3.3/4
M.687,
1.1.14
M.1034,
10.12
M.1035
10.1

Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y,
except
MEGA cells
No

M.819-1,
10.3
M.819-1,
10.4
M.819-1,
10.5

Yes Y
No
Yes Y
No
Yes Y
No

Source

Meets?*

Table 18.Generic Requirements and Objectives Relevant to the

Evaluation of Candidate Radio Transmission Technologies
IMT-2000 Item Description
Obj/Req
Radio interfaces and subsystems, network related performance requirements

* Explanation is requested when the candidate SRTT checks the No box.

** The source Recommendation suggests numerical limits for the overall delay, but provides no guidance about the measurement
techniques. Moreover there is an apparent inconsistency with ITU-T Rec. G.114, where the value of 40 ms is indicated as the
objective value. These issues are addressed in a liaison statement sent to the relevant ITU groups. Until TG 8/1 receives a
response and resolves this issue, proponents should submit candidates providing delay values using the methodology specified in
Rec. ITU-R M.1225.

115

Security comparable to that of PSTN/ISDN

Obj

Support mobility, interactive and distribution services

Req

M.687-1
4.4
M.816 6

Support UPT and maintain common presentation to users

Obj

M.816 4

Voice quality comparable to the fixed network (applies to both mobile and
fixed service)

Req

Support encryption and maintain encryption when roaming and during

handover
Network access indication similar to PSTN (e.g. dialtone)

Req

Meet safety requirements and legislation

Req

Meet appropriate EMC regulations

Req

M.819-1
Table 1,
M.1079
7.1
M.1034
11.3
M.1034
11.5
M.1034
11.6
M.1034
11.7
M.1034
12.1.2
M.1034
12.1.4
M.1034
12.2.2
M.1034
12.2.3
M.1034
12.2.5

Req

Support multiple public/private/ residential IMT-2000 operators in the same Req

locality
Support multiple mobile station types
Req
Support roaming between IMT-2000 operators and between different IMT- Req
2000 radio interfaces/ environments
Support seamless handover between different IMT-2000 environments such Req
that service quality is maintained and signalling is minimized
Simultaneously support multiple cell sizes with flexible base location, support Req
use of repeaters and umbrella cells as well as deployment in low capacity
areas
Support multiple operator coexistence in a geographic area
Req
Req
Support different spectrum and flexible band sharing in different countries
including flexible spectrum sharing between different IMT-2000 operators
(see M.1036)
Support mechanisms for minimizing power and interference between mobile Req
and base stations
Support various cell types dependent on environment (M.1035 10.1)
Req
High resistance to multipath effects

Req

Support appropriate vehicle speeds (as per 7)

NOTE: applicable to both terrestrial and satellite proposals
Support possibility of equipment from different vendors

Req
Req

Offer operational reliability as least as good as 2nd generation mobile systemsReq

Ability to use terminal to access services in more than one environment,
desirable to access services from one terminal in all environments
End-to-end quality during handover comparable to fixed services

Obj
Obj

Support multiple operator networks in a geographic area without requiring

Obj
time synchronization
Layer 3 contains functions such as call control, mobility management and
Obj
radio resource management some of which are radio dependent. It is desirable
to maintain layer 3 radio transmission independent as far as possible
Obj
Desirable that transmission quality requirements from the upper layer to
physical layers be common for all services
Reason for no: It is possible to trade transmission quality against capacity.
Furthermore, different services have different inherent requirements on the
transmission quality. This is in line with the discussion which took place
within ITU-R TG 8/1 WG4 on this topic.

M.1034
12.2.5
M.1034
12.2.8

Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No

Yes Y
No
Yes Y
No
yes
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No

Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
M.1035 8 Yes Y
No
M.1034
12.2.8.3
M.1034
12.2.9
M.1034
12.3.1
M.1034
12.3.2
M.1034
12.1.3
M.1034
12.3.5
M.1035
7.1

M.1035
8.1

Yes
No N

116

The link access control layer should as far as possible not contain radio
Obj
transmission dependent functions
Traffic channels should offer a functionally equivalent capability to the ISDN Obj
B-channels
Continually measure the radio link quality on forward and reverse channels Obj
Facilitate the implementation and use of terminal battery saving techniques

Obj

Accommodate various types of traffic and traffic mixes

Obj

Application of IMT-2000 for fixed services and developing countries

Repeaters for covering long distances between terminals and base stations, Req
small rural exchanges with wireless trunks etc.
Withstand rugged outdoor environment with wide temperature and humidity Req
variations
Provision of service to fixed users in either rural or urban areas
Obj
Coverage for large cells (terrestrial)

Obj

Support for higher encoding bit rates for remote areas

Obj

Additional satellite- component specific requirements and objectives

Links between the terrestrial and satellite control elements for handover and Req
exchange of other information
Take account for constraints for sharing frequency bands with other services Obj
(WARC-92)
Compatible multiple access schemes for terrestrial and satellite components Obj
Service should be comparable quality to terrestrial component as far as
possible
Use of satellites to serve large cells for fixed users

Obj

Key features (e.g. coverage, optimization, number of systems)

Obj

Radio interface general considerations

Req

Doppler effects

Req

Obj

M.1035
8.3
M.1035
9.3.2
M.1035
11.1
M.1035
12.5
M.1036
1.10

Yes Y
No
Yes Y,
No
Yes Y
No
Yes Y
No
Yes Y
No

M.819-1
Table 1
M.819-1
Table 1
M.819-1
4.1
M.819-1
7.2
M.819-1
10.1

Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No
Yes Y
No

M.818-1
3.0
M.818-1
4.0
M.818-1
6.0
M.818-1
10.0
M.819-1
7.1
M.1167
6.1
M.1167
8.1.1
M.1167
8.1.2

Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
No

Table 19.Subjective Requirements and Objectives Relevant to the

Evaluation of Candidate Radio Transmission Technologies*
IMT-2000 Item Description
Fixed Service - Power consumption as low as possible for solar and other sources

Obj/Req
Req

Minimize number of radio interfaces and radio sub-system complexity, maximize

commonality (M.1035 7.1) Y
Minimize need for special interworking functions Y

Req

Minimum of frequency planning and inter-network coordination and simple resource

management under time-varying traffic Y
Support for traffic growth, phased functionality, new services or technology evolution Y

Req

Facilitate the use of appropriate diversity techniques avoiding significant complexity if

possible Y
Maximize operational flexibility

Req

Designed for acceptable technological risk and minimal impact from faults

Req

When several cell types are available, select the cell that is the most cost and capacity
efficient

Obj

Req

Req

Req

Source
M.819-1
Table 1
M.1034
12.2.1
M.1034
12.2.4
M.1034
12.2.6
M.1034
12.2.7
M.1034
12.2.10
M.1034
12.2.11
M.1034
12.2.12
M.1034
10.3.3

117

Minimize terminal costs, size and power consumption, where appropriate and consistent
with other requirements

10.

Obj

M.1036
1.12

118

ANNEX C CAPACITY AND COVERAGE RESULTS

10.1

Introduction

The performance of the proposed RTT (UTRA FDD mode) has been evaluated by means of computer simulations.
This evaluation is carried out based on the methods and conditions described in ETSI Technical Report 101 112,
Annex B. The evaluated test cases are contained in Table 1. The values have been taken from Attachment 7 of ITUR Circular Letter 8/LCCE/47. In that table it is said that the lowest bit rate and the highest bit rate possible by the
RTT under test should be evaluated. The cases that will be provided for the submission in June 1998 are indicated
with the word <high> in Table 20. The reason is limited amount of time. However the essential test cases should
have been covered since the high priority test cases coincide with the minimum performance capabilities as defined
in Annex 6 of the Circular Letter. The following test data rates have been made chosen from the exhaustive list of
test cases in the Circular Letter:
For the speech service the 8 kbps bearer have been selected.
Only the Long Constrained Delay (LCD) data service has been evaluated. The short delay data service is thus
excluded. The reason is that there are no specific delay requirements defined in the ITU-R test services which
could give any guidance about what level of delay those definitions assume.
The data packet services are called UDD (Unconstrained Delay Data) in Table 1 and is modelled as a packet
service with ARQ protection and no loss of data is expected over the radio link. Hence, there is no need to
specify a Bit Error Ratio (BER) requirement.
The performance evaluation consists of two stages: the link level simulation and the system level simulation. Each
stage includes both uplink (UL) simulations and downlink (DL) simulations. This document describes the detailed
conditions and assumptions for each stage. Note that the results provided will be subject for further refinement and
analysis during the evaluation phase and may also be subject to the ongoing development of the radio interface
specification.
Table 20. Simulation cases and priorities
Indoor (A), 3km/h

Pedestrian (A), 3 km/h

Vehicular (A), 120 km/h

8 kbps <high>

8 kbps <high>

8 kbps <high>

LCD

64 kbps

64 kbps

64 kbps

BER = 10 -6

2048 kbps <high>

384 kbps <high>

144 kbps <high>

UDD

64 kbps

64 kbps

64 kbps

2048 kbps <high>

384 kbps <high>

144 kbps <high>

Speech
BER = 10 -3

10.2
10.2.1

Implementation OF Simulations
Link-Level Simulations

10.2.1.1
Simulation Model
In the simulations, sampling was made at chip level. Fast power control is included in all simulations. In actual
systems, power control commands are sent on the return channel, i.e. the uplink power control commands are sent on
the downlink and vice versa. In the simulations, random errors with a certain error probability are added to the power
control commands. To find the appropriate values for this error probability of (Transmit Power Control) TPC
symbols, the errors of TPC symbols are collected under the simulation condition that provides predetermined BER
for information bit stream, e.g. BER of 10-3 for speech. The error probabilities of 4% and 1% are used for speech
and high-speed data respectively for the UL and DL. The power amplifier is not modelled; i.e. an ideal power
amplifier is assumed.
A fixed searcher is used in the receiver; i.e. the receiver knows the delay of all rays and picks up the energy of some
rays using a fixed set of fingers in the RAKE. This is further discussed in the section describing the channel models.

119

The channel estimation is based on several pilot groups and the different groups are multiplied by a weighting factor.
There are several possible choices of weighting factor and all choices have not been evaluated for all test cases due
to the limited amount of time. Further studies will be undertaken in order to determine if one of the choices is
superior to the others.
All interference is modelled as additive white Gaussian noise.
Table 21. Simulation parameters and methods for UL
Channel estimation method

Channel estimation value is based on the present pilot group and pilot
groups before and after the present slot. The different pilot groups are
multiplied by a weighting factor. In the parameter list for the different
test cases the vector alpha contains the weight factors for the pilot
groups. The weighting factor for the present slot corresponds to the
element in the middle of the alpha vector.

SIR estimation method

S: Channel estimation value per each Rake finger is calculated as

average of pilot symbols from one pilot group. S is the sum of the
powers of channel estimates from different fingers.
I: Interference is assumed constant in the link level simulations

Channel model

According to M.1225 Annex 2 (Annex B of ETSI TR 101 112 ).

V=3,120 km/h (c=3*108, fc=2 GHz)

Number of RAKE fingers

Indoor, Outdoor to indoor and pedestrian: 2 fingers/branch

Vehicular: 4 fingers/branch

Searcher

Fixed delays, see Figure 73

Sampling rate

Chip level sampling (1 sample/chip)

PC dynamic range

80 dB

PC delay

1 slot (0.625 ms)

PC symbol error

4% random error for speech

1% random error for high speed data

Interference from other users

Modelled as AWGN

Eb/No scaling

Eb will be calculated as the received power for each information bit.

The following items will be calculated as overhead: pilot, TPC, TFI,
Outer coding synch info, CRC, tail, Convolutional Coding, RS coding,
repetition, block number in the case of packet.

Table 22. Simulation parameters and methods for DL

Channel estimation method

SIR estimation method

Channel estimation value is based on the present pilot group and pilot
groups before and after the present slot. The different pilot groups are
multiplied by a weighting factor. In the parameter list for the different
test cases the vector alpha contains the weight factors for the pilot
groups. The weighting factor for the present slot corresponds to the
first element that is one in the alpha vector.
Each pilot symbol will be processed through coherent detection and
maximum ratio combining with the channel estimation value of its
own block and several blocks before own block. The blocks are
multiplied by a weighting factor and the different weights are
collected in the vector alpha.
Vehicular-A: 2 blocks are used, alpha = (0.6,1.0)
Indoor-A: 3 blocks are used alpha = (0.3, 0.8, 1.0)

120

Channel model
Number of RAKE fingers
Sampling rate
PC dynamic range
PC delay
Searcher
Interference from other users
TPC bit error
Eb/No scaling

Pedestrian-A: the same as indoor-A

S: S is calculated as a square of the average of the results from the
coherent detection and the maximum ratio combining of pilot
symbols within the pilot block.
I: Interference is calculated as an exponentially weighted average of
the variance of the results from coherent detection and maximum
ratio combining of the pilot symbols within the pilot block (A
forgetting factor of 0.99 is used in the exponentially weighted
averaging.)
According to M.1225 Annex 2 (Annex B of ETSI TR 101 112).
v=3,120 km/h (c=3x108, fc=2GHz)
Indoor, Outdoor to indoor and pedestrian: 2
Vehicular : 4
Chip level sampling
Unlimited
1 slot (0.625 ms)
Fixed delays, see Figure 73
Modelled as AWGN
4% random error for speech
1% random error for high speed data
Eb will be calculated as the received power for each information bit.
Following items will be calculated as overhead: pilot, TPC, TFI,
Outer coding synch info, CRC, tail, Convolutional Coding, RS
coding, repetition, block number in the case of packet.

10.2.1.2
Channel Models
The channel models given in M.1225 Annex 2, also used in ETSI TR 101 112, cannot be used right away, since the
time resolution of the simulation model is one sample. For the simulations the following model was used:
Each ray is split into two rays, one to the sample to the left and one to the sample to the right. The power of these
new rays is such that the sum is equal to the original power, and the power of each of the new rays is proportional to
the (1-normalised distance to the original ray). Finally, the power of all rays on one sample are added up and
normalised. This yields a model with a number of independently Rayleigh fading rays on the sampling instants.

121

In the simulations the sampling time is equal to the chip time, resulting in the channel models in Figure 73 that were
used in simulations.
Outdoor to indoor and pedestrian A
0

10

10

Average power [dB]

Average power [dB]

Indoor office A
0

15
20

15
20

25

25

30

30

35
0

100

200
300
Relative delay [ns]

400

500

35
0

100

10

10

15
20

30

30

1000
1500
2000
Relative delay [ns]

20

25

20
25

500

500

15

25

35
0

400

Vehicular B

Average power [dB]

Average power [dB]

Vehicular A

200
300
Relative delay [ns]

2500

3000

35
0

10
15
Relative delay [us]

Figure 73. Modified channel models used in the simulations.

The rays picked up by the RAKE receiver are marked with o in Figure 73, while other rays are marked with x.
No special link simulations were made for soft handover situations. In a soft handover the result from two single
connection RAKEs are combined. For the Vehicular case this would mean 8 RAKE fingers. However, the number of
RAKE fingers can be lowered in soft handover without affecting the performance, so 4 - 6 fingers should be
sufficient.

10.2.2

System-Level Simulations

10.2.2.1
Simulation Environment
The simulation environments are described in ETSI TR 101 112 Annex B. Implementation assumptions are
described below.
The Indoor office environment characterises a three floors office building where users are moving (3 km/h) between
an office room to the corridor or vice versa. The base stations (60 base stations all using Omni-directional antennas)
are deployed in every second office room. No wall propagation loss was assumed, only between floors.
The Outdoor to indoor and pedestrian deployment environment is a Manhattan-like environment with the block size
of 200 m and low speed (3 km/h) users. The environment consists of 72 base stations and are located as described in
Annex B of ETSI TR 101 112. The base stations are using Omni-directional antennas and are deployed 10 m above
ground, which is below the rooftops. The radio propagation going above rooftops is also included in the system
simulation model. The street width is 30 m and it is assumed that the pedestrians are moving in the middle of the
street.
The Vehicular environment is classic macro environments with site-to-site distance of 6 km (1.5 km site-to-site is
used for UDD144 and LCD144). Three-sectored sites are used, i.e. each site is serving three sectors (cells). The
speed of the mobile stations is 120 km/h. Wrap around is used in order to make an infinite cell plan, i.e. there are no
border effects in the simulations. The BS transmit power is limited to 20 W, including common control channels.

122

10.2.2.2
Downlink Orthogonality
The downlink will not be perfectly orthogonal due to multipath propagation. The downlink orthogonality factor, i.e.
the fraction of the total output power that will be experienced as intra-cell interference, has been calculated for the
different environments and is presented in Table 23. An orthogonality factor of zero corresponds to a perfectly
orthogonal downlink, while a factor of one is a completely non-orthogonal downlink. As seen in the table below,
40% of the power transmitted from the own cell will act as intra-cell interference in the Vehicular environment.
Table 23. Orthogonality factor for the environments different propagation models.
Propagation model
Indoor office A
Outdoor to indoor and
pedestrian A
Vehicular A

Orthogonality factor
0.10
0.06
0.40

The orthogonality factor has been derived in the following way:

Two simulations were made one with white Gaussian noise and one with intra-cell interference. The BER was then
plotted as a function of Eb/No and Eb/Io respectively. These curves may differ significantly, where the Eb/Io curve is
to the left of the Eb/No curve. A difference of 10 dB means that a given Eb/Io gives the same BER as Eb/No = Eb/Io +
10. Consequently, a certain Io in the system simulations is equivalent to having 10 dB less No in the link-level
simulations. Hence, it is possible to say that the orthogonality removes 90% of the interference, or in other words an
orthogonality factor of 10% is obtained (10% of the interference remains).
10.2.2.3
Multiple downlink scrambling codes
In the downlink all users in a cell (sector) use the same scrambling code. Hence, all users share the available
channelisation codes in the OVSF code-tree. This means that the channelisation codes in the downlink is a much
more limited resource in the downlink than in the uplink, where each user has its own scrambling code and therefore
can utilise all the codes in the code-tree. On the other hand, with only one scrambling code in the downlink, the
degree of orthogonality between users (and physical channels) will be high. The orthogonality depends on the
multipath profile of the channel.
Since orthogonality in the downlink helps increasing downlink capacity, one could draw the conclusion that the
current solution with one orthogonal code set in the downlink is the right one. Also, since CDMA systems are
interference limited, the limited number codes should probably not be a problem. However, there are situations
where the limited number of codes in the downlink can be too small.
The are two ways to increase the number of downlink channelisation codes:
1. Use multiple downlink scrambling codes (use multiple code-trees as described above).
2. Change the channel coding scheme in order to increase the spreading factor (the number of channelisation codes is
equal to the spreading factor).
The current simulation results utilize the first approach only. The users have been allocated randomly between
different code sets within the cell. Further, it is assumed that the orthogonality factor within of the code set is same as
the values stated in Table 23 and that no orthogonality between different code sets exists.
10.2.2.4
Soft / Softer Data Combining
For the Indoor office and the Outdoor to indoor and pedestrian environment soft handover is used between base
stations. This means that the uplink C/I (or SIR = PGC/I) is calculated as selection diversity and the downlink as
maximum ratio combining (a sum of the received C/I from each base station). For the Vehicular environment softer
handover is used, i.e. the mobile is connected to several sectors belonging to the same site, which will affect the
calculation of the uplink C/I. Therefore the uplink C/I for all sectors belonging to one site is calculated as maximum
ratio combining. Soft handover in the Vehicular environment is treated as regular selection diversity.
The softer handover data combining (maximum ratio combining) is performed on Layer 1 in the UTRA/FDD mode.
Softer handover is used only in the Vehicular environment. In the uplink (within one site) and downlink the SIR
during softer handover is modelled as:

SIRcombined =

SIR

sec tor

sec tors

The combined downlink (maximum ratio combining) SIR during soft handover is modelled as:

SIR DL ,combined =

SIR

sec tors

sec tors

The combined uplink (selection diversity) SIR during soft handover is modelled as:

123

SIRUL ,combined = max( SIRcombined )

sites

10.2.2.5
Increase in TX Power due to Power Control
One effect of the fast power control is that the transmitted power from each mobile will vary with time, and this can
cause an increase in the average background interference power.
For the speech service the average transmitted powers increase is used when calculating the interference to other
cells (the power increase will not affect the own cell). A good model of the power increase is perfect tracking of the
fast fading. This assumption is valid only for the 3 km/h cases (Indoor office and Outdoor to indoor and pedestrian).
The power increase in the Vehicular environment is negligible since the power control cannot track the fading, and is
therefore not included in the system simulations.
For the UDD simulations fast fading values from the link-level simulations are used in the system-level simulator to
adjust the output power of the transmitters for each frame. This means that for each frame a new fading value will be
used when calculating the gain matrix (including path loss, shadow fading and fast fading).
10.2.2.6
Radio Resource Management
Fast SIR based power control is assumed in both uplink and downlink, and the powers of the transmitters are
balanced to meet the averaged SIR during one frame.
Soft/softer handover is used for the circuit-switched services. The soft/softer handover algorithm simply connects the
strongest, based on path loss (excluding fast fading), base stations within the handover window. The soft/softer
handover window threshold is set to 3 dB and the algorithm is executed every 0.5 second and the maximum active
set size is two. No significant performance improvement is expected by having an active set size of three or more in
these environments. Measurement errors are not included. No soft handover is currently used in the packet
simulations; the user simply connects to the strongest base station. Some simulation cases also use C/I based soft
handover, which means that the handover decision is based on path loss and uplink interference, i.e. the algorithm
tries to minimise the MS transmit power.
About 5% of the total BS power is allocated to downlink common control channels, which will interfere all users in
the system. I.e. we assume in the system level simulations that all common channels are acting as non-orthogonal
channels, which is a bit pessimistic since only the SCH is non-orthogonal to the users within the same cell.
For the UDD service dedicated channel packet transmission is used. No random access / forward access signalling is
included in the results.
We assume that a RLC block can be re-transmitted in the next frame, i.e. that the ACK/NACK channel is error free
and infinitely fast.
10.2.2.7
Performance Measures
Circuit-Switched Services
The circuit-switched services have been evaluated by means of dynamic system simulations. The performance
measure of the speech (8 kbps, 50% voice activity) and LCD services is that 98% of the users are satisfied. A user is
satisfied if all three of the following constraints are fulfilled:
1. The user does not get blocked when arriving to the system.
2. The user has sufficiently good quality more than 95% of the session time. The quality threshold is defined as
BER =10-3 (speech) or BER = 10-6 (LCD).
3. The user does not get dropped. A speech user is dropped if BER > 10-3 during 5 s and e.g. for the LCD 384 a
user is dropped if BER > 10-6 during 26 s.
Packet Services
The performance measure of the packet services is that 98% of the users are satisfied. A user is satisfied if all three
of the following constraints are fulfilled:
1. The user does not get blocked.
2. The user does not get dropped.
The active session throughput2 shall not be below 14.4 kbps (UDD 144), 38.4 kbps (UDD 384) or 204.8 kbps (UDD
2048) if a user should be satisfied. Taking the invert will show the delay per bit that is allowed.
The time waiting on ACK/NACK (i.e. when the transmitter buffer is empty) is not included when calculating the
active session throughput. If the data packet that shall be transmitted has fewer bits than can be transmitted in a
frame, dummy bits (or rather dummy blocks) are added. These dummy bits are not included when calculating the
session throughput, however they will increase the interference in the system. A data packet will be divided into data
2

The active session throughput is defined as the ratio of correctly received bits during the entire session and the
session length excluding the time when there is nothing to transmit (i.e. empty input buffer).

124

blocks of 340 bits (304 information bits) for the uplink. Several blocks are then put into a frame, e.g. 8 blocks per
frame for the UDD 384 service.

10.3
10.3.1

UTRA/FDD Results
Link-Level Simulations

The Eb/No values presented here are the actual Eb/No values that are needed in the receiver to achieve the
corresponding BER, FER and BLER. The Eb/No values include all overhead, i.e. the DPCCH (Dedicated Physical
Control Channel: pilot symbols, power control bits, TFI) and overhead on the DPDCHs (Dedicated Physical Data
Channels) such as CRCs, block numbers and tail bits for the convolutional code. In other words, the Eb value
contains all energy needed to transmit one information bit. Energy from common broadcast channels is not included
in the link-level results.
After coding of the DPDCH rate matching is applied, using puncturing or repetition. On the DPCCH rate matching is
always performed using repetition. The rate matching used for the different services are given below. The
interleaving for the different services is specified in the tables below as x*y. This should be interpreted as vectors of
x bits are read in and vectors of y bits are read out at the transmitter according to a conventional block-interleaver
scheme.
The detailed simulation parameters are shown in Tables 24 to 29, and the link results are presented in Tables 30 to
32.
Note that in the results and parameters below concatenated coding using Reed-Solomon and convolutional codes
have been used for services that require bit error rates of around 10-6 after forward error correction. Turbo coding
has been studied as an alternative. First results using the turbo encoder in Figure 74 have indicated gains of around 2
dB at BER 10-6, when using turbo codes instead of concatenated coding (with similar complexity).

Figure 74. Example of turbo encoder considered for UTRA.

125

Table 24. Parameters for Speech (8kbps) in UL

Physical channel rate

32 kbps

32 kbps

Info/CRC/tail bit per frame

80/16/8

160/16/8 20ms

Convolutional coding rate

1/3

1/3

Repetition

8 bits/10 ms (312 ->320)

88 bits/20ms (552->640)

Interleaver

10 ms, 16*20

20 ms, 20*32

Pilot/TPC/TFI bits per slot

6/2/2 or 7/3/0

6/2/2 or 7/3/0

Antenna receiver diversity

On

On

Channel estimation

Indoor, Outdoor to indoor

and pedestrian: 5 pilot
groups, alpha = (1,1,1,1,1)
Vehicular A: 3 pilot groups,
alpha = (0.4,1,0.4)

Indoor, Outdoor to indoor

and pedestrian: Present slot
and 8 previous averaged
Vehicular: Present slot and
the previous one averaged

-3

-3

DPCCH/DPDCH power [dB]

Table 25. Parameters for LCD Services in UL

Source rate

64 kbps

144 kbps

384 kbps

2048 kbps

Physical channel
rate

256 kbps

512 kbps

1024 kbps

1024 kbps * 6

Information bits

5120 bits
(80 ms)

11520 bits
(80ms)

30720 bits
(80 ms)

163840 bits
(80 ms)

RS coding

(36,32)

(36, 32)

(36,32)

(36,32)

Symbol
Interleaver

80 ms, 36*20

80 ms, 36*45

80 ms, 36*120

80 ms, 36*640

Outer coding
process synch.
Info

3 bits per subframe

(subframe=720
bits)

3 bits per subframe (sub-frame

= 810 bits)

3 bits per subframe

(subframe=720
bits)

3 bits per subframe

(subframe=720
bits)

CRC

13 bits per subframe

(subframe=720
bits)

13 bits per subframe (sub-frame

= 810 bits)

13 bits per subframe

(subframe=720
bits)

13 bits per subframe

(subframe=720
bits)

Tail

8 bits per frame

(1 sub-frame = 1
frame, 10 ms)

8 bits per frame

(2 sub-frames = 1
frame, 10 ms)

8 bits per frame

(6 sub-frames = 1
frame, 10 ms)

8 bits per frame

(32 sub-frames =
1 frame, 10 ms)

1/3

1/3

1/3

1/3

328 bits/10 ms
(2232 -> 2560)

140 bits/10 ms
(4980 -> 5120 )

3032 bits/10 ms
(13272 -> 10240)

9240 bits/10 ms
(70680 -> 61440

Convolutional
coding rate
Repetition

126

per 6 code
channels)
Bit Interleaver

10 ms, 16*160
bits

10 ms, 16*320

10 ms, 16*640

10 ms, 16*640
per one code
channel

Pilot/TPC/TFI
bits per slot

6/2/2

6/2/2

6/2/2

6/2/2

On

On

On

On

-9

-9 relative one
DPDCH code

Antenna receiver
diversity
Channel
estimation

3 pilot groups, alpha = (0.4,1,0.4)

DPCCH/DPDCH
power [dB]

-9

-9

Table 26. Parameters for UDD Services in UL

Source rate

64 kbps

144 kbps

384 kbps

2048 kbps

30.4 kbps

60.8 kbps

243.2 kbps

486.4 kbps

128 kbps
(sf_DPDCH = 32)

256 kbps
(sf_DPDCH = 16)

1024 kbps
(sf_DPDCH = 4)

1024 kbps
(sf_DPDCH = 4)

304 bits

304 bits

304 bits

304 bits

# blocks per frame

16

CRC

16

16

16

16

Block number

12

12

12

12

Tail

1/3

1/3

1/3

1/3

Rate matching

Repetition
520 bits/10ms
(2040 -> 2560)

Repetition
520 bits/10ms
(2040 -> 2560)

Repetition
520 bits/10ms
(2040 -> 2560)

Puncturing
6080 bits/10ms
(16320 -> 10240)

Interleaving

10 ms or 20ms

10 ms or 20 ms

10 ms

10 ms

6/2/2

6/2/2

6/2/2

6/2/2

On

On

On

On

Information bit rate

Physical channel rate

Block size

Convolutional coding rate

Pilot/TPC/TFI bits per

slot
Antenna receiver diversity
Channel estimation

DPCCH/DPDCH power
[dB]

Indoor, Outdoor to indoor and pedestrian: Present slot and 7 previous averaged
Vehicular: Present slot and the previous one averaged
-6

-6

-12

-12

127

Table 27. Parameters for Speech (8 kbps) in DL

Physical channel rate

32 ksps

32 ksps

Info/CRC/tail bit per frame

80/16/8

160/16/8 20ms

Convolutional coding rate

1/3

1/3

Repetition

8 bits/10 ms (312 -> 320)

24 bits/20ms (552 -> 576)

Interleaver

10 ms, 16*20

20 ms, 32*18

Pilot/TPC/TFI bits per slot

8/2/0

8/2/0

Antenna receiver diversity

Off

Off

Channel estimation

Indoor, Outdoor to indoor and pedestrian: 4 pilot groups

alpha = (0.6,1.0,1.0,0.6)
Vehicular: 6 pilot groups alpha = (0.3,0.8,1.0,1.0,0.8,0.3)

Table 28. Parameters for LCD Services in DL

Source rate
Physical channel
rate
Information bits

64 kbps
64ksps

144 kbps
256ksps

384 kbps
1024ksps

2048 kbps
1024ksps x 4

5120(80ms)

11520(80ms)

RS coding
Symbol
interleaver
CRC

(36, 32)
36x20

(36, 32)
36x45

5120(1B)x6
(80ms)
(36, 32) per 1B
36x20 per 1B

5120(1B)x32
(80ms)
(36, 32) per 1B
36x20 per 1B

13 bit

13 bit per subframe

(subframe=6480bit)

Tail

8 bit

8 bit per 8frame

(2subframe=8 frame)

Convolutional
coding rate
Rate matching

1/3

1/3

13 bit per
subframe
(1B=1subframe)
8 bit per 8frame
(6subframe=8
frame)
1/3

13 bit per
subframe(1B=1su
bframe)
8 bit per 8frame
(32subframe=8
frame)
1/3

65 bits repetition
(80ms)

326 bits puncturing

(80ms)

776 bits repetition

(80ms)

Bit interleaver

80[ms], 128x136

80[ms], 128x302

Pilot/TPC/TFI bit
per slot
DPCCH/DPDCH
power [dB]

8/2/0

254 bits
repetition
(80ms)
80[ms],
128x814
16/2/0

Channel
estimation
Antenna receiver
diversity

0 dB

0dB

0 dB

80[ms], 128x4336

3 dB

Indoor, Outdoor to indoor and pedestrian: 4 pilot groups alpha = (0.6,1.0,1.0,0.6)

Vehicular: 6 pilot groups alpha = (0.3,0.8,1.0,1.0,0.8,0.3)
On

On

On

On

128

Table 29. Parameters for UDD Services in DL

Source rate

64 kbps

144 kbps

384 kbps

2048 kbps

Information bit rate

30.4 kbps

60.8kbps

243.2kbps

486.4kbps

Physical channel rate

64ksps

128ksps

512ksps

1024ksps

Block size

304 bits

304 bits

304 bits

304 bits

#blocks per frame

16

CRC

16

16

16

16

Block number

12

12

12

12

Tail

Convolutional coding rate

1/3

1/3

1/3

1/3

Rate matching

Repetition
8 bits/ 10ms
(2040 -> 2048)
10[ms] 16x128

Interleaver

Repetition
4 bits/10 ms
(1020 ->1024)
10[ms] 16x64

10[ms] 16x510

10[ms] 16x1020

Pilot/TPC/TFI bit per slot

8/2/0

Channel estimation

16/2/0

Indoor, Outdoor to indoor and pedestrian: 4 pilot groups alpha = (0.6,1.0,1.0,0.6)

Vehicular: 6 pilot groups alpha = (0.3,0.8,1.0,1.0,0.8,0.3)

Antenna receiver diversity

On

On

On

Table 30. Link results for speech.

Eb/No @ BER = 10-3
[dB]
(UL / DL)
4.8 / 6.7

Service

Environment

Speech
(8 kbps, 50% VA)
10 ms interleaving

Indoor(A),
3km/h
Pedestrian (A),
3km/h
Vehicular (A),
120km/h

6.8 / 6.4 / 8.8 (no TFI)

Speech
(8 kbps, 50% VA)

Indoor(A),
3km/h

3.5 / 3.3 / 6.0 (no TFI)

20 ms interleaving

Pedestrian (A),
3km/h
Vehicular (A),
120km/h

4.8 / 6.8

3.6 / 3.3 / 6.1 (no TFI)

6.3 / 6.2 / 7.9 (no TFI)
5.6 / - (no TFI & PC
step = 0.25 dB)
The shaded columns represent the cases with high priority.

On

129

Table 31. Link results for LCD services.

Service

Environment

LCD 64

Eb/No @ BER = 10-6

[dB]
(UL / DL ant. div.)
3.8 / -

Indoor (A),
3km/h
Pedestrian (A),
3.3 / 3km/h
Vehicular (A),
5.2 /
120km/h
LCD 144
Vehicular (A),
3.2 / 2.5
120km/h
LCD 384
Pedestrian (A),
1.9 / 1.1
3km/h
LCD 2048
Indoor (A),
2.5 / 1.6
3km/h
* The shaded columns represent the cases with high priority.
Table 32. Link results for UDD services.
Service

Environment

UDD 64

Indoor (A),
3km/h
Pedestrian (A),
3km/h
Vehicular (A),
120km/h

UDD 144

Vehicular (A),
120km/h

Eb/No @ BLER = 106

[dB]
(UL / DL ant. div.)
1.4 / 1.2
1.4 / 1.2
3.8 / 3.0
3.5 / - (20 ms interl.)
3.3 / - (20 ms interl. &
PC step = 0.25 dB)
3.0 / 2.9
2.7 / - (20 ms)
2.6 / - (20 ms interl. &
PC step = 0.25 dB)

UDD 384

Pedestrian (A),
0.4 / 0.1
3km/h
UDD 2048
Indoor (A),
0.5 / 0.1
3km/h
* The shaded columns represent the cases with high priority.

10.3.2

System-Level Simulations

Dynamic system simulations have been performed for three different types of services in three different
environments described in Annex B of ETSI TR 101 112. In these simulations all base stations are assumed to be
equipped with one 4.096 Mcps UTRA/FDD carrier using 5 MHz carrier spacing (assuming 3 carriers within 15
MHz). It is likely that the concept will perform better if a larger bandwidth is used for higher data rates due to a
better interference averaging). Therefore all results of higher data rate services shall be regarded as pessimistic
results. Also, the simulations of the UDD services have only used a fixed bit-rate radio bearer, which will also
decrease the performance of the UDD services.
10.3.2.1
Circuit-Switched Services
Four circuit-switched services, speech, LCD 144, LCD 384 and LCD2048, have been evaluated by means of
dynamic system simulations. The performance measure of the speech (8 kbps, 50% voice activity) and LCD services
is that 98% of the users are satisfied. No admission control has been used; therefore no users are blocked due to high
interference level. Also, the simulation results show that cell capacity in all cases is limited by the requirement that a
satisfied user must have sufficiently good quality more than 95% of the session time or the blocking criteria has been
reached and not by the dropping criteria.
The UTRA/FDD concept uses fast power control also in downlink. This means that slow moving users can
compensate for the fast channel fading, hence no substantial diversity gain from connecting more base stations (i.e.

130

increase the maximum number of active set) is seen. Connecting more base stations will only increase the required
capacity of base station to base station controller transmission. Users moving with high speed do not require good
tracking of the fast channel fading due to the gain from coding and interleaving.
Speech and LCD results are presented in Table 33. The shaded columns represent the cases with high priority as
defined in Table 20. At this moment of the submission of this document, only these cases with high priority are
simulated. The other cases are left for further investigation. Please note that in Table 33 the term cell is defined as
an area covered by a sector.
The speech service is evaluated using 50% voice activity. However, the DPCCH is transmitted with constant bit-rate
(and energy) independent of the speech user information rate (8 kbps or 0 kbps information bit-rate). Therefore, the
spectrum efficiency will increase more than 25-30% if a voice activity of 100% is used, due to the decreased
DPCCH (relative) overhead.
10.3.2.2
Packet Services
Three different packet data services have been evaluated: UDD 144, UDD 384 and UDD 2048. The performance
measure of the packet services is that 98% of the users are satisfied.
As mentioned before the UDD services are evaluated using a fix rate bearer, e.g. 60.8 kbps for UDD 144 and 480
kbps for UDD 2048. Better interference averaging will be achieved if higher chip-rate or variable rate is used, i.e.
this will improve the results for the high data rates.
The results for the UDD services are shown Table 33. The shaded columns represent the cases with high priority as
defined in Table 20. At the moment of the submission, only the cases with high priority have been simulated. The
other cases are left for further investigation.

10.3.3

Capacity results

Table 33 shows the capacity results. The table includes also the link level results that have been used in the system
simulations.

131

Table 33. Summary of simulation results. The voice activity is 50% for the speech service. Note that UDD bearer
bit rates are not the same as the source bit rates specified for the UDD service.
Service

Environment

Source bit
rate

Eb/No [dB]
(UL / DL)

Speech

Indoor (A),
3km/h
Pedestrian (A),
3km/h
Vehicular (A),
120km/h
Indoor (A),
3km/h

8 kbps

20 ms
interl.

LCD

3.3 / 6.0

Cell capacity
[Erlang/carrier/cell]
(UL / DL)
165 / 883

Spectrum efficiency
[kbps/MHz/cell]
(UL / DL)
132 / 70

8 kbps

3.3 / 6.1

154 / 1574

123 / 125

8 kbps

5.6 / 7.9

107 / 885

86 / 70

2.5 / 1.6

0.35 / 0.516 / -

144 / 207 / -

1.9 / 1.1

3.0 / 6.07
4.36 / -

230 / 461
330 / -

144 kbps

3.2 / 2.5

64 kbps

1.4 / -

204 / 198
204 / 210
227 / -

2048 kbps

0.5 / 0.1

64 kbps

1.4 / -

384 kbps

0.4 / 0.1

64 kbps

3.8 / -

144 kbps

3.0 / 2.9

7.1 / 6.98
7.1 / 7.34
979 / (parallel sessions)
51 / 824
(parallel sessions)
1529 / (parallel sessions)
91 / 1354
(parallel sessions)
729 / (parallel sessions)
59 / 85
(parallel sessions)

64 kbps
2048 kbps

Pedestrian (A),
3km/h

64 kbps
384 kbps

Vehicular (A),
120km/h

UDD

Indoor (A),
3km/h

Pedestrian (A),
3km/h

Vehicular (A),
120km/h

64 kbps

280 / 453
356 / 449 / 668
168 / 202 / 290

* The shaded columns represent the cases with high priority.

10.4

Coverage Analysis (Link Budget Calculation)

In the following pages link budgets are presented for the simulated test cases. The link budgets follows the link
budget template in ETSI TR 101 112 (UMTS 30.03) which is based on the link budget template as contained in
ITU-R M.1225, and also presents some range calculations using concept optimized parameters.

10.4.1

Basic Assumptions

Since it is the average transmitter power per traffic channel that is specified in UMTS 30.03, power control is
included in the link-level simulations to find the coverage. However, this means that the transmitted power can be
increased due to the power control, and this is compensated for in the row Power control TX power increase.
3

Capacity at 2% blocking (code limited case, interference limited at 99 Erlang/carrier/cell).

Two code sets are used.

Capacity at 2% blocking (code limited case, interference limited at 91 Erlang/carrier/cell).

C/I based soft handover.

Four code sets are used.

Capacity at 2% blocking (code limited case, interference limited at 9.0 Erlang/carrier/cell).

Estimated capacity based on the higher bit-rate service.

132

The TX power increase is dependent on the environment and service. For speech and LCD soft handoff is assumed,
while UDD services assume no soft handoff. The values used are presented in Table 34.
Table 34. TX power increase in different environments.
Environment
Speech & LCD
Speech & LCD
UDD
UDD
Uplink [dB]
Downlink [dB]
Uplink [dB]
Downlink [dB]
Indoor office
0
2
2
4
Outdoor to indoor and
0
2
2
4.5
pedestrian
Vehicular
0
0
0
0
The handoff gain and log-normal fade margin were calculated for 95% area coverage with a shadowing correlation
of 50%. Values can be found in Table 35.
Table 35. Log-normal fade margins and handoff gains.
Environment

Indoor office
Outdoor to indoor and
pedestrian
Vehicular

[dB]

12
10

3.0
4.0

10

3.76

Log-normal
fade margin
[dB]
15.4
11.3
11.3

Handoff gain
(soft handoff)
[dB]
6.1
5.0
5.0

Handoff gain
(hard handoff)
[dB]
5.9
4.7
4.7

Please note that the total TX EIRP is not computed (only one user is assumed). Also, the coverage analysis is done
for an unloaded system. This means that the RX interference density is zero (set to -1000 dBm/Hz in the tables).

10.4.2

Concept Optimized Parameters

An alternative link budget is presented below the link budget according to ETSI TR 101 112 (UMTS 30.03), in
which the antenna gains and TX powers are modified to more reasonable values.
The specified three sector antenna in the Vehicular environment has a gain of only 13 dBi, which is rather low. A
more reasonable value of 17 dBi has been used. The mobile antenna gain is specified as 0 dBi for all services and
environments. It is expected that a mobile station handling the high bit rates will not be used next to the ear. This is
taken into account by increasing the gain to 2 dBi.
The average TX powers specified in UMTS 30.03 are quite low, especially for high bit rate services. Higher values
are proposed (DL/UL): Indoor office A 13/10 dBm, Outdoor to indoor and pedestrian A 23/20, Vehicular A 30/24
dBm.

133

Downlink

Uplink

Downlink

Uplink

Downlink

Uplink

Test environment

Indoor

Indoor

Pedestr.

Pedestr.

Vehicular

Vehicular

Multipath channel class

Mobile speed

3 km/h

3 km/h

3 km/h

3 km/h

120 km/h

120 km/h

Test service

Speech

Speech

Speech

Speech

Speech

Speech

20 ms int

20 ms int

20 ms int

20 ms int

20 ms int

20 ms int

Note
Bit rate

bit/s

8000

8000

8000

8000

8000

8000

Average TX power per traffic ch.

dBm

10

20

14

30

24

Maximum TX power per traffic ch.

dBm

10

20

14

30

24

Maximum total TX power

dBm

10

20

14

30

24

Cable, conn. and combiner losses

dB

TX antenna gain

dBi

10

13

TX EIRP per traffic channel

dBm

10

28

14

41

24

Total TX EIRP

dBm

10

28

14

41

24

RX antenna gain

dBi

10

13

Cable and connector losses

dB

Receiver noise figure

dB

Thermal noise density

dBm/Hz

-174

-174

-174

-174

-174

-174

RX interference density

dBm/Hz

-1000

-1000

-1000

-1000

-1000

-1000

Total effect. noise + interf. density

dBm/Hz

-169

-169

-169

-169

-169

-169

Information rate

dBHz

39.0

39.0

39.0

39.0

39.0

39.0

Required Eb/(No+Io)

dB

6.0

3.3

6.1

3.3

7.9

5.6

RX sensitivity

dB

-124.0

-126.7

-123.9

-126.7

-122.1

-124.4

Power control TX power increase

dB

2.0

0.0

2.0

0.0

0.0

0.0

Handoff gain

dB

6.1

6.1

5.0

5.0

5.0

5.0

Explicit diversity gain

dB

Other gain

dB

Log-normal fade margin

dB

15.4

15.4

11.3

11.3

11.3

11.3

Maximum path loss

dB

122.7

121.4

143.6

142.4

156.8

153.1

Maximum range

717.2

649.1

773.5

721.9

5787.3

4613.9

Coverage efficiency

km /cell

1.6

1.3

1.9

1.6

21.8

13.8

Maximum TX power per traffic ch.

dBm

13

10

23

20

30

24

TX antenna gain

dBi

10

17

RX antenna gain

dBi

10

17

Maximum path loss

dB

125.7

127.4

146.6

148.4

160.8

157.1

Maximum range

902.9

1028.7

919.3

1019.7

7393.7

5894.6

Coverage efficiency

km /cell

2.6

3.3

2.7

3.3

35.5

22.6

Concept optimized parameters

134

Downlink

Uplink

Downlink

Uplink

Downlink

Uplink

Test environment

Indoor

Indoor

Pedestr.

Pedestr.

Vehicular

Vehicular

Multipath channel class

Mobile speed

3 km/h

3 km/h

3 km/h

3 km/h

120 km/h

120 km/h

Test service

LCD 2048

LCD 2048

LCD 384

LCD 384

LCD 144

LCD 144

Note
Bit rate

bit/s

2048000

2048000

384000

384000

144000

144000

Average TX power per traffic ch.

dBm

10

20

14

30

24

Maximum TX power per traffic ch.

dBm

10

20

14

30

24

Maximum total TX power

dBm

10

20

14

30

24

Cable, conn. and combiner losses

dB

TX antenna gain

dBi

10

13

TX EIRP per traffic channel

dBm

10

28

14

41

24

Total TX EIRP

dBm

10

28

14

41

24

RX antenna gain

dBi

10

13

Cable and connector losses

dB

Receiver noise figure

dB

Thermal noise density

dBm/Hz

-174

-174

-174

-174

-174

-174

RX interference density

dBm/Hz

-1000

-1000

-1000

-1000

-1000

-1000

Total effect. noise + interf. density

dBm/Hz

-169

-169

-169

-169

-169

-169

Information rate

dBHz

63.1

63.1

55.8

55.8

51.6

51.6

Required Eb/(No+Io)

dB

1.6

2.5

1.1

1.9

2.5

3.2

RX sensitivity

dB

-104.3

-103.4

-112.1

-111.3

-114.9

-114.2

Power control TX power increase

dB

2.0

0.0

2.0

0.0

0.0

0.0

Handoff gain

dB

6.1

6.1

5.0

5.0

5.0

5.0

Explicit diversity gain

dB

Other gain

dB

Log-normal fade margin

dB

15.4

15.4

11.3

11.3

11.3

11.3

Maximum path loss

dB

103.0

98.1

131.8

127.0

149.6

142.9

Maximum range

158.3

108.7

391.9

297.3

3734.6

2477.7

Coverage efficiency

km /cell+
B66

0.1

0.0

0.5

0.3

9.1

4.0

Maximum TX power per traffic ch.

dBm

13

10

23

20

30

24

TX antenna gain

dBi

10

17

Concept optimized parameters

RX antenna gain

dBi

10

17

Maximum path loss

dB

108.0

106.1

136.8

135.0

155.6

148.9

Maximum range

232.4

200.9

522.6

471.2

5392.9

3577.9

Coverage efficiency

km /cell

0.2

0.1

0.9

0.7

18.9

8.3

135

Downlink

Uplink

Downlink

Uplink

Downlink

Uplink

Test environment

Indoor

Indoor

Pedestr.

Pedestr.

Vehicular

Vehicular

Multipath channel class

Mobile speed

3 km/h

3 km/h

3 km/h

3 km/h

120 km/h

120 km/h

Test service

UDD 2048 UDD 2048 UDD 384

UDD 384

UDD 144

UDD 144

Note
Bit rate

bit/s

486400

486400

243200

243200

60800

60800

Average TX power per traffic ch.

dBm

10

20

14

30

24

Maximum TX power per traffic ch.

dBm

10

20

14

30

24

Maximum total TX power

dBm

10

20

14

30

24

Cable, conn. and combiner losses

dB

TX antenna gain

dBi

10

13

TX EIRP per traffic channel

dBm

10

28

14

41

24

Total TX EIRP

dBm

10

28

14

41

24

RX antenna gain

dBi

10

13

Cable and connector losses

dB

Receiver noise figure

dB

Thermal noise density

dBm/Hz

-174

-174

-174

-174

-174

-174

RX interference density

dBm/Hz

-1000

-1000

-1000

-1000

-1000

-1000

Total effect. noise + interf. density

dBm/Hz

-169

-169

-169

-169

-169

-169

Information rate

dBHz

56.9

56.9

53.9

53.9

47.8

47.8

Required Eb/(No+Io)

dB

0.1

0.5

0.1

0.4

2.9

3.0

RX sensitivity

dB

-112.0

-111.6

-115.0

-114.7

-118.3

-118.2

Power control TX power increase

dB

4.0

2.0

4.5

2.0

0.0

0.0

Handoff gain

dB

5.9

5.9

4.7

4.7

4.7

4.7

Explicit diversity gain

dB

Other gain

dB

Log-normal fade margin

dB

15.4

15.4

11.3

11.3

11.3

11.3

Maximum path loss

dB

108.5

104.1

131.9

128.1

152.7

146.6

Maximum range

242.3

172.8

396.1

318.2

4500.0

3097.3

Coverage efficiency

km /cell

0.18

0.09

0.49

0.32

13.15

6.23

24

Concept optimized parameters

Maximum TX power per traffic ch.

dBm

13

10

23

20

30

TX antenna gain

dBi

10

17

RX antenna gain

dBi

10

17

Maximum path loss

dB

113.5

112.1

136.9

136.1

158.7

152.6

Maximum range

355.6

319.4

528.2

504.4

6498.2

4472.6

Coverage efficiency

km /cell

0.40

0.32

0.88

0.80

27.43

12.99