ETSI TR 145 903 V13.0.0 (2016-01)

TECHNICAL REPORT
Digital cellular telecommunications system (Phase 2+);
Feasibility study on Single Antenna Interference
Cancellation (SAIC) for GSM networks
(3GPP TR 45.903 version 13.0.0 Release 13)

R
GLOBAL SYSTEM FOR
MOBILE COMMUNICATIONS
3GPP TR 45.903 version 13.0.0 Release 13 1 ETSI TR 145 903 V13.0.0 (2016-01)

Reference
RTR/TSGG-0145903vd00
Keywords
GSM
ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: +33 4 92 94 42 00  Fax: +33 4 93 65 47 16

Siret N° 348 623 562 00017 - NAF 742 C
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88

Important notice
The present document can be downloaded from:
http://www.etsi.org/standards-search
The present document may be made available in electronic versions and/or in print. The content of any electronic and/or
print versions of the present document shall not be modified without the prior written authorization of ETSI. In case of any
existing or perceived difference in contents between such versions and/or in print, the only prevailing document is the
print of the Portable Document Format (PDF) version kept on a specific network drive within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, please send your comment to one of the following services:
https://portal.etsi.org/People/CommiteeSupportStaff.aspx
Copyright Notification
No part may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying
and microfilm except as authorized by written permission of ETSI.
The content of the PDF version shall not be modified without the written authorization of ETSI.
The copyright and the foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2016.
All rights reserved.
TM TM TM
DECT , PLUGTESTS , UMTS and the ETSI logo are Trade Marks of ETSI registered for the benefit of its Members.
TM
3GPP and LTE™ are Trade Marks of ETSI registered for the benefit of its Members and
of the 3GPP Organizational Partners.
GSM® and the GSM logo are Trade Marks registered and owned by the GSM Association.
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 2 ETSI TR 145 903 V13.0.0 (2016-01)
Intellectual Property Rights
IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (https://ipr.etsi.org/).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This Technical Report (TR) has been produced by ETSI 3rd Generation Partnership Project (3GPP).
The present document may refer to technical specifications or reports using their 3GPP identities, UMTS identities or
GSM identities. These should be interpreted as being references to the corresponding ETSI deliverables.
The cross reference between GSM, UMTS, 3GPP and ETSI identities can be found under
http://webapp.etsi.org/key/queryform.asp.
Modal verbs terminology
In the present document "shall", "shall not", "should", "should not", "may", "need not", "will", "will not", "can" and
"cannot" are to be interpreted as described in clause 3.2 of the ETSI Drafting Rules (Verbal forms for the expression of
provisions).
"must" and "must not" are NOT allowed in ETSI deliverables except when used in direct citation.
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 3 ETSI TR 145 903 V13.0.0 (2016-01)
Contents
Intellectual Property Rights . 2
Foreword . 2
Modal verbs terminology . 2
Foreword . 5
Introduction . 5
1 Scope / objectives . 7
2 References . 7
3 Abbreviations . 7
4 Network scenarios for SAIC evaluation . 8
5 Interference modelling . 11
5.1 Introduction . 11
5.2 Interference statistics . 11
5.3 Synchronous link level models . 14
5.3.1 Interferer levels . 14
5.3.2 Delay distributions . 16
5.3.3 Frequency offset distributions. 17
5.4 Asynchronous link level models . 17
5.4.1 Burst structure . 18
5.4.2 Time-offset modelling . 18
5.4.3 Power control . 19
5.4.4 Phase transition . 20
5.4.5 Guard period and power ramping . 20
5.4.6 DTX . 21
5.5 Summary . 21
6 SAIC Link Level Characterisation . 21
6.1 Introduction . 21
6.2 Link level performance . 22
6.2.1 Results for exemplary link models . 22
6.2.2 Additional results . 24
6.3 Link-to-system interface . 24
7 SAIC system level characterization . 25
7.1 Introduction . 25
7.2 Link-to-system mapping . 26
7.3 System level simulator . 26
7.3.1 Satisfied user definition . 27
7.4 System level simulation results . 28
7.4.1 System capacity for 100% SAIC mobile penetration . 28
7.4.1.1 Configuration 1 – unsynchronized network . 28
7.4.1.2 Configuration 2 – synchronized network . 28
7.4.1.3 Configuration 2 – unsynchronized network . 29
7.4.1.4 Configuration 3 – synchronized network . 30
7.4.1.5 Configuration 3 – unsynchronized network . 31
7.4.1.6 Configuration 4 – unsynchronized network . 32
7.4.2 Impact of SAIC Mobile Penetration . 33
7.4.3 Additional results . 37
7.4.3.1 Effect of antenna patterns and Quality of Service (QoS) on system capacity . 37
7.4.3.2 System performance for Configuration 1, another perspective . 39
7.4.3.3 Impact of 8-PSK interference on GMSK SAIC performance . 40
7.5 The effect of SAIC on GPRS performance . 40
7.6 Summary and conclusions . 44
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 4 ETSI TR 145 903 V13.0.0 (2016-01)
8 SAIC field trials . 45
8.1 Asynchronous network field trial . 46
8.2 Synchronous network field trial . 46
9 Test considerations . 47
9.1 Introduction . 47
9.2 Discussion . 48
9.3 Summary . 51
10 Signalling considerations . 51
10.1 Logical binding of receiver performance to protocol version . 52
10.2 Release-independent indication of receiver performance: Classmark 3 IE . 52
10.3 Release-independent indication of receiver performance: MS Radio Access Capability IE . 53
10.4 Summary . 55
10.5 References . 55
11 Conclusions . 55
11.1 Specification impacts . 56
11.1.1 Core specifications . 57
11.1.2 Testing specifications . 57
Annex A: Change history . 58
History . 59

ETSI
3GPP TR 45.903 version 13.0.0 Release 13 5 ETSI TR 145 903 V13.0.0 (2016-01)
Foreword
rd
This Technical Report has been produced by the 3 Generation Partnership Project (3GPP).
The contents of the present document are subject to continuing work within the TSG and may change following formal
TSG approval. Should the TSG modify the contents of the present document, it will be re-released by the TSG with an
identifying change of release date and an increase in version number as follows:
Version x.y.z
Where:
x the first digit:
1 presented to TSG for information;
2 presented to TSG for approval;
3 or greater indicates TSG approved document under change control.
y the second digit is incremented for all changes of substance, i.e. technical enhancements, corrections,
updates, etc.
z the third digit is incremented when editorial only changes have been incorporated in the document.
Introduction
This document studies the feasibility of utilising Single Antenna Interference Cancellation (SAIC) as a means of
increasing the downlink spectral efficiency of GSM networks.
SAIC is a generic name for techniques, which attempt to cancel or suppress interference by means of signal processing
without the use of multiple antennas. The primary application is the downlink, where terminal space and aesthetics
typically preclude the use of multiple antennas.
Clause 1 of this document defines the scope and objectives of this feasibility study. Clause 4 defines the network
scenarios that have been defined to evaluate SAIC performance in GSM networks. These scenarios are representative
of typical GSM deployments worldwide today. Clause 5 presents the interference statistics associated with the network
scenarios defined in Clause 4. These interference statistics are developed via system simulations, and are defined in
terms of the distributions of the parameters which are critical to understanding SAIC performance. These critical
parameters include;
• The Carrier to Interference plus noise Ratio (CIR)
• The Dominant to rest of Interferer Ratio (DIR)
• The other interferer ratios, which define the relative power of the dominant co-channel interferer to each of the
other considered interferers
• The delay between the desired signal and each of the interferers.
It is important to understand the network statistics of these key parameters since most SAIC algorithms can only cancel
one interferer, and their effectiveness in doing this is affected by the 'remaining' interference, and delays between the
desired signal and the interferers.
In Clause 6, candidate SAIC algorithms are evaluated at the link level based on the interference statistics defined in
Clause 5. Both 'long-term average' and per burst results are generated. The long-term average results represent the
classical way of looking at link performance via link simulations, defining the Bit Error Rate (BER) and Frame Error
Rate (FER) averaged over the entire simulation run as a function of the CIR. This is the type of performance that is
typically specified in the GSM standards. However, to develop a system capacity estimate, it is necessary to define the
link performance on a per burst basis. To this end, Clause 6 also defines the average BER over the burst as a function
of the burst CIR and burst DIR. This burst performance is used to develop a link-to-system level mapping. This
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 6 ETSI TR 145 903 V13.0.0 (2016-01)
mapping is used in Clause 7 to develop voice capacity and data throughput estimates for both conventional and SAIC
receivers. The voice capacity gain and data throughput gain for SAIC is then deduced from these estimates.
Clause 8 describes the field trials that have been conducted using an SAIC prototype Mobile Station (MS). Clause 9
addresses testing considerations for SAIC capable MSs, while Clause 10 defines a couple of signalling options for
identifying an MS as being SAIC capable. Finally, Clause 11 provides the relevant conclusions that can be drawn from
this feasibility study, the most important of which is the conclusion that SAIC is a viable and feasible technology, which
will support significant voice capacity gains for both synchronous and asynchronous networks when applied to GMSK
modulation. In addition, modest increases in GPRS data throughput are also supported for the types of data traffic
considered. Clause 11 also identifies those clauses of the core and testing specifications that will be impacted by the
inclusion of an SAIC capability.

ETSI
3GPP TR 45.903 version 13.0.0 Release 13 7 ETSI TR 145 903 V13.0.0 (2016-01)
1 Scope / objectives
The objective of this document, as defined in the work item [2], is to determine the potential of SAIC in typical network
layouts. This includes study of the following aspects:
a) Determine the feasibility of SAIC for GMSK and 8PSK scenarios under realistic synchronized and non-
synchronized network conditions. Using a single Feasibility Study, both GMSK and 8PSK scenarios will be
evaluated individually.
b) Realistic interference statistics including CIR (Carrier to Interference plus noise Ratio) and DIR (Dominant-to-
rest of Interference Ratio) levels and distributions based on network simulations and measurements, where
possible.
c) Robustness against different training sequences.
d) Determine method to detect/indicate SAIC capability.
2 References
The following documents contain provisions, which, through reference in this text, constitute provisions of the present
document.
• References are either specific (identified by date of publication, edition number, version number, etc.) or
non-specific.
• For a specific reference, subsequent revisions do not apply.
• For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including
a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same
Release as the present document.
[1] ETSI TR 101 112 v3.2.0 (1998-04), "Universal Mobile Telecommunications System (UMTS);
Selection procedures for the choice of radio transmission technologies of the UMTS".
[2] 3GPP TSG-GERAN TDOC GP-022891: "Work Item Description, Single Antenna Interference
Cancellation", Sophia Antipolis, France, 18-22 November 2002.
[3] 3GPP TSG-GERAN SAIC Workshop TDOC GAHS-030009: "Network level simulation scenarios
and assumptions for SAIC", Atlanta, USA, 8-9 January 2003.
[4] 3GPP TSG-GERAN SAIC Workshop TDOC GAHS-030005: "Scenarios and Modelling
Assumptions for SAIC in GERAN", Atlanta, USA, 8-9 January 2003.
[5] 3GPP TSG-GERAN SAIC Workshop TDOC GAHS-030002: "Single antenna interference
cancellation - evaluation principles and scenarios", Atlanta, USA, 8-9 January 2003.
[6] 3GPP TSG-GERAN SAIC Workshop TDOC GAHS-030020: "Interference Characterization for
SAIC Link Level Evaluation", Seattle, USA, 4-5 March 2003.
[7] 3GPP TSG-GERAN SAIC Workshop TDOC GAHS-030022: "Link Level model for SAIC",
Seattle, USA, 4-5 March 2003.
Additional references are noted in the individual clauses of this document
3 Abbreviations
ACI Adjacent Channel Interference
AMR Adaptive Multi Rate
BEP Bit Error Probability
BER Bit Error Rate
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 8 ETSI TR 145 903 V13.0.0 (2016-01)
BLER Block Error Rate
BTS Base Transceiver Station
CDF Cumulative Distribution Function
C/I Carrier-to-Interference Power Ratio
cdfs cumulative distribution functions
CINR Carrier to Interference-plus-Noise Ratio
DIR Dominant-to-rest Interference Ratio
DPC Downlink Power Control
DTX Discontinuous Transmission
EFL Effective Frequency Load
FEP Frame Error Probability
FER Frame Error Rate
FL Frequency Load
FR Full Rate
FTP File Transfer Protocol
GMSK Gaussian Minimum Shift Keying
GPRS General Packet Radio Service
HR Half Rate
IE Information Element
MMS Multimedia Messaging Service
MS Mobile Station
PDF Probability Distribution Function
PSK Phase-Shift Keying
QoS Quality of Service
SAIC Single Antenna Interference Cancellation
TSC Training Sequence
4 Network scenarios for SAIC evaluation
A multi-step approach was taken to evaluate SAIC performance in realistic network scenarios. This approach consisted
of first determining relevant interference statistics based on the network scenarios described in this clause. These
interference statistics were then used to determine the link level performance at the GSM burst level. From this link
level characterization, link-to-system mapping tables were developed, which were then used in system level simulations
to determine the voice and data capacity gains provided by SAIC capable MSs. The network scenarios used in these
simulations were discussed and agreed to as part of SAIC Workshop #1.
It was agreed that the network scenarios, also referred to as configurations in this document, should represent typical
GERAN networks at the time frame when operators would be deploying SAIC capable MSs. The goal was to try to
make the interference statistics as realistic as possible, while trying to keep the overall complexity of the simulations
reasonable. As a result of [3], [4], and [5], the following parameters are considered to be the major issues which affect
the interference statistics:
• Frequency Hopping scheme
• Reuse (also adjacent channel reuse) and cell radius
• Regularity of the network (different cell sizes, different number of TRXs per cell, hotspots)
• Propagation conditions, including network topology (street corner effects, shadowing from buildings/hills
etc.)
• Downlink Power Control (DPC) scheme
• Channel coding, mainly if quality-based DPC is used; schemes with less coding requires higher
transmission powers
• Penetration of different MSs/bearers in the network
• SAIC MS penetration: power levels, higher tolerated load/interference for SAIC MSs, but the non-SAIC
MS must not be negatively impacted
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 9 ETSI TR 145 903 V13.0.0 (2016-01)
• Packet-switched connections to support GPRS and EGPRS, which are characterized by short connection
times, asymmetry, bursty traffic, multiplexing of several users on the same time slot, and often lack of DPC
• Legacy non-AMR (mainly EFR) mobiles: higher transmit powers, less robustness
• Level of synchronization in the network
• Mobility: speed distribution of the mobiles affects the interference pattern
Going into the study, it was believed that SAIC would support larger gains in tighter reuse networks, as the interference
becomes more and more limiting to system performance. Similarly, the higher the load, the more interference to cancel.
However, interference scenarios are more complex with a higher load, so the interference cancellation algorithms may
be less efficient. Finally, SAIC techniques generally give the largest gains in synchronized networks. These initial
observations were found to be true, for the most part as is shown in clause 7, which provides a characterization of the
system level performance of SAIC.
Two tables define the network scenario assumptions. Table 4-1 defines operator or configuration specific assumptions,
while table 4-2 defines parameters common to all of the configurations.  Both tables were derived from [3], [4], [5],
and discussed as part of the SAIC Workshop #1. The four configurations defined in Table 4-1, and the common
parameters defined in Table 4.2 are described in detail in clause 7.
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 10 ETSI TR 145 903 V13.0.0 (2016-01)
Table 4-1
Configuration specific network scenario assumptions
Parameter Value Unit Comment
Configuration 1 - Asynchronous
Frequency 900 MHz
Bandwidth 7.8 MHz
Reuse 4/12 (BCCH)
3/9 (TCH)
Hopping Baseband
Voice Codec AMR 12.2 FR
Blocking 2 %
Modulation Source/Interferer
GMSK/GMSK
GMSK/8PSK
Cell Radius 500 m
Configuration 2 – Sync & Async
Frequency 1900 MHz
Bandwidth 1.2 MHz
Reuse 1/1 (TCH)
Hopping Random RF
Voice Codec AMR 5.9 FR/HR
Frequency Load 20, 40 (FR) %
10, 20 (HR) %
Modulations Source/Interferer
GMSK/GMSK
GMSK/8PSK
8PSK/GMSK
8PSK/8PSK
Cell Radius 1000 m
Configuration 3 – Sync & Async
(Optional)
Frequency 900 MHz
Bandwidth 2.4 MHz
Reuse 1/1 (TCH)
Hopping Random RF
Voice Codec AMR 5.9 FR/HR
Frequency Load 40, 70 (FR) %
25, 40 (HR) %
Modulation Source/Interferer
GMSK/GMSK
Cell Radius 750 m
Configuration 4 - Asynchronous
Frequency 900 MHz
Bandwidth 7.2 MHz
Reuse 1/3 (TCH)
Hopping Random RF
Voice Codec AMR 12.2 FR
Blocking 2 %
Frequency Load 30 %
Modulation Source/Interferer
GMSK/GMSK
GMSK/8PSK m
Cell Radius 300
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 11 ETSI TR 145 903 V13.0.0 (2016-01)
Table 4-2
Common network scenario assumptions
Parameter Value Unit Comment
Sectors (cells) per site
Sector antenna pattern UMTS 30.03

Propagation model UMTS 30.03  Pathloss
exponent, MCL
Per 30.03
Log-normal fading standard deviation 6 (900) dB
8 (1900) dB
Correlation distance 110 m
Adjacent channel interference attenuation 18 dB Carrier +/- 200
KHz
Handover margin 3 dB
Mobile speed TU3 and TU50 km/h
Mean Call length 90 sec.
Minimum Call Length 5 sec.
Voice activity 60% Includes SID
signalling.
DTX Enabled
Link adaptation Disabled
BTS output power 20 W
Power control RxQual/RxLev
Dynamic Range 14 dB
Step Size 2 dB
Noise figure 10 dB Reference
temperature
25c
Inter-site Lognormal Correlation Coefficient 0
Channel Allocation Random
See clause 7.5 Web-browsing
Traffic data models for GPRS
& FTP/MMS
5 Interference modelling
5.1 Introduction
When assessing the link and system level performance it is important to base the performance investigations on realistic
link level models. Especially for SAIC receivers previous studies have demonstrated that the SAIC link level
performance for the same interference level will vary significantly for different link level models [GP-030276].
Therefore a lot of work has been done in the SAIC feasibility study to define realistic models and the outcome of this
work is recaptured in this clause.
Defining realistic link level models is clearly impossible without investigating the interference statistics seen by mobiles
when operating in different network scenarios. Thus an important part of the modelling work has been analysis of
network traces generated by network simulators for the four different network configurations defined in clause 4.
To types of link level models have been derived one for synchronous network configurations and one covering
asynchronous networks. The latter is an extension of the model derived for synchronous networks taking into account
such effects as delay, power control, DTX, etc.
5.2 Interference statistics
In GSM/EDGE the performance of the mobiles in interference limited scenarios have traditionally been evaluated for a
single interfering signal at a high input level where the sensitivity performance of the mobile will have no or very little
influence. This can be described by the conventional CIR (Carrier to Interference Ratio):
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 12 ETSI TR 145 903 V13.0.0 (2016-01)
C
CIR =
I + N
where C is the power of the carrier, I the power of an interfering signal (co- or adjacent channel interference) and N the
thermal noise . Although widely used, for evaluation, this ideal one interferer scenario happens very rarely in practice
especially when the network is highly loaded. When using e.g. AMR a high frequency load can be expected and
consequently the mobiles will receive interference from a number of base stations at the same time. This can easily be
introduced in the above definition of the CIR:
C
CIR =
I + N
∑ k 0
k
I can be both co- and adjacent channel interference (for the adjacent channel interference a realistic ACP (Adjacent
k
Channel Protection) shall be used e.g. ACP=18dB).
For a small number of interfering base stations the performance of a conventional receiver will be identical for the two
definitions, but for a SAIC mobile the performance (interference cancellation capability) will depend upon the
distribution of the interferer powers. An initial, simple measure to capture this is the Dominant to rest of Interference
Ratio (DIR), which is the power of the dominant interferer to the sum of the powers of the rest of the interferers plus N .
o
This ratio is defined as:
I
max
DIR =
I − I + N
∑ k max 0
k
where I is the average power of the dominant interfering signal (co- or adjacent channel interference). When only a
max
single interferer is active, as in the standard interference test case in 45.005, then the DIR will be identical to the I/N of
the received interfering signal. Although the standard interference test case is widely used it has been demonstrated in a
number of contributions that this test case does not reflect a realistic scenario for a SAIC mobile [GAHS-030017][
GAHS-030018][ GAHS-030022].
In [GAHS-030008] a new measure called DIR was introduced in the link level modelling discussion. The DIR
2 2
measure is defined as:
I
max 2
DIR =
I − I − I + N
∑ k max max 2 0
k
and basically it can be used to investigate the validity of using a simple two cochannel interferer model when evaluating
the SAIC link level performance. In TSG GERAN #13 the DIR measure was included in a number of studies and the
initial conclusion was that more than two cochannel interferers are needed in the SAIC link level model [GP-030159,
GP-030276].
Figures 5-1 through 5-3 are examples of interferer statistics for network configuration 2 . The figures clearly
demonstrate how the interferer statistics in a network are much more complicated than the single interferer scenario
currently tested in 45.005. The DIR and DIR statistics clearly demonstrate the need to define link level models having
multiple interferers.
CIR is also referred to in this document as CINR = Carrier to Interference plus Noise Ratio
The figures have been taken from [GAHS-030017] but similar figures have been presented in [GAHS-030022] and [GAHS-030018].
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 13 ETSI TR 145 903 V13.0.0 (2016-01)

C.D.F. plot over CINR
excl. ACI
incl. ACI
−10 −5 0 5 10 15 20 25 30 35
CINR [dB]
Figure 5.1 The CIR cdfs observed by a MS operating in network configuration 3 [GAHS-030017].

C.D.F. plot over DIR (incl. ACI)
CINR < 10 dB
CINR < 0 dB
−10 −5 0 5 10 15 20
DIR [dB]
Figure 5-2 The DIR cdfs observed by a MS operating in network configuration 3 [GAHS-030017].

ETSI
C.D.F. [%] C.D.F. [%]
3GPP TR 45.903 version 13.0.0 Release 13 14 ETSI TR 145 903 V13.0.0 (2016-01)

C.D.F. plot over DIR (incl. ACI)
CINR < 10 dB
CINR < 0 dB
−10 −5 0 5 10 15
DIR [dB]
Figure 5-3 The DIR cdfs observed by a MS operating in network configuration 3 [GAHS-
030017].
5.3 Synchronous link level models
Early link level investigations for SAIC demonstrated a higher link level gain when using a synchronous link level
configuration compared to an asynchronous one. Consequently it was decided to develop link level models for both
types of networks focusing initially on the synchronous mode , which will be described in this clause.
5.3.1 Interferer levels
Having identified the need to have multiple interferers in the link level model the necessary number of interferers and
their levels have to be estimated. During the SAIC Adhoc #2 a procedure for the estimation was agreed based on
investigations made in document [GAHS-030018] and [GAHS-030022]. From network traces the cdfs of a number of
co- and adjacent channel interferers plus the residual interference were derived. Examples of the cdfs can be seen in
Figure 5-4 and Figure 5-5. In the estimation process only bursts having a CIR < 10dB have been taken into account
because SAIC algorithms are expected to have the largest link level gain for low CIR. The mean power level of each
interferer was chosen as the observed median value, with all ratios defined with respect to the dominant cochannel
interferer. For example, in configuration 3 the ratio of the dominant interferer, I to the second dominant interferer I is
1 2
4 dB as shown in Figure 5-5. The final agreed-to numbers are listed in Table 5-1, where the following interferers are
defined: three discrete co-channel interferers, one discrete adjacent channel interferer, one residual cochannel interferer
and one residual adjacent channel interferer . The numbers for the adjacent channel interference are assumed measured
after a receive filter having an attenuation of 18dB. Thus, in the channel model the power level should be 18dB higher
than shown in the table. For configuration 1 the values have been derived in [GP-031203], for configuration 2 and 3 the
values were derived at the SAIC Adhoc #2 and finally for configuration 4 the values have been agreed as the average of
the values from [GP-031289] and [GP-031203].

Only burst wise synchronization is assumed.
In Table 5-1, the dominant interferer is also referred to as Ic1, while the remaining discrete co-channel interferers are referred to as Ic2 and Ic3. The
discrete adjacent channel interferer is referred to as Ia, while the residual interferers are referred to as Icr and Iar, respectively.
ETSI
C.D.F. [%]
3GPP TR 45.903 version 13.0.0 Release 13 15 ETSI TR 145 903 V13.0.0 (2016-01)

Config 3 (CIR<10dB)
Config 2 (CIR<10dB)
Co-channel Model Order = 3
Co-channel Model Order = 3
CDF - Co+Adj-Chan Intf Pow Ratio
CDF - Co+Adj-Chan Intf Pow Ratio
(dB)
(dB)
i1-i2 i1-i2
i1-i3 i1-i3
0.8 i1-i4 0.8 i1-i4
i1-i5
i1-i5
i1-ir i1-ir
0.6 i1-aci-1 i1-aci-1
0.6
i1-aci-r i1-aci-r
0.4
0.4
0.2
0.2
-10 -5 0 5 10 15 20 25 30
-10 -5 0 5 10 15 20 25 30
Co+Adj-Chan. Intf. Pow. Ratio
Co+Adj-Chan. Intf. Pow . Ratio
(dB)
(dB)
Figure 5-4 cdfs of interferer powers for Figure 5-5 cdfs of interferer powers for
estimation of link level model for network estimation of link level model for network
configuration 2 [GAHS-030024]. configuration 3 [GAHS-030024].

For the modelling of residual co- and adjacent channel interference an AWGN source is filtered using the 8PSK
modulation filter (linearized GMSK pulse) specified in 45.004 clause 3.5. The filtering is done to ensure the correct
spectral properties. The residual adjacent channel interference is applied with half the power on each side of the carrier
i.e. for configuration 2 two residual adjacent channel interferers being offset ±200kHz from the carrier and having
power level 0dB should be included.
During the initial investigation of SAIC a number of companies observed that the performance of most SAIC
algorithms is degraded when the interferer has a TSC included compared to use of the standard GMSK-modulated
random sequence defined in 45.005 [GP-020822]. Therefore an important part of the link level modelling is to include
TSCs for all except the residual interferers i.e. the interferers generally have a normal burst structure. Apart from the
dominant cochannel interferer, the TSC is taken from a uniform distribution including all eight TSCs defined in 45.002.
In an optimized network it is expected that TSC collisions to some extent can be avoided for the dominant interferer and
therefore TSC0 is not included for the dominant interferer.
When performing link level analysis the fading is an important part of the modelling and as can be seen in Table 5-1 all
except the three residual interferers are subject to fading. Fading is not applied on the residual interferers because these
are used to model interference from a number of BTSs each having independent fading. Thus the power variations of
the residual interference will be small and are thus, neglected in the link level model.
Table 5-1 Interferer levels for network configuration 1-4.

Link Parameter Configuration 1 Configuration 2 40% Configuration 3 70% Configuration 4
Load Load
Desired signal, C
TSC TSC0 TSC0 TSC0 TSC0
Fading
Dominant Coch. Interf.
TSC Random TSC Random TSC excluding Random TSC excluding Random TSC excluding
excluding TSC0 TSC0 TSC0 TSC0
The 18dB adjacent channel protection has been taken into account.
ETSI
Cumulative Density
Function
Cumulative Density
Function
3GPP TR 45.903 version 13.0.0 Release 13 16 ETSI TR 145 903 V13.0.0 (2016-01)
Fading
nd
2 Strongest Coch.
Interf.
Ic1/Ic2
10 dB 6 dB 4 dB 9 dB
TSC
Random TSC Random TSC Random TSC Random TSC
Fading
rd
3 Strongest Coch
Interf.
Ic1/Ic3
20 dB 10 dB 8 dB 17 dB
TSC
Random TSC Random TSC Random TSC Random TSC
Fading
Residual Coch. Interf.
(filtered AWGN)
Ic1/Icr
TSC - 9 dB 5 dB 20 dB
No Fading NA NA NA NA
Dominant Adj. Interf.
6 15 dB 14 dB 14 dB 16 dB
Ic1/Ia
Random TSC Random TSC Random TSC Random TSC
TSC
Fading
Residual Adj. Interf.
(filtered AWGN)
Ic1/Iar 20 dB 15 dB 14 dB 21 dB
TSC NA NA NA NA
No Fading
5.3.2 Delay distributions
Even in a synchronized network the mobile station will receive interference from the different BTSs at various delays
due to the distance to the interfering sites. Although most SAIC receivers are expected to be robust to delays less than
10 symbols even small delays can affect the correlation properties between different TSCs and therefore the
performance of both conventional and SAIC receivers.
Based on network traces, modelling of delay in the synchronous link level models has been investigated by Motorola
for the four network configurations. The outcome of these studies is the delay model summarized in this clause.
Using a delay resolution of 0.2 symbols, and the observation that delays in the four configurations are limited to the
range [-2,+5] symbols, the discrete delay distribution can be approximated as:
1. for delay less than 0, for k=1 to 10, the probability P()k of delay equal to -0.2k is:

After the Rx filter assuming an 18dB ACP.
ETSI
3GPP TR 45.903 version 13.0.0 Release 13 17 ETSI TR 145 903 V13.0.0 (2016-01)
k
P()kA=−p(1 p)
11 1
2. for delay greater than 0, for k=1 to 25, the probability P()k of delay equal to 0.2k is:
k
P()kA=−p (1 p )
22 2
3. for zero delay:
P(0) = A
The parameters to be used for the different configurations can be seen in Table 5-2.
Table 5-2 Summary of delay model parameters.
p p A A A
Configuration
1 2 0 1 2
Configuration 1 @2% blocking 0.9 0.7 0.5602 0.5 2
Configuration 2@40% 0.37 0.09 0.2157 0.1274 0.8555
Configuration 3@70% 0.7 0.26 0.4005 0.1658 0.7433
Configuration 4@30% 0.95 0.25 0.1106 0.1874 1.1742

The model demonstrates that the carrier and the interferers often are synchronized when received by the mobile station.
5.3.3 Frequency offset distributions
Frequency offset is inevitable in practical implementations and consequently also needed in the SAIC link level model
[GP-032246]. When a mobile station is connected to a BTS it is synchronized in frequency to this serving BTS.
Therefore the mobile station will not detect if the carrier of this BTS is offset compared to a correct carrier frequency.
Although synchronized some frequency jitter due to inaccuracy of the frequency estimation procedure will exist in
practice. It has been agreed not to include this vendor specific frequency jitter in the model but clearly each vendor has
to include their own model when performing simulations.
However, the frequency offset has to be included for each of the three discrete co-channel, and the one discrete adjacent
channel interferers having a value that includes the fixed offset of the serving BTS . For each of these interferers the
frequency offset will be varying on burst-by-burst basis due to frequency hopping and the fact that the interference in
the model comes from a number of BTSs all having different offset. The mean value of these offsets is assumed to be
0Hz (plus the fixed frequency offset of the serving BTS ) with a standard deviation of either 17 Hz for 850/900 MHz
operation, or 33Hz for 1800.
...