ETSI TR 103 554 V1.1.1 (2018-08)

TECHNICAL REPORT
Rail Telecommunications (RT);
Next Generation Communication System;
LTE radio performance simulations and
evaluations in rail environment

2 ETSI TR 103 554 V1.1.1 (2018-08)

Reference
DTR/RT-0047
Keywords
LTE, railways
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
https://portal.etsi.org/TB/ETSIDeliverableStatus.aspx
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.

© ETSI 2018.
All rights reserved.
TM TM TM
DECT , PLUGTESTS , UMTS and the ETSI logo are trademarks of ETSI registered for the benefit of its Members.
TM TM
3GPP and LTE are trademarks of ETSI registered for the benefit of its Members and
of the 3GPP Organizational Partners.
oneM2M logo is protected for the benefit of its Members. ®
GSM and the GSM logo are trademarks registered and owned by the GSM Association.
ETSI
3 ETSI TR 103 554 V1.1.1 (2018-08)
Contents
Intellectual Property Rights . 5
Foreword . 5
Modal verbs terminology . 5
Executive summary . 5
Introduction . 6
1 Scope . 7
2 References . 7
2.1 Normative references . 7
2.2 Informative references . 7
3 Abbreviations . 8
4 Assumptions and parameters for simulations and evaluations . 9
4.1 Introduction . 9
4.2 Simulation tools . 10
4.3 Scenarios . 10
4.4 Bandwidth and transmit power . 11
4.4.1 Bandwidths . 11
4.4.2 Transmit powers . 11
4.5 Antenna diagrams . 12
4.5.1 Antenna diagrams at the base station . 12
4.6 Radio propagation aspects . 12
4.6.1 Radio propagation model . 12
4.6.2 Conclusion . 13
4.7 Frequency reuse scheme . 14
4.8 Summary . 14
4.9 Outcomes of the simulations . 16
5 Simulation results . 16
5.1 Results set 1 . 16
5.1.1 Description . 16
5.1.2 Specific assumptions and parameters . 18
5.1.3 Results . 19
5.1.4 Notes and remarks . 21
5.2 Results set 2 . 22
5.2.1 Description . 22
5.2.1.1 Lab setup high level description . 22
5.2.1.2 Lab setup: 3GPP RF Channel Emulator . 23
5.2.1.3 Lab setup: FRMCS Traffic Generator and Analyzer. 24
5.2.2 Specific assumptions and parameters . 24
5.2.3 Results . 24
5.2.4 Notes and remarks . 24
5.3 Results set 3 . 24
5.3.1 Description . 24
5.3.2 Specific assumptions and parameters . 25
5.3.3 Results . 26
5.3.4 Notes and remarks . 27
6 Results evaluation . 28
6.1 Analysis . 28
6.1.1 General . 28
6.1.2 Overheads analysis . 28
6.1.2.1 General . 28
6.1.2.2 IP stack, PDCP and RLC overheads . 29
6.1.2.3 Physical layer overheads . 29
ETSI
4 ETSI TR 103 554 V1.1.1 (2018-08)
6.1.2.4 Link-level comparison . 29
6.1.3 Train speed impact . 30
6.1.4 Neighbouring cells interference impact . 32
6.2 Identified system limitations . 32
7 Conclusion . 33
Annex A: Theoretical peak throughput for LTE . 34
Annex B: Throughput curves for simulation results set 1 . 35
Annex C: Data Throughput Measurements for results set 3 . 45
Annex D: Antenna diagrams . 47
Annex E: Change history . 53
History . 54

ETSI
5 ETSI TR 103 554 V1.1.1 (2018-08)
Intellectual Property Rights
Essential patents
IPRs essential or potentially essential to normative deliverables 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.
Trademarks
The present document may include trademarks and/or tradenames which are asserted and/or registered by their owners.
ETSI claims no ownership of these except for any which are indicated as being the property of ETSI, and conveys no
right to use or reproduce any trademark and/or tradename. Mention of those trademarks in the present document does
not constitute an endorsement by ETSI of products, services or organizations associated with those trademarks.
Foreword
This Technical Report (TR) has been produced by ETSI Technical Committee Railway Telecommunications (RT).
Modal verbs terminology
In the present document "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.
Executive summary
In order to assess 3GPP LTE radio performance in a rail environment, three scenarios have been defined: Rural, Hilly
and Urban, representing various radio conditions typical to rail environment. Each scenario has been defined with its
radio parameters, load condition and train speeds.
UIC and E-UIC spectrum bands have been assumed, with bandwidth of 1,4 MHz, 3 MHz and 5 MHz, corresponding to
possible deployments with LTE and GSM-R co-existence and deployment with a standalone LTE.
Three different studies are described. One is based on simulation with a software chain tool using a Monte-Carlo
statistical approach, including multiple cells in a linear deployment along the track. The two others are based on
laboratory radio test bench, featuring hardware communication devices and wireless channel emulators, but not taking
into account multiple cells interferences.
The present document includes results from software chain tool study and from one of the two laboratory radio test
bench study.
In the present document, only results for LTE using a channel bandwidth of 1,4 MHz with maximum UE power of
23 dBm in the 900 MHz band are provided. A set of initial conclusions has been drawn from these partial results;
however a final conclusion will need the completion of the analysis with results for 3 MHz and 5 MHz channel
bandwidths and for maximum UE powers of 26 and 31 dBm. Furthermore, the impact of using a TDD mode in other
frequency bands will need to be added to the report.
ETSI
6 ETSI TR 103 554 V1.1.1 (2018-08)
Introduction
The present document outlines the study conducted within TC RT on LTE radio performance simulations and
evaluations in rail environment.
ETSI
7 ETSI TR 103 554 V1.1.1 (2018-08)
1 Scope
3GPP LTE radio access is one candidate for the radio access technology to be used for the Future Rail Mobile
Communications System (FRMCS). In the present document, the term FRMCS refers -unless stated otherwise- to the
radio part of the communication system.
The present document is intended to:
• Define the simulation parameters relevant to rail environment relating to 3GPP LTE radio performance. This
includes in particular operating frequency bands, bandwidths, deployment scenario (inter-site distance), and
antenna characteristics, transmit powers and channel models, along with relevant metrics to be evaluated.
• Collect and analyse the simulation results of an LTE system in the rail environment.
• Identify limitations of an LTE system in the rail environment.
Radio performance evaluation of an LTE system could be done by simulation, through software and processing
resources only, or through a test bench incorporating pieces of equipment emulating parts of the chain, e.g. the RF. In
both cases, it is important to align the parameters and the assumptions made in the simulation and in the evaluation
chain to be able to reflect better a deployment in a rail environment, and to better compare and understand the
simulation and the evaluation results.
2 References
2.1 Normative references
Normative references are not applicable in the present document.
2.2 Informative references
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
referenced document (including any amendments) applies.
NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee
their long term validity.
The following referenced documents are not necessary for the application of the present document but they assist the
user with regard to a particular subject area.
[i.1] ETSI TS 145 005 (V14.4.0) (04-2018): "Digital cellular telecommunications system (Phase 2+)
(GSM); GSM/EDGE Radio transmission and reception (3GPP TS 45.005 version 14.4.0
Release 14)".
[i.2] ETSI TS 136 104 (V14.7.0) (04-2018): "LTE; Evolved Universal Terrestrial Radio Access
(E-UTRA); Base Station (BS) radio transmission and reception (3GPP TS 36.104 version 14.7.0
Release 14)".
[i.3] ETSI TS 136 101 (V14.7.0) (04-2018): "LTE; Evolved Universal Terrestrial Radio Access
(E-UTRA); User Equipment (UE) radio transmission and reception (3GPP TS 36.101
version 14.7.0 Release 14)".
[i.4] Recommendation ITU-R M.2135-1 (12-2009): "Guidelines for evaluation of radio interface
technologies for IMT advanced".
[i.5] IST-4-027756 Winner II D1.1.2 V1.2 Winner II Part I: "Channel Models", European Commission,
Deliverable IST-WINNER D.
ETSI
8 ETSI TR 103 554 V1.1.1 (2018-08)
[i.6] Ikuno, J. Colom, Martin Wrulich, and Markus Rupp.: "Performance and modelling of LTE
H-ARQ." Proc. International ITG Workshop on Smart Antennas (WSA 2009), Berlin, Germany
2009.
[i.7] ETSI TS 136 211 (V14.6.0) (04-2018): "LTE; Evolved Universal Terrestrial Radio Access
(E-UTRA); Physical channels and modulation (3GPP TS 36.211 version 14.6.0 Release 14)".
[i.8] Recommendation ITU-R M.1225 (1997): "Guidelines for evaluation of radio transmission
technologies for IMT-2000".
[i.9] European Integrated Railway Radio Enhanced Network System Requirements Specification,
UIC CODE 951, GSM-R Operators Group, December 2015.
3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
ACS Adjacent Channel Selectivity
AMC Adaptive Modulation and Coding
AWGN Additive White Gaussian Noise
BS Base Station
BTS Base Transceiver Station
BW Bandwidth
CDF Cumulative Distribution Function
CDL Clustered Delay Line
COST Cooperation of Scientific and Technical
CP Cyclic Prefix
DL Down Link
EIRENE European Integrated Railway radio Enhanced NEtwork
eNB evolved Node B
ETU Extended Typical Urban model
E-UTRA Evolved UMTS Terrestrial Radio Access
FDD Frequency Division Duplex
FEC Forward Error Correction
FRMCS Future Rail Mobile Communications System
FSTD Frequency Switched Transmit Diversity
GSM Global System for Mobile communications
GSM-R Global System for Mobile communication for Railway application
HARQ Hybrid Automatic Repeat-Request
HO Hand Over
HST High Speed Train
IMT International Mobile Telecommunications
IP Internet Protocol
ISD Inter Site Distance
ISI Inter-Symbol Interference
ITU-R Internail Telecommunication Union - Radiocommunication sector
LOS Line Of Sight
LTE Long Term Evolution
MAC Media Access Control
MCS Modulation and Coding Scheme
MIMO Multiple Input, Multiple Output
MISO Multiple Input, Single Output
MOS Mean Opinion Score
MRS Mobile Relay Station
NLOS Non Line Of Sight
OFDM Orthogonal Frequency Division Multiplexing
PBCH Physical Broadcast Channel
PDCCH Physical Downlink Control Channel
PDCP Packet Data Convergence Protocol
PDP Power Delay Profile
PER Packet Error Rate
ETSI
9 ETSI TR 103 554 V1.1.1 (2018-08)
PHY PHYsical layer
PUCCH Physical Uplink Control Channel
QAM Quadrature Amplitude Modulation
QCI QoS Class Identifier
RB Resource Block
REC Railways Emergency Call
RF Radio Frequency
RLC Radio Link Control
RT Rail Telecommunications
SFBC Space-Frequency Block Coding
SGW Serving Gateway
SIMO Single Input, Multiple Output
SINR Signal to Interference-plus-Noise Ratio
SISO Single Input, Single Output
SNR Signal to Noise Ratio
SRS System Requirement Specification
TC Technical Committee
TCP Transmission Control Protocol
TDD Time Duplex Division
UDP User Datagram Protocol
UE User Equipment
UIC Union Internationale des Chemins de fer
UL Up Link
UMTS Universal Mobile Telecommunications System
USB Universal Serial Bus
4 Assumptions and parameters for simulations and
evaluations
4.1 Introduction
In the scope of this study, the following points are addressed:
• Simulations take into account railway specifics
• Simulations are flexible in order to simulate different system configurations, parameter settings and scenarios
• Consideration of different carrier band-widths (at least 1,4, 3 and 5 MHz)
• Consideration of TDD and FDD duplex modes
• Consideration of different subscriber and train densities and distributions
• Considerations of FRMCS system parameters (e.g. Cyclic Prefix)
• Different power classes of FRMCS equipment
• Different antenna radiation patterns and tilts
• SISO, SIMO, MISO und MIMO
• Different installation heights of antennas
• Different distances and densities of fixed transmitter equipment (eNB)
• Different specified and appropriate coding and modulation schemes
• Different 3GPP Releases (e.g. LTE: ≥ 13) to take into account new features, e.g. performance improvements
for high speed.
ETSI
10 ETSI TR 103 554 V1.1.1 (2018-08)
4.2 Simulation tools
Software simulations are made at radio level, i.e. above the physical layer as depicted in Figure 1. Overheads like pilots
and cyclic prefixes are taken in to account, but not the overheads that are added by layers above PHY, in particular
PDCP and IP headers.
Other simulations, e.g. hardware simulations and laboratory tests, could have a reference point at application level.
UE
Hardware simulation and laboratory test
reference point
Application Application
Upper Core Network
BS
layers
PDCP PDCP
RLC RLC
MAC MAC
Software simulation
reference point
PHY PHY
RF RF
Figure 1: Reference point for the software simulations
4.3 Scenarios
The objective is to define the minimum number of scenarios which cover the majority of the radio environment.
Three scenarios have been retained: Urban, Rural, and Hilly. Urban is relative to areas where train density is high, but
move at moderate speed. Rural scenario typically intends to model high speed lines. Hilly scenario intends to handle
more complex situations from radio propagation point of view, with in particular extensive multi-path propagation.
Tunnels are complex scenarios, since they depend widely on tunnel shape and tunnel/train relative geometry. They are
not considered in this study as they would require a more long and thorough work.
Only train-ground communications are considered in this study. Handset or shunting area scenarios are for further
study.
Whether it is possible to have several antennas on trains roof tops and what could be their characteristic needs further
discussions.
ETSI
11 ETSI TR 103 554 V1.1.1 (2018-08)
4.4 Bandwidth and transmit power
4.4.1 Bandwidths
Three scenarios are considered, on bandwidths of 1,4 MHz, 3 MHz and 5 MHz in the UIC and E-UIC bands, as
depicted in Figure 2:
1) Scenario 1 considers GSM-R in UIC band as per today, with the addition of a 1,4 MHz LTE carrier in the
upper part of E-UIC band. This scenario corresponds to a migration phase, with co-existence of both GSM-R
and LTE systems.
2) Scenario 2 assumes is an extension of scenario 1 with a LTE carrier extended to 3 MHz in the E-UIC band.
3) Scenario 3 assumes a deployment with no GSM-Rand one LTE 5 MHz carrier in UIC band, overlapping the
E-UIC band.
E-UIC band UIC band
DL
918 919.5 920.9 921 925
(MHz)
Scenario 1: Co-existence with GSM-R
LTE 1.4 MHz GSM-R
UL
873 874.5 875.9 876 880
(MHz)
DL
918 921 925
(MHz)
Scenario 2: Co-existence with GSM-R
LTE 3 MHz
GSM-R
and extented LTE carrier
UL
873 876 880
(MHz)
DL
918 920 925
(MHz)
LTE 5 MHz Scenario 3: Overlapping LTE carrier
UL
873 875 880
(MHz)
Figure 2: Carriers and bandwidths in the deployment scenarios considered
Scenario 1 is of highest priority.
4.4.2 Transmit powers
Transmit power in the E-UIC band is subject to limitations in case of FRMCS system deployment uncoordinated with
commercial systems operating in neighbouring bands.
The method to compute the maximum transmit power derives the impact from the adjacent channel selectivity related
specifications (wideband blocking and narrow band blocking), takes into account applicable effects (0,8 dB
desensitization, slope of the filtering, etc.) as well as corrections resulting from spurious emissions from base station
transmission and from UE. ACS (Adjacent Channel Selectivity) has been found as not relevant for this study.
Summary of the acceptable maximum transmit power of a FRMCS system in case of uncoordinated deployment is
shown in Table 1.
ETSI
12 ETSI TR 103 554 V1.1.1 (2018-08)
Table 1: FRMCS acceptable transmitted power at eNB connector taking into account
impact of BS Tx spurious emissions and Noise Rise from UE
FRMCS 1,4 MHz channel centre
918,7 920,3
frequency (MHz)
Standard under consideration in
Multi- Multi-
UMTS LTE UMTS LTE
adjacent bands Standard Standard
FRMCS acceptable Tx power (dBm) 24,2 22,2 22,2 48,8 45,8 48,8

In coordinated scenario, the maximum transmit power at 918,7 MHz can be the same than at 920,3 MHz.
4.5 Antenna diagrams
4.5.1 Antenna diagrams at the base station
Different types of antennas are deployed depending on the area. For the study, two different antennas are selected: One
with a horizontal beam angle of 65°, devoted to Non Line Of Sight (NLOS) situations - typically hilly terrains and
urban areas, and one more directive, with a horizontal beam angle of 30°, more suited to Line Of Sight (LOS) situations
- typically rural areas.
Antenna characteristics are summarized in Table 2 and an extended description is provided in Annex D.
Table 2: Summary of base station antenna patterns
Horizontal Vertical Gain Polarization Usage
Polarization Polarization
65° 7° 18 dB ±45° NLOS
30° 8,5° 20,5 dB ±45° LOS/NLOS

4.6 Radio propagation aspects
4.6.1 Radio propagation model
Simulations have to be based on railway specific time-variant channel impulse responses of the radio channel in order
to take into account multi-path radio propagation and Doppler-effects.
Four families of standards have been considered:
1) Okumura-Hata, Cost 207-GSM, COST 231 models and GSM specified models (see [i.1])
2) ITU-R 1997 for IMT 2000 (see [i.8]) and LTE specified scenarios (see [i.2] and [i.3])
3) ITU-R for IMT advanced (see [i.4])
4) Winner II (see [i.5])
Recent propagation models and multipath profiles have been aimed at being used for wireless systems with a small or
medium range. This is coherent since 3G and 4G standards have been developed for capacity rather than for coverage.
Early defined models such as COST 207 or 231 were derived at a time when coverage was the main priority rather than
high speed operation which is of particular significance within the scope of this study.
Most relevant parameters in rail environment are then:
• Frequency range
• Delays in Cluster Delay Line models
• Geometry, most of models are considering 1,5 m for handheld User Equipment
• Inter Site Distances (ISD)
ETSI
13 ETSI TR 103 554 V1.1.1 (2018-08)
• LOS scenarios are using Ricean factor with high domination of the direct path
Characteristics of models are summarized in the following Table 3, discrepancies are highlighted in red.
Table 3: Summary of model characteristics
Railway Okumura-Hata, ITU-R ITU-R IMT Winner II
current COST 207- IMT 2000 advanced
GSM COST 231
Propagation Frequency Band 8 150 to 2 000 MHz Rural: 450 MHz Rural: 2 GHz to
aspects range (900 MHz) 1 500 MHz to 6 GHz 6 GHz
Inter Site Up to 12 km Range up to Max = 1 732 m 20 km for Rural MRS 1 to 2 km
Distance 100 km (RMa) 20 km for Rural
(see note) (see note)
Path LOS, Ricean Ricean Factor = ETU has no LOS, LOS,
clearance < 3 dB 0 dB air direct path, HST Ricean factor Ricean factor
has only direct = 6 dB = 6 dB
path
Delayed paths Up to 20 µs HTx: up to Max delay = Max delay Max delay
20 µs 5 µs = 0,22 µs (not in < 0,5 µs (not in
line with 20 km line with 20 km
ISD) ISD)
Train speed 360 km/h, Max = 250 km/h Max = 350 km/h Max = 350 km/h Max = 350 km/h
projection to in R 1, no with double
500 km/h double Doppler Doppler
Geometry Base Station 10 to 45 m 30 to 200 m Up to 35 m 20 to 70 m
Δhb = 0 to
Antenna 50 m, i.e. up to
Height
46 m for 4 m
train antenna
height
Train Antenna 4 m to 4,5 m 1 to 10 m 1,5 m 1,5 m / 2,5 m

Height
NOTE: Delays are shorter than what can be expected with such ISD.

Indeed, propagation and geometry parameters that are deemed particularly relevant for Railways are summarized
below.
Table 4: Main characteristics of Railway context
Frequency range Band 8 (900 MHz)
Inter Site Distance Up to 12 km
Propagation aspects Path clearance LOS, Ricean < 3 dB
Delayed paths Up to 20 µs
Train speed 360 km/h, projection 500 km/h
Base Station Antenna Height 10 m to 45 m
Geometry
Train Antenna Height 4 m to 4,5 m

The Ricean factor taken here corresponds to worst case scenario. In actual deployments, higher values could be
encountered, leading to more favourable channel conditions.
4.6.2 Conclusion
Okumura-Hata models and COST 207-GSM COST 231 family (see [i.1]) are taken as the basis.
ETSI
14 ETSI TR 103 554 V1.1.1 (2018-08)
4.7 Frequency reuse scheme
In LTE radio, the frequency band is split in Resource Blocks (RB) which can be allocated individually to UEs by the
base station scheduler for each frame. All LTE cells may operate on the same frequency band; however, to mitigate
interference from neighbouring LTE cells, one technique is to coordinate RB allocations among cells. One possible
coordination scheme is fractional frequency reuse, which consists for example in allocating different RBs among two
neighbouring cells to cell edge UEs, while still allocating all the RBs (at a reduced power) for cell centre UEs (see
Figure 3). This can be seen as a frequency reuse factor 1 for cell centre UEs, and a frequency reuse factor > 1 (equal
to 2 in Figure 3 example) for cell edge UEs. Hence, not all RBs are allocated to cell edge UEs, but this is compensated
by a better SINR for those blocks.

Figure 3: Example of fractional frequency reuse for rail deployment
Results should indicate which kind of Fractional Frequency Reuse techniques is used.
4.8 Summary
Table 5 sums up all the parameters.
Table 5: Summary of evaluation parameters
Environment/scenario Rural/Urban
Railway shape and LOS/NLOS propagation Rural:
Straight: LOS
Curves: NLOS
(2 separate sets of results)
Hilly:
NLOS only
Urban:
NLOS only
Carrier Frequency (DL/UL) (MHz) 875,2/920,2 (for 1,4 MHz bandwidth)
874,5/919,5 (for 3 MHz bandwidth)
877,5/922,5 (for 5 MHz bandwidth)
Bandwidth (MHz) 1,4 (mandatory)
3 (optional)
5 (optional)
Inter-site distance (ISD) (km) Rural: 8
Urban: 2 and 4
BS antenna height (m) 18 (urban) - 30 (rural)
Train antenna height (m) 4,5
Tower to track distance (m) 15
Neighbour cells load Rural: 4 trains (2 in each direction)
High speed: 2 trains (1 in each direction)
Urban:
- 6 trains (3 in each direction)
- 4 trains (2 in each direction)
See note 1
ETSI
15 ETSI TR 103 554 V1.1.1 (2018-08)
Environment/scenario Rural/Urban
Train speeds (km/h) Urban: 80
Rural: 350
Hilly: 160
DL max power (dBm) In UIC-band:
46 before feeder (output of the BTS)
3 dB feeder loss
In E-UIC band (see clause 4.4.2)
22 or 46 (output of the BTS)
3 dB feeder loss
UL max Power (dBm) 23
See note 2
UL Power Control For instance: Open loop full compensation to be mentioned along with the
results.
Channel Estimation For instance: Real channel estimation - Frequency-domain Wiener 1-D -
Time interpolation - to be mentioned along with the results.
Link Channel Model
Tap delay lines Based on ETSI TS 145 005 [i.1]
Urban area 6 taps
Rural area 6 taps
Clustered delay lines Hilly terrain 12 taps
Channels for different antennas are not correlated.

ETSI TS 145 005 [i.1] channel models are Tapped Delay Line models.
Other Models (Recommendation ITU-R M.2135-1 [i.4]) provide additional
small scale parameters (Angles of arrival/departure (AoA/AoD) of the rays).
To take into account some small scales parameters, ETSI TS 145 005 [i.1]
channel models can be combined with the AoA/AoD provided in ITU-R
models. Since the number of taps in ETSI TS 145 005 [i.1] models
(6 taps/12 taps) is generally different from ITU-R models, AoA/AoD from
ITU-R models corresponding to the strongest first 6/12 taps are considered
for this hybrid channel model.
Path Loss Model (propagation model) Urban: Okumura-Hata
(LOS/NLOS effect is only taken into account in link channel model through
Rice coefficient distribution for the first tap)
Rural: Hata sub-urban
(LOS/NLOS effect is only taken into account in link channel model through
Rice coefficient distribution for the first tap)
Hilly: Hata sub-urban
Shadowing standard deviation (dB) Urban: Okumura-Hata
8 dB (NLOS only)
Rural: Hata sub-urban
6 dB in LOS, 8 dB in NLOS
Hilly: 8 dB (NLOS only)
Noise (dBm) -121,4
See note 3
Cyclic prefix Rural: Extended prefix
Urban: Normal prefix
Hilly: Extended prefix
Fractional frequency reuse technique To be mentioned along with the results.
Antenna pattern eNB/Antenna gain See clause 4.5.1
See note 4
eNB antenna downtilt (°) To be mentioned along with the results.
Antenna pattern UE/antenna gain One antenna: Omnidirectional/0dBi - Vertical polarization
Two antennas: Vertical polarization, > 10 λ separation
See note 5
MIMO schemes DL: 2x1, 4x1
DL: 2x2, 4x2
UL: 1x2, 1x4
NOTE 1: The aggregate data traffic per cell is 100 %.
NOTE 2: It is considered that the UE antenna gain compensates the feeder loss.
NOTE 3: Corresponds to thermal noise in a Resource Block of 180 kHz.
NOTE 4: In rural environment with straight line railway shape, the 30° HP antenna is assumed.
NOTE 5: For the antenna gain, see note of 'UL max Power' parameter.
ETSI
16 ETSI TR 103 554 V1.1.1 (2018-08)

4.9 Outcomes of the simulations
Output metrics need to include at least throughputs for DL and UL under the following conditions:
• Peak Data Rate
• Average
• 5 %-tile cell edge. This metric corresponds to the worst case of radio propagation conditions at the worst
position in the cell (maximum throughput experienced by the 5 % of trains with worst throughput)
NOTE: This 5 %-tile cell edge (or Worst Cell Edge) differs from coverage specification as defined in EIRENE
SRS ([i.9]), in which the specified GSM-R radio coverage probability is 95 % in each location intervals of
100 m.
Worst cell edge is 5 %-tile on every location starting from the hand over point, and therefore the associated data
throughput corresponds to a much more severe criteria than the one used in EIRENE specification ([i.9]).
5 Simulation results
5.1 Results set 1
5.1.1 Description
The simulator used for this result set is a software chain tool using a Monte-Carlo statistical approach. It simulates a
complete LTE PHY layer, i.e. it operates at 'Software simulation reference point' as defined in Figure 1.
The simulator considers multiple cells in a linear deployment along the track and encompasses link-level simulation as
well as system-level simulation.
Link level simulations allow to compute the bit error rate and packet/block error rate (PER) of the radio transmission
scheme, including detailed simulation of modulation and coding, MIMO scheme, channel estimation, small-scale fading
effects and AWGN. However, link level simulation does not include any effect of large-scale fading, i.e. distance-
dependent path-loss and shadowing, which impacts the (experienced) Signal-to-Noise Ratio (SNR) as well as the inter-
cell interference level.
System level simulations are required in order to quantify the impact of inter-cell interference on the system throughput
at cell level.
The simulation tool comprises then:
Step 1: Link level simulation
1) Computation of the PER vs. Signal-to-Interference-plus-Noise Ratio (SINR) for N different transmission
i
schemes (characterized by a specific modulation, coding rate, and MIMO scheme) that results in link level
throughputs T , i=1,…,N (assuming AWGN interference).
i
2) For each transmission scheme i and each SINR value, computation of the resulting throughput T (SINR)
res,i
taking into account PER as
T =×T 1− PER SINR
()()
res, i i i
3) For each SINR, storage in a look-up table of the maximum resulting throughput as shown in Figure 4 among
all transmission schemes (modulation, coding rate, MIMO) as a result of ideal link adaptation to large-scale
channel properties:
ETSI
17 ETSI TR 103 554 V1.1.1 (2018-08)
T()SINR = arg max T()SINR
()
max res,i
i
Step 2: System level simulation
1) For many drops of User Equipments (UEs) and many large-scale channel realizations (including large-scale
fading statistics), computation of the resulting SINR for each UE:
- A drop is a realization of UE positions within the cells. These positions are randomly drawn under the
constraints of the scenario of interest. For instance, the UE distribution depends on UE density.
2) From all the drops, computation of the Cumulative Density Function (CDF) of the throughput by using the
obtained SINR values as inputs in the look-up table T (SINR) obtained in the link-level evaluation step.
max
Figure 4: Maximum resulting throughput example for a given transmission scheme
and UE speed (link level simulation)
Antenna patterns are taken into account together with antenna down-tilt in the system level step. Large-scale fading
statistics follows a log-normal distribution.
In this railway environment, a straight railway line is assumed, with trains moving on both directions (see Figure 5).
The Inter Site Distance (ISD) is set depending on the scenario, i.e. ISD is set to 8 km for rural and to 2 km for urban, as
required in clause 4.8.
Each train embeds one UE and train positions are drawn following a uniform random distribution ensuring the train
density requirement for each scenario, i.e. 1 train per cell in each direction in high velocity train scenario, 2 in rural
scenario and 3 in urban scenario. These train positions form a train position set, each set corresponding to a UE drop.
A worst-case interference level is assumed: all active cells are fully loaded in both UL and DL, i.e. transmission occurs
over the whole bandwidth. DL interference experienced by the train in the serving cell depends on its position. UL
interference in the serving cell depends on the position of the trains in neighbour cells.
In total, 1 600 train positions sets have been considered during simulations, with 400 channel models realizations per
set.
From system-level simulations, the cell average spectral efficiency and the cell-edge throughput (e.g. the 5 %-tile
throughput) are computed. For getting the 5 %-tile throughput, the throughput CDF at any position of a track is
computed. This is different from the cell-edge throughput computed in 3GPP, which is the cell 5 %-tile throughput
taken over the entire cell coverage. The resulting curve allows evaluating the 5 %-tile data throughput at the worst
position of the train on the track.
ETSI
18 ETSI TR 103 554 V1.1.1 (2018-08)

(b)
(a)
Figure 5: Railway line configuration and inter-cell interference (a: downlink; b: uplink)
5.1.2 Specific assumptions and parameters
• Link channel model: As foreseen in clause 4.7, link level simulations combine Power Delay Profiles (PDP)
taken from ETSI TS 145 005 [i.1] and geometrical aspects (angle of arrival and angle of departure of the rays)
of Clustered Delay Lines (CDL) from Recommendation ITU-R M.2135-1 [i.4].
• Channel estimation includes a time-interpolation between consecutive subframes. It introduces a small
processing delay (0,07 ms in downlink with 2 transmit antennas, 0,14 ms with 4 transmit antennas and 0,29 ms
in uplink) but lowers Doppler effect.
• In DL, the MIMO schemes that have been chosen for the simulations are transmission schemes providing
transmit diversity (see clause 6.3.4.3 of ETSI TS 136 211 [i.7]), as they are more robust to the high train
velocity:
- The transmit diversity scheme with two transmit antennas is the Alamouti Space Frequency Block Code
(SFBC) applied on two adjacent sub-carriers (spatial diversity of 2 N with N number of receive
R R
antennas).
- The transmit diversity scheme with four transmit antennas is a combination of Alamouti SFBC and
).
Frequency Switched Transmit Diversity (FSTD) on four adjacent sub-carriers (spatial diversity of 4 N
R
• In UL, single-antenna transmission only is considered (SIMO) (diversity gain of N ).
R
• Transmit power in DL is 43 dBm taking into account a 3dB feeder loss.
• Transmit power in UL is 23 dBm.
• Antenna tilt is 3 degrees downtilt, if not stated otherwise.
• Bandwidth is 1,4 MHz, centred at 875,2/920,2 MHz.
• Rice factor for rural model (high speed scenario) is 0,4475 dB. There is no line of sight component in the other
models.
• Large scale shadowing standard deviation is 4 dB in Rural model, and 8 dB for Urban and Hilly models.
Frequency reuse scheme
The simulations do not implement a fractional frequency reuse algorithm. Separate results are provided for different
frequency reuse factors (hard frequency reuse), leading to a strong decrease of offered throughput in cell centres for
frequency reuse > 1.
However, with a fractional frequency reuse algorithm, the throughput results with frequency reuse 2 or 3 will be the
ones cell edge UEs could experience, while frequency reuse 1 results should be considered for cell centre UEs.
ETSI
19 ETSI TR 103 554 V1.1.1 (2018-08)
5.1.3 Results
Simulations have been made for the different scenarios foreseen in clause 4 and considering frequency reuse factors of
1, 2 and 3.
The different scenarios simulated are the following.
Table 6: Scenarios summary
ISD Neighbour cell load
Scenario name Model and speed
(km) (trains)
Urban Urban (NLOS, 80 km/h) 2 6
High Density Urban (
...