Deterministic Modeling and Stochastic Analysis for
Channel in Composite High-Speed Railway
Scenario
Jingya Yang
∗
,BoAi
∗†
,KeGuan
∗
, Danping He
‡
,RuisiHe
∗
, Bei Zhang
∗
, Zhangdui Zhong
∗
, Zhuyan Zhao
§
,
Deshan Miao
§
, and Hao Guan
§
∗
State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, 100044, Beijing, China
†
Engineering University of Armed Police Force, Dept. Telecommunication Engineering, 710086, Xi’an, China
‡
ETSI y Sistemas de Telecomunicaci
´
on, Technical University of Madrid, 28031, Madrid, Spain
§
Nokia, Beijing, China
E-mail: jyyang@bjtu.edu.cn
Abstract—The rapidly time-varying channel in high-speed
railway poses tough design challenges, which necessitates the
research of accurate channel models. Existing researches focus on
the isolated high-speed railway scenarios, and mainly deal with
path loss, shadowing fading, Ricean K-factor and delay spread.
However, few studies have been done in Doppler domain. In this
paper, a deterministic channel model that employs ray-tracing
algorithm is presented. The proposed deterministic modeling
approach is applied in composite high-speed railway scenario
rather than isolated one. The scenario is flexibly reconstructed
through SketchUp. The simulation results are validated by the
corresponding data measured. The channel characteristics in
Doppler domain and the effect of Doppler shift are statistically
analyzed based on deterministic channel model. The transition
regions in the composite scenario are emphatically investigated,
and the results are compared with those of prior studies.
Keywords—Composite scenario, high-speed railway, propaga-
tion channel, ray-tracing model, 3D reconstruction
I. INTRODUCTION
The ongoing development of railway communication sys-
tems requires a deep understanding of the high-speed railway
(HSR) channel [1]. The outdoor channel between the base
station (BS) and the dedicated mobile relay station (MRS) on
the surface of the train is mainly studied in this paper.
Several measurement campaigns have been conducted in
different HSR environments for the outdoor channels, e.g.,
the viaduct scenario, the cutting scenario, the train station
scenario, etc [2]. These studies mainly focused on path loss,
shadowing fading, Ricean K-factors and delay spread in an iso-
lated scenario, and thus ignored fine details about the transition
from one scenario to another scenario as well as small-scale
parameters in Doppler domain. Thus, the non-stationarity of
HSR rapidly time-varying channels is not fully demonstrated.
This work is supported by Nokia, the NNSF of China under Grant
U1334202, 863 project under Grant 2014AA01A706, NNSF under Grant
61501021, the Natural Science Base Research Plan in Shaanxi Province of
China under Grant no. 2015JM6320, State Key Lab of Rail Traffic Control
and Safety Project under Grant RCS2015ZZ001, and National Natural Science
Foundation of China under Grant 61501020. Corresponding author: Bo Ai,
e-mail: boai@bjtu.edu.
However, the research challenge of HSR outdoor channels
is exactly the non-stationarity characteristic [1]. The non-
stationarity of propagation channel in HSR environments is
mainly caused by the fact that the MRS with high velocity
has experienced various scenarios over a short period of time.
The high velocity leads directly to the dispersion in Doppler
domain. In a real HSR environment, scenarios are successive,
and radio propagation environment is composed of different
scenarios, such as, viaducts, cuttings, tunnels, etc. Hence, it
is highly desired to investigate the channel non-stationarity
characteristics in composite scenario. In [3], the path loss was
analysed in composite scenarios. Nevertheless, more channel
changes are still missing.
Deterministic channel modeling based on ray-tracing (RT)
algorithm is a common modeling approach in HSR (e.g. [4],
[5]). However, the weakness of the RT algorithm is apparent
in large and complicated scenarios. Since the 3D scenario
reconstruction was purely manual processes [5] or based on
electronic map, it may expend much manpower and funds. Fur-
thermore, the authors in [6] used laser scanners and cameras
building 3D geometry database. In [7], the OpenStreetMap
data is employed, which is few for HSR scenarios yet. Ref.
[8] automatically reconstructs 3D scenario based on machine
learning algorithm that seems a demanding work.
A novel 3D scenario reconstruction is presented in this
paper. We obtain the design drawings from China Railway Sur-
vey and Design Institute, and then import them to SketchUp
for reconstructing 3D scenario. The export of SketchUp is
linked with RT algorithm for deterministic modeling (see Sec-
tion II). The deterministic modeling for propagation channel in
composite HSR scenario is validated by data measured in the
same scenario (Section III). One drawback of deterministic
modeling is the high computational effort. For simplicity, a
stochastic analysis is provided by introducing the key channel
properties in HSR composite scenario.
978-1-5090-1698-3/16/$31.00 ©2016 IEEE