research article propagation and wireless channel...

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2013 Article ID 981281 11 pageshttpdxdoiorg1011552013981281

Research ArticlePropagation and Wireless Channel Modeling Development onWide-Sense Vehicle-to-X Communications

Wenyi Jiang Ke Guan Zhangdui Zhong Bo Ai Ruisi He Binghao ChenYuanxuan Li and Jia You

State Key Laboratory of Rail Traffic Control and Safety Beijing Jiaotong University Beijing 100044 China

Correspondence should be addressed to Ke Guan myeconehotmailcom

Received 22 June 2013 Revised 6 September 2013 Accepted 11 September 2013

Academic Editor Cesar Briso Rodrıguez

Copyright copy 2013 Wenyi Jiang et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The need for improving the safety and the efficiency of transportation systems has become of extreme importance In this regardthe concept of vehicle-to-X (V2X) communication has been introduced with the purpose of providing wireless communicationtechnology in vehicular networks Not like the traditional views the wide-sense V2X (WSV2X) communications in this paper aredefined by including not only vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications but also train-to-X(T2X) communications constituted of train-to-train (T2T) and train-to-infrastructure (T2I) communications All the informationrelated to the wide-sense V2X channels such as the standardization scenarios characters andmodeling philosophies is organizedand summarized to form the comprehensive understanding of the development of the WSV2X channels

1 Introduction

Over the past few years both the V2X and T2X have gainedpopularity in their attempts to improve road safety andrailway safety respectively As shown in Figure 1 in order toform the comprehensive understanding V2X and T2X arecollected together to constitute the complete conceptmdashwide-sense vehicle-to-X (WSV2X) The idea behind the WSV2Xcommunications is the deployment of wireless communica-tion technology in vehicular and railway networks In thismanner the vehicles trains and infrastructures build up awireless network enabling them to exchange controlling andtraffic information such as road obstacles accidents and soforth via the wireless communication links

Following are the situation of the standardization and theapplication related to WSV2X

(i) Intelligent transport systems (ITSs) [1] for V2X com-munications ITSs are advanced applications whichwithout embodying intelligence as such aim to pro-vide innovative services relating to different modes oftransport and traffic management and enable varioususers to be better informed and make safer morecoordinated and ldquosmarterrdquo use of transport networksIn order to realize the ITSs Institute of Electrical and

Electronics Engineers (IEEE) and EuropeanTelecom-munications Standards Institute (ETSI) have selectedthe same MAC and PHY layers for road traffic safetyapplications IEEE 80211p [2] and ETSI ITS G5 [3]In the USA in 1999 the Federal CommunicationsCommission (FCC) allocated 75MHz of licensedspectrum from 585 to 5925GHz as part of the ITSto use for Dedicated Short Range Communications(DSRC) [4] Hence IEEE 80211p part of theWirelessAccess in Vehicular Environments (WAVE) initiative[5] is developed to operate at this 59GHz bandwith 75MHz bandwidth and seven 10MHz channelsIn Europe ldquoITS-G5 mode of operationrdquo is definedby ETSI ES 202 663 (European profile standard forPHY and MAC layer of 5GHz ITS) operating atthe 59GHz band with 30MHz bandwidth and five10MHz channels Systems based on IEEE 80211p aswell as alternative systems are also being developed intheUSA EuropeanUnion and someAsian countries

(ii) For T2X communications the three main stan-dard systems are communication based train con-trol (CBTC) system [6] global system for mobilecommunications railway (GSM-R) [7] andTerrestrial

2 International Journal of Antennas and Propagation

Wide-sense vehicle-to-X (WSV2X)

Vehicle-to-X (V2X)

Vehicle-to-infrastructure Train-to-infrastructureVehicle-to-vehicle

Train-to-X (T2X)

Train-to-train (V2I) (V2V) (T2T) (T2I)

Figure 1 Block diagram of the constitution of the concept of wide-sense vehicle-to-X (WSV2X)

Trunked Radio (TETRA formerly known as Trans-European Trunked Radio [8]) (i) As one part ofEuropean Train Control System (ETCS) [9] CBTCis a railway signalling system that makes use ofthe telecommunications between the train and trackequipment for the trafficmanagement and infrastruc-ture control Bymeans of the CBTC systems the exactposition of a train is known more accurately thanwith the traditional signalling systems This results ina more efficient and safe way to manage the railwaytraffic Normally CBTC works at 24GHz (ii) As asubsystem of European Rail Traffic Management Sys-tem (ERTMS) [10] GSM-R is an international wire-less communications standard for railway commu-nication and applications used for communicationbetween train and railway regulation control centersGSM-R is based on GSM and thus works at the900MHz band In EIRENE-MORANE specificationsGSM-R can guarantee performance at speeds up to500 kmh without any communication loss (iii) Asa standard of ETSI TETRA is a professional mobileradio and two-way transceiver (colloquially known asa walkie talkie) specification TETRA was specificallydesigned for use by government agencies emergencyservices (police forces fire departments and ambu-lance) for public safety networks rail transportationstaff for train radios transport services and the mili-tary Normally TETRA works at the 400MHz band

(iii) So far there is no official standard for T2T commu-nications but on the basis of the thought of the ad hocintervehicle communication many academic effortshave been made to achieve the direct T2T commu-nications The most representing system is calledRailway Collision Avoidance System (RCAS) [11]studied by Germen Aerospace Center (DLR) Usingthe global satellite navigation system GALILEO thissystem determines and broadcasts information aboutposition movement vector to other trains around inits coverage for collision detection When a collisionis detected the train transmits prewarning messagesto others to avoid the accident

(iv) All the standard systems mentioned above are nar-rowband systems or at least are only implemented atthe stage of the narrowband However various wide-band systems such as Long Term Evolution for

Railway (LTE-R) [12 13] and LTE for V2X commu-nications are researched and under industrial con-sideration Railway operators do benefit from theevolution in public networks Fulfilling train controlrequirements GSM-R is based on the mature GSMtechnology This success is expected to be repeatedwith LTELTE-R [14] International Union of Rail-ways (UIC) has commenced to generate LTE-Rrequirements [15] In [16] the authors discuss severaltechnical challenges for LTE to be used for railroadcommunications because of fast mobility of trainup to 430 kmh Since signalling and train controlsystem is related to public safety very tight reliabilityrequirement needs to be met Correspondingly thewideband channel for train communication systemshas carried extensive research [17] Authors in [18]emphasize the feasibility of a smooth evolution fromGSM-R to LTE-R According to the authors it will notaffect the system while the subcarrier interval in LTEis 15 kHz above the 11 kHz recommended Authors in[19] propose a combination of wireless technologiesto provide broadband to the train communicationsystems As clarified in [20] the Doppler effect isalways a matter of study in this work Apart from theeffect from the academy many industrial companieshave been pushing the standardization of LTE-Rsuch as the work presented by Nokia Siemense in[14] by Alcatel-Lucent in [21] and by Huawei in [22]In the near future the time of the wideband WSV2Xcommunications can be expected to come For thepreparation the wideband channel characteristicsof WSV2X communications are under research by alarge number of scholars

As the basis of WSV2X systems the propagation andchannel characterization are always a fundamental topic withhigh research interest In the rest of the paper all the relatedinformation such as the scenarios characters and modelingphilosophies of the propagation and wireless channel ofWSV2Xwill be summarized to offer a panorama of the recentprogress and the current state of the artThe aim of this paperis to help the designers of communications systems to gainan comprehensive understanding of the pertinent channelcharacteristics and propagation researchers to assess wherethe most urgent requirements for further work lie

International Journal of Antennas and Propagation 3

Urban streets

Cutting

Viaduct


Vehicle-to-X (V2X) Train-to-X (T2X)

Highways

Rural streets

Suburban streets

Urban

Water

Rural

Suburban

Mountain

Tunnel

Station

Propagation mechanisms

Few scatterers present so the direct ray and reflection from the road surface rail surface or the water surface (much stronger than the reflection

from concrete surface vegetation or soil) dominate the wave propagation making typical

LOS propagation scenarios

Cutting walls can form a big container with rich reflection and scattering Diffraction owing to the terrain changes and scattering from the surface of

mountain could be the main multipath components Not many buildings present in these scenarios

There are few or no buildings in these two scenarios The vegetation could serve as some

scatterers but the main propagation mechanisms are still LOS and reflection from road surface or

rail surface

The channel appears strong multipath due to the presence of buildings Walls of the tunnel or the

station generate rich reflection This gives the chance to employ the waveguide theory to explain the propagation In the station scenario the huge

awnings are usually designed to stop the rain fromreaching the passengers and the trains These awnings

have a big chance to block the LOS

Figure 2 Typical scenarios of WSV2X channels The scenarios in the same light colorful blocks are similar or comparable

2 Scenarios of WSV2X Channels

Channel characteristics of WSV2X channels are dominatedby the properties of the scenarios in which the cars trainsand infrastructures communicate with each other As shownin Figure 2 there are four main scenarios for the V2Xchannels (highways suburban streets rural streets and urbanstreets defined by [23]) and nine main scenarios for theT2X channels (viaduct water suburban cutting mountainrural tunnel urban and station defined by [24]) Theinfluence of these scenarios on the channel characteristics hasbeen studied by many researchers For instance the channelcharacteristics of V2X in the highways are well summarizedby the authors of [23]Themultipath in the rural streets fromhills [25] and forests [26] is studied as well A preliminarytable of some narrowband channel characteristics of T2X isgiven by [24] For the tunnel scenario the propagationmech-anism changes along with the different distance betweentransmitter (Tx) and receiver (Rx) Thus the segmentation-based thought for modeling the propagation inside tun-nels has been developed and purchased by many scholars[27ndash35]

As shown in Figure 2 the scenarios in the blocks withthe same color are comparable They have similarities on

the characters of the environments so the channels areexpected to show similar properties However up to nowthe authors cannot find any open publication of doingthe comparative study on these channels in the similarscenarios even though each single scenario has already beenwell researched independently Some common senses of thepropagation mechanisms in every group of comparable V2Xand T2X scenarios are summarized in Figure 2 to give somerough inspirations of the joint analysis of the V2X and T2Xscenarios

In the future the combination of the similar V2X andT2X scenarios is expected In this way a unified theory ofinterpreting and modeling the propagation in the WSV2Xchannels can come trueThus the corresponding researcherscan use the same methodology and philosophy to character-ize the channel fulfill the network planning and design thesystem of WSV2X communication systems

3 Comparison of WSV2X Channels withCellular Communication Channels

Characteristics of WSV2X channels differ from those of tra-ditional cellular communications channelsThese differences


can be summarized in accordance with the following dimen-sions

(i) Heights of Tx and Rx the relation of heights of Txand Rx in T2I channels is very similar to cellularcommunication (CC) channelsmdashTx is considerablyhigher than Rx However the Tx and the Rx inV2X channel and T2T channel are normally at thesame height and in similar environments (peer-to-peer communications) This leads to different con-stitutions of the propagation mechanisms in thesechannels (i) LOS in CC channel and T2I channelthe Tx is much higher than the Rx So comparedwith V2X and T2T channels the LOS is relativelyeasier to be kept in CC channel and T2I channel(ii) Diffraction in CC channel and T2I channel thewave mainly propagates in the vertical plane so theobstructs are the roof tops of buildings top of thecutting walls and terrain changes Diffraction losscan be effectively estimated by the algorithms basedon raster database such as the classic multiedgemodels and their improved versions [37] Howeverthe propagation in V2X and T2T channels mainlyhappens in the horizontal plane the diffraction lossowing to street corners and buildings cannot bepredicted by raster-based algorithms vector-basedalgorithms such as geometrical theory of diffraction(GTD) and uniform theory of diffraction (UTD) arerequired (iii) Distributions of scatterers normally thearea around the Tx in CC channel and T2I channelis free of scatterers but in V2X and T2T channelsscattering and reflection can come from the scatterersaround both the TX and the RX

(ii) Frequency of communication compared withthe main frequency bands for CC channel (700ndash2100MHz) and T2X channel (400MHz 900MHzand 24GHz) the carrier frequency of V2X channelwhich is at 59GHz is very high Hence highersignal attenuation occurs in the V2X channel andspecific propagation processes like diffraction are lessefficient

(iii) Distance between Tx and Rx in the typical CCand T2I channels the distance of communicationis 1ndash3 km however in V2X and T2T channels thedistance over which communications can take placeis normally dozens of meters or hundreds of metersThis makes the V2X and T2T channels more localand therefore the scatters far from Tx and Rx can beignored

(iv) Nonstationarity in V2X and T2T channels both Txand Rx as well as many scatterers (other vehicles ortrains) are dynamic while in CC channels only oneof the Tx or Rx is moving andmoving scatterers haveless relative importance In T2I channels the speedof high-speed train can achieve 350 kmh whichis dramatically faster than the velocity of the userin the CC channel As a consequence the channelfluctuations inWSV2X channels are faster so that the

commonly used assumptions on stationarity usuallyare not valid The WSV2X channel is the typicalnonstationary channel that is the channel statisticschange within a rather short period of time [23]

4 WSV2X Channel Characterization

In the WSV2X channel all the propagation effects aresubsumed into the channel impulse response (CIR) whichcan be treated as the superposition of the contributions byall multipath components (MPCs) The CIR is time-variantsince the channel changes as Tx Rx and scatterers aremovingaround A complete description of the channel is thus givenby the time variant CIR [38] In order to facilitate the process-ing several metrics are extracted to provide more condensedcharacterization of the WSV2X channel Generally speakingdifferent systems require different metrics to support thechannel characteristics In this paper we summarize all thetypical metrics by classifications of the domains they belongto and the systems they serve

Figure 3 offers a panorama of metrics for the WSV2Xchannel classified in accordance with the antenna configura-tion bandwidth of the system and different domains All theparameters can be assigned to totally six domains

(i) loss and fading path loss shadow fading and small-scale fading

(ii) time domain coherence time and stationary time(iii) frequency domain coherence bandwidth and station-

ary bandwidth(iv) doppler domain Doppler shift and Doppler spread(v) delay domain RMS delay spread(vi) angular domain angle of arrival (AoA) and angle of

departure (AoD)

It can be seen that for the narrow-band single input singleoutput (SISO) systems the knowledge of path loss shadowfading small-scale fading coherence time stationary timecoherence bandwidth stationary bandwidth and Dopplershift is enough to characterize the channel But for the wide-band SISO systems Doppler spread and RMS delay spreadshould be compensated When the system evolves from SISOto multi-input multioutput (MIMO) the information of theangular domain such as AoA and AoD is mandatory to 5characterize the spatial correlations of the subchannels

Note that the stationary time and the stationary band-width are the specific parameters exclusively for the non-stationary channel Not like the coherence time and thecoherence bandwidth which are extracted from the scatteringfunction the stationary time and the stationary bandwidthmeasure the amount of Doppler and delay correlationrespectively They characterize the local scattering function(LSF) variation with respect to time and frequency In theWSV2X channel the channel statistics are valid only for ashort period of time (ldquoregion of stationarityrdquo determinedby the stationary time and the stationary bandwidth) Forinstance in each region of stationarity the Doppler spectrumof the first delay tap can be different because theDoppler shift


AntennaSISO MIMO

Bandwidth ofsystem

system

Narrowband Wideband

Loss andfading

Time domain

Frequencydomain

Doppler domain Doppler shift Doppler spread

RMS delay spreadDelay domain

Angulardomain

Path

Coherence

loss

timeStationary

time

Angle ofarrival(AoA)

Angle ofdeparture

(AoD)

Coherencebandwidth

Stationarybandwidth

Shadow Small-scalefading fading

Figure 3 Panorama of metrics for WSV2X channel classified in accordance with the antenna configuration and bandwidth of the systemand different domains Stationary time and stationary bandwidth (in the block with light red color) are the two specific parameters in thenonstationary channel

1

08

06

04

02

01086

Time (s)420

Del

ay (120583

s)

Figure 4 Example of the time-varying power delay profile (average squared magnitude of impulse response) in the WSV2X channel It canbe found that the delay of the first taps varies because the Tx and Rx approach each other firstly and move away from each other later TheLOS tap experiences fading the gap in the delay domain between strong components (clusters) and the LOS tap changes with time and thereis splitting of clusters over time as well

of the LOS component can change [23] Figure 4 (which is themeasured results in [39 40]) gives an example of the time-varying power delay profile (average squared magnitude ofimpulse response) in the WSV2X channel It can be foundthat the delay of the first taps varies because the Tx and Rxapproach each other firstly and move away from each otherlater The LOS tap experiences fading the gap in the delaydomain between strong components (clusters) and the LOStap changes with time and there is splitting of clusters overtime as well [23]

Up till now most of the metrics given by Figure 3 in theV2X channel have been well studied by many researcherssuch as the metrics given by Table 1 in [23] where thewideband parameters such as the delay spread and the

Doppler spread have been extracted from extensivemeasure-ments However there is no deep insight or comprehensiveunderstanding of these characters for wide-bandMIMOT2Xcommunication systems The authors of [24] offer the nar-rowband parameters for T2X channels in high-speed railwayat 930MHz but all the wideband parameters are still absentSome researchers have started to do some efforts such asthe wideband channel characterization of LTE for high-speedrailway at 235GHz with the tapped delay line model [41]and the calculation of the stationary interval in high-speedrailway scenario [42]The ergodicity and the reliability of theexperimental data and corresponding parameter extractionsare still quite limited more works in different scenarios atvarious frequencies are expected


Table 1 Channel characteristics of T2X V2X and standard scenarios

Category Scenarios Path lossexponent

Standard deviationof shadowing [dB] 119870-factorNakagami-119898 Delay spread (mean)

[ns]T2X Viaduct 2ndash4 2ndash4 119870 38ndash83 dB 192V2X Highway 18ndash4 40ndash61 119870 33ndash43 dB 30ndash340V2X Rural 179ndash4 mdash mdash 20ndash150Standard Suburban 24ndash40 4ndash8 119870 9 dB 59ndash84

Standard Ruralmacrocell 22ndash25 4ndash8 119870 7 dB 368ndash421

Standard Moving networksBS-MRS rural 19ndash22 4 119870 7 dB 40ndash55

T2X Cutting 25ndash4 3ndash5 119870 188 dB mdashV2X Suburban 21ndash39 mdash 119870 21 dB 20ndash150

StandardIndoor to

Outdooroutdoor toindoor

2 7 119870 32 dB 34ndash40

Standard Typical urbanmacrocell 26 4ndash8 119870 7 dB 854

T2X Tunnel 18ndash3 5ndash8 119898 08ndash54 15ndash40T2X Station 3ndash5 3ndash5 119870 05ndash54 dB mdashV2X Urban 161 42ndash67 119870 3ndash67 dB 3ndash1100Standard Indoor officeresidential 19 3-4 119870 7 dB 142ndash43

Standard Typical urbanmicrocell 23ndash4 3-4 119870 9 dB 37ndash74

Standard Large indoorhall 14ndash38 3-4 119870 2 dB 237ndash408

Table 1 summarizes some channel parameters of T2XV2X and standard scenarios (WINNER) according to theresearch results of [23 41 43ndash47] respectively Not likethe traditional way of researching the scenarios of differentsystems independently the similar or comparable scenariosof T2X V2X and standard scenarios are collected togetherin Table 1 There are three preliminary categories

(i) ldquoRelatively open scenariosrdquo this category is define bythe scenarios with a relative wide open space but fewscatterers or strong reflectors The T2X viaduct V2Xhighway V2X rural standard suburban standardrural macrocell and standard moving networks fallinto this category As shown in Table 1 the path lossexponent in this category is slightly larger than 2similar to the case of two-ray model This is becausethere are few scattering or strong reflections from thesurrounding space but the reflection from the railsurface or the road surface still contributes the powerand therefore makes the channel close to the two-raymodel Since the LOS can be kept in most of the casesof this category the Rician 119870-factor is the largest inthe three categories Furthermore the delay spread isrelatively large because the space is wide

(ii) ldquoSemiclosed scenariosrdquo this category indicates thatthe scenario is surrounded by some walls buildingsor terrain but still has some free or open space Thiscategory is in the middle of the three categories so

the values of the channel characteristics are basicallybetween the ldquorelatively open scenariosrdquo and ldquorelativelyclosed scenariosrdquo

(iii) ldquoRelatively closed scenariosrdquo this category mainlyincludes the limited-space and closed scenarios suchas the T2X tunnel T2X station V2X urban stan-dard indoor officeresidential standard typical urbanmicrocell and standard large indoorhall Since thesescenarios are mainly in a limited space the delayspread is relatively small Moreover the tunnel wallsstation structures and building walls constitute arelatively closed space which generates and containsthe multipath waves and finally results in a relativelysmall Rician 119870-factor Last but not least most of thescenarios in this category have the path loss exponentsmaller than 2 this indicates that certain waveguideeffect is established by the surrounding walls oftunnels stations and buildings This makes the pathloss even smaller than the free space propagation

This is the first effort to break the barrier of the V2XT2X and standard scenarios As shown in Table 1 the newclassification integrates the scenarios of different systems andredivides the categories according to the physical structureand propagation mechanisms In the future the comparativestudy is expected to gain a deeper insight to the channels invarious scenarios and finally form a comprehensive conceptof this work


5 Modeling Approach for WSV2X Channel

There are primarily three types of channel modelingapproaches forWSV2X channel deterministic channel mod-els stochastic channel models and geometry-based stochas-tic models

51 Deterministic Channel Model for WSV2X ChannelsDeterministic channel models are based on ray-tracing tech-niques which model the propagation channel in a specificlocation using the geographical and morphological informa-tion from a database This kind of modeling approach waspioneered byWiesbeck [48ndash50] Normally the 3D ray-opticalapproach covers the direct path specular reflections anddiffuse scattering Specular reflections can be calculated bythe image method [51] up to the 119899th order but in practiceonly the first and second order reflections are considered inorder to limit the computation time The diffuse scatteringis taken into account on surfaces seen by both the Tx andthe Rx Thus all surfaces of the structures in the scenariosuch as buildings trees vehicles and traffic signs are dividedinto tiles and single scattering processes can be determinedFor each ray full-polarimetric antenna patterns are used andchannel polarization matrices are computed The output ofthe deterministic channel models for each communicationlink is a time-variant CIR ℎ(120591 119905) which well characterizes thefrequency-selective channel and can be expressed as

ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)

where the 119896-th multipath component at time 119905 is formulatedby an amplitude 119886

119896(119905) a delay 120591

119896(119905) and an additional phase

shift 120593119896(119905) As the car or the train is running the number of

multipath components (119873(119905)) is also time variant Since theamplitude and the phase term for each multipath componentcan be combined in a complex coefficient 119860

119896(119905) = 119886

119896(119905) sdot

119890119895(2120587119891120591119896(119905)+120593119896(119905)) (1) can be rewritten as

ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)

By involving the polarization 119860119896(119905) can be expressed by

119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)

where the superscript 119867 denotes the Hermitian transpose119875119896denotes the complex channel polarization matrix that

includes the reflection or scattering losses of the consideredpath 997888

119890 (120579Tx119896 120593Tx119896) and 997888119890 (120579Rx119896 120593Rx119896) are the complex

polarization vectors for the Tx and the Rx respectively thatcover the gain and the polarization of the antennas at theangles of departure (120579Tx119896 120593Tx119896) and the angles of arrival(120579Rx119896 120593Rx119896) for the 119896-th multipath ray 119871

119896(119905) includes the

propagation loss and the phase shift based on the delay 120591119896(119905)

of the considered path For the direct ray 119875119896is the identity

matrix whereas for reflected and scattered rays of the order119899 119875119896is derived by [52] using

119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)

where 119877(120593) is the rotation matrix that can be expressed as

119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)

120593Tx119896 denotes the angle between the normal vector of theincidence plane and the field strength component 119864

120579at the

first reflection or scattering point and 120593Rx119896 denotes the anglebetween the scattering or reflecting plane and 119864

120579at the last

reflection or scattering point for the 119896-th multipath compo-nent respectively For the case of the multiple reflections orscattering the term 120593

119894119896denotes the angle between119864

120579and the

normal vector of the incidence plane at the 119894th reflection orscattering point for the 119896-th multipath ray 119877

119894119896denotes the

reflection or scattering matrix which is given by

119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)

where the elements 119903perp119894119896

and 119903119894119896

are the reflection or scatteringcoefficients of the surface belonging to the 119894th reflection orscattering point of the 119896-thmultipath component for perpen-dicular and parallel components of the electric field respec-tively For reflection processes these elements are determinedbased on the Fresnel equations for scattering processes thesevalues depend on the radar cross-section (RCS) of the object1205771119894119896

and 1205772119894119896

are cross-polarization coupling coefficientswhich are normally set to zero More detailed informationabout the deterministic modeling approach can be found in[53 54] Figures 5(a) and 5(b) offer an example of a ray-tracing map of the realistic high-speed railway scenario thepartial enlarged drawing for direct reflected and scatteringpaths in one snapshot Figure 5(c) gives a ray-tracing map ofa typical urban V2V channel

52 Stochastic ChannelModels Thestochastic channelmodeldoes not determine the impulse response in a specificlocation Rather it models the propagation characteristicsby assuming stochastic distributions of scatterers Basicallythe stochastic channel models can be divided into twotypes narrowband stochastic channel models and widebandstochastic channel models

Narrowband stochastic models do not concern the fre-quency selectivity of the channel but rather focus on char-acterization of the fading statistics as well as the DopplerspectrumNot like the Jakes spectrum in the CC channel witha ldquobathtubrdquo shape the Doppler spectrum in V2V channels ismore smoother [55 56] More efforts and interesting resultsare given in [57 58]

Wideband stochastic channel models create the statisticsof the received power with a certain delay Doppler shift and


Rx

Tx

(a)

(c)(b)

Figure 5 Example of (a) a ray-tracing map of the realistic high-speed railway scenario (b) partial enlarged drawing for direct reflected andscattering paths in one snapshot (c) and a ray-tracing map of a typical urban V2V channel

angle of arrival Based on the wide-sense stationary uncorre-lated scattering (WSSUS) assumption [59] the tapped delayline model is developed and widespread used for cellularsystem simulations This approach describes the CIR bymeans of a finite impulse response filter with a number ofdiscrete taps each of which is fading according to a pre-scribed probability density function and Doppler spectrumIn particular the IEEE 80211 p channel models employ the6-tap and 12-tap models developed by Ingram and coworkers[60 61] in different types of environments Since each tapcontains multiple paths and each path can have a differenttype of Doppler spectrum almost arbitrary Doppler spectracan be synthesized for each tap Due to the low complexityof the implementation tapped delay line models are widelyused even though they may suffer from less accurate repre-sentation of the nonstationarity in WSV2X channels

53 Geometry-Based Stochastic Models Geometry-basedstochasticmodeling (GSCM) is similar to the stochasticmod-eling but it uses a simplified ray tracing along with randomscatterers The GSCM is widely used in MIMO channelmodeling [38] Generally the GSCM models can be dividedinto the regular-shaped GSCMs (RS-GSCMs) and irregularshaped GSCMs (IS-GSCMs) [62] It depends on whether theeffective scatterers distribute on a regular shape (one-ringtwo-ring and ellipses) or an irregular shape determined bythe environment RS-GSCMs are mathematically tractableand used for theoretical analysis However the IS-GSCM caneasily handle the nonstationarity of WSV2X channel Boththe RS-GSCM and IS-GSCMmodels have been researched inthe V2X channel and the T2I channel Figure 6 gives sketchesof the geometry-based stochastic model of T2I channel [36]in the cutting scenario and V2V channel [26] with scatterersin realistic positions

54 Summary and Selecting a Suitable Modeling ApproachEach of the modeling methods offered above has specificadvantages and disadvantages

(i) Deterministic channel models offer themost accuratesimulation of the realistic channel model includingthe nonstationarity of the channel naturally Howeverthey require highly accurate topographical databasesand are numerically intensive to process Moreoverthe high resolution digital elevation model (DEM)for the ray-tracing method is very expensive and thislimits the application of the deterministic models

(ii) Stochastic channel models particularly the tappeddelay linemodels can be parameterized in a relativelyflexible way to describe channels in different scenar-iosHowever the implementations of the tappeddelayline models especially in the IEEE 80211 p channelmodels do not consider the nonstationarity of thechannel This is the main drawback of the applicationof such kind of modeling method to the WSV2Xchannel and therefore more and more researcherssuch as the authors in [63] are making efforts tointroduce the nonstationarity to the stochastic chan-nel models

(iii) Geometry-based stochastic models can reflect therealistic behavior of the WSV2X channel and implic-itly describe the nonstationarity of the channel Theyrequire more computation time than stochastic chan-nelmodels to realize aCIR but normally are stillmucheasier than the deterministic models

Hence it is hard to say which modeling method isthe most outstanding The choice always strongly dependson the concrete requirements of the application and thedevelopment of each method For instance the stochasticmodelingmethods can be a very simple and promising choiceto realize the channel if the channel does not change fast orthe nonstationarity can be successfully involved in the imple-mentation The deterministic modeling method can be moreproper if the network planning or the system level simulationhas a very high accurate requirement and more effectiveacceleration techniques such as the parallel processing and


BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax

dTrarrp(t) dprarrR(t)

dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)

Figure 6 (a) Sketch of the geometry-based stochastic model of T2I channels in the cutting scenario with scatterers in realistic positions [36](b) Sketch of the geometry-based stochastic model of V2V channels [26]

backface culling [64] are implemented According to thecurrent level the geometry-based stochastic model can betreated as a middle course of the other two kinds of modelingmethods in the simulation but it cannot be used for thenetwork planning or the site-specific simulation

6 Conclusion

Traditional concepts of V2X and T2X are summarizedtogether to constitute a more general conceptmdashwide-senseV2X (WSV2X) This paper offers an overview of the devel-opment of the WSV2X channels such as the standard-ization scenarios characters and modeling approachesThis paper integrates the common senses of V2X T2Xand standard scenarios and forms a new panorama of thescenarios classified by the similar physical characters andpropagation mechanisms This effort is made to give somerough inspirations of the joint research of the V2X and T2Xscenarios Following this thought it is expected to interpretandmodel theWSV2X channels in a uniformway As a result

the channel characterization network planning and systemdesign of WSV2X communication systems can be realizedby using the same methodology and philosophy Howeverbesides these results the answers of more open questions arestill missing For instance the developments of the V2X andT2X channels are not balanced how can we compensate thecorresponding research in T2X channels to fulfill a panoramaof theWSV2X channel How can we describe the nonstation-arity of the WSV2X channel in an easy way How can webreak the limitations of eachmodeling approaches or developsome hybrid models integrating the advantages of variousmodeling methods So many open questions combined withthe increasing importance of WSV2X communications willsurely make the WSV2X channels an exciting and promisingresearch field in the next years

Acknowledgments

The authors express their thanks to the support from theNNSF of China under Grant U1334202 NNSF of China


under Grant 61222105 Beijing Municipal NSF under Grant4112048 the Key Grant Project of Chinese Ministry ofEducation (no 313006) Project of StateKey Lab underGrantsRCS2012ZT013 RCS2011K008 and RCS2011ZZ008 and KeyProject for RailwayMinistry ofChina underGrant 2012X008-A

References

[1] T Vaa M Penttinen and I Spyropoulou ldquoIntelligent transportsystems and effects on road traffic accidents state of the artrdquo IETIntelligent Transport Systems vol 1 no 2 pp 81ndash88 2007

[2] IEEE Std 802 11-2007 ldquoPart 11 weireless lAN medium accesscontrol (MAC) and physical layer (PHY) specificationsrdquo IEEEStd 802 11 2007

[3] ETSI ES 202 663 ldquoIntelligent transport systems europeanprofile standard on the physical and medium access layer of5GHz ITSrdquo Draft Version 0 0 6 October 2009

[4] U S Federal Communications Commission ldquoReport andorderrdquo Tech Rep FCC 03-324 2003

[5] IEEE Std 1609 4 ldquoIEEE standard for wireless access in vehicu-lar environments (WAVE)-multi-channel operationrdquo February2011

[6] ldquoIEEE standard for communications-based train control(CBTC) performance and functional requirementsrdquo December1999

[7] httpwwwuicassofr[8] ETSI ETR 300-3 (2000-02) ldquoTerrestrial trunked radio

(TETRA) voice plus data (V+D) designersrsquo guide part 3Direct Mode Operation (DMO)rdquo 1st edition 2000

[9] H Zhou and B Luo ldquoDesign and budget analysis of RF receiverof 58 GHz ETC readerrdquo in Proceedings of the InternationalConference on Communication Technology (ICCT rsquo10) pp 354ndash357 Nanjing China November 2010

[10] ERTMSETCSmdashClass 1 FFFIS for Eurobalise Subset-026 Issue2 4 1 September 2007

[11] C R Garcıa A Lehner T Strang and K Frank ldquoChannelmodel for train to train communication using the 400MHzbandrdquo inProceedings of the 2008 IEEE 67thVehicular TechnologyConference pp 3082ndash3086 Singapore May 2008

[12] 1st Year Research Report on Railroad Wireless CommunicationSystem Korea Railroad Research Institute Seoul Republic ofKorea 2011

[13] K Guan Z Zhong and B Ai ldquoAssessment of LTE-R using highspeed railway channel modelrdquo in Proceedings of the 3rd Inter-national Conference on Commnunications and Mobile Coput-ing pp 461ndash464 Qingdao China 2011

[14] P TibergGSM-Railway Yesterday Today and Tomorrow NokiaSiemens Networks Banebranchen Copenhagen Denmark2009

[15] httpwwwuicorg[16] R Y Kim J S Kwak and H C Hwang ldquoTechnical chal-

lenges of railroad communications using long term evolutionrdquoin Proceedings of the IEEE International Conference on ICTConvergence (ICTC rsquo12) pp 1ndash2 Jeju Repablic of Korea 2012

[17] M Aguado I Liedo Samper M Berbineau and E Jacob ldquo4Gcommunication technologies for train to ground communi-cation services LTE versus WIMAX a simulation studyrdquo inProceedings of the 9th World Congress on Railway Research pp1ndash10 Lille France 2011

[18] T Gao and B Sun ldquoA high-speed railway mobile communica-tion system based on LTErdquo in Proceedings of the InternationalConference on Electronics and Information Engineering (ICEIErsquo10) pp 414ndash417 Kyoto Japan August 2010

[19] J Zhang Z Tan Z Zhong and Y Kong ldquoA multi-mode multi-band and multi-system-based access architecture for high-speed railwaysrdquo in Proceedings of the IEEE 72nd Vehicular Tech-nology Conference Fall (VTC rsquo10) pp 1ndash5 Ottawa CanadaSeptember 2010

[20] UIC GSM-R Procurement Guide Union Internationale desChemins de Fer Paris France 2007

[21] O Andre LTE and Its Applications in Railways Networks andthe New Economy Alcatel-Lucent Cambridge UK 2010

[22] ldquoHuawei eyes European marketrdquo September 2011 httpwwwrailwaygazettecomnewsbusinesssingle-viewviewhuawei-eyes-european-markethtml

[23] A F Molisch F Tufvesson J Karedal and C F Meck-lenbrauker ldquoA survey on vehicle-to-vehicle propagation chan-nelsrdquo IEEE Wireless Communications vol 16 no 6 pp 12ndash222009

[24] B Ai R He Z Zhong et al ldquoRadio wave propagation scenepartitioning for high-speed railsrdquo International Journal onAntennas and Propagation vol 2012 Article ID 815332 7 pages2012

[25] L Cheng B E Henty D D Stancil F Bai and P MudaligeldquoMobile vehicle-to-vehicle narrow-band channel measurementand characterization of the 59GHz dedicated short rangecommunication (DSRC) frequency bandrdquo IEEE Journal onSelected Areas in Communications vol 25 no 8 pp 1501ndash15162007

[26] J Karedal F Tufvesson N Czink et al ldquoA geometry-basedstochastic MIMO model for vehicle-to-vehicle communica-tionsrdquo IEEE Transactions on Wireless Communications vol 8no 7 pp 3646ndash3657 2009

[27] K Guan Z Zhong B Ai and C Briso ldquoPropagation mecha-nism analysis before the break point inside tunnelsrdquo in Proceed-ings of the IEEE 74th Vehicular Technology Conference pp 1ndash5San Francisco Calif USA 2011

[28] KGuan Z Zhong B Ai andC Briso-Rodrıguez ldquoPropagationmechanism modelling in the near region of circular tunnelsrdquoIET Microwaves Antennas and Propagation vol 6 no 3 pp355ndash360 2012

[29] K Guan Z Zhong B Ai and C Briso-Rodrıguez ldquoModelingof the division point of different propagation mechanismsin the near-region within arched tunnelsrdquo Wireless PersonalCommunications vol 68 no 3 pp 489ndash505 2013

[30] K Guan Z Zhong B Ai R He Y Li and C Briso ldquoProp-agation mechanism modeling in the near-region of arbitrarycross-sectional tunnelsrdquo International Journal of Antennas andPropagation vol 2012 Article ID 183145 11 pages 2012

[31] K Guan Z Zhong B Ai and C Briso-Rodrıguez ldquoResearchof propagation characteristics of break point Near zone andfar zone under operational subway conditionrdquo in Proceedingsof the 6th International Wireless Communications and MobileComputing Conference (IWCMC rsquo10) pp 114ndash118 Caen FranceJuly 2010

[32] K Guan Z Zhong B Ai and C Briso-Rodrıguez ldquoMeasure-ment and modeling of subway near shadowing phenomenonrdquoin Proceedings of the 5th International ICST Conference onCommunications and Networking in China (ChinaCom rsquo10) pp1ndash5 Beijing China August 2010


[33] K Guan Z Zhong B Ai and C Briso ldquoNovel hybrid prop-agation model inside tunnelsrdquo in Proceedings of the IEEE 75thVehicular Technology Conference pp 1ndash5 Yokohama Japan2012

[34] K Guan Z Zhong B Ai et al ldquoComplete propagation modelstructure inside tunnelsrdquo Progress in Electromagnetics Researchvol 141 pp 711ndash726 2013

[35] K Guan Z Zhong B Ai R He and C Briso-Rodriguez ldquoFive-zone propagation model for large-size vehicles inside tunnelsrdquoProgress in Electromagnetics Research vol 138 pp 389ndash4052013

[36] B Chen and Z Zhong ldquoGeometry-based stochastic modelingfor MIMO channel in high-speed mobile scenariordquo Interna-tional Journal of Antennas and Propagation vol 2012 ArticleID 184682 6 pages 2012

[37] K Guan Z D Zhong B Ai and T Kuerner ldquoSemi-deter-ministic propagation modeling for high-speed railwayrdquo IEEEAntennas andWireless Propagation Letters vol 12 pp 789ndash7922013

[38] P Almers E Bonek A Burr et al ldquoSurvey of channel andradio propagation models for wireless MIMO systemsrdquo EurasipJournal on Wireless Communications and Networking vol 2007Article ID 19070 2007

[39] T Abbas J Karedal F Tufvesson A Paier L Bernado and A FMolisch ldquoDirectional analysis of vehicle-to-vehicle propagationchannelsrdquo in Proceedings of the IEEE 73rd Vehicular TechnologyConference (VTC rsquo11) pp 1ndash5 Budapest Hungry May 2011

[40] J Nuckelt T Abbas F Tufvesson C Mecklenbrauker LBernado and T Kurner ldquoComparison of ray tracing and chan-nel-sounder measurements for vehicular communicationsrdquo inProceedings of the IEEE 77th Vehicular Technology Conference(VTC rsquo13) pp 1ndash5 Dresden Germany June 2013

[41] L Liu C Tao J Qiu et al ldquoPosition-basedmodeling forwirelesschannel on high-speed railway under a viaduct at 235GHzrdquoIEEE Journal on Selected Areas in Communications vol 30 no4 pp 834ndash845 2012

[42] B Chen Z Zhong and B Ai ldquoStationarity intervals of time-variant channel in high speed railway scenariordquo China Commu-nications vol 9 pp 64ndash70 2012

[43] H Wei Z Zhong K Guan and B Ai ldquoPath loss models inviaduct and plain scenarios of the high-speed railwayrdquo in Pro-ceedings of the 5th International ICST Conference on Commu-nications and Networking in China (ChinaCom rsquo10) pp 1ndash5Beijing China August 2010

[44] K Guan Z Zhong J I Alonso and C Briso-Rodrıguez ldquoMea-surement of distributed antenna systems at 24GHz in a realisticsubway tunnel environmentrdquo IEEE Transactions on VehicularTechnology vol 61 no 2 pp 834ndash837 2012

[45] Y Guo J Zhang C Tao L I U Liu and L Tian ldquoPropagationcharacteristics of wideband high-speed railway channel inviaduct scenario at 2 35GHzrdquo Journal of Modern Transporta-tion vol 20 no 4 pp 206ndash212 2012

[46] G Acosta-Marum andM A Ingram ldquoSix time- and frequency-selective empirical channel models for vehicular wirelessLANsrdquo IEEE Vehicular Technology Magazine vol 2 no 4 pp4ndash11 2007

[47] IST WINNER II D1 1 2 ldquoWINNER II channel modelsrdquo httpwwwist-winnerorgWINNER2-DeliverablesD112v11pdf

[48] J Maurer T Fugen andWWiesbeck ldquoNarrow-band measure-ment and analysis of the inter-vehicle transmission channel at52 GHzrdquo in Proceedings of the IEEE 55th Vehicular Technology

Conference (VTC rsquo02) pp 1274ndash1278 Birmingham UK May2002

[49] J Maurer T M Schafer and W Wiesbeck ldquoA realistic descrip-tion of the environment for inter-vehicle wave propagationmodellingrdquo inProceedings of the IEEE 54thVehicular TechnologyConference (VTC rsquo01) pp 1437ndash1441 Atlantic NJ USA October2001

[50] J Maurer T Fugen T Schafer and W Wiesbeck ldquoA new inter-vehicle communications (IVC) channel modelrdquo in Proceedingsof the IEEE 60th Vehicular Technology Conference (VTC rsquo04) pp9ndash13 Los Angeles Calif USA September 2004

[51] J W McKown and R L Hamilton Jr ldquoRay tracing as a designtool for radio networksrdquo IEEE Network vol 5 no 6 pp 27ndash301991

[52] A Maltsev R Maslennikov A Lomayev A Sevastyanov andA Khoryaev ldquoIEEE 802 11-090431r0rdquo April 2009

[53] J Nuckelt M Schack and T Kurner ldquoDeterministic andstochastic channel models implementedin a physical layersimulator for Car-to-X communicationsrdquo Advances in RadioScience vol 9 pp 165ndash171 2011

[54] K Guan Z Zhong B Ai and T Kurner ldquoDeterministic pro-pagation modeling for the realistic high-speed railway envi-ronmentrdquo in Proceedings of the IEEE 77th Vehicular TechnologyConference (VTC rsquo13) pp 1ndash5 Dresden Germany 2013

[55] W C Jakes Ed Microwave Mobile Communications WileyNew York NY USA 1974

[56] A S Akki and FHaber ldquoA statisticalmodel ofmobile-to-mobileland communication channelrdquo IEEE Transactions on VehicularTechnology vol 35 no 1 pp 2ndash7 1986

[57] A S Akki ldquoStatistical properties of mobile-to-mobile landcommunication channelsrdquo IEEE Transactions on VehicularTechnology vol 43 no 4 pp 826ndash831 1994

[58] F Vatalaro ldquoDoppler spectrum in mobile-to-mobile communi-cations in the presence of three-dimensional multipath scatter-ingrdquo IEEE Transactions on Vehicular Technology vol 46 no 1pp 213ndash219 1997

[59] P A Bello ldquoCharacterization of randomly time-variant linearchannelsrdquo IEEE Transactions on Communications vol 11 pp360ndash393 1963

[60] G Acosta andM A Ingram ldquoModel development for the wide-band expressway vehicle-to-vehicle 24GHz channelrdquo in Pro-ceedings of the IEEE Wireless Communications and NetworkingConference (WCNC rsquo06) pp 1283ndash1288 Las Vegas Nev USAApril 2006


[62] C Wang X Cheng and D Laurenson ldquoVehicle-to-vehiclechannel modeling and measurements recent advances andfuture challengesrdquo IEEE Communications Magazine vol 47 no11 pp 96ndash103 2009

[63] A Chelli and M Patzold ldquoA non-stationary MIMO vehicle-to-vehicle channel model derived from the geometrical streetmodelrdquo in Proceedings of the IEEE 74th Vehicular TechnologyConference (VTC rsquo11) pp 1ndash6 San Francisco Calif USASeptember 2011

[64] T Kurner and M Schack ldquo3D ray-tracing embedded into anintegrated simulator for car-to-X communicationsrdquo in Proceed-ings of the 20th URSI International Symposium on Electromag-netic Theory (EMTS rsquo10) pp 880ndash882 Berlin Germany August2010

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks




Vehicle-to-X (V2X)

Vehicle-to-infrastructure Train-to-infrastructureVehicle-to-vehicle

Train-to-X (T2X)

Train-to-train (V2I) (V2V) (T2T) (T2I)

Figure 1 Block diagram of the constitution of the concept of wide-sense vehicle-to-X (WSV2X)

Trunked Radio (TETRA formerly known as Trans-European Trunked Radio [8]) (i) As one part ofEuropean Train Control System (ETCS) [9] CBTCis a railway signalling system that makes use ofthe telecommunications between the train and trackequipment for the trafficmanagement and infrastruc-ture control Bymeans of the CBTC systems the exactposition of a train is known more accurately thanwith the traditional signalling systems This results ina more efficient and safe way to manage the railwaytraffic Normally CBTC works at 24GHz (ii) As asubsystem of European Rail Traffic Management Sys-tem (ERTMS) [10] GSM-R is an international wire-less communications standard for railway commu-nication and applications used for communicationbetween train and railway regulation control centersGSM-R is based on GSM and thus works at the900MHz band In EIRENE-MORANE specificationsGSM-R can guarantee performance at speeds up to500 kmh without any communication loss (iii) Asa standard of ETSI TETRA is a professional mobileradio and two-way transceiver (colloquially known asa walkie talkie) specification TETRA was specificallydesigned for use by government agencies emergencyservices (police forces fire departments and ambu-lance) for public safety networks rail transportationstaff for train radios transport services and the mili-tary Normally TETRA works at the 400MHz band

(iii) So far there is no official standard for T2T commu-nications but on the basis of the thought of the ad hocintervehicle communication many academic effortshave been made to achieve the direct T2T commu-nications The most representing system is calledRailway Collision Avoidance System (RCAS) [11]studied by Germen Aerospace Center (DLR) Usingthe global satellite navigation system GALILEO thissystem determines and broadcasts information aboutposition movement vector to other trains around inits coverage for collision detection When a collisionis detected the train transmits prewarning messagesto others to avoid the accident

(iv) All the standard systems mentioned above are nar-rowband systems or at least are only implemented atthe stage of the narrowband However various wide-band systems such as Long Term Evolution for

Railway (LTE-R) [12 13] and LTE for V2X commu-nications are researched and under industrial con-sideration Railway operators do benefit from theevolution in public networks Fulfilling train controlrequirements GSM-R is based on the mature GSMtechnology This success is expected to be repeatedwith LTELTE-R [14] International Union of Rail-ways (UIC) has commenced to generate LTE-Rrequirements [15] In [16] the authors discuss severaltechnical challenges for LTE to be used for railroadcommunications because of fast mobility of trainup to 430 kmh Since signalling and train controlsystem is related to public safety very tight reliabilityrequirement needs to be met Correspondingly thewideband channel for train communication systemshas carried extensive research [17] Authors in [18]emphasize the feasibility of a smooth evolution fromGSM-R to LTE-R According to the authors it will notaffect the system while the subcarrier interval in LTEis 15 kHz above the 11 kHz recommended Authors in[19] propose a combination of wireless technologiesto provide broadband to the train communicationsystems As clarified in [20] the Doppler effect isalways a matter of study in this work Apart from theeffect from the academy many industrial companieshave been pushing the standardization of LTE-Rsuch as the work presented by Nokia Siemense in[14] by Alcatel-Lucent in [21] and by Huawei in [22]In the near future the time of the wideband WSV2Xcommunications can be expected to come For thepreparation the wideband channel characteristicsof WSV2X communications are under research by alarge number of scholars

As the basis of WSV2X systems the propagation andchannel characterization are always a fundamental topic withhigh research interest In the rest of the paper all the relatedinformation such as the scenarios characters and modelingphilosophies of the propagation and wireless channel ofWSV2Xwill be summarized to offer a panorama of the recentprogress and the current state of the artThe aim of this paperis to help the designers of communications systems to gainan comprehensive understanding of the pertinent channelcharacteristics and propagation researchers to assess wherethe most urgent requirements for further work lie


Urban streets

Cutting

Viaduct



Highways

Rural streets

Suburban streets

Urban

Water

Rural

Suburban

Mountain

Tunnel

Station









rail surface


























departure (AoD)




AntennaSISO MIMO

Bandwidth ofsystem

system

Narrowband Wideband

Loss andfading

Time domain

Frequencydomain



Angulardomain

Path

Coherence

loss

timeStationary

time


Angle ofdeparture

(AoD)

Coherencebandwidth

Stationarybandwidth



1

08

06

04

02

01086

Time (s)420

Del

ay (120583

s)













StandardIndoor to


2 7 119870 32 dB 34ndash40















ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Urban streets

Cutting

Viaduct



Highways

Rural streets

Suburban streets

Urban

Water

Rural

Suburban

Mountain

Tunnel

Station









rail surface


























departure (AoD)




AntennaSISO MIMO

Bandwidth ofsystem

system

Narrowband Wideband

Loss andfading

Time domain

Frequencydomain



Angulardomain

Path

Coherence

loss

timeStationary

time


Angle ofdeparture

(AoD)

Coherencebandwidth

Stationarybandwidth



1

08

06

04

02

01086

Time (s)420

Del

ay (120583

s)













StandardIndoor to


2 7 119870 32 dB 34ndash40















ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation






















departure (AoD)




AntennaSISO MIMO

Bandwidth ofsystem

system

Narrowband Wideband

Loss andfading

Time domain

Frequencydomain



Angulardomain

Path

Coherence

loss

timeStationary

time


Angle ofdeparture

(AoD)

Coherencebandwidth

Stationarybandwidth



1

08

06

04

02

01086

Time (s)420

Del

ay (120583

s)













StandardIndoor to


2 7 119870 32 dB 34ndash40















ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










AntennaSISO MIMO

Bandwidth ofsystem

system

Narrowband Wideband

Loss andfading

Time domain

Frequencydomain



Angulardomain

Path

Coherence

loss

timeStationary

time


Angle ofdeparture

(AoD)

Coherencebandwidth

Stationarybandwidth



1

08

06

04

02

01086

Time (s)420

Del

ay (120583

s)













StandardIndoor to


2 7 119870 32 dB 34ndash40















ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation

















StandardIndoor to


2 7 119870 32 dB 34ndash40















ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













ℎ (120591 119905) =

119873(119905)

sum

119896=1

119886119896 (119905) sdot 119890

119895(2120587119891120591119896(119905)+120593119896(119905)) sdot 120575 (120591 minus 120591119896 (119905)) (1)


119896(119905) a delay 120591




119896(119905) = 119886

119896(119905) sdot


ℎ (120591 119905) =

119873(119905)

sum

119896=1

119860119896 (119905) sdot 120575 (120591 minus 120591

119896 (119905)) (2)


119860119896 (119905) = 119871

119896 (119905) sdot997888119890 (120579Rx119896 120593Rx119896)

119867sdot 119875119896sdot997888119890 (120579Tx119896 120593Tx119896) (3)









119875119896= 119877 (120593

119899minus119898119896) sdot

119899minus2

prod

119898=0

(119877119899minus119898119896

sdot 119877 (120593119899minus119898119896

)) sdot 1198771119896

sdot 119877 (120593Tx119896)

(4)


119877 (120593) = (cos120593 sin120593

minus sin120593 cos120593) (5)


120579at the


120579at the last



120579and the




119877119894119896

= (119903perp119894119896

1205771119894119896

1205772119894119896

119903119894119896

) (6)


and 119903119894119896


and 1205772119894119896






Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Rx

Tx

(a)

(c)(b)










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










BS

Velocity

Scatterers

Y

Z

X

[Xmin Xmax ]

WDI

Ymin

minusYmin

(a)

ΩTp(t)

ΩRp(t)Wroad

WDI

xmin

xR(t) yR(t)

xp(t) yp(t)

120592p

120592R

120592T

x

y

0

xq yq

xr yr

xmax


dTrarrR(t)

y2DIy2SD

y1DI

y1SD xT(t) yT(t)

(b)



6 Conclusion



Acknowledgments




References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











References





































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article propagation and wireless channel...

Documents