Research On V2X Wireless Channel Modeling And Positioning Method In Intelligent Transportation Scenarios

With the rapid development of society and economy,traffic problems are becoming more and more serious,such as traffic congestion,traffic accidents and lower travel rate,etc.Vehicular Ad Hoc Network(VANET)is the core component of intelligent transportation system,it is supported by information technology,and the high-speed,low-latency communication and traffic situational awareness among various elements(human,vehicle,road,cloud)in network can not only guarantee traffic safety and improve transportation efficiency,but also lay a solid foundation for unmanned driving and traffic decision control.Vehicle-to-Everything(V2X)is the main communication mode of VANET,and its channel is different from the traditional cellular channel,with the characteristics of high working frequency band,low antenna height and fast vehicle movement,etc.Through V2 X communication,it can provide the required information for vehicle localization,therefore it is necessary to research on V2 X channel modeling and vehicle localization.At present,the research on V2 X channel modeling is mainly focused on urban scenarios,and there are fewer wireless channel models in tunnel and highway environments,and most of the models in tunnel scenarios do not consider the scattering components of the tunnel walls.Currently,the more adopted vehicle localization technique is the multi-base station localization technique,but this method has the problems of time synchronization and inadequate infrastructure.For the tunnel scenarios,if this technology is used,there are also problems of dense base stations and high system redundancy.In order to solve these problems,this paper investigates V2 X channel modeling and vehicle localization methods in tunnel and highway scenarios.The main research work in this paper is as follows:(1)Aiming at the problem that it is difficult to separate multipath high-resolution parameters of V2 X wireless signals in the dense multipath propagation environments,a SpaceAlternating Generalized Expectation-Maximization(SAGE)algorithm based on Autoregressive Moving Average(ARMA)filter is proposed in this paper.The proposed algorithm is applicable to the channel parameter estimation in the dense multipath environment.This method uses ARMA filter to preprocess the Diffuse Multipath Component(DMC)and combines the quasi-maximum likelihood SAGE algorithm to sequentially and iteratively estimate the parameters of each path and DMC.The problem of difficult separation of multipath parameters due to missing rank of signal covariance matrix in dense multipath environment is solved.In this paper,the effectiveness and performance of the new method are verified by both computer simulation and field measurement experiments.Compared with the traditional SAGE algorithm and the SAGE algorithm based on the Autoregressive(AR)process(AR-SAGE),the results all show that the proposed method in this paper can estimate multipath parameters more accurately and effectively than the traditional method in dense multipath scenarios.(2)Aiming at the demand for wireless channel characteristics and models for the testing of VANET technologies in typical traffic scenarios,large-scale and small-scale channel fading characteristics in tunnel and highway scenarios are analyzed,and wireless channel models for Vehicle-to-Vehicle(V2V)communication are established.Firstly,based on the high-precision wireless channel measurement equipment,the 5.2 GHz wireless channel were measured in the tunnel and highway scenarios.Based on the measured channel data,the models of path loss,shadow fading,distribution of received signal amplitude,Rice K factor,Root Mean SquareDelay Spread(RMS-DS),and Tapped Delay Lines(TDL)were developed.In addition,a physical optics-based channel model of the tunnel scenarios is developed considering the direct component,the reflected and scattered components from the tunnel walls,and the reflected components from the surrounding vehicles.The validity of the model proposed in this paper is verified by the measured data.The experimental results show that the proposed model can accurately describe the variation of received power during V2 V communication in the tunnel environment.(3)Aiming at the problem that the existing connectivity models for VANET rarely consider the effects of traffic flow and small-scale fading on connectivity,the connectivity probabilities between any two vehicles under Weibull and Nakagami-m fading channels are derived,and the connectivity models for VANET under different fading are established.The effects of largescale fading parameters,Weibull fading parameters,Nakagami-m fading parameters,vehicle density,transmit power,Signal to Noise Ratio(SNR)threshold,and adjacent vehicle order on the connectivity of VANET are analyzed.(4)Aiming at the problem that traditional satellite navigation positioning methods are difficult to achieve accurate vehicle positioning in tunnels,a single base station vehicle positioning method for the rectangular tunnel scenario is proposed.Based on the actual structural characteristics of the tunnel,virtual base stations are established.According to the virtual base station technology,the single reflection path from the tunnel walls can be transformed into the direct path from the virtual base station,which simplifies the complexity of localization.When the propagation between the base station and the target vehicle is Lineof-Sight(LOS)propagation,the location of the target vehicle is estimated using Two Step Weighted Least Squares(TSWLS)by combining the Time-of-Arrival(TOA)information of the LOS path and the two paths from the virtual base stations.When the propagation between the base station and the target vehicle is Non Line-of-Sight(NLOS)propagation,the location of the target vehicle is estimated using Unscented Particle Filter(UPF)algorithm based on the TOA and Direction-of-Arrival(DOA)information of the propagation paths from the two virtual base stations.The performance of the algorithm is verified by simulation.The simulation results show that the proposed positioning algorithm in this paper can effectively estimate the location of the target vehicle and has high positioning accuracy.In summary,this paper studies the channel parameter estimation algorithm,channel model,VANET connectivity and vehicle localization methods,etc.It provides a new algorithm for channel parameter estimation in dense multipath environment,a new idea for channel modeling in tunnel scenario,a new model for VANET connectivity analysis,and a new method for vehicle localization in tunnel.This paper can provide a scientific basis for the design and optimization of the VANET system.