Research On Channel Estimation Based On LTE-V2X For Internet Of Vehiclies

Intelligent transportation system(ITS)is the core of future transportation development,and it is also the foundation to support unmanned and intelligent transportation such as vehicle-road coordination,autonomous driving,and vehicle formation.In-vehicle wireless communication under the Internet of vehicles is an important part of the efficient operation of ITS and a key technology to ensure traffic safety and solve urban congestion problems.In order to ensure quality of service(Qo S)for the Internet of Vehicles communication,channel estimation and channel equalization processing are needed on the wireless receiver side to combat the problem of channel fading and noise interference under vehicle-to-everything(V2X)communication.Long term evolution(LTE)-V2 X uses Zadoff-Chu(ZC)pilot sequences.In the V2 X scenario,the wireless channel will be subject to noise interference,and time/frequency domain selective fading caused by multipath and Doppler effects.In order to solve the problems of noise,non-stationary,fast time-varying and inter-carrier interference(ICI)in V2 X channels,this paper designs improved channel estimation algorithms on pilot and data symbols on the basis of traditional channel estimation algorithms.And the proposed algorithms achieve certain performance gain.The main work is as follows.Firstly,for the V2 X wireless channel,the time/frequency selective fading and nonstationary characteristics are analyzed,and the time-varying multipath channel model and the channel model considering ICI are given.Several traditional channel estimation algorithms on pilot symbols and data symbols are introduced respectively.Through simulation analysis,the advantages and disadvantages of traditional channel estimation algorithms are compared,and the shortcomings of traditional channel estimation algorithms are pointed out.Then,aiming at the problem that the traditional discrete fourier transform(DFT)channel estimation algorithm cannot eliminate the time-domain noise in the cyclic prefix(CP),an improved DFT channel estimation algorithm based on the ZC sequence is proposed.The proposed algorithm uses the autocorrelation characteristics of the ZC sequence to estimate the channel multipath delay by setting the detection threshold.And the multipath delay is used in the DFT algorithm to suppress the time domain noise in the CP.The computational complexity and system simulation results show that the proposed algorithm can effectively improve the channel estimation performance on pilot symbols by setting a reasonable multipath delay detection threshold under low computational complexity.What’s more,aiming at the problem of time-varying channel tracking on data symbols,a channel estimation method based on joint iterative decision feedback and time-varying channel impulse response(CIR)fitting is proposed.Using the principle of constellation point mapping of the transmission signal,the transmission signal is piloted for feedback channel estimation by the decision feedback channel estimation method,and the joint iterative processing method solves the problem of fast time-varying and nonstationary channel.Furthermore,the time-varying CIR fitting is performed by polynomial interpolation to obtain the complete CIR in each symbol and eliminate the influence of ICI.Simulation results show that compared with traditional channel interpolation and decision feedback algorithms,the proposed algorithm not only has better estiamtion accuracy,but also has better algorithm robustness under different Doppler frequency shifts.