Research On Channel Estimation In Internet Of Vehicles Scenario

Nowadays,the intelligent transportation system based on the Internet of Vehicles(IOV)has drawn wide attention.In the scenario of IOV,the communication receiver is fast-moving,complex in channel and less demanding in transmission delay.To ensure the quality of good communication,it is of critical importance to select the appropriate channel model and estimation method to satisfy the requirements of IOV.Therefore,the present study focuses on the estimation model and method of fast time-varying channel in IOV system.It begins with a comparison of the physical layer technologies between LTE-V2 X and the Dedicated Short Range Communication(DSRC).Then,the characteristics of channel fading in the scenario of IOV are involved followed by a IOV channel model based on the MIMO-OFDM.With a detailed analysis of the Linear Model(LM)and the Basis Expansion Model(BEM)in the fast time-varying channel model,this study has found that the BEM has more advantages in the scenario of IOV.In addition,this study also concerns a simulated analysis of the performance of different basis functions.Considering the sparsity of BEM based on MIMO-OFDM,this study further proposes a SAMP restricting algorithm that is variable-step by investigating the adaptive matching pursuit restricting algorithm with unknown sparsity in compressed sensing.With the generalized Dice method to replace the inner product calculation,this algorithm measures the similarity of atoms.Besides,it also improves the speed of approximation by setting the step size in sections.The results of the simulation show that with the increase of sparsity,the mean square error(MSE)is minimum when the restricting algorithm is used to recover the sparse signal and the MSE performance is also minimum when the channel has Doppler frequency shift.Finally,this study extends the BEM sparse model of single antenna channel to the MIMO-OFDM system,which is used to estimate the channel of IOV combined with the improved SAMP restricting algorithm.Results of the simulation comparison show that there is an apparent improvement in the performance of the estimation method when different shifts are involved.