Research On The Acquisition Of Channel State Information Based On Massive MIMO System Of High Speed Train

With the development of high-speed train,people have put forward higher requirements for the quality of service in high mobility scenarios,which requires accurate Channel State Information(CSI).Massive multiple-input multiple-output(MIMO)technology is considered to be the main solution due to its advantage of significantly improving the channel capacity and spectral efficiency as well as the potential to reduce the influence of Doppler effect.However,the significant Doppler frequency shift and the large antenna array still seriously restrict the CSI accuracy.In order to overcome the challenges brought by high velocity,two channel estimation algorithms suitable for high mobility scenarios are proposed from the perspectives of time-frequency(TF)domain and doppler-delay(DD)domain.The theoretical analysis of convergence are given and a large number of simulations have verified their performance.The proposed algorithm has accurate channel estimation accuracy in various communication environments and can adapt to maneuverability scenarios at different velocities.The main contents of this paper are as follows:(1)From the perspective of TF domain,a spatial information-assisted joint uplink and downlink channel parameter tracking scheme is proposed for the fast time-varying channel.Firstly,a sparse model of received signal is constructed,each column of the observation matrix corresponds to a channel path.Secondly,the real-time CSI acquisition problem is transformed into a channel parameter tracking problem in Sparse Bayesian Learning(SBL)form,and each channel parameter was iteratively solved by adapting MM algorithm.Finally,a significant gradient based fast searching scheme is proposed to solve the multi-peak optimization problem of angle surrogate functions.(2)From the perspective of DD domain,a joint channel estimation and signal detection scheme is proposed based on the DD domain channel which has a cyclic matrix form.Firstly,the Orthogonal Time Frequency Space(OTFS)is adopted to modulate the channel to the DD domain and establish the input-output signal relationship of the DD domain.Secondly,the immune algorithm is used to solve the initial channel estimation problem and a signal detection scheme based on Message Passing(MP)is proposed.Finally,based on the cyclic matrix characteristic of DD domain channel,a low-complexity joint optimization scheme of detection signal and channel estimation is proposed.