Research On Channel Estimation Algorithms Based On Compressed Sensing In Massive MIMO Systems

Massive Multiple-Input Multiple-Output(Massive MIMO)is considered as the core technology of 5G mobile communication because it is equipped with large-scale antenna array,makes full use of diversity technology and spatial multiplexing technology,also can significantly improve data transmission efficiency without consuming additional spectrum resources.Whether the gain of Massive MIMO system can be used effectively depends on the accuracy of channel state information(CSI),so the quality of channel estimation is a hot research fields of Massive MIMO system.With the increasing number of antennas at the receiver and transmitter,the problems of pilot overhead and algorithm complexity bring severe challenges to channel estimation for Massive MIMO systems.Compressed sensing(CS)technology provides an effective way.The reconstruction algorithm can be used to estimate the CSI by utilizing the sparsity of channel matrix in the transform domain.In order to further understand the above problems,this paper studies the system pilot pollution and data transmission channel characteristics,and the work contents are as follows:Channel estimation for multi-cell multi-user Massive MIMO systems,this paper establishes a multi-cell channel model,studies the channel matrix parameters and joint sparsity under pilot pollution,proposes an adaptive dual threshold channel estimation algorithm.The idea of Proportion-Integral-Derivative(PID)is introduced into the piece-wise matching orthogonal pursuit algorithm.In each iteration,the threshold of selecting atoms can be adaptively adjusted according to the current reconstruction error,and multiple atoms with strong correlation with the original signal can be selected efficiently at one time.In addition,considering the residual distribution characteristics,the iteration stop threshold is dynamically modified by calculating the residual relationship of two adjacent iterations,which can more truly reflect the CSI when the prior conditions are unknown.Simulation results show that this algorithm can improve the accuracy of reconstructed signal by adaptive searching for the best iteration threshold and stop threshold when the number of users changes.Channel estimation for high-mobility scenarios,this paper considers an OTFS frame under integer Doppler condition,analyzes the sparsity of channel in the delay-Doppler domain,and proposes a dynamic variable step size channel estimation algorithm.By calculating the change rate of residual energy in two iterations,the algorithm dynamically adjusts the step size according to the different state of the current stage,so as to avoid over estimation and improve the estimation accuracy.Simulation results show that the proposed algorithm can achieve accurate channel state information with low pilot overhead.