Massive MIMO-OFDM Channel Estimation Algorithm Based On Compressed Sensing

As the next-generation wireless communication system,5G mainly utilizes a Massive Multiple-Input Multiple-Output technology in which a large number of antennas are arranged at the receiving and transmitting ends of the communication,thereby effectively utilizing the spatial freedom to realize the double increase of energy efficiency and spectrum efficiency of information transmission.Accurate acquisition of channel state information is a prerequisite for achieving high performance of a Massive MIMO system.However,a pilot-based linear channel estimation method requires that the length of the pilot sequence is proportional to the number of antennas,the number of antennas increases,and the pilot overhead increases.In this paper,aiming at the problem of large pilot overhead of the linear channel estimation algorithm in Massive MIMO-OFDM systems,a channel estimation algorithm with reduced pilot overhead is proposed.This paper first studies the principle of channel estimation algorithms such as Least Square,Minimum Mean Square Error,and Scaled Least Squares in MIMO systems.By giving the mean square error of various algorithms,the optimal pilot structure corresponding to the three algorithms is deduced.According to the error function,the effect of the number of antennas on the channel estimation pilot overhead is explained.Secondly,on the basis of illustrating the common sparseness of the Massive MIMO-OFDM time-domain channel,the channel estimation is performed using the recovery algorithm in Compressed Sensing technology.Firstly,the channel estimation is performed using the Orthogonal Matching Pursuit algorithm and the Subspace Pursuit algorithm in CS.These two algorithms need to know the channel sparsity in advance,but it is difficult to achieve in practical situations.To solve this problem,Adaptive Block Subspace Pursuit algorithm,which can adaptively acquire channel sparsity,is proposed.The simulation results show that the CS-based channel estimation algorithm can solve the problem of large pilot overhead.At the same time,the proposed ABSP algorithm is less than the OMP algorithm and SP algorithm in the channel estimation error.Finally,the beam-domain channel of Massive MIMO-OFDM system in the far-field scenario is also sparse.According to this property,CS is used to perform channel estimation on the beam domain.Regularization Orthogonal Matching Pursuit algorithm is used to estimate the channel.Based on this algorithm,an Adaptive Regularized Subspace Pursuit algorithm that can estimate the channel with unknown sparsity is proposed.The simulation results show that the performance of the proposed ARSP algorithm is better than the ROMP algorithm and the pilot overhead is reduced compared to the LS algorithm.