Sparse Channel Estimation Algorithms In Massive MIMO System

Massive MIMO system is considered as one of the key technologies in 5G mobile communication system,it can effectively improve data transmission rate and link reliability by deploying a large number of antennas at the base station end.In order to maximize the theoretical performance advantages of Massive MIMO system,accurate channel state information is critical.Considering the spatial common sparsity of virtual angle domain in Massive MIMO system,Particle Swarm Optimization-Stagewise Orthogonal Matching Pursuit algorithm is proposed in this paper,by introducing Particle Swarm Optimization into StOMP algorithm,realize adaptive search of iteration threshold in StOMP algorithm,when the sparsity of channel matrix changes,it can achieve adaptive channel estimation.According to the space-time common sparsity of time-delay domain in Massive MIMO system,we can acquire Channel State Information in a joint space-time system based on structured compressed sensing.Specifically,we firstly transmit non-orthogonal pilot signals at the base station ends,then,according to the structured sparsity of Massive MIMO channels,an Adaptive Structured Subspace Pursuit algorithm based on Dice coefficients is proposed at the user side,it estimates the channel impulse response associated with multiple OFDM symbols on the user side,Dice coefficients take replace of the inner product operation between vectors in the algorithm,this algorithm improves the accuracy of atomic recognition and screening.Thus,the estimation accuracy of Massive MIMO channel matrices in time-delay domain is improved,and the pilot overhead of the system is reduced..