Channel Estimation Methods Based On Adaptive Compressed Sensing For Mm Wave Massive MIMO Systems

Millimeter-wave massive MIMO technology has been widely considered as one of the key technologies for 5G for that the combination of millimeter-wave technology and massive MIMO can achieve the dramatic improvements in capacity and spectral efficiency.In order to take advantage of the millimeter-wave massive MIMO technology,it is necessary to effectively estimate the channel.But channel estimation is challenging for millimeter-wave massive MIMO systems,because the number of antennas is much larger than the number of radio frequency chains and the signal-to-noise ratio(SNR)is low before beamforming.Thus,it is necessary to design efficient millimeter-wave MIMO channel estimation methods.Traditional channel estimation methods,such as least-squares algorithm and minimum mean square error algorithm,are not suitable for millimeter-wave massive MIMO systems due to excessive pilot overhead.Because millimeter wave massive MIMO channel has sparse characteristics in angle domain,the channel estimation problem can be modeled as a sparse signal recovery problem,and then,the problem of channel estimation can be solved by compressed sensing algorithms.Moreover,the channel estimation method based on adaptive compressed sensing can achieve better performance in low SNR.This paper focuses on the design of efficient channel estimation methods based on adaptive compressed sensing.First,this paper compares the performance of the channel estimation method based on adaptive compressed sensing and that based on compressed sensing.Secondly,in order to solve the problem that the adaptive compressed sensing algorithm is complex when the codebook space is large,this paper proposes a codebook design method based on iterative grid for adaptive compressed sensing.Finally,this paper proposes a criterion of minimizing the average cross-correlation value of the sensing matrix,and optimizes the antenna distribution to reduce the average cross-correlation value of the sensing matrix.The simulation results show that the adaptive compressed sensing algorithm can achieve better channel estimation performance at low SNR.Furthermore,the proposed codebook optimization algorithm and optimized antenna distribution method can significantly reduce the complexity of the adaptive compressed sensing algorithm and improve the performance of channel estimation.