Mass MIMO Channel Estimation Based On Compressed Sensing

With the rapid development of virtual reality technology,driverless technology,Internet of things,high-definition video,telemedicine and other technologies,4G network can not meet people's needs in communication capacity and transmission speed.Therefore,5g communication technology came into being.In the process of research and development of 5G technology,massive MIMO technology has become a key technology in 5G communication technology because of its large number of antennas and the characteristics of spatial beam multiplexing,which can greatly improve the capacity of communication system.In order to realize the huge advantages of massive MIMO,it is very important to obtain the channel state information(CSI)accurately.However,due to the large number of antennas in the massive MIMO system,the channel dimension increases dramatically.The traditional channel estimation method will generate huge pilot overhead and increase the computational complexity.Therefore,it is necessary to explore a new and effective channel estimation scheme.In recent years,the theory of compressed sensing has been widely used in many fields,such as signal processing and so on,because of its accurate reconstruction performance for sparse signal compression.Considering the characteristics of signal sparsity,the theory of compressed sensing provides a new scheme for massive MIMO channel estimation.In this paper,we mainly do the following work for the research of massive MIMO channel estimation based on compression sensingFirstly,this paper studies the theory of compressed sensing,compressed sensing algorithm,and analyzes the shortcomings of traditional channel estimation algorithm.By studying the compressed sensing algorithm,aiming at the unknown signal sparsity and the influence of different fixed step size on the reconstruction accuracy of the compressed sensing algorithm,and combining the advantages of similar algorithms,a massive MIMO channel estimation scheme based on the improved variable step size regular backtracking SAMP algorithm(YSAMP)is proposed.The algorithm does not need to take the sparsity of the signal as a known condition,and can adaptively adjust the step size according to the results of each iteration of the algorithm,so as to ensure the relatively high reconstruction accuracy under a relatively low number of iterations.The simulation results show that the algorithm has advantages in massive MIMO channel estimation.Secondly,on the basis of the above research,considering the influence of shadow fading,large-scale fading and noise on the transmission signal in the process of massive MIMO mobile communication,the signal is seriously distorted,which reduces the accuracy of channel estimation.In addition,the existing compression sensing algorithm does not effectively reduce the noise of the signal.In this paper,the noise reduction backtracking SAMP algorithm(NrSAMP)is proposed.In the multi cell,multi-user massive MIMO channel scenario,this algorithm is used to estimate the channel,and compared with other improved algorithms in reconstruction accuracy,which proves that this algorithm can obtain better accuracy under the same conditions.Therefore,the algorithm has practical significance.