Millimeter Wave Channel Estimation Based On Compressed Sensing

As one of the key technologies of 5G communication technology,millimeter wave technology can improve the system capacity of communication system and relieve the pressure of spectrum resources.It is very important to make full use of the advantages of millimeter wave spectrum resources and accurately obtain Channel State Information(CSI)of millimeter wave system.Because of the unique fading characteristics of millimeter wave,it is necessary to study a new channel estimation method.In this paper,the millimeter wave system and its channel estimation algorithm are studied,and the further details are as follows:(1)Study the transmission characteristics of millimeter wave and find the transceiver structure suitable for its system,and establish the mathematical model of millimeter wave system and millimeter wave sparse channel model.(2)Study the millimeter wave sparse channel estimation and compressed sensing sparse reconstruction,and apply compressed sensing theory to channel estimation.By simulation comparison,it is found that compressed sensing algorithm is more suitable for high dimensional sparse channel estimation of millimeter wave system.(3)Propose feedback re-selection algorithm for compressed sensing channel estimation.To solve the problem that the reconstruction speed and reconstruction precision of similar algorithms cannot be both taken into account,the proposed Ts-StOMP algorithm introduces the feedback idea to improve the estimation accuracy and uses Sigmoid function to establish a nonlinear relationship between the simulation times and the number of selected atoms,by which the goal of fast approximation of large step length and accurate approximation of small step length.The simulation results show that the Ts-StOMP algorithm improves the estimation accuracy and estimation speed.(4)Propose a sparse adaptive compressed sensing channel estimation algorithm.In terms of the defect of the lack of adaptive ability of the proposed Ts-StOMP algorithm in millimeter wave channel estimation with different sparsity,the optimization ability of the PSO algorithm is utilized to input the corresponding optimal threshold parameters at different sparsity,thus improving the estimation performance of the algorithm.Simulation results show that the proposed algorithm can adapt the sparsity and improve the estimation performance.