Research On Channel Estimation Algorithm For Millimeter Wave Massive MIMO System

Massive Multiple Input Multiple Output(MIMO)system is extensively studied as one of the most promising technologies for the fifth generation communication system,since the system employed with a considerable number of antennas has the potential to obtain larger array gain than conventional MIMO system.Because of the abundant spectral bandwidth and higher data rates,millimeter wave(mm Wave)communication has also become one of the most enabling technologies in future wireless communication system.By taking advantage of both technologies,Massive MIMO at mm Wave can provide great beamforming gain to compensate the propagation loss caused by high-frequency mm Wave.In mm Wave MIMO systems,accurate estimation of channel state information(CSI)is essential for the design of hybrid beamforming processors.However,due to the considerable number of antennas,channel estimation algorithm for conventional MIMO with high computational complexity is infeasible in mm Wave system.This thesis uses the sparse characteristics of mm Wave channel to design a channel estimation algorithm based on compressive sense(CS)for mm Wave MIIMO system.The thesis firstly studies the channel estimation algorithm for mm Wave MIMO system based on block sparsity adaptive CS.Based on the non-zero elements distribution of the mm Wave channel in the angular domain,a Generalized Block Sparsity Adaptive Matching Pursuit(G-BSAMP)algorithm is designed.This algorithm can estimate the CSI when the number of scattering paths in the channel is unknown.Simulation results show that G-BSAMP algorithm proposed in this thesis can effectively estimate the CSI of mm Wave MIMO system without knowing the number of effective paths in the mm Wave channel.Then,the thesis studies the channel estimation algorithm for mm Wave MIMO system based on CS under continuous frames.Because rate of change of the channel in the indoor low-speed mm Wave wireless communication scenario is relatively low,the channel vectors in adjacent frames have the similar sparse characteristics.A channel estimation algorithm in continuous frames based on Time Domain Sparsity Adaptive Matching Pursuit(TD-SAMP)is proposed.In this channel estimation algorithm,the system uses the CSI estimated in the previous frame as the a priori information for the channel estimation in the current frame,to realize the channel estimation for the mm Wave system in the continuous frames.Simulation results show that the TD-SAMP algorithm can effectively estimate the CSI of mm Wave channel in continuous frames.Finally,the thesis studies a channel estimation algorithm for the multi-panel mm Wave MIMO system.In multi-panel mm Wave MIMO system,the interval between the panels is negligible,so the angles of departure(Ao Ds)between different antenna panels are approximately the same.Based on this structural feature,Generalized Joint Orthogonal Matching Tracking(G-JOMP)algorithm and Optimized Joint Orthogonal Matching Tracking(O-JOMP)algorithm is designed.Simulation results show that the G-JOMP algorithm and O-JOMP algorithm can effectively estimate the CSI of multipanel mm Wave MIMO systems.