The Research Of Hybrid Beamforming Technology Based On Digital-Analog Hybrid Architecture

With the emerging technologies such as virtual reality and autonomous driving,along with the increase in traffic of mobile communication data services,higher requirements are also placed on the transmission rate of wireless communication systems.The future 5G communication system aims to provide users with Gbit/s transmission rate.However,the conventional low-frequency band cannot provide sufficient transmission bandwidth,and the high-frequency millimeter band has gradually become a research hotspot in academic circles.Massive Multiple-Input Multiple-Output(MIMO)technology,as one of the key technologies of 5G mobile communication,can improve the spectrum efficiency of the system by configuring a large-scale antenna array at the base station and serving multiple users.Combining millimeter wave communication with massive MIMO systems is a key direction in the research of 5G communication technology.This thesis first introduces the channel model of millimeter-wave,the conventional MIMO system which based on millimeter-wave channel includes an full-digital beamforming system and a full-analog beamforming system.Since the base station side needs to use as many radio frequency(RF)chains as the antennas in the all-digital beamforming architecture,this solution is only applicable to small-scale antenna architectures.The lack of amplitude control on a full-analog beamforming array leads to much worse performance than an full-digital beamforming system.Therefore,the hybrid beamforming system is more suitable for the millimeter wave MIMO system,which can improve the spectrum efficiency while reducing the system complexity.The conventional full-digital based beamforming methods for MIMO systems are introduced and analyzed in Chapter 3,including Zero-Forcing(ZF)beamforming,minimum mean square error(MMSE)beamforming,block Block Diagonalization(BD)beamforming,Signal to Leakage and Noise Ratio(SLNR)beamforming,and their performance are compared horizontally,then the advantages and limitations of these algorithms are analyzed.In Chapter 4,the hybrid beamforming algorithms for single-user scenario are introduced.And then,an improved algorithm for hybrid beamforming based on DFT codebooks is proposed with orthogonal matching pursuit criterion,the proposed scheme shows its advantage in computational complexity reduction.This thesis also improves the conventional phase extraction based hybrid beamforming,balancing its performance and complexiy.Chapter 5 first introduces the hybrid-beamforming technique in multi-user scenarios.And then,two improved hybrid beamforming algorithms for multi-user are presented.One is based on conjugate transposion,another one is based on singular value decomposion,both of which optimize the precoding matrix for the tranceiver jointly.The proposed schemes could maximize the dianonal elements of the equivalent channel matrix for digital beamforming and obtain the analog precoding matrix by simply conjugate transposition operation.Compared with conventional schemes,the proposed algorithmes achieves more precoding gain for the equivalent channel and provide higher achievable rates.