Research On Hybrid Beamforming Technology In Millimeter Wave Multi-user MIMO System

The short wavelength of millimeter wave makes it possible to arrange large scale array antennas in a very small area,achieving high beamforming gain and making up for the serious path loss of millimeter wave transmission.So massive MIMO and beamforming technologies are the key technologies in millimeter wave communication.In millimeter wave massive MIMO system,analog beamforming can only support the transmission of single data stream.It is simple to implement,but the system spectrum efficiency is too low.If full digital precoding is used,although multi-stream can be transmitted and the best performance can be obtained theoretically,it will face the problems of high power consumption and high complexity.The hybrid beamforming technology in millimeter wave massive MIMO system combines the advantages of analog beamforming and digital precoding,which achieves a good balance of complexity and performance.In this paper,the hybrid beamforming technology in millimeter wave multiuser MIMO system is studied.The research content and work achievements are as follows:1.The full digital precoding scheme based on SVD decomposition in millimeter wave single user system is studied.It is considered as the theoretical bound for the performance comparison of hybrid beamforming.Based on this,the hybrid beamforming scheme based on OMP is studied.By utilizing the sparsity of the channel,the analog beamformer and the analog combiner are constructed with the array response of the antenna to achieve a good performance close to the full digital precoding.2.In the millimeter wave multi-user MIMO system,when the transmitter is a hybrid beamforming architecture and the receiver of each user is a multi-antenna single data stream transmission,for the problem that the PZF hybrid beamforming algorithm cannot achieve the optimal system performance in baseband precoding processing,this paper proposes a hybrid beamforming algorithm based on particle swarm optimization.The algorithm takes the expression of each user's SINR as the objective function to search the optimal baseband precoding vector,which can achieve higher spectral efficiency and better bit error rate.3.In the millimeter wave multi-user MIMO system,when the transceivers are both hybrid beamforming architecture and multi-stream transmission,this paper first introduces a hybrid beamforming algorithm based on block diagonalization,which can effectively eliminate the multi-user interference and transmit multi-user data with multi-stream.In order to improve the precoding gain of this algorithm,this paper proposes a hybrid beamforming algorithm based on successive optimization of SLNR.This algorithm continuously optimizes the interference and leakage matrix of each user,and gradually reduces the power leaked to other users.By maximizing the SLNR of each user,the multi-user interference is suppressed,and the spectral efficiency and bit error rate performance of the system are improved.