Research On Millimeter-wave Beamforming For Downlink Multiuser Transmission

With the development of electronic information technology,more applications that are new appear in the daily work and life of users.In the future,mobile communication network will also face unprecedented challenges.Hybrid beamforming,as one of the key technologies of millimeter wave(mmWave)massive multiple-input multiple-output(MIMO)technology,can effectively improve the overall spectrum efficiency of the system and reduce energy consumption.In recent years,the research on intelligent reflecting surfaces(IRSs)has also become a hot spot.The IRSs can receive the input signal from the base station in real time and reflect the received signal to the users.Therefore,through the joint design of beamforming at the IRSs and the base station,the signal power received by the remote user is enhanced and the coverage of mmWave signal is improved,which is also the key technology in the mmWave massive MIMO systems.This paper first introduces the mmWave massive MIMO and its advantages,as well as the research status of hybrid beamforming,and then focuses on the research of hybrid beamforming algorithm with high spectral efficiency and low computational complexity in mmWave massive MIMO systems,which includes the following two aspects:1.This paper focuses on the mmWave downlink multiuser MIMO system which supports multi data stream transmission per user.In the above scenario,aiming at the problem of strong correlation between users’ channels,this paper proposes a hybrid beamforming algorithm based on channel correlation cancelation.From the perspective of channel correlation,the overall spectral efficiency of the system is decomposed into a series of sub rate optimization problems,and the correlation terms are extracted from the sub problems for optimization.Finally,block diagonalization are used in the equivalent channel to eliminate the interference between users and data streams.Simulation results show that the performance of the algorithm is better than the existing algorithms,especially when there is a high correlation between user channels.2.In mmWave communication systems,the existence of channel occlusion affects the quality of wireless communication.IRSs assisted mmWave communication can effectively solve this problem.Due to the difficulty in joint design of hybrid beamforming at base station and reflection beamforming at IRSs,this paper proposes a multi-user beamforming neural network(MBNN),which can capture the features of signals.MBNN has the advantages of adapting to the changes of channel environment and low computational complexity.Simulation results show that the performance of the algorithm is close to the upper bound of spectral efficiency.