Research On Performance Of Massive MIMO Beam Selection Algorithm Based On NOMA

Millimeter-wave Massive MIMO,which has emerged in recent years,places high requirements on the power consumption of the transmitter hardware.Therefore,industry and academia have turned their attention to the study of beam selection algorithms in the field of beamspace.In addition,considering that non-orthogonal multiple access technology has broad application prospects in future cellular mobile networks,it can solve the problem that the number of users in the beamspace is greater than the number of radio frequency links.In this paper,NOMA and beam selection algorithm are regarded as the focus.Firstly,the traditional MIMO multi-path channel model is given,and a system model of beamspace is transformed by a matrix of orthogonal bases.In this paper,the widely used S-V channel model is used,and the channel matrix converted to beamspace has sparseness.Since the NOMA solves the specific problems in the beamspace model,the serial interference cancellation technology and power allocation algorithm in NOMA are introduced.The performance problems such as spectrum efficiency and energy efficiency of the NOMA-based beam selection algorithm are studied.Since the channel matrix in the beam space is sparse,different beam selection algorithms can be selected to reduce the transmission link and increase the energy efficiency while ensuring that the communication quality is hardly affected.Based on this,the application of NOMA solves the problem of more users than the number of beams.firstly,the channel matrix is transformed into the beamspace,and the dimension of the channel matrix is reduced by the beam selection algorithm.Then the user is clustered according to the amplitude of the channel vector component,and the improved zero-forcing precoding based on singular value decomposition is used to derive the signal expression received by the user.The serial interference cancellation technique and distributed power allocation algorithm are then used to maximize the system and rate while ensuring user fairness.Combining NOMA with different beam selection algorithms to analyze the spectrum efficiency and energy efficiency of different combinations.Finally,an improved user pairing method is studied to compare the method of clustering users according to the magnitude of channel vector components and discuss whether it can improve system performance when different beam selection algorithms are applied.After the beam selection algorithm is completed,the clustering is performed according to the correlation coefficient and gain difference between the user channel vectors,and whether the performance indicators such as spectrum efficiency are improved by the simulation.In addition,based on this model,the improved beam selection algorithm is studied.Before the beam selection,the user clusters according to the amplitude of the channel vector component,and then performs precoding and beam selection based on the channel matrix formed by strong user channel vectors in the cluster.Since the dimensions of the matrix used to perform the beam selection will be reduced,the total number of beams eventually selected will also be reduced,and the energy efficiency will also change with this.This article will simulate and analyze the performance of the improved algorithm and provide a reasonable analysis through the simulation curve and provide a reasonable algorithm solution for the actual communication system according to different needs.