Research On Single User Hybrid Precoding For Millimeter Wave Massive MIMO System

During these years,with the rapid development of 5G and future mobile communication technologies,higher transmission rate,higher delay resolution and higher spatial resolution,millimeter wave communication has gradually been the critical technology to increase data rate in future communication systems.As one of the key technical methods to improve performance and reduce interference in terms of millimeter-wave massive MIMO systems,precoding technology is trending to be a research hotspot.Firstly,the background of mm Wave Massive MIMO system is introduced by this work.The prior work has proposed a precoding algorithm which combines orthogonal matching pursuit(OMP)algorithm and bird swarm algorithm(BSA).Derived from this,this work substitutes BSA with Quantum genetic algorithm(QGA)as the core method.The characteristic of QGA is to search and analyze global optimal value rapidly,it can search the most satisfactory array response vector obtained by multiplying with the residual matrix,so that it can estimate candidate matrix with respect of OMP algorithm precisely.This scheme is consistent with the BSA algorithm,however,in the process of selecting initial population size,the scale required by QGA is much smaller than that of BSA.The simulation results show that the system spectrum efficiency of the algorithm compared with the existing OMP algorithm is improved.Secondly,in order to reduce the system complexity of the hybrid precoding technology,a low-complexity hybrid precoding and combination scheme based on Array Response Vectors(ARV)is improved by this work.In order to maximize the spectrum efficiency in the hybrid precoding and combination scheme based on the array response vector,the decoupling optimization between the analog and baseband domains is applied,and the incremental continuous selection of the array response vector is used to generate the radio frequency(RF)precoder and the combination matrix,after that,it utilizes the SVD decomposition of the equivalent channel to obtain the digital precoding matrix and the digital combination matrix.Due to that the scale of equivalent channel dimension is low,the complexity of matrix and the spectrum efficiency of the system is low.In terms of improved algorithm,the digital precoding matrix and the digital combination matrix are solved by applying the least squares rule column by column,and the digital module composition scheme of the transceiver is given.By the comparison,the spectral efficiency of the system is stronger than the mixed precoding based on the array response vector.The coding and combination scheme is slightly more complicated.Finally,through simulation of traditional OMP algorithm,ARV algorithm,optimal hybrid precoding and improved hybrid precoding algorithm,the effectiveness of the scheme proposed by this work is proved.