Research And Design Of Hybrid Precoding Schemes In Mm Wave Massive MIMO Systems With RF Distortion

With the massive access of wireless terminals,wireless communication technology faces the challenge of lack of spectral resources.Due to its abundant spectral resources,millimeter wave(mm Wave)has become the preferred band for the current 5-th generation mobile networks(5G)and even the next generation mobile networks.The short-wavelength characteristic of mm Wave greatly increase the number of antennas that can be deployed in a specific spatial range,so as to integrate massive MIMO(Multiple-Input Multiple-Output)system.In addition,the large gain brought by dual-polarized antenna arrays can be used to compensate for the harsh characteristics of the transmission environment of mm Wave channel.Therefore,the mm Wave massive MIMO system combining mm Wave and massive MIMO technology has become a research focus of 5G.In order to reduce the cost and power consumption of communication transceivers,the radio frequency(RF)front-end architecture with one RF chain connecting multiple antennas is adopted by mm Wave massive MIMO system,so the hybrid precoding technology for this new architecture has been widely studied.However,at the same time,there are constraints in the manufacturing process,technical standards and cost of mm Wave RF components,which lead to the problem of RF distortion in the transceivers of the mm Wave massive MIMO system that cannot be ignored.For this problem,this thesis studies the design of the hybrid precoding scheme under the influence of RF distortion in the transmitter/receiver of the mm Wave massive MIMO system.The main work is as follows:Firstly,this thesis focuses on the problem of in-phase and quadrature-phase(IQ)imbalance at the transmitter in the downlink,and designs a hybrid precoding scheme based on a widely-linear real-valued signal model to suppress the impact of IQ imbalance on mm Wave massive MIMO system performance.Complex-real transformation is utilized to transform the received signal model containing conjugate interference into a widely linear real-valued signal model.Based on the orthogonal matching pursuit algorithm,the analog precoding matrix and digital precoding matrix are designed iteratively in complex and real domains respectively.The simulation results show that the proposed scheme can effectively improve spectral efficiency and bit error rate performance.Secondly,this thesis investigates the adverse impact of residual hardware impairments on the system performance of the transceivers of the mm Wave massive MIMO system.In order to minimize the mean square error between the received signal and transmitted signal,a hybrid precoding/combining scheme based on manifold optimization is proposed.First,the closed-form solutions of the digital precoding and digital combining matrices of the transceivers are derived based on the optimization problem respectively.Then the optimal solutions of the analog precoding matrix and combining matrices are obtained by dealing with the constant modulus constraint on the Riemannian manifold respectively.The simulation results show that the proposed scheme effectively suppresses the impact of residual hardware impairments on the millimeter-wave massive MIMO system,and significantly improves the performance of the system.