Research On Cooperative MIMO Precoding Algorithm For Millimeter Wave System

Millimeter-wave communication is one of the key technologies in 5G communication system,which has abundant spectrum resources and can meet the needs of large bandwidth,low delay and high speed in future mobile communication.The short wavelength of millimeter wave leads to serious path loss in the process of propagation.However,the path loss of millimeter wave in space propagation can be compensated by the gain of large antenna arrays.Cooperative multi-point transmission technology can reduce inter-cell interference and improve system throughput.In traditional microwave cooperative communication system,all-digital precoding technology is used to achieve cooperative multi-point transmission.Full-digital precoding requires a radio frequency link for each antenna.This will bring high hardware cost and power overhead in massive antenna system,and is not suitable for the millimeter wave communication system.Therefore,the research of cooperative MIMO precoding algorithm for millimeter wave communication system is very important.The cooperative MIMO precoding algorithm for millimeter-wave system under two main cooperative transmission modes in cooperative transmission technology is studied and designed in thesis.Firstly,a hybrid precoding algorithm based on generalized low-rank approximation is designed in multi-base station cooperative transmission scenarios.The algorithm aims at maximizing the array gain from base station to user.The analog precoding matrix is obtained by the iteration algorithm of generalized low-rank matrix approximation,and then semi-static cooperation is carried out according to the equivalent channel gain obtained from the analog precoding matrix.The digital precoding are obtained by singular value decomposition of the baseband equivalent channel of the cooperative set.The simulation results show that the spectral efficiency performance of the proposed algorithm is close to that of the unconstrained all-digital optimal precoding algorithm,which realizes the joint cooperative transmission between base stations in millimeter-wave communication system and improves the spectral efficiency of cell edge users.Secondly,there are inter-cell interference and inter-user interference in the cooperative mode scenario of cooperative scheduling/beamforming.A precoding algorithm based on the improved SLNR idea is designed.The algorithm designs analog precoding matrix to maximize the equivalent channel gain.Then the digital precoding design is carried out in the baseband digital domain using the equivalent channel.In the design of digital precoding,considering signal leakage after baseband digital combine precoding at the receiving terminal,the traditional SLNR precoding algorithm is improved to obtain digital precoding.In addition,in view of the problem that SLNR precoding algorithm can not guarantee the fairness of the system,an improved WSLNR precoding algorithm is proposed.Similarly,after the analog precoding being obtained,a weighting factor is designed according to the equivalent channel quality information including the user digital combine precoding matrix,so that the edge users with poor equivalent channel quality are least interfered by signal leakage from other users,thus increasing the rate of edge users with poor equivalent channel quality.The simulation results show that the performance of the improved SLNR precoding algorithm is higher than that of the traditional SLNR hybrid precoding algorithm.The improved SLNR precoding algorithm can effectively solve inter-cell interference and inter-user interference in the cell and improve the sum rate of the system.The proposed improved WSLNR algorithm improves the rate of edge users and increases the fairness of the system.