Hybrid Precoding Technology For Millimeter Wave Massive MIMO System

With the development of wireless data services,people urgently need to find a suitable spectrum range to meet increasing demand.Therefore,millimeter wave communication has been widely concerned in academic circles.Due to the high path loss and penetration loss of millimeter wave,the beamforming of antenna plays a key role in establishing and maintaining robust communication links.In 5G communication system,beamforming in millimeter-wave massive multi-input multiple-output(MIMO)communication systems is achieved through hybrid precoding technology,which combines high-dimensional analog phase shifters and power amplifiers with low-dimensional digital signal processing unit.Hybrid precoding technology reduces the cost and power consumption of millimeter wave communication,which conforms to 5G energy-saving design vision.This thesis mainly studies the hybrid precoding technology and its design scheme in millimeter-wave massive MIMO systems.The main work is as follows:1.For the single-user transmission scenario of millimeter-wave massive MIMO system,in order to improve the spectral efficiency of the system and ensure the energy efficiency of the system,this thesis proposes a hybrid precoding algorithm based on alternative iterative singular value decomposition(Alt-ISVD).Before using the alternating minimization architecture,it is proved that the phase angle of the ordered right singular vector of the channel matrix can be used to initialize the analog precoder,thereby avoiding the complicated optimization process of searching for the approximate optimal analog precoding matrix.The simulation results show that the performance of Alt-ISVD algorithm exceeds the single-stream optimal column by column optimization algorithm when the signal to noise ratio(SNR)is 5d B,and about 3.2 bit/s/Hz higher than that of alternating semidefinite relaxation(Alt-SDR)algorithm,while the complexity is lower.2.This thesis extends the scene of single-user transmission scenarios to multiple users in millimeter-wave massive MIMO system.In order to eliminate interference among users and reduce the transmission overhead of a multi-user transmission system,a hybrid precoding algorithm based on discrete fourier transformation(DFT)codebook combining block diagonalization(BD)and SVD is proposed in this thesis.First,the analog precoder is obtained by predefining DFT codebook and maximizing the channel gain.Then BD algorithm is used to eliminate the interference among users.Finally,SVD algorithm is used to obtain higher gain.Through the simulation comparison,the performance of the proposed algorithm is slightly better than Hybrid Minimum Mean Square Error algorithm based on SVD(Hy-MMSE-SVD),while the complexity is greatly reduced without distinct performance,compared with Hy-BD algorithm based on column by column optimization.Futhermore,the algorithm has higher energy efficiency than fully connected multi-user orthogonal matching pursuit(OMP)algorithm.