Research On Hybrid Precoding In Millimeter Wave Massive MIMO Systems

5G wireless communication system can significantly improve mobile data transmission speed,and massive MIMO technology is one of the key technologies of 5G.Millimeter wave has also received extensive attention due to its rich frequency band resources.Millimeter wave can reduce the physical size of antenna array,and the high antenna array gain provided by massive MIMO system can help do overcome the serious path loss of millimeter wave signal.The combination of the two can achieve an order of magnitude growth of system capacity.Therefore,millimeter wave massive MIMO technology has been established as one of the core solutions to support 5G wireless system.In millimeter wave massive MIMO system,due to the large number of antennas at the receiver and transmitter,the hardware cost and power consumption of the digital precoding scheme are too high.Considering the cost problem in practical application,this paper mainly focuses on the research of digital analog hybrid precoding technology,which is a compromise between performance and cost.First of all,a hybrid precoding scheme in the millimeter wave massive MIMO system is proposed with partially connected stucture at the transmitter and digital combining at the receiver.In this scheme,the digital precoder and combiner are designed based on the SVD of the equivalent channel matrix,while the optimal analog precoder is obtained by blocking the equivalent channel to maximize spectral efficiency according to the block diagonalization.Considering the hardware cost and power consumption,the precoding scheme with partially connected stucture at the receiver is further proposed.Otherwise,the analog precoder and combiner are iteratively updated.The simulation results show that the proposed scheme which focuses on the digital combining at the receiver has better performance,while the precoding scheme focuses on the hybrid combining with partially connected structure at the receiver has advantages in the power consumption and hardware cost.Then,a millimeter wave massive MIMO system with fully connected structure is studied and analyzed.Two schemes are proposed for single user receiving and multi-user receiving respectively.The first scheme is a hybrid precoding scheme based on SVD-SIC algorithm for single user reception.The analog precoding is designed by maximizing the effective channel gain in advance.On this basis,the sum rate maximization problem is decomposed into a series of sub array optimization problems,and the digital precoding is designed by SVD in turn.The second scheme is a hybrid precoding scheme based on conjugate transposition for multi-user reception.The AOD of the maximum gain path between the user and the channel is calculated,and the corresponding vectors are selected from the antenna array to design the analog combining of each user independently.Then the analog precoding is designed by the nature of conjugate transposition with the objective function of maximizing the effective channel gain.The digital part is designed by ZF or MMSE precoding.Simulation results show that the proposed hybrid precoding scheme based on SVD-SIC algorithm has advantages in performance,while the hybrid precoding scheme based on conjugate transposition has advantage in complexity.