Research On Two-tier Precoding In Multi-cells Massive MIMO System

Massive multi-input and multi-output(Massive MIMO)is one of the key technologies of the next generation wireless communication system.It uses large-scale antenna array technology to greatly increase the capacity of communication system.The two-tier precoding technology is a way to eliminate interference in massive MIMO downlink.It can effectively reduce the complexity of precoding by assigning users with similar spatial characteristics to the same group through JSDM technology and eliminating component interference and intra group interference in stages.However,with the increase of the system,especially in the multi-cell system,the problem of the precoding complexity surge still exists.In view of the above problems,this paper focuses on two-stage precoding in multi-cell systems.First,this paper introduces the large-scale MIMO system and its precoding theory.Several common nonlinear precoding and Linear Precoding algorithms are briefly introduced,and their application prospects in multi cell systems are analyzed.The JSDM precoding algorithm and system performance parameters of joint groups processing(JGP)and per groups processing(PGP)are mainly derived.How to reduce the dimension of channel matrix is explained and the selection of grouping scheme and precoding process in practical application are summarized.Then,this paper studies the precoding problem in massive MIMO multi-cell systems.A two-tier precoding scheme based on Givens-QR transform is proposed.In this scheme,the user distribution information is shared by multi-cell joint processing,and the interference channel is processed by QR decomposition based on givens transform.In the first stage,the zero space of channel interference matrix is obtained,and a precoding matrix is generated to eliminate the interference between adjacent cells.In the second stage,the zero space of different user channels in the cell is calculated,and a post precoding matrix is generated to eliminate the interference in the cell.The theoretical analysis shows that the complexity of the proposed scheme is significantly lower than that of the traditional BD algorithm.The simulation results show that the bit error rate of the scheme is reduced and the reachable rate is improved.Finally,this paper studies the multi cell precoding scheme in time-varying channel.A compensation tracking algorithm based on precoding is proposed.The optimization problem of precoding is mapped to a lattice manifold.The optimal solution of the precoding matrix is derived by using the minimum iteration algorithm on the Grassmann manifold.Derivation of the optimal solution of the precoding matrix by using the minimum iteration algorithm on the Grassmann manifold and a compensation term is introduced to estimate the change of target signal subspace.In the case of time-varying channel covariance matrix,the tracking ability of the precoder is enhanced.Theoretical analysis shows that the precoding scheme proposed in this chapter has low algorithm complexity.The simulation results show that the bit error rate of the scheme is reduced and the reachable rate is improved.