Research On Hybrid Precoding Technology In Millimeter-Wave MIMO Systems

As one of the alternative frequency bands for 5G network,millimeter waves can cope with the current challenges due to bandwidth shortage.Moreover,it can also meet the growing demand of people for communication rate.In millimeter-wave Massive multi-input multi-output(MIMO)systems,traditional digital precoding schemes require a separate RF link for each antenna,which is difficult to implement in practical applications.A hybrid precoding scheme can effectively solve this problem and reduce the number of RF links.This thesis mainly studies on the hybrid precoding problem in millimeter-wave massive MIMO systems,and aims at the pure digital precoding algorithm to further improve the performance of the hybrid precoding algorithm.The main research contents are as follows.1.In order to further improve the performance of single-user point-to-point millimeter-wave massive MIMO(SU-MIMO)communication system,and reduce the complexity of the algorithm,this thesis proposes a hybrid precoding algorithm based on alternating iterative matrix decomposition(Alt-IMD).The algorithm first initializes the analog precoding matrix;then obtains the digital precoding matrix by simulating the precoding matrix and the optimal precoding matrix,updates the analog precoding matrix by extracting the phase information of the pseudo pure product of the optimal pure digital precoding matrix and the digital precoding matrix.The algorithm updates the digital precoding matrix with the updated analog precoding matrix;finally,it converges through multiple iterations.The simulation results show that the performance of the proposed Alt-IMD algorithm is better than the traditional Orthogonal Matching Pursuit(OMP)hybrid precoding algorithm.Moreover,compared to the Phase Extraction Alternating Minimum(PE-AltMin)algorithm,the complexity is reduced by about 47%.2.In order to more effectively suppress the interference between multiple users in multi-user millimeter-wave massive MIMO(MU-MIMO)systems,and improve the spectrum efficiency of the system,this thesis proposes an alternating iterative matrix decomposition based on block diagonalization(AltIMD-BD)scheme.The scheme divides the hybrid precoding design into two phases.Firstly,on the RF side,using the proposed Alt-IMD algorithm to design the simulation merge matrix for each user.Secondly,the analog precoding matrix can be designed by the equal gain transmission(EGT)method.Finally,constructing the equivalent channel matrix,fully considering the user space,and designing the digital precoding matrix by using the improved BD algorithm at the baseband end.Theoretical analysis and simulation results show that the AltIMD-BD algorithm proposed in this thesis is closer to the performance of the optimal pure digital precoding algorithm.