Research On Hybrid Precoding Algorithm For Millimeter Wave Massive MIMO System

Massive multiple input multiple output(MIMO)technology has significant effect on improving system performance,millimeter wave provides a number of unauthorized spectrum resource for modern communication system,and millimeter wave can miniaturize the antenna array because of the short wavelength,which greatly reduces the hardware cost.Combining millimeter wave with massive MIMO technology can not only effectively improve the transmission rate of the system,but also greatly increase the transmission bandwidth.In millimeter wave massive MIMO systems,hybrid precoding technology can effectively improve the spectral efficiency of the system,at present there are two main types of hybrid precoding structure,namely fully connection structure and partially connection structure,in this paper,hybrid precoding algorithm with better spectral efficiency is designed under these two structures.The specific research contents are as follows:(1)Design a spectral efficient hybrid precoding algorithm based on improved orthogonal matching pursuit(OMP)in a single user millimeter wave massive MIMO fully connected structure system.This algorithm decouples the analog precoding matrix and digital precoding matrix,first,for the design of the analog precoding matrix,the array response vector is selected as its candidate matrix to reduce the complexity and interference between data streams;then,in order to solve the digital precoding matrix,use the regularized zero forcing(RZF)method to process the equivalent matrix constructed by the channel matrix and the analog precoding matrix,which greatly reduces the influence of noise and the influence between data streams;finally,by observing the simulation results,it can be concluded that the proposed hybrid precoding algorithm can achieve better spectral efficiency under perfect and imperfect channel state information(CSI),which verifies the value of the proposed algorithm in practical applications.(2)In the partially connected structure system,a spectral efficient hybrid precoding algorithm based on geometric mean decomposition(GMD)is designed.The algorithm jointly solves the digital precoding matrix and the analog precoding matrix,and reduces restrictions on hybrid precoding matrices,first the algorithm decomposes the spectral optimization problem into several sub-problems,then constructs an intermediate matrix and performs GMD on it,which can avoid complicated calculation processes;next optimizes each sub-problem to achieve the problem of maximizing the spectral efficiency of the system;finally the simulation results indicate that the proposed hybrid precoding algorithm can obtain higher spectral efficiency under both perfect CSI and imperfect CSI.