Intelligent Reflecting Surface-assisted Signal Processing Technology In Millimeter Wave Communication System

With the commercialization of 5G communication networks,the research of the sixth generation of mobile communications has been gradually carried out.Aiming at the problem of limited signal transmission distance and coverage in millimeter wave wireless communications,it is proposed to use intelligent reflectors to assist millimeter wave communications.In this paper,we consider channel estimation for intelligent reflecting surface-assisted millimeter wave communication systems,and joint active and passive precoding design for intelligent reflecting surface-assisted millimeter wave communications.In practical applications,intelligent reflectors are passive and do not have signal processing capabilities,which brings greater challenges to the channel estimation for intelligent reflecting surface-assisted millimeter wave communication systems.The existing channel estimation methods usually have relatively large pilot overhead and implementation complexity.Therefore,how to estimate the channel state information for intelligent reflecting surface-assisted millimeter wave communication systems is a subject worthy of study.In this paper,we use the sparse scattering characteristics of the millimeter wave channel to find the sparse representation of the cascaded channel,then transform the channel estimation into a sparse signal recovery problem,and use a bipartite graph encoding and decoding scheme to solve it.On the other hand,the joint beamforming optimization of the base station and the intelligent reflector is one of the important research directions of the intelligent reflecting surface-assisted communication system.Under the condition that the channel state information of the base station is known,the joint optimization problem is decoupled into two sub-problems,and the two variables are optimized alternately to maximize the spectral efficiency of the receiver.And then,we discussed the joint beamforming optimization problem based on estimated channel.Firstly,the channel estimation under the broadband intelligent reflecting surfaceassisted millimeter wave communication system is studied.We use algorithms based on sparse bipartite graph to estimate channel,and represent the sparse channel from the base station to the intelligent reflecting surface and the intelligent reflecting surface to the user as a cascaded channel,and give the sparse representation of the cascaded channel in the angle domain.First,we encode the sparse channel through a set of bipartite graphs,and then use a simple decoding to obtain the cascaded channel state information,that is,the position and amplitude of the non-zero entries of the sparse matrix.The algorithm can be obtained accurately with less complexity.Secondly,the optimization problem of passive and active precoding design is studied.The original highly coupled non-convex optimization problem is decoupled into two subproblems,and the solution of the phase shift matrix is discussed under the condition of known channel state information.Finally,the joint beamforming optimization problem based on estimated channel is studied.Assuming that only the information of the line-ofsight link of the channel is known,the phase shift matrix with wider beam coverage is obtained.