Research On Sum Rate Optimization And Channel Estimation Of Intelligent Reflecting Surface-assisted Multi-user Communication Systems

Nowadays,wireless communication has become a necessary part of modern social life.Wireless communication networks are facing serious challenges such as surging traffic and diverse demands.In response to these trends,intelligent reflecting surface(IRS)has been proposed and has received a lot of attention from academia and industry in recent years due to its low power consumption,cheap manufacturing and easy deployment.Typically,IRS is made up with huge number of reflective units,which can be individually configured.It could produce additional phase shift to the reflected signal by jointly controlling multiple reflective units.Based on this,the system can use IRS devices to actively adjust the electromagnetic characteristics of the signal,thus assisting the transmission of information and realizing the concept of "intelligent radio environment".By combining with traditional wireless communication systems,IRS can provide performance gains to the system.But in the meanwhile,this also brings a series of problems that need to be addressed.In this thesis,we focus on the sum-rate optimization and the channel estimation problems in IRS-assisted multi-user communication systems.Firstly,this thesis investigates sum-rate optimization problem for an IRS-assisted uplink multi-user communication system using non-orthogonal multiple access(NOMA)technology.We formulate the IRS phase matrix design problem.Then,we propose an iterative computation scheme using alternating directional multiplier method(ADMM)due to its low complexity and fast convergence property.Finally,simulation results show the relationship between the sumrate and the parameters such as the transmit power of the device,the number of devices,as well as the number of reflective units.Compared with existing studies in the literature using the semidefinite relaxation(SDR)algorithm,the proposed scheme in this thesis does not require ignoring the nonconvex constraints,while ensuring the accuracy of the solution.The obtained results illustrate that our proposed scheme can efficiently obtain the IRS phase matrix parameters and achieve good sum-rate performance.Secondly,this thesis investigates the channel estimation issue in IRS-assisted multi-user systems.The passive IRS devices lead to a serious challenge in channel estimation.However,considering the characteristic that all users share the common channel between BS-IRS in the system,a two-step channel estimation scheme is proposed based on the compressed sensing(CS)theory.Finally,simulation results show the relationship between the NMSE and the factors such as the pilot overhead,the transmit power of the device,and the number of reflective units.The obtained results show that our proposed scheme can guarantee the channel estimation accuracy while reducing the pilot overhead compared with the existing estimation schemes in the literature.