Intelligent reflecting surface (IRS) has been widely studied in recent years, it has emerged as a new technology which can reflect the incident signal by intelligently configuring the reflection elements, thus changing the signal propagation environment, enhancing the signals users desire and suppressing the interference between users. In this paper, we study an IRS aided multi-users wireless communication where the base station (BS) sends a variety of signals, each user receives desired signals. In order to guarantee the fairness of wireless communications, we need to maximize the minimum rates of users, subject to the power constraint of BS and the phase constraint of IRS. Prior works on IRS mainly consider optimizing BS beamforming and IRS passive beamforming, this paper also aims to optimize the IRS location. The considered problem is shown to be non-convex, we decompose the problem into two subproblems, transforming the two subproblems into a lower bound problem and using alternating optimization (AO) and successive convex approximation (SCA) to solve them, respectively. Finally, the two subproblems are optimized alternately to make the objective function value converge in an acceptable range. Simulation results verify the convergence results of our proposed algorithm, and the performance improvement compared with the benchmark scheme in wireless communication system.
翻译:智能反射面(Intelligent reflecting surface,IRS)已经被广泛地研究和应用,在近年来成为一项新技术。IRS 通过智能地配置反射元素,反射入射信号,从而改变信号传播环境,增强用户所需的信号并抑制用户之间的干扰。本文研究了在多用户无线通信的情况下,基站(Base Station,BS)通过IRS辅助传输各种信号,使用户接收所需要的信号。为了保证无线通信的公平性,需要最大化用户最小速率,同时满足BS的功率约束和IRS的相位约束。之前的IRS研究主要考虑BS波束成形和IRS被动波束成形的优化,本文还在此基础上考虑了IRS位置的优化。考虑到问题为非凸问题,我们将问题分解为两个子问题,将两个子问题转化为下界问题,并使用交替优化和连续凸近似来分别解决这两个子问题。最后,这两个子问题交替优化,使目标函数值收敛于可接受范围。仿真结果验证了我们提出的算法的收敛性,并证明了与无线通信系统基准方案相比的性能改进。