智能反射面使能认知多播鲁棒波束赋形设计  

Robust Beamforming Design for Cognitive Multicast Enabled by Intelligent Reflective Surface

作　　者：李建森[1] LI Jiansen(Southwest China Institute of Electronic Technology,Chengdu 610036,China)

机构地区：[1]中国西南电子技术研究所,成都610036

出　　处：《微处理机》2024年第2期22-25,共4页Microprocessors

基　　金：军委科技委基础计划加强重点基础研究项目(2019-JCJQ-ZD-050-00)。

摘　　要：针对智能反射面使能认知多播系统波束赋形设计面临不精确信道状态信息的问题,提出一种联合认知基站主动与IRS被动波束赋形的鲁棒优化方法。先利用连续凸近似方法、S-procedure和舒尔补引理处理不精确CSI导致的非凸约束,再利用交替优化解耦决策变量,进而提出基于交替优化和罚凸凹过程的次优方法。仿真结果表明所提算法收敛快,在信道估计存在一定误差时仍可用以部署IRS并显著提高认知多播网络性能。In order to solve the problem of inaccurate channel state information of the beamforming design for cognitive multicast system enabled by the intelligent reflective surface,a robust optimization method for joint base station active and IRS passive beamforming is proposed.S-procedure and Shure complement lemma are used to deal with the non-convex constraints caused by inaccurate CSI,alternate optimization is adopted to decouple the decision variables,and then a suboptimal method based on alter-nating optimization and penalty convex and concave process is proposed.Simulation results show that the proposed algorithm converges quickly,and even if there is a certain error in channel estimation,the de-ployment of IRS and the use of the proposed algorithm can still significantly improve the performance of cognitive multicast network.

关 键 词：智能反射面 认知无线电 多播 不完美信道状态信息 波束赋形 

分 类 号：TN929.5[电子电信—通信与信息系统]