无线电通信技术2025，Vol.51Issue(1)：36-44,9.DOI:10.3969/j.issn.1003-3114.2025.01.005
基于RIS辅助下XL-MIMO系统联合天线选择和波束成形
Joint Antenna Selection and Beamforming in RIS-assisted XL-MIMO Systems
摘要
Abstract
Energy Efficiency(EE)is a key metric to measure the performance of 5G/6G wireless communication systems,and Extremely Large-Scale Multiple-Input-Multiple-Output(XL-MIMO)and Reconfigurable Intelligent Surface(RIS)are considered to be effective ways to improve this metric.Due to the existence of near-field propagation and spatial non-stationarity in XL-MIMO system,this leads to some antennas having a reduced effect on system performance,consequently increasing the energy consumption.To en-hance energy efficiency under the power limitations of Base Station(BS)transmission power and RIS amplitude,an alternating optimi-zation-based solution is proposed.First,the antenna set is re-expressed using the l0 norm,making the total power consumption a con-tinuous function that is easier to solve.Subsequently,an optimal solution is achieved by iteratively optimizing BS beamforming matrix and RIS phase shift matrix in an alternating manner.To mitigate the increased algorithmic complexity introduced by auxiliary variables,a deep learning-based approach is developed.The method employs a three-stage network architecture comprising a phase-shift network,a beamforming network,and an antenna selection network,coupled with a custom loss function designed to maximize EE,enabling effi-cient joint optimization.According to simulation results,the proposed deep learning approach enhances system energy efficiency while significantly reducing computational complexity compared to traditional alternating optimization methods.关键词
超大规模多输入多输出/智能超表面/联合天线选择和波束成形/交替优化/无监督深度学习Key words
XL-MIMO/RIS/joint antenna selection and beamforming/alternating optimization/unsupervised deep learning分类
信息技术与安全科学引用本文复制引用
褚宏云,贾帅..基于RIS辅助下XL-MIMO系统联合天线选择和波束成形[J].无线电通信技术,2025,51(1):36-44,9.基金项目
国家自然科学基金(62401467) （62401467）
陕西省自然科学基金(2022JQ-635) National Natural Science Foundation of China(62401467) （2022JQ-635）
Shaanxi Provincal Natural Science Foundation of China(2022JQ-635) （2022JQ-635）
