通信与信息技术Issue(4)：1-6,6.
基于DDPG的光控阵波束赋形智能控制方法
The intelligent beamforming control method for optical phased arrays based on DDPG
摘要
Abstract
To address the issues associated with traditional beamforming methods in optical phased array antennas—such as com-plex excitation control,difficulty in sidelobe suppression,and limited adaptability—an intelligent beam optimization model based on deep reinforcement learning is developed.Utilizing the Deep Deterministic Policy Gradient(DDPG)algorithm,a closed-loop state-action-reward mechanism is designed to enable adaptive amplitude adjustment of elements in a 4×4 two-dimensional array.Under a carrier fre-quency of 38GHz,simulation results demonstrate that the proposed model significantly improves the maximum sidelobe level from 14.40 dB(under uniform excitation)to 30.40dB,achieving a gain of 16.00dB.These results highlight the efficiency and feasibility of the method in sidelobe suppression and beam control.关键词
强化学习/波束赋形/光控阵/主旁瓣抑制比Key words
Reinforcement learning/Beamforming/Optically controlled array/Main-to-Sidelobe ratio分类
信息技术与安全科学引用本文复制引用
江东洋,叶佳,闫连山,余晓,邓炯斌..基于DDPG的光控阵波束赋形智能控制方法[J].通信与信息技术,2025,(4):1-6,6.基金项目
中国国家重点研发计划(项目编号:2022YFB2802701)国家自然科学基金(项目编号:U23A20376和62271422)四川省杰出青年科学基金(项目编号:2024NSFJQ0016) （项目编号:2022YFB2802701）
