Joint Active and Passive Beamforming Design in Intelligent Reflecting Surface(IRS)-Assisted Covert Communications:A Multi-Agent DRL Approach

扫码查看

原文链接

NETL
NSTL
万方数据

外文摘要：Intelligent Reflecting Surface(IRS),with the potential capability to reconstruct the electromag-netic propagation environment,evolves a new IRS-assisted covert communications paradigm to eliminate the negligible detection of malicious eavesdroppers by coherently beaming the scattered signals and sup-pressing the signals leakage.However,when multi-ple IRSs are involved,accurate channel estimation is still a challenge due to the extra hardware complex-ity and communication overhead.Besides the cross-interference caused by massive reflecting paths,it is hard to obtain the close-formed solution for the op-timization of covert communications.On this basis,the paper improves a heterogeneous multi-agent deep deterministic policy gradient(MADDPG)approach for the joint active and passive beamforming(Joint A&P BF)optimization without the channel estima-tion,where the base station(BS)and multiple IRSs are taken as different types of agents and learn to en-hance the covert spectrum efficiency(CSE)cooper-atively.Thanks to the'centralized training and dis-tributed execution'feature of MADDPG,each agent can execute the active or passive beamforming inde-pendently based on its partial observation without re-ferring to others.Numeral results demonstrate that the proposed deep reinforcement learning(DRL)ap-proach could not only obtain a preferable CSE of legit-imate users and a low detection of probability(LPD)of warden,but also alleviate the communication over-head and simplify the IRSs deployment.

外文关键词：

covert communicationsdeep reinforce-ment learningintelligent reflecting surface

作者：

Gao Ang、Ren Xiaoyu、Deng Bin、Sun Xinshun、Zhang Jiankang

展开 >

作者单位：

School of Electronic Information,Northwestern Polytechnical University,Xi'an,710072,China

Key Laboratory of Near Ground Detection and Perception Technology,Wuxi 214035,China

Department of Computing and Informatics,Bournemouth University,UK

基金：

Key Laboratory of Near Ground Detection and Perception TechnologyKey Laboratory of Near Ground Detection and Perception TechnologyShaanxi and Taicang Keypoint Research and Invention ProgramShaanxi and Taicang Keypoint Research and Invention Program

项目编号：

614241422040661424142101012021GXLH-01-15TC2019SF03

出版年：

2024

DOI：

10.23919/JCC.fa.2023-0548.202409

中国通信(英文版)

CSTPCD

影响因子：0.463

ISSN：1673-5447

年,卷(期)：2024.21(9)