Scholarlink

With the advancement of sixth-generation (6G) wireless communications, wireless endogenous security has become a hot topic. To address this, we can leverage Reconfigurable Intelligent Surface (RIS) to design integrated communications and security (JCAS) paradigms, where the dual functions of communication and security mutually benefit each other. In this paper, we consider an ICAS system that can simultaneously transmit data and generate secret keys. Specifically, the optimization problem aims to maximize energy efficiency concerning both the data transmission rate and the key generation rate by jointly optimizing the phase-shifts of the RJS and its ON/OFF state. Given the uncertainty of wireless environments, we propose a model-free deep reinforcement learning (DRL) algorithm to determine the optimal RIS configuration strategy. Simulation results demonstrate that the proposed algorithm significantly enhances the energy efficiency of the ICAS system and outperforms existing benchmark methods.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

RIS-Assisted Energy-Efficient Communication and Key Generation with Reinforcement Learning

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yuze Yao,
  • Yue Li,
  • Junjie Chang,
  • Guorong Yang,
  • Xincheng Xia,
  • Ning Gao

摘要

With the advancement of sixth-generation (6G) wireless communications, wireless endogenous security has become a hot topic. To address this, we can leverage Reconfigurable Intelligent Surface (RIS) to design integrated communications and security (JCAS) paradigms, where the dual functions of communication and security mutually benefit each other. In this paper, we consider an ICAS system that can simultaneously transmit data and generate secret keys. Specifically, the optimization problem aims to maximize energy efficiency concerning both the data transmission rate and the key generation rate by jointly optimizing the phase-shifts of the RJS and its ON/OFF state. Given the uncertainty of wireless environments, we propose a model-free deep reinforcement learning (DRL) algorithm to determine the optimal RIS configuration strategy. Simulation results demonstrate that the proposed algorithm significantly enhances the energy efficiency of the ICAS system and outperforms existing benchmark methods.