Scholarlink

In the multi-hop network relay selection problem, how to balance The balance between performance and cost is a critical consideration. Traditional optimal selection schemes are centralized and require all local channel information during the relay process. This paper uses reinforcement learning to address the relay selection problem, achieving high performance at a lower cost while considering communication security issues. The experimental findings reveal that, in a multi-hop network with eavesdroppers, the proposed scheme attains an end-to-end rate that is closer to the optimal, while also offering lower computational complexity and signaling overhead when compared to other optimized decentralized strategies and the optimal scheme.

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

Research on Multi-hop Secure Communication Based on Reinforcement Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Yuyang Xie

摘要

In the multi-hop network relay selection problem, how to balance The balance between performance and cost is a critical consideration. Traditional optimal selection schemes are centralized and require all local channel information during the relay process. This paper uses reinforcement learning to address the relay selection problem, achieving high performance at a lower cost while considering communication security issues. The experimental findings reveal that, in a multi-hop network with eavesdroppers, the proposed scheme attains an end-to-end rate that is closer to the optimal, while also offering lower computational complexity and signaling overhead when compared to other optimized decentralized strategies and the optimal scheme.