Relay algorithms in power line and wireless dual-media cooperative communication exhibit distinct advantages and disadvantages under random channels, and the algorithm that achieves the best performance can be determined through an adaptive relay strategy. However, the theoretical performance analysis of multiple relay algorithms for mixed fading and impulse noise encounters practical difficulties, such as lack of closed-form expressions and high complexity of the approximation process. This study proposes an adaptive relay strategy selection algorithm based on Q-learning, which can maximize the mutual information in the system. The feedback of the amount of mutual information is obtained by learning the dual-media cooperative communication environment of the time-varying fading channel to determine the optimal relay algorithm and power allocation factor. This study also analyzes the influence of the average signal-to-noise ratio, relay position, and power line channel parameters on the performance of the algorithm in terms of mutual information and outage probability and implements Q-learning-based relay strategy selection. Simulation results show that the proposed algorithm can obtain a quasi-optimal system performance in terms of the above two measures.

电力线和无线双媒体协作通信中的中继算法在随机信道下表现出明显的优缺点，可以通过自适应中继策略确定性能最佳的算法。然而，针对混合衰落和脉冲噪声的多种中继算法的理论性能分析遇到了实际困难，例如缺乏闭式表达式和逼近过程的高复杂度。本研究提出了一种基于Q-learning的自适应中继策略选择算法，可以最大化系统中的互信息。通过学习时变衰落信道的双媒体协作通信环境，确定最优中继算法和功率分配因子，得到互信息量的反馈。本研究还从互信息和中断概率方面分析了平均信噪比、中继位置和电力线信道参数对算法性能的影响，并实现了基于 Q-learning 的中继策略选择。仿真结果表明，所提出的算法在上述两种措施下都能获得准最优的系统性能。和电力线信道参数在互信息和中断概率方面对算法性能的影响，并实现了基于 Q-learning 的中继策略选择。仿真结果表明，所提出的算法在上述两种措施下都能获得准最优的系统性能。和电力线信道参数在互信息和中断概率方面对算法性能的影响，并实现了基于 Q-learning 的中继策略选择。仿真结果表明，所提出的算法在上述两种措施下都能获得准最优的系统性能。