基于深度强化学习的无线传感器网络分簇算法  

Wireless sensor network clustering algorithm based on deep reinforcement learning

作　　者：肖祥[1] XIAO Xiang(Fujian College of Water Conservancy and Electric Power,Yongan 366000,China)

机构地区：[1]福建水利电力职业技术学院,福建永安366000

出　　处：《无线互联科技》2024年第24期122-124,共3页Wireless Internet Science and Technology

摘　　要：鉴于无线传感器网络中拓扑频繁变动挑战分簇稳定性,引发数据传输冗余与能耗激增,文章提出基于深度强化学习的分簇算法。该算法融合能量、位置与密度因素,利用深度强化学习结合传感器能量模型,精准聚类节点并优化簇间路径,实现高效分簇策略。仿真验证显示,相较于对比方法,该算法显著提升网络生存时间约40%,能够有效遏制能耗,显著延长无线传感器网络的生命周期,展现了其在复杂环境中的优越性能与实用价值。In view of the frequent topology changes in wireless sensor networks challenge the stability of clustering,which leads to data transmission redundancy and energy consumption surge,a clustering algorithm based on deep reinforcement learning is proposed.By integrating energy,position and density factors,deep reinforcement learning combined with sensor energy model is used to accurately cluster nodes and optimize inter-cluster paths to achieve efficient clustering strategies.Simulation results show that compared with the comparison method,the research algorithm can significantly improve the network lifetime by about 40%,effectively curb energy consumption,and significantly extend the life cycle of wireless sensor networks,demonstrating its superior performance and practical value in complex environments.

关 键 词：深度强化学习 无线传感器 网络分簇 节点信任度 数据聚类 

分 类 号：TN711[电子电信—电路与系统]