传感技术学报2025，Vol.38Issue(5)：917-922,6.DOI:10.3969/j.issn.1004-1699.2025.05.022
面向多Sink无线传感网络负载均衡的深度强化学习算法
Deep Reinforcement Learning Algorithm for Load Balancing of Multi-Sink Wireless Sensor Networks
摘要
Abstract
Sink nodes are usually located at the edge or center of the network,which are used to receive data and communicate with ex-ternal networks.When there are multiple central nodes,it is easy to cause load imbalance issues.In order to improve the load balancing and operation efficiency of wireless sensor networks,a deep reinforcement learning algorithm for load balancing in multi-sink wireless sensor networks is proposed.The energy consumption status of wireless sensor networks is analyzed,which is taken as a constraint,Markov decision process is used to analyze the network load allocation problem,a network load balancing model with energy constraints is constructed,the agent is trained through deep reinforcement learning algorithm,and the optimal load allocation strategy is selected based on the current state in the MDP model.The simulation results show that the load balancing factor value of the proposed algorithm is as high as 3 200,the average deviation of network nodes is below 1.5 J,the network transmission delay is always below 0.5 s,and the number of dead nodes is below 3.5,indicating good load balancing ability.关键词
无线传感器网络/负载均衡/深度强化学习/多Sink/马尔科夫决策过程Key words
wireless sensor network/load balancing/deep reinforcement learning/multi sink/Markov decision-making process分类
计算机与自动化引用本文复制引用
张伟华,王海英..面向多Sink无线传感网络负载均衡的深度强化学习算法[J].传感技术学报,2025,38(5):917-922,6.基金项目
河南省科技厅科技攻关项目(232102220010) （232102220010）
郑州商学院新工科创新融合团队项目(2021-CXTD-05) （2021-CXTD-05）
