摘要
水声传感器网络(UASNs)存在传输可靠性低、高能耗和长传输时延等问题。针对该问题,提出了一种蚁群辅助强化学习的机会路由(ACO-RL-OR)协议。在该协议中,首先设计了蚁群优化(ACO)算法辅助强化学习的全局决策函数,将ACO算法中的信息素和强化学习中的Q值综合作为动作策略以确定全局最优路由。然后建立有效的奖励表达式,并通过蚂蚁在所走的路径上遗留的信息素,找到路由跳数最少的全局最优路径。最后将Q值引入节点的保持时间中,提高协议实时性。仿真结果表明:与其他的水声路由协议相比,所提协议在包交付率(PDR)、能耗、网络寿命和端到端时延方面等均表现良好。
Underwater acoustic sensor networks(UASNs)have problems such as low transmission reliability,high energy consumption,and long transmission delays.To address these problems,a ant colony-aided reinforcement learning opportunistic routing(ACO-RL-OR)protocol is proposed.In this protocol,firstly,a global decision function for ant colony optimization(ACO)algorithm-aided is designed,and the pheromone in the ACO algorithm and the Q value in reinforcement learing are combined as action strategies to determine the globally optimal route.Then,an effective reward expression is established,and the globally optimal path with the minimum number of routing hops is found through the pheromones left by the ants on the path taken.Finally,the Q value is introduced into the holding time of nodes to improve the real-time performance of the protocol.Simulation results show that compared with other underwater acoustic routing protocols,the proposed protocol performs well in terms of packet delivery rate(PDR),energy consumption,network lifetime,and end-to-end delay.
作者
冯晓美
张育芝
李梦凡
韩翔
FENG Xiaomei;ZHANG Yuzhi;LI Mengfan;HAN Xiang(School of Communication and Information Engineering,Xi’an University of Science and Technology,Xi’an 710054,China)
出处
《传感器与微系统》
北大核心
2025年第12期57-63,共7页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(61801372)
陕西省教育厅科研计划项目(22JK0454)。
关键词
蚁群优化算法
强化学习
机会路由
可靠性
低能耗
ACO algorithm
reinforcement learing
opportunistic routing
reliability
low energy consumption
