基于Q学习的无线传感网自适应容侵覆盖优化方法  

Self-adaptive intrusion tolerance coverage optimization method for wireless sensor networks based on Q-learning

作　　者：夏云芝 王书林 范晓萱 耿艺璇 唐潇 邓贤君[1] XIA Yunzhi;WANG Shulin;FAN Xiaoxuan;GENG Yixuan;TANG Xiao;DENG Xianjun(School of Cyber Science and Engineering,Hubei Key Laboratory of Distributed System Security,Hubei Engineering Research Center on Big Data Security,Huazhong University of Science and Technology,Wuhan 430074,China;School of Computer Science and Technology,Huazhong University of Science and Technology,Wuhan 430074,China;Wuhan Research Institute of Posts and Telecommunications,Wuhan 430073,China;School of Mechanical Engineering,University of South China,Hengyang 421001,Hunan China)

机构地区：[1]华中科技大学网络空间安全学院,分布式系统安全湖北省重点实验室,湖北省大数据安全工程技术研究中心,湖北武汉430074 [2]华中科技大学计算机科学与技术学院,湖北武汉430074 [3]武汉邮电科学研究院,湖北武汉430073 [4]南华大学机械工程学院,湖南衡阳421001

出　　处：《华中科技大学学报（自然科学版）》2023年第2期25-31,共7页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(62272182,61871209,61901210,41904163);中央高校基本科研业务费专项资金资助项目(2022JYCXJJ024);湖南省自然科学基金资助项目(2020JJ5483);湖南省教育厅优秀青年项目(21B0428)。

摘　　要：针对无线传感器节点易受干扰和入侵的特点,为保障网络覆盖和安全性,基于信任管理和Q学习提出了一种新的自适应容侵覆盖优化方法.利用信任管理机制从节点的历史通信行为、能量、信息权重等方面评估节点的信任值,并根据节点信任值对受干扰或恶意节点的通信半径和感知半径进行调整,降低网络安全风险;同时,采用Q学习对网络覆盖进行优化,对节点进行训练,满足网络覆盖要求.所提方法具有较好的自适应能力,可有效地识别受干扰或恶意节点,提升网络覆盖率,且可快速收敛达到纳什均衡.Considering wireless sensor nodes are vulnerable to interference and intrusion,to ensure network coverage and security,a new self-adaptive intrusion tolerance coverage optimization method for wireless sensor networks(WSNs)based on Q-learning was proposed.The proposed method used trust management to evaluate the trust value of nodes from the aspects of historical communication behavior,energy,information weight,etc.According to the node trust value,the communication radius and sensing radius of interfered or malicious nodes were adjusted to reduce the network security risk.Furthermore,Q-learning was used to optimize the network coverage and nodes were trained to meet the coverage requirements.The proposed method can effectively identify interfered or malicious nodes and quickly converge to reach Nash equilibrium,which is self-adaptive and significantly improves the network coverage.

关 键 词：无线传感器网络(WSNs) 信任管理 Q学习 覆盖优化 容侵 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置] TN929.5[自动化与计算机技术—控制科学与工程]