基于可靠性度量的无线传感网络恶意节点检测算法  

Reliability metric-based detection algorithm of compromised sensor nodes in wireless sensor networks

作　　者：宋三华[1] SONG Sanhua(College of Information Engineering,HuangHuai University,Zhumadian 463000,China)

机构地区：[1]黄淮学院信息工程学院,河南驻马店463000

出　　处：《中国测试》2018年第7期148-152,共5页China Measurement & Test

基　　金：河南省科技厅基金项目(162102110119)

摘　　要：在无线传感网络中二值事件的分布式检测系统中,局部的传感节点统计过程容易遭受恶意节点的安全袭击,而检测二值事件的准确性依赖于传感节点的可靠性。为此,提出基于可靠度量的恶意节点的检测算法RDICS识别恶意节点,并减少这些恶意节点的融合权值,进而削弱它们对最终决策的贡献。在设置融合权值时,先计算基于所有传感节点的融合中心决策与单个传感节点决策的不一致性,然后再计算包含除此传感节点的决策外的所有传感节点的融合决策与此传感节点决策的不一致性。再依据这两个参数,估计传感节点的可靠性,进而设置融合权值。最后,实验数据表明,RDICS算法的检测率优于现存的RBDA算法。In the distributed detection of a binary event in wireless sensor networks(WSNs),the statistical process of local sensor nodes is vulnerable to security attacks of compromised sensor nodes.The detection precision of a binary event strongly depends on the reliability of these sensor nodes in the network.Therefore,the reliability metric-based detection algorithm to identify compromised sensor nodes(RDICS)is proposed to identify the compromised sensor nodes and decreases their fusion weights so as to reduce their contributions towards the final decision.In computing the fusion weights,firstly the inconsistency between the fusion center’s decision where all the sensor nodes contributions are considered and the single node decision should be calculated,and then the inconsistency between the fusion center’s decision where the single node contributions is excluded and the single node decision should be calculated.Finally,according to these two parameters,the reliability of the sensor nodes is estimated,and the fusion weights are set.Experiment results show that the proposed RDICS algorithm outperforms the existing RBDA algorithm.

关 键 词：无线传感网络 二值事件 融合中心 恶意节点 检测率 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]