一种物联网传感器故障节点检测方法设计  

Design of a sensor fault node detection method for Internet of things

作　　者：王建文 裴祥喜 崔炳德 王海晖[2] WANG Jian-wen;PEI Xiang-xi;CUI Bing-de;WANG Hai-hui(Department of Computer Science and Information Engineering Hebei University of Water Resources and Electric Engineering,Hebei Cangzhou 061001 China;School of Computer Science&Engineering,Wuhan Institute of Technology,Hubei Wuhan 430205,China)

机构地区：[1]河北水利电力学院计算机科学与信息工程学院,河北沧州061001 [2]武汉工程大学计算机科学与工程学院,湖北武汉430205

出　　处：《计算机仿真》2022年第3期354-357,480,共5页Computer Simulation

基　　金：河北省高等学校科学研究计划青年基金项目(QN2016204)。

摘　　要：采用目前算法对物联网传感器节点进行故障标定时,没有对节点信息进行聚类处理,导致算法存在虚警率高、计算复杂度高和标记范围小的问题。提出基于BDPCA聚类的物联网传感器节点故障标记算法,采用BDPCA聚类方法对物联网传感器节点信息进行聚类处理,并在聚类过程中对节点信息进行了零均值归一化处理。结合区分函数和区分矩阵在粗糙集理论的基础上对不同类别的节点信息进行知识约简处理,并通过贝叶斯决策理论实现物联网传感器节点的故障标记。实验结果表明,所提算法的虚警率低、计算复杂度低、标记范围广。Recently,during the calibration of sensor node faults in the Internet of things,the traditional methods ignore the clustering of node information,resulting in a high false alarm rate,complex calculation,and narrow marking range.This paper proposes a fault marking algorithm for sensor nodes in the Internet of things based on BDPCA clustering.According to the BDPCA clustering method,the sensor node information of the Internet of things was clustered.Meanwhile,during the clustering process,the node information was normalized by zero mean.Based on the discernibility function and discernibility matrix,the node information of different categories was simplified.The fault marking of sensor nodes in the Internet of things was realized based on Bayesian decision theory.The results show that the algorithm has a low false alarm rate,simple calculation,and wide marking range.

关 键 词：BDPCA聚类 物联网 传感器节点 粗糙集理论 故障标记 

分 类 号：TP393.02[自动化与计算机技术—计算机应用技术]