基于最大似然估计与朴素贝叶斯的WSN故障检测

TP393; WSN中的故障节点导致网络的数据传递延迟与能耗增加,同时可引起网络拥塞等问题,对此提出一种基于最大似然估计与朴素贝叶斯分析器的WSN故障节点诊断与定位算法.首先,从数据包的协议部分提取大量特征作为训练数据集,从中估算边际概率并建立朴素贝叶斯分类器,使用最大似然估计估算条件概率.检测阶段则通过判断传输延迟是否满足阈值条件来决定可疑节点,然后使用朴素贝叶斯分类器检测故障节点,最终将节点成功进行分类....

Full description

Saved in:
Bibliographic Details
Published in电子技术应用 Vol. 41; no. 7; pp. 114 - 117
Main Author 敬明旻 肖莉 杨传书
Format Journal Article
LanguageChinese
Published 中国石化石油工程技术研究院信息与标准化研究所,北京,100101 2015
Subjects
故障诊断
无线传感器网络
最大似然估计
朴素贝叶斯分类器
最大似然估计
故障诊断
fault detection
wireless sensor network
朴素贝叶斯分类器
无线传感器网络
Naive Bayes classifier
maximum likelihood estimation
Online AccessGet full text
ISSN0258-7998
DOI10.16157/j.issn.0258-7998.2015.07.032

Cover

More Information
Summary:TP393; WSN中的故障节点导致网络的数据传递延迟与能耗增加,同时可引起网络拥塞等问题,对此提出一种基于最大似然估计与朴素贝叶斯分析器的WSN故障节点诊断与定位算法.首先,从数据包的协议部分提取大量特征作为训练数据集,从中估算边际概率并建立朴素贝叶斯分类器,使用最大似然估计估算条件概率.检测阶段则通过判断传输延迟是否满足阈值条件来决定可疑节点,然后使用朴素贝叶斯分类器检测故障节点,最终将节点成功进行分类.
Bibliography:Fault nodes in WSN lead to longer transmit delay and more energy consumption, and lead to problems such as network congestion, aimed at that, a new maximum likelihood and Naive Bayes classifier based fault diagnosis and location algorithm for WSN is proposed. Firstly, a large amount of features are abstracted from the protocol part of the data package, and the marginal probability is estimated and the Naive Bayes classifier is set up, and the condition probability is estimated by maximum likelihood estimation. In the detection phase, the transmission time are compared with threshold to adjust the fault node, then, the Naive Bayes classifier is used to detect the fault nodes, at last, the nodes are classified successfully.
Jing Mingmin, Xiao Li, Yang Chuanshu （Research Institute of Petroleum Engineering, Research Department of Information and Standardization Beijing 100101, China）
maximum likelihood estimation;Naive Bayes classifier;fauh detection;wireless sensor network
11-2305/TN
ISSN:0258-7998
DOI:10.16157/j.issn.0258-7998.2015.07.032