摘要： 针对大规模无线传感器网络(WSN)中故障检测准确率降低，并产生较大通信负载的问题，根据传感器节点的时空相关性特点，提出一种基于簇的分布式传感器故障检测算法。通过邻居节点间的数据交换和互相测试检测簇首节点，并以正常簇首节点作为参照诊断故障节点。性能分析与实验结果表明，在大规模WSN中，该算法具有良好的故障检测能力和较低的通信负载，在邻居节点数较少、节点故障率较高的情况下，能达到98%以上的故障检测准确率，并保持较低的能耗水平。

关键词: 无线传感器网络, 传感器故障检测, 分布式, 大规模网络, 分簇, 时空相关性

Abstract: The fault detection in large scale Wireless Sensor Network(WSN) has low performance in accuracy and can bring high communication load. Considering temporal and spatial correlation in WSN, a novel cluster-based sensor fault detection algorithm is proposed. The algorithm provides a way to identify faulty sensors by finding normal cluster heads and using its as reference to infer detection statuses of its neighbors. Performance analysis and experimental results show that the algorithm has good fault detection ability and lower communication load, it can achieve over 98% fault detection accuracy and keep lower energy consumption even in the case of smaller neighboring nodes and higher node’s failure ratio.

Key words: Wireless Sensor Network(WSN), sensor fault detection, distributed, large scale network, clustering, temporal and spatial correlation

中图分类号: 

• TP391