LEDFD: A Low Energy Consumption Distributed Fault Detection Algorithm for Wireless Sensor Networks

Abstract

Detection of faulty nodes and network energy saving have become the hottest research topics. Furthermore, current fault detection algorithms always pursue high detection performance but neglect energy consumption. In order to obtain good fault detection performance and save the network power, this paper proposes a low energy consumption distributed fault detection algorithm (LEDFD), which takes full advantage of temporally correlated and spatially correlated characteristics of the sensor nodes. LEDFD utilizes the temporally correlated information to examine some faulty nodes and then utilizes the spatially correlated information to examine the nodes that have not been detected as faulty through exchanging information among neighbor nodes to determine those nodes' state. Because LEDFD takes the data produced by nodes themselves to detect certain types of faults, which means nodes need not exchange information with their neighbor nodes during the entire detection process, the energy consumption of networks is efficiently reduced. Experimental results show that the algorithm has good performance and low energy consumption compared with current algorithms.