Provided by: Science and Development Network (SciDev.Net)
Fault detection is vital to wireless sensor networks since node death is a typical fault. One of the central challenges is to design a detection algorithm which has good performance in energy efficiency. In this paper, the authors propose an intelligent fault detection algorithm based on numerical taxonomy. Firstly, all nodes are divided into clusters according to their geographical distribution. Besides, the sink node uses numerical taxonomy to process the measurements and detect faulty node. Simulation results validate that this algorithm outperforms NDHN and will consume less energy.