Fault Diagnosis Algorithm for WSN Based on Clustering and Credibility

Fault diagnosis is one of the challenging problems in wireless sensor network (WSN). This paper proposes a fault diagnosis algorithm based on clustering and credibility (FDCC). Firstly, the network is divided into several clusters according to both geographic positions and measurements of sensor nodes for the purpose of improving the accuracy of network diagnostic result. The process of clustering can be divided into five phases: region division, head selection, coarse clustering, coarse cluster merge and cluster adjustment. Then, in order to further improve the accuracy of diagnostic result, a credibility model based on historical diagnostic result and remaining energy is established for each neighbor node. At last, nodes with higher credibility are selected to participate in diagnostic process. Simulation results show that the proposed algorithm can guarantee higher diagnostic accuracy.