Normal Vector Based Fault-Tolerant Event Boundary Detection in Wireless Sensor Networks

This paper presented a distributed fault-tolerant event boundary detection scheme for wireless sensor networks. In this paper, the node coordinate and its reading were mapped on a 3D mesh surface, By comparing the node normal vector got by Normal Vector Based Fault-Tolerant Event Boundary Detection (NVBD) with a predefined threshold, the event boundary nodes are found. This approach was composed of two major tasks: (1) detecting the faulty node and then replacing its detection value with its neighbors´ weighted median value, (2) computing node normal vector using its neighbors´ detection values. Then we compared it with a threshold to judge if it is a boundary node. The results of numerical simulations show that the NVBD scheme can achieve a substantial improvement on detection performance.