Fault Diagnosis for Wireless Sensor Network´s Node Based on Hamming Neural Network and Rough Set

To accurately diagnose node fault in wireless sensor network (WSN) can improve long-distance service of nodes in WSN, assure reliability of information transfer and prolong lifetime of WSN. In this paper, a novel method of fault diagnosis for node of WSN was brought forward. First, attribute reduction for decision-making of fault diagnosis could be founded based discernibility matrix in rough set theory. Furthermore, a set of model for node´s fault diagnosis in WSN could be built through classification algorithm based on attribute matching. Finally, a set of method for fault classification was founded by hamming network. The result of simulation shows that characteristics of this method are as follows: high veracity of diagnosis, a little expenditure of communication, low energy consumption and strong robustness.