Abstract:

For facilitating the efficient routing planning, node management for upper computers or sink nodes, longdistance node service and prolonging the lifetime of wireless sensor network （WSN）, a new method for node fault diagnosis in WSN is proposed. The attribute reduction for decisionmaking of fault diagnosis is found based on the discriminate matrix in rough set theory. The algorithms of node fault classification are proposed based on attribute matching. The fault can be located and diagnosed individually for each node and module in WSN. Simulations show that the proposed method yields high diagnosis accuracy with low communication cost and energy consumption. It is suitable for wireless sensor networks with stringing energy limits.

Key words: fault diagnosis, wireless sensor network, rough set theory, discriminate matrix, attribute reduction