Study on distributed data mining model in Wireless Sensor Networks

Aiming at the severe energy and computing resource constraints of Wireless Sensor Network (WSN), based on rough set theory and ART2 network, a distributed data mining model for WSN is proposed. This model poses a three-layer MLP for data aggregation in the clustered sensor network. And the input layer neuron and the first layer neuron are located in every cluster member, while the second layer neuron and the output layer neuron are located in every cluster head. The features of the training samples were extracted to build up the decision table; the rough set theory was applied to reduce the decision table. Finally, the reduced decision attributes were used to construct ART2 neural network classification data. Constructed data mining algorithm can be integrated in each sensor network node. Simulation results prove data dimension is reduced and data redundancy is eliminated after the raw-data is processed by data mining algorithm, and the communication traffic is decreased and the life of WSN is extended.