Data aggregation of WSN based on neighborhood support and BP neural network

To improve wireless sensor network (WSN) reliability、robustness and reduce error and redundant information transmission. A dynamic clustering BP neural network Data Aggregation algorithm based on neighborhood support is proposed. The dynamic clustering and the selection of cluster head are based on the state of the node, the neighborhood support and the residual energy of the node. At the same time, in order to reduce communications in WSN, the cluster head uses three-layer neural network for feature extraction of monitoring data, and then sends the feature value to the sink node. Simulation results show that this algorithm can not only improve the reliability and robustness of WSN, but also reduce data redundancy and extend the network life compared with the classic clustering protocol LEACH.