WSN data fusion approach based on improved BP algorithm and clustering protocol

Network energy consumption is a critical factor determining the WSN development speed. In this paper, a novel approach of neural network and wireless sensor network combination for the inner data integration is adopted in order to effectively improve data transmission efficiency, reduce network energy consumption. Firstly, a kind of clustering protocol called CNN-LEACH based on Hamming network and a kind of optimization algorithm called SMPSO-BP based on neural network are proposed. Then, the CNN-LEACH clustering routing protocol integrated with SMPSO-BP optimization algorithm is applied in the WSN data fusion process. The above-mentioned protocol and algorithm under different scenarios are simulated and compared on the NS2 platform. The result show that, SMPSO-BP algorithm has improvement in convergence and CNN-LEACH protocol really balance the energy consumption of the network load to some extent. Finally, Their combination reduce the redundant data in WSN and the energy consumption of senor node and prolong the network lifetime.