A multi-level data fusion algorithm for wireless sensor networks

Data fusion is an important technology in Wireless sensor network (WSN) to enhance data collection, integration, and transmission. In this paper, we propose a multi-level, real-time data fusion algorithm for wireless sensor networks. Based on Variable Step Size Least Mean Square Error (VSS LMSE) and Dempster-Shafer (DS) Evidence Theory, this algorithm realizes the real-time acquisition and processing of multi sensor data in a parallel manner and implements complementary optimization of data from the distributed WSN. The results of our experiment study show that our method can lead to fast convergence and with small errors. By effectively eliminating redundant data without sacrificing the performance of a WSN, this algorithm can potentially reduce the energy consumption of the whole network and prolonging the network lifetime.