A Novel Cluster-Based Data Fusion Algorithm for Wireless Sensor Networks

The characteristics such as high reliability, high scalability, fault tolerance, low cost and rapid deployment make wireless sensor networks (WSN) useful in many military and civilian applications. The limitation of sensor nodes energy necessitates that energy saving is a key issue in WSN. This paper proposes a novel cluster-based data fusion algorithm for WSN. It divides network into grids, and selects one cluster-head from each grid by considering residual energy of nodes, afterward, each non-cluster-head node selects its cluster-head basing on the signal strength of advertisements sent by cluster-heads to form clusters, finally, every cluster-head fuses the data received from its member nodes and transmits the fused data to base station. This algorithm conduces to uniform cluster-heads distribution and rational cluster formation, consequently, achieves better performance on both energy efficiency and lifetime of WSN.