Energy-efficient and Low Package Loss Clustering in UAV-assisted WSN using Kmeans++ and Fuzzy Logic

Clustering method has attracted wide attention in the area of wireless sensor networks (WSNs) due to its significant contributions to balance the loads and prolong the lifetime of sensor nodes. This paper focuses on the clustering method in UAV-assisted WSNs, in which UAV can fly to a sensor node to collect data and then fetch the collected data to the nearby base station. We firstly execute the clustering procedure using Kmeans++ considering the uneven distributions of sensor nodes in the real world, which is different from LEACH and grid divide method typically used in previous clustering methods. Secondly, we apply the fuzzy logic method to dynamically choose cluster heads according to the remaining energy and storage of sensor nodes. Simulation results show that our method performs better on the lifetime prolongations and package loss reductions compared to LEACH, Voronoi divide algorithm, and grid divide algorithm.