Improved K-harmonic means in wireless sensor networks

A wireless sensor network (WSN) is a network comprising of autonomous sensors that are spatially distributed which are used to monitor the physical or the environmental conditions. Through propagating in its network, the data is being transferred from every sensor to a base station. In our work, we have implemented K-Harmonic means clustering algorithm in a network simulator. K-harmonic means (KHM) is one of the center-based clustering algorithm similar to K-means (KM), but uses the harmonic averages of the distances from each sensor node to the cluster centers defining its performance function with the aim of improving its efficiency and overcoming KM's one of the major drawbacks that is, its high dependency on the initial identification of sensor nodes that represent the clusters. Also, the energy and distance have been taken into consideration while clustering the nodes. Simulation results are obtained by comparing the performance of KHM with LEACH and KM, demonstrating its efficiency.