Evolving a Hybrid K-Means Clustering Algorithm for Wireless Sensor Network Using PSO and GAs
Alaa F. Sheta, Basma Solaiman · 2015
Wireless Sensor Networks (WSN) became an essential component in many real-life applications such as military, smart energy, commercial, health and many others. However, WSN still suffer many problems related to energy consumption. Clustering found to be an effective technique to solve the energy consumption problem for WSN by avoiding long distance communication. In this paper, we explore our initial idea on developing a hybrid clustering algorithm which has two folds 1) Use the K-Means unsupervised learning algorithms to select the sensors belonging to each cluster using an arbitrary number of clusters 2) Use Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) separately to select the best CHs. We name these two algorithms as KPSO and KGAs. The developed hybrid algorithms are tested over number of experiments with various layouts. KPSO provided better results compared to the KGAs.