Energy-Efficient Clustering in Wireless Sensor Networks: A Multi-Objective Approach Using PSO and Fuzzy Logic

Energy efficiency is one of the biggest challenges in WSNs; typically, these networks are plagued with several problems such as uneven distribution of energy amongst nodes, hotspots, and clustering with big disparities. This work tries to some extent solve some of these problems through a prototype: Efficient Wireless Sensor Network with Multi-Objective Clustering, or EEMOC, which is a new framework that exploits the power of Particle Swarm Optimization to realize optimal cluster head selection. In addition, it uses the FIS in the dynamic selection of a cluster radius to maximize its effectiveness in networking. EEMOC exploits the fuzzy logic benefits in governing uncertainty in data and optimizes the sizing of clusters, thereby controlling hotspots and optimizing the energy distribution scheme. Simulation results show EEMOC to be far superior to traditional clustering algorithms and may significantly extend the network lifetime with the utmost saving in energy under various operating conditions. The technique which integrates PSO with FIS provides a robust platform for improving WSN performance.