Fuzzy Based Dynamic Clustering in Wireless Sensor Networks

In recent times the incorporation of Wireless Sensor Network (WSN) with Internet of Things (IoT) has become more conscientious for the researchers. The collection of enormous amount of homogenous sensor nodes forms the Wireless Sensor Network. These sensor nodes have restricted battery power and memory and so the limited amount of energy is considered as the major issue. To overcome this issue several mechanisms were proposed, among them clustering is a popular way which minimizes the consumption of energy in the sensor nodes and thus the life span of the Wireless Sensor Network can be increased. Grouping the sensor nodes in an energy efficient and distributed approach is considered as the important issue in clustering. Hotspots problem and energy hole problem are the problems faced by the non-distributed clustering. So in order to triumph over these issues, a Fuzzy Based Dynamic Clustering (FDC) in Wireless Sensor Network is proposed. Also a new fitness function for Particle Swarm Optimization (PSO) is proposed which discovers the possible cluster heads The hotspots problem and energy hole problem is overcome by the Fuzzy Inference System (FIS) which chooses unique radius for each cluster head thus unequal clustering is formed. A fair comparison is done between this proposed algorithm and some existing algorithms. The simulation results obtained reveals that our proposed algorithm increases the lifetime and has better energy efficiency.