Fuzzy Logic Based Smart Cluster Head Selection and Enhanced Particle Swarm Optimization for Energy Optimization in WSNs

Abstract

Wireless Sensor Networks (WSNs) are the fundamental part of the majority of applications like environmental monitoring, industrial automation and healthcare. These networks are highly power-constrained in operation and hence, resource-efficient protocols are designed in a manner that the network remains on for a prolonged duration period. Maximum Optimal Cluster Head (CH) selection is the easiest WSN problem of achieving even power consumption and enhanced network performance. Existing cluster protocols cause energy distribution imbalance. Thus, network stability is decreased. In this paper, Fuzzy Logic (FL)-based Cluster Head selection strategy and Enhanced Particle Swarm Optimization (EPSO) algorithm are utilized with the objective of reducing WSNs power consumption. Fuzzy Logic (FL) controller approximates the parameters like residual energy, base station distance and node population with the objective of making optimum CH decisions. Concurrently, the EPSO algorithm adaptively updates inertia weight and hybrid mutation strategies to rectify the optimization process for evading premature convergence and local optimum issues. Long-term simulation is performed to compare the newly introduced FL-EPSO scheme with other existing network lifetime, energy and Packet Delivery Ratio (PDR)-based clustering schemes. Results show that the proposed FL-EPSO model reduces energy consumption distribution, network life and communication efficiency to a remarkable degree. Adaptive CH selection mechanism provides real-time adjustment ability, thus making the model highly suitable for practical WSN applications. Experimental results show that FL and EPSO combined is a highly effective and scalable method of energy-aware clustering in WSNs. The hybrid models are suitable candidates to design and optimize for the future in order to have optimal performance.