Evolutionary Optimisation of Energy-Efficient Communication in Wireless Sensor Networks

摘要: Many real-world problems can be efficiently optimised using a multi-objective function—as these are simultaneously improved multiple objectives, which most often preclude each other. A single-objective function incorporating all information required to solve the problem appears appropriate, but not without penalties of slow convergence and difficulty in obtaining best fitness function. This paper therefore implements hybrid evolutionary system that minimises penalties. We conscript two distance functions, improve communication between sensor nodes cluster heads (CHs), CHs sink or base station. These functions then mainstreamed into globally defined genetic algorithm (GA). Important parameters established by GA topology preserved serve variety modified particle swarm optimisation (PSO) models, discover how suitable they reshape process. Simulation results revealed robustness our proposed framework, as framework enabled consistent coverage clustering topology. The could maintain good diversity genealogy across population generations, clustered network presented stable structure such mobile do unnecessarily exceed global boundary. PSO-fitness guaranteed particles maintained shortest possible within (population) space. Furthermore, PSO with Time Varying Inertia Weight Constriction factor (PSO-TVIW–C) achieved tremendous improvements overall performance is effective solving minimisation wireless networks (WSNs).