A Smart Fuzzy Metaheuristic Energy Optimisation Framework for Heterogeneous Wireless Sensor Networks

The proposed work introduces an innovative approach to energy management in heterogeneous wireless sensor networks (HWSNs) by integrating fuzzy logic, metaheuristic optimization algorithms, and fuzzy c-means (FCM) clustering. The system aims to optimize energy consumption, enhance adaptability, and improve overall network sustainability and efficiency. The system utilizes FCM clustering to organize sensor nodes into clusters based on similarities in data patterns concerning residual energy, node degree deviation and distance to centrality for cluster head selection and uses residual energy of the node, packet load size and distance to BS for routing path selection. Fuzzy logic is used to address uncertainties and automate factors that play a role in enhancing adaptability, maximizing operational lifetimes, and ensuring balanced energy distribution among heterogeneous nodes. This is extended to a hybrid adaptive neuro-fuzzy inference system (ANFIS) and particle swarm optimization (PSO) algorithm that combines the learning capabilities of ANFIS with the optimization capabilities of PSO to dynamically create the optimized rule base used for energy distribution amongst the heterogeneous nodes in the network. The objectives of the system include maximizing network operational lifetime, minimizing energy consumption, and ensuring balanced energy distribution thereby enhancing overall network performance. Simulation experiments, reflecting real-world HWSN scenarios, demonstrate the effectiveness of the system in extending operational lifetimes, and enhancing overall network performance. By intelligently integrating fuzzy logic, FCM clustering, and the hybrid ANFIS-PSO algorithm, this project presents a promising solution for fortifying the sustainability and efficiency of heterogeneous wireless sensor networks.