Hybrid Neuro-Fuzzy and Swarm Optimization-Based Energy-Efficient Routing for Large-Scale IoT Networks

Abstract

The proliferation of large-scale Internet of Things (IoT) networks has raised various concerns about energy usage, routing efficiency, scalability and lifetime. However, traditional routing protocols like LEACH, HEED, PEGASIS, and RPL can face challenges in maintaining optimal performance in dense IoT networks, as they tend to consume significant amounts of energy, have high communication overheads, and are less flexible in adapting to changing network conditions. For overcoming these issues, the current paper presents Hybrid Neuro-Fuzzy and Swarm Optimization (HNF-SO) routing protocol that combines Neuro-Fuzzy intelligence with Whale Optimization Algorithm (WOA) to improve routing decision making and global route optimization. The proposed architecture is based on hierarchical clustering network in which a Neuro-Fuzzy engine is used to calculate the energy efficiency of all considered routing candidates, to calculate the distance to the sink, to calculate the node density and to calculate the traffic load, while the WOA is used to identify the most energy-efficient cluster heads and routing paths. A large-scale IoT network scenario was used throughout the extensive simulations, and the protocol was assessed based on the energy consumption, network lifetime, residual energy, packet delivery rate (PDR), throughput, delay, and scalability. The results show that the proposed HNF-SO significantly improves the packet delivery performance, lowers the energy consumption, provides the long lifetime of the network, and can operate without any disruption even with the increase of the network size compared with the traditional routing protocols. The key contribution of this work is the development of an intelligent, scalable and energy-efficient solution for large scale deployments of IoTs by combining the algorithms of Neuro-Fuzzy decision making and swarm based optimization into a single routing framework.