ReLeC-based clustering and multi-objective optimization for efficient energy optimization in IoT networks

Abstract

In response to the escalating demand for energy-efficient wireless sensor networks (WSNs) within the expanding Internet of Things (IoT) landscape, we introduce ReLeC-MO, a novel protocol that integrates the ReLeC clustering algorithm with multi-objective optimization. Leveraging reinforcement learning-based clustering, ReLeC optimizes network topology to enhance energy efficiency. Multi-objective optimization further refines this process by identifying non-dominated solutions on the Pareto front, facilitating a balanced trade-off between network lifetime, energy consumption, and data transmission quality. Our comprehensive simulations reveal the remarkable performance improvements achieved by ReLeC-MO over existing techniques. Specifically, ReLeC-MO demonstrates a 39% reduction in delay, a 50% decrease in energy consumption, and a 25% increase in throughput, showcasing its efficacy in enhancing both energy efficiency and network performance. It also increases network lifetime by 20%, surpassing the latest existing model. Furthermore, its implementation in MATLAB ensures ease of replication and adaptation across diverse IoT applications.