ReLeC‐MEO: Reinforcement Learning‐Based Clustering With Multi‐Objective Efficient Optimization for Energy‐Efficient IoT Networks

ABSTRACT In response to the increasing need for energy‐efficient wireless sensor networks (WSNs) in the quickly evolving Internet of Things (IoT) arena, we introduce ReLeC‐MEO, a new protocol that combines the ReLeC clustering approach with multi‐objective efficient optimization. ReLeC‐MEO improves energy efficiency by using clustering based on reinforcement learning to optimize network design. By finding non‐dominated solutions on the Pareto front, multi‐objective optimization enhances this procedure even more and guarantees a just trade‐off between data transmission quality, energy consumption, and network lifetime. Numerous simulations verify that ReLeC‐MEO works noticeably better than current techniques. In comparison to baseline protocols, it specifically achieves a 42.9% reduction in latency, a 51.6% drop in energy consumption, and a 35% increase in throughput. It also outperforms the next best protocol by 20.4% in terms of network longevity.