Energy-Aware Clustering and Routing for Wireless Sensor Networks Using Enhanced Osprey Optimization and Multi-Agent Reinforcement Learning

Abstract

Wireless Sensor Networks (WSNs) deployed in large-scale monitoring environments often experience high energy consumption, unstable communication, and reduced quality of service due to inefficient clustering and routing mechanisms. To address these challenges, this work presents an energy-aware communication framework for improving network efficiency and stability in large-scale WSNs. Initially, sensor nodes are organized using Fuzzy C-Means (FCM) clustering integrated with Sailfish Optimization (SFO) to improve cluster formation and communication efficiency. For cluster head selection, an Enhanced Osprey Optimization Algorithm (EOOA) is employed, where the key novelty lies in the integration of adaptive fitness weighting and dynamic exploration-exploitation control to improve cluster head stability, balance energy utilization, and extend network lifetime. Furthermore, Multi-Agent Reinforcement Learning (MARL)-based routing is applied to achieve energy-efficient and reliable communication by reducing delay, congestion, and transmission overhead. The proposed framework enhances energy efficiency, scalability, network stability, and overall communication performance in large-scale wireless sensor networks.