A Lightweight and Adaptive Energy-Aware Clustering Protocol for Wireless Sensor Networks

Abstract

Energy efficiency remains a critical challenge in Wireless Sensor Networks (WSNs) due to limited, non-rechargeable node batteries. This paper presents a lightweight and adaptive clustering protocol that improves energy utilization and network stability by jointly evaluating residual energy, local neighborhood connectivity (node degree), and proximity to the base station for Cluster Head (CH) selection. An adaptive energy threshold dynamically excludes energy-constrained nodes from election, and a closed-form, energy-dependent probabilistic direct-transmission mechanism supports isolated nodes. MATLAB Monte Carlo simulations (20 independent runs, paired statistical testing) against both the classical LEACH baseline and a modern energy-and-distance-aware benchmark (OEE-WCRD) show a statistically significant improvement in the First Node Dead (FND) metric over LEACH under both uniform (+74.4%, p < 0.05) and non-uniform (+54.8%, p < 0.05 ) deployments, and a statistically significant FND improvement over OEE-WCRD under non-uniform deployment (+7.3%, p < 0.05), which is the realistic scenario this protocol's degree-aware design specifically targets.