LEACH-G: Local-Support-Aware Sequentially Inhibited Cluster-Head Election for Energy-Efficient Wireless Sensor Networks

Abstract

Wireless sensor networks (WSNs) are frequently deployed in unattended environments where battery replacement is difficult. Low-Energy Adaptive Clustering Hierarchy (LEACH) reduces redundant transmissions through clustering and cluster-head rotation, but its cluster-head election remains highly random and can select low-energy nodes, generate spatially crowded cluster heads, and introduce repeated topology-related control overhead. This paper presents LEACH-G, a lightweight and interpretable improvement of LEACH for static center-base-station WSNs. The proposed protocol caches the inter-node topology at the base station during initialization and updates only residual-energy states in later rounds. A local support factor is designed to characterize the structural support capability around each candidate node by jointly considering candidate residual energy, neighboring residual energy, and distance-weighted neighborhood relationships. A sequential inhibition mechanism then selects cluster heads one by one and suppresses nearby candidates after each selection, thereby reducing local cluster-head crowding. MATLAB simulation under a 100 m x 100 m center-base-station scenario with 100 uniformly distributed homogeneous nodes shows that LEACH-G delays the first node death to round 723 and achieves an energy efficiency of 1507.36 packets/J. Compared with classical LEACH, the first node death time and energy efficiency are improved by 34.64% and 9.88%, respectively. Compared with a no-policy random cluster-head baseline, the first node death time is improved by 42.04%. The results indicate that local-support-aware and sequentially inhibited cluster-head election can improve early-stage stability and effective energy utilization while preserving the simplicity of the LEACH framework.