Multivariate Lifetime Prediction Model for Energy Efficient Region-Based Wireless Sensor Network

Abstract

In wireless sensor networks (WSNs), optimising energy efficiency while maintaining coverage and managing resource constraints remains a critical challenge. This paper introduces a novel Region-Based Multilevel Energy Efficiency Protocol (RBMEEP), which innovatively partitions the network into regions and sub-regions to enhance energy utilisation through optimised clustering and communication with the base station (BS). Unlike conventional protocols, RBMEEP significantly extends network lifetime, outperforming the Stable Election Protocol (SEP). The novelty lies in the integration of a Regression Prediction Model (RPM), which accurately predicts network lifetime based on node density and packet size. Simulation results demonstrate the model's high prediction accuracy, with up to 99.94% in smaller network areas and 99.87% in larger areas. This predictive capability allows for adaptable and efficient WSN design, tailored to specific user requirements. The proposed approach presents a significant advancement in extending the operational life of WSNs, offering a robust solution for energy and coverage optimisation. This work not only improves the theoretical understanding of WSN energy efficiency but also provides a practical framework that can be deployed in real-world scenarios.