Objective-oriented intelligent charging coordinator for on-demand WRSNs

Abstract

Abstract While wireless energy transfer offers a practical solution to the problem of energy scarcity in Wireless Sensor Networks (WSNs), the efficient management of Wireless Rechargeable Sensor Networks (WRSNs) in smart environments still faces two main inhibiting challenges: the inefficiency of single-charger architectures in large-scale deployments (lack of scalability) and the absence of adaptive, real-time scheduling mechanisms. Motivated by this, this paper introduces the Objective-Oriented Intelligent Charging Coordinator (OOICC), an innovative scheduling and charging resource allocation framework designed to overcome these limitations through a comprehensive two-layer strategy. By fundamentally rethinking network management, OOICC first dynamically determines the optimal number of Mobile Chargers (MCs) required for scalable coverage, moving beyond the conventional limitations of single chargers. Second, it establishes a hybrid offline-online decision-making architecture that synergistically combines strategic planning with operational adaptability. At the core of OOICC lies a formally defined multi-objective optimization framework that incorporates designer-specified characteristics into a comprehensive ten-fold objective function. This ensures the balanced pursuit of critical metrics such as charging latency, node survival rate, and network throughput. The coordinator operates through an intelligent workflow whereby the offline phase employs fuzzy clustering guided by a metaheuristic optimizer to create an optimal initial network partition and MC assignment. This framework actively interacts with an online phase, where an intuitive fuzzy logic engine performs context-aware charging scheduling by processing live network data in real-time. Through extensive simulations, OOICC demonstrates superior and consistent performance compared to established benchmarks. The method achieves significant improvements in charging response time, node survival rate, energy consumption efficiency, network stability, and data throughput. Quantitatively, the proposed framework outperforms the TSFM, CFMCRS, FLCSD, ESS, and NJNP methods by 8%, 48%, 68%, 73%, and 89%, respectively, confirming its efficacy as a robust and scalable solution for next-generation WRSN management.