A Long-Distance First Matching Algorithm for Charging Scheduling in Wireless Rechargeable Sensor Networks

Abstract

In large wireless rechargeable sensor networks (WRSNs), the limited battery capacity of sensor nodes and finite network lifetime are commonly considered as performance bottlenecks. Previous works have employed wireless mobile vehicles (vehicles) to charge sensor nodes (nodes), but they face limitations in terms of low speed and offroad terrain. The rapid development of wireless charging drones (drones) brings a new perspective on charging nodes; nevertheless, their use is limited by small capacity batteries and cannot cover large regions alone. Most existing works consider the charging of nodes only with vehicles or drones. However, these solutions may not be robust enough, as some nodes’ energy will have run out before vehicles’ or drones’ arrival. Considering the merits and demerits of vehicles and drones comprehensively, we propose a novel WRSN model whose charging system integrates one vehicle, multiple drones and one base station together. Moreover, a charging strategy named long-distance first matching (LDFM) algorithm to schedule the vehicle and multiple drones collaboratively is proposed. In the proposed scheme, drones that are carried by the vehicle start from the base station. According to distance and deadline of nodes with charging requests, LDFM prioritizes nodes with the longest matching distance for allocation to drones. As a result, the proposed scheme aims to minimize the moving distance of charging scheduling of the WCV on premise of satisfying charging requests with the cooperation of WCVs and drones. Our proposed scheme is thus designed to maximize the efficiency of drone usage and shares the charging burden of the vehicle, which makes WRSNs work well in large and complex terrain regions, such as a hill, natural disaster areas or war zones. Simulation results confirm that our proposed scheme outperforms hybrid scheme in previous work with respect to total number of charging nodes and network energy consumption. Especially with heavy traffic load, the proposed scheme can charge more than 10% additional nodes compared to the hybrid. Moreover, the proposed scheme achieves a reduction of over 50% in the moving distance compared to the hybrid.