A hybrid charging scheme for efficient operation in wireless sensor network

The adoption of Wireless Sensor Networks (WSN) is persistently increasing over various applications. Being an integral part of the Internet of Things, prolonged sustainability of a sensor node is highly demanded. In this perspective, energy depends on supporting high-end sustainability factors in the WSN environment. In many recent studies, wireless transfer of energy concepts has contributed to ensuring on-demand charging of rechargeable sensor nodes to facilitate a better data aggregation process. A review of existing studies on the wireless transfer of energy exhibits adopting the clustering approach, scheduling-based approach, approaches with ambiguous assumptions, and sophisticated techniques. Most recently, it has been noticed that learning-based schemes have contributed to facilitating better charging strategy; however, they are also associated with iterative learning operation that adversely affects the operational cost of charging. Therefore, the proposed scheme has introduced a novel and simplified hybrid charging scheme considering the presence of normal and critical sensor nodes in the deployment area. A reinforcement learning scheme is used for optimizing the policy construct toward formulating a better charging strategy. At the same time, the study implements Type-I and Type-II chargers based on their static location and mobility, respectively. With an extensive simulation environment, the proposed scheme has been benchmarked with the existing charging scheme to find that the proposed hybrid scheme offers reduced operational cost, enhanced battery span, and higher sustainability in contrast to the current system.