An Efficient Scheduling Scheme for Semi-On-Demand Charging in Wireless Rechargeable Sensor Networks

It is one of invariable main challenges in charging scheduling for WRSNs to improve charging efficiency by using charging capability of wireless charging vehicles (WCVs) efficiently. Proactive charging-enabled on-demand charging scheme is developed through solving such challenges. While replying to on-demand charging requests (CRs) preferentially, SoC scheme includes proactive charging for the potential Bottleneck Nodes (pBNs), although sensor nodes do not generate CRs, so long as WCV has redundant capability. The existing scheme, however, not only fails to predict the pBNs exactly owing to use of a fixed deadline threshold, but also selects the proactive charging nodes randomly among the predicted pBNs, thus leaving space for improving charging and network performance further. This paper proposes a new SoC scheduling algorithm using FAHP-VWA and Q-Learning to solve these issues. The proposed algorithm applies FAHP-VWA to assign the weights to multi-criteria characterizing the pBNs and then predict the pBNs being homologous to WCV's charging capability based on these weights exactly. Then, in optimal proactive charging node selection with Q-Learning, these weights are also used to design the reward function and select the most suitable pBNs that would be included in a charging round. Extensive experiment results verify that the proposed scheme improves the whole charging and network performance in comparison to other existing scheme.