Long-Term throughput optimization in WPCN with battery-powered devices

Wireless powered communication networks are becoming an effective solution for improving self sustainability of mobile devices. In this context, a hybrid access point transfers energy to a group of nodes, which use the harvested energy to perform computation or transmission tasks. While the availability of the wireless energy transfer mechanism opens up new frontiers, it also requires an appropriate choice of the network parameters (e.g., transmission powers, transmission duration, amount of transferred energy, etc.) in order to achieve high performance. In this work, we study the throughput optimization problem in a system composed of an access point which recharges the batteries of two devices at different distances. In the literature, the main focus so far has been on slot-oriented optimization, in which all the harvested energy is used in the same slot in which it is harvested. However, this approach may be strongly suboptimal because it does not exploit the possibility to store the energy and use it at a later time. Thus, instead of considering the slot-oriented case, we address the long-term maximization. This assumption greatly increases the optimization complexity, as it requires to consider, e.g., the channel state realizations, its statistics and the batteries time evolution. Our objective is to find the best scheduling scheme, both for the energy transferred by the access point and for the data sent by the two nodes. We discuss how to perform the optimization and show that the slot-oriented policies proposed so far are strongly sub-optimal in the long-term case. Our scenario can be considered as a first step toward the study of more complex and distributed schemes in wireless energy-transfer scenarios in the presence of battery-powered nodes.