부산대학교 도서관

Small cells (SCs) mounted on top of the unmanned aerial vehicles (UAVs) are a promising solution to boost the capacity in hotspot areas. However, the adoption of UAV-SCs involves the planning of their missions over time, which includes the scheduling of recharging actions of each UAV-SC at ground sites. Typically, the energy needed to recharge UAV-SCs is derived from the grid, which can be coupled with microgeneration exploiting renewable energy sources (e.g., solar panels). In this architecture, the energy that is locally produced can be either sold to the grid or used to recharge the UAV-SCs. On the other hand, when the energy from microgeneration is insufficient for recharging the UAV-SCs, additional energy can be bought from the grid. In this paper, we investigate the trade-off between maximizing the throughput provided by the UAV-SCs over a set of areas, maximizing energy sold to the grid, and maximizing energy bought from the grid. The proposed model, MaxUAVProfit, is designed to (i) plan the UAV-SCs missions as a sequence of positions and actions in 3D space vs. time, (ii) manage the grid-connected microgeneration, and (iii) control the amount of throughput received by each hotspot. We then evaluate the MaxUAVProfit in a realistic scenario, which is based on the measurement of real cellular metrics and a realistic UAV-SC energy consumption model. Our findings demonstrate the superiority of the MaxUAVProfit with respect to other competing solutions, which include either optimization of microgeneration or maximization of the area throughput.