WinCross: Find the Energy-efficient Crossing Window for UAV with Joint Optimization of Path and Speed

Data collection from a series of IoT ground nodes is a common scenario for Unmanned Aerial Vehicles (UAVs) applications in the wireless sensor network area in recent years. However, a key limitation lies in the energy efficiency of UAV due to the very limited battery capacity in contrast to the long service time demands. Thus, it's highly desired to reduce the energy consumption of UAV by optimizing its path and speed without affecting the completion of data collection tasks under the transmission time requirements. In this paper, we propose WinCross, a novel algorithm that jointly optimizes the path and speed of UAV with minimized energy consumption. WinCross can first derive an energy-efficient crossing window that the UAV should pass through, during the trajectory scheduling. It utilizes a proposed original projection method that can map a 3D problem (GN location and transmission ranges on the 2D plane, plus the minimum transmission time on the temporal dimension) to a 2D problem, that can reduce the optimization complexity remarkably. Secondly, WinCross can further optimize the energy consumption by classifying all feasible crossing window cases and scheduling the corresponding trajectory and velocity respectively. Based on an iterative approach, WinCross can complete all transmission tasks while minimizing the energy consumption. Experiments demonstrate that WinCross can save as much as 22.2% energy over baseline.