Optimal Continuous Trajectory Design for UAV-Assisted Reliable Communication Operating with Finite Blocklength Codes

Abstract:
In this paper, we consider an unmanned aerial vehicle (UAV) assisted low latency communication scenario with a single ground user. We assume the UAV operates at a fixed altitude and has fixed starting and ending points for providing the low latency service to the user. Within the finite blocklength (FBL) regime and based on a practical probabilistic line-of-sight (LoS) model, we formulate an achievable throughput maximization problem under a reliability constraint. With a given UAV operation period, we target at optimizing the continuous UAV trajectory. To optimally address the problem, we particularly define an artificial potential field (APF) and correspondingly transform the UAV trajectory design problem to a mechanical problem for optimizing the mechanical rope shape, which can be efficiently solved based on the physical principles for equilibrium state. Afterwards, the optimal UAV trajectory is equivalently reconstructed from the optimal rope solution, while the optimality is reserved. Our proposed solution is finally validated via simulations.