Throughput-aware path planning for UAVs in D2D 5G networks

Unmanned Ariel Vehicles (UAVs) face increasing challenges in obtaining sensory data and transferring them to the user even before the completion of their flight for time-critical processing. Traditionally bounded by only area coverage and battery capacity, UAVs now need to meet network QoS requirement when streaming data. The emergence of 5G Device-to-Device (D2D) Networks enables high speed network communication for UAVs to transfer data via D2D links during a flight. The planning of UAV flight paths is now subject to both battery capacity and network quality of service (QoS) constraints. In this paper, we focus on the path planning for UAVs, which stream data to a data receiver machine, under the constraints of full area coverage and network throughput. We present a mathematical model to formulate the issue as a combinatorial optimization problem that attempts to minimize the flight cost of multiple UAVs covering the entire area. We show that the problem is NP-hard, therefore propose a heuristic method to derive the number of UAVs and determine their flight paths. The solution is proved to be bound by K.OPT. We conduct simulations to evaluate how the size of the area and the maximal flight distance of a UAV affect the number of UAVs needed, and how the D2D channel parameters affect the link throughputs.

Automated summary

This paper focuses on the path planning for UAVs during flight to meet the requirements of data transmission and network communication. A mathematical model is proposed to solve the combinatorial optimization problem of multiple UAVs covering the entire area, and the solution is theoretically and empirically evaluated.