UAV-Assisted Edge Computing with 3D Trajectory Design and Resource Allocation

Pengle Wen,Xiaoyan Hu,Kai-Kit Wong

DOI: https://doi.org/10.1109/vtc2023-fall60731.2023.10333504

2023-01-01

Abstract:With the explosive increase in computing demands and the rise of portable wearable devices, the concept of mobile-edge computing (MEC) has emerged and attracted a lot of attention from both academia and industry. Unmanned Aerial Vehicle (UAV) as flexible moving platform has been wide adopted as an edge computing server to help ground users compute their intensive tasks. Although UAV-assisted edge computing is capable to enhance the computing performance, there are still many challenges in this system, including UAV 3D trajectory design, the allocation of UAV computational resources and the communication time allocation between users and UAV. In this article, we try to solve these challenges in a UAV-assisted edge computing system, aiming at minimizing the completion time of computing users’ tasks. Specially, we propose a combination algorithm of the alternating optimization method and the bisection search method to minimize the delay of the whole system. The whole algorithm can be described in two iterative steps. In the first step, with given total number of time slot N assuming each slot with fixed length, we check whether the current N can satisfy the computational demands of the whole system through the alternating optimization algorithm to obtain the computational and time allocation. In the second step, we use the resource allocation results obtained in the first step to choose whether to increase or decrease N via the bisection search method. Then we repeat the first and second steps until we find the the smallest N that best fits the current computational demand. Extensive experimental results demonstrate that our proposed algorithm greatly reduces the users’ task completion time in comparison with traditional benchmarks. In addition, the convergence of the proposed algorithm can be guaranteed.