arrow
返回

Utility-Aware UAV Deployment and Task Offloading in Multi-UAV Edge Computing Networks

delete2024-04-15
delete9
PRE
AI
Z
Zhufang Kuang
H
Haobin Wang
李洁 封面图
李洁 (Jie Li) *
F
Fen Hou
DOI:10.1109/JIOT.2023.3344570delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Unmanned aerial vehicle (UAV)-enabled mobile-edge computing (MEC) is expected to provide low-latency, ultrareliable, and highly robust network services to improve user service experience. In this article, the UAV deployment, task offloading, and resource allocation problem is investigated in a multi-UAV-enabled MEC system with task-intensive region. UAVs as edge servers to provide computing services for ground terminal devices (TDs). The time-sensitive tasks of TDs can be computed locally or offloaded to UAVs. The goal is to improve the utility of tasks, i.e., maximize the number of tasks offloaded to UAVs under conditions of ensuring a desired task computed success rate and satisfying the energy and latency constraints. The jointly optimizing problem of the 3-D deployment, elevation angle, computational resource allocation of the UAV, and task offloading decision is formulated. To this end, a two-layer optimization approach is proposed to solve the formulated problem. Specifically, the upper layer decides the UAV position, elevation angle, and transmission power of TDs based on the actual ground situation. The lower layer determines the computational resource allocation of UAVs and the task offloading decision based on the optimized results derived from the upper layer. Through the two-layer joint optimization, our goal is finally achieved. Simulation results demonstrate that our proposed algorithm effectively improves the number of tasks offloaded to UAVs and the task completion rate simultaneously with the flexible UAV deployment and well-designed task offloading strategy.
Keyword:
Convex optimization (CO)
differential evolution
mobile-edge computing (MEC)
multi-unmanned aerial vehicle (UAV)
resource allocation
task offloading

期刊

IEEE Internet of Things Journal 封面图
IEEE Internet of Things Journal
IF:
8.9
论文数:
1.4W
被引数:
7.8W

机构

U
University of Macau
学者数:
1.1W
论文数: 1.3W
被引数: 2.0W
引用论文

引用论文

err分享
err收藏
学者 查看更多内容