物联网学报 ›› 2021, Vol. 5 ›› Issue (1): 27-35.doi: 10.11959/j.issn.2096-3750.2021.00190

所属专题: 边缘计算

• 专题:物联网边缘智能与雾计算 • 上一篇    下一篇

多无人机辅助移动边缘计算中的任务卸载和轨迹优化

嵇介曲, 朱琨, 易畅言, 王然   

  1. 南京航空航天大学,江苏 南京210016
  • 修回日期:2021-02-06 出版日期:2021-03-30 发布日期:2021-03-01
  • 作者简介:嵇介曲(1993- ),女,博士,南京航空航天大学博士生,主要研究方向为无人机辅助通信、移动边缘计算、云计算、工业互联网、主动缓存、无人机安全通信等
    朱琨(1984- ),男,博士,南京航空航天大学教授,主要研究方向为5G和超5G网络中资源管理和调度、无线虚拟化、自组织网络、无人机辅助通信等
    易畅言(1989- ),男,博士,南京航空航天大学教授,主要研究方向为无线通信网络、移动边缘计算、云计算、工业互联网、网络资源管理和调度等
    王然(1989- ),男,博士,南京航空航天大学副教授,主要研究方向为无线通信网络、智能电网中智能管理和控制、网络性能分析、电动汽车互联等
  • 基金资助:
    国家自然科学基金资助项目(62071230);国家自然科学基金资助项目(62061146002)

Joint task offloading and trajectory optimization for multi-UAV assisted mobile edge computing

Jiequ JI, Kun ZHU, Changyan YI, Ran WANG   

  1. Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Revised:2021-02-06 Online:2021-03-30 Published:2021-03-01
  • Supported by:
    The National Natural Science Foundation of China(62071230);The National Natural Science Foundation of China(62061146002)

摘要:

提出了一个多无人机辅助移动边缘计算系统,系统部署了多个配备计算资源的无人机为本地资源有限的移动用户提供计算卸载机会。每个用户的计算任务分为两部分:一部分卸载到关联的无人机进行计算,剩余部分在本地进行计算。旨在联合优化一个有限周期内的无人机轨迹和用户调度来最小化所有用户间最大的总时延。所提出的问题是一个混合整数非凸优化问题,为了有效求解该问题,通过引入一些辅助变量将其等价地转化为一种更易处理的形式,然后提出了一个惩罚凹凸过程的算法来求解转化后的问题。数值结果表明,与其他基准方案相比,提出的联合优化方案能获得显著的性能增益。

关键词: 无人机, 移动边缘计算, 轨迹设计, 用户调度

Abstract:

An unmanned aerial vehicle (UAV)-assisted mobile edge computing system was proposed in which multiple UAVs equipped with computing resources were employed to provide computation offloading opportunities for mobile users with limited local resources.The computing tasks of each user can be divided into two parts.One portion was offloaded to its associated UAV for computing and the remaining portion was processed locally.It was aimed at minimizing the sum of the maximum delay among all user devices by jointly optimizing the user scheduling and the UAV trajectory in a finite period.The proposed problem was a mixed-integer non-convex optimization problem.To facilitate solving this problem, it was equivalently converted into a more tractable problem by introducing some auxiliary variables, and then a penalty concave-convex procedure algorithm was proposed to solve the converted problem.Simulation results show that the proposed joint optimization scheme achieves significantly better performance than other benchmark schemes.

Key words: unmanned aerial vehicle, mobile edge computing, trajectory design, user scheduling

中图分类号: 

No Suggested Reading articles found!