多无人机协同陆地设施辅助移动边缘计算的系统能耗最小化方法  

Energy Minimization for Multi-UAVs Cooperative Ground Access Points Assisted Mobile Edge Computing

作　　者：陈阳[1] 皮德常[1] 代成龙[1,2] 李本田 王碧 薛乔 CHEN Yang;PI De-chang;DAI Cheng-ong;LI Ben-tian;WANG Bi;XUE Qiao(College of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing,Jiangsu 211100,China;School of Artificial Intelligence and Computer Science,Jiangnan University,Wuxi,Jiangsu 214028,China;School of Computer Science and Engineering,Southeast University,Nanjing,Jiangsu 211189,China;Department of Electronics and Information Engineering,The Hong Kong Polytechnic University,Hong Kong 999077,China)

机构地区：[1]南京航空航天大学计算机科学与技术学院,江苏南京211100 [2]江南大学人工智能与计算机学院,江苏无锡214028 [3]东南大学计算机科学与工程学院,江苏南京211189 [4]香港理工大学电子与资讯工程学院,中国香港999077

出　　处：《电子学报》2023年第4期984-992,共9页Acta Electronica Sinica

基　　金：国家科技创新2030“新一代人工智能”重大项目(No.2021ZD0113103)。

摘　　要：无人机作为移动基站辅助边缘计算可为用户设备提供广泛的服务范围和额外计算能力,本文提出一种多无人机协同陆地设施辅助边缘计算的系统.该系统将多架无人机作为移动基站,来协同多个陆地设施对移动用户提供计算卸载服务.系统分为局部计算模型、无人机计算模型、陆地设施计算模型以及无人机盘旋能耗模型.目的是优化多个无人机的位置和用户的卸载决策使得系统总体能耗最小.为求解该问题,提出一种多子群驱动的均衡优化算法.该方法基于两个子种群演化交互,集成了变异和种群重启机制,具有良好的优化能力.仿真实验表明,提出的算法能更好降低系统能耗.UAV(Unmanned Aerial Vehicle/Drones)-assisted mobile edge computing can provide extensive coverage and additional computing power to user devices.In this paper,we study a system of multi-UAVs collaborative ground facil-ities-assisted mobile edge computing.The system provides offloading computing for user equipment through multiple UAVs in collaboration with multiple ground facilities.The system is divided into the local computing model,UAVs com-puting model,Ground computing model,and UAVs energy consumption model.The objective is to optimize the UAVs'lo-cations and user offloading decisions to minimize the system energy consumption.The system energy minimization is a large-scale mixed integer optimization problem.To solve the problem,we propose a multi-subgroup driven equilibrium op-timizer.The algorithm incorporates two subgroup evolutionary interactions,mutation and population restart mechanisms.Experiments show that the proposed algorithm can better reduce the system energy consumption compared with several oth-er swarm intelligence algorithms.

关 键 词：无人机辅助边缘计算 能耗最小化 均衡优化 多子群驱动 群智能优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]