作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2021, Vol. 47 ›› Issue (2): 19-25. doi: 10.19678/j.issn.1000-3428.0058085

• 热点与综述 • 上一篇    下一篇

移动边缘计算中的卸载决策与资源分配策略

杨天, 杨军   

  1. 宁夏大学 信息工程学院, 银川 750021
  • 收稿日期:2020-04-16 修回日期:2020-06-05 出版日期:2021-02-15 发布日期:2020-05-21
  • 作者简介:杨天(1995-),男,硕士研究生,主研方向为移动边缘计算、普适计算;杨军(通信作者),教授。
  • 基金资助:
    赛尔网络下一代互联网技术创新项目(NGII20161206)。

Offloading Decision and Resource Allocation Strategy in Mobile Edge Computing

YANG Tian, YANG Jun   

  1. School of Information Engineering, Ningxia University, Yinchuan 750021, China
  • Received:2020-04-16 Revised:2020-06-05 Online:2021-02-15 Published:2020-05-21

摘要: 为在移动边缘计算服务器计算资源有限的情况下最小化系统总成本,提出一种多用户卸载决策与资源分配策略。优化任务执行位置选择和计算资源分配过程,对基于精英选择策略的遗传算法在编码、交叉、变异等操作方面进行改进,设计联合卸载决策与资源分配的improve-eGA算法。实验结果表明,与All_local、All_offload、RANDOM和CGA等算法相比,improve-eGA在迭代次数、任务周期数、任务传输数据量等影响因素下系统总成本均为最低,验证了所提策略的有效性。

关键词: 移动边缘计算, 计算资源, 卸载决策, 资源分配, 遗传算法

Abstract: In order to minimize the total system cost when the computing resources of Mobile Edge Computing(MEC) servers are limited,this paper designs a multi-user offloading decision and resource allocation strategy.The strategy jointly optimizes the selection of task execution location and the allocation of computing resources,and improves the encoding,crossover and mutation parts of the Genetic Algorithm(GA) based on the elite selection strategy(e-GA).On this basis,improve-eGA algorithm is designed combining offloading decision and resource allocation.Experimental results show that,compared with the ALL_Local algorithm,ALL_Offload algorithm,RANDOM algorithm and Conventional Genetic Algorithm(CGA),etc.,improve-eGA has the smallest total system cost under the influence of the iteration number,CPU working frequency,transferred data size of tasks,etc.,which verifies the validity of the proposed strategy.

Key words: Mobile Edge Computing(MEC), computing resource, offloading decision, resource allocation, Genetic Algorithm(GA)

中图分类号: