摘要
针对移动边缘计算(MEC),提出了一种基于机器学习的随机任务迁移算法,通过将任务划分为可迁移组件和不可迁移组件,结合改进的Q学习和深度学习算法生成随机任务最优迁移策略,以最小化移动设备能耗与时延的加权和.仿真结果表明,该算法的时延与能耗加权和与移动设备本地执行算法相比节约了38. 1%.
For mobile-edge computing(MEC),a machine learning-based stochastic task offloading algorithm was proposed.By dividing the task into offloadable components and unoffloadable components,the improved Q learning and deep learning algorithm were used to generate the optimal offloading strategy of stochastic task,which minimized the weighted sum of energy consumption and time delay of the mobile devices.The simulation results show that the proposed algorithm saves the weighted sum of energy consumption and time delay by 38.1%,compared to the local execution algorithm.
作者
孟浩
霍如
郭倩影
黄韬
刘韵洁
MENG Hao;HUO Ru;GUO Qian-ying;HUANG Tao;LIU Yun-jie(Beijing Advanced Innovation Center for Future Internet Technology,Beijing University of Technology,Beijing100124,China;State Key Laboratory of Networking and Switching Technology,Beijing University of Posts and Telecommunications,Beijing100876,China)
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2019年第2期25-30,共6页
Journal of Beijing University of Posts and Telecommunications
基金
北京市科技新星计划项目(Z151100000315078)
国家科技重大专项项目(2018ZX03001019-003)
国家高技术研究发展计划(863计划)项目(2015AA015702)
关键词
移动边缘计算
随机任务迁移
机器学习
时延
移动设备能耗
mobile-edge computing
stochastic task offloading
machine learning
delay
mobile device’s energy consumption
