摘要
针对物联网边缘计算中计算卸载的高能耗、高时延以及移动智能设备自身条件不足等问题,提出一种基于binAD算法的边缘计算任务卸载决策。以最大化移动终端设备系统效用为目标,将任务执行计算能耗和时延加权和定义为系统效用的优化函数。将二进制人工蜂群算法和二进制差分进化算法相结合得到binAD算法,其中为扩大可行解范围、提高全局搜索能力,将最个体引入变异操作中,将交叉概率因子修改为适应性交叉概率因子。使用binAD算法解决优化问题。仿真结果表明,binAD算法在能耗、时延、系统效用方面均具有优越性。
Aiming at the problem of high energy consumption and high time delay of computing offload in edge computing of the Internet of Things,as well as the problem of insufficient conditions of mobile intelligent devices,a decision of edge computing task offload based on binAD algorithm was proposed.With the goal of maximizing the system utility of mobile terminal equipment,the weighted sum of task execution energy consumption and time delay were defined as the optimization function of system utility.Binary artificial bee colony algorithm and binary differential evolution algorithm were combined to obtain binAD algorithm.To expand the range of feasible solutions and improve the global search ability,the optimal individual was introduced into the mutation operation,and the crossover probability factor was modified into the adaptive crossover probability factor.The binAD algorithm was used to solve the optimization problem.The simulation results show that binAD algorithm has advantages in energy consumption,time delay and system utility.
作者
王泽
郭荣佐
WANG Ze;GUO Rong-zuo(College of Computer Science,Sichuan Normal University,Chengdu 610101,China)
出处
《计算机工程与设计》
北大核心
2023年第8期2289-2296,共8页
Computer Engineering and Design
基金
国家自然科学基金项目(11905153、61701331)。
关键词
物联网
边缘计算
计算卸载
卸载决策
二进制人工蜂群算法
二进制差分进化算法
binAD算法
internet of things
edge computing
computing offloading
offloading decision
binary artificial bee colony algorithm
binary differential evolution algorithm
binAD algorithm