基于混合果蝇算法的计算卸载方法  

Computation Offloading Decision Strategy Based on Hybrid Fruit Fly Optimization Algorithm

作　　者：杨子轩 张文柱[1] 程鹏 谢书翰 YANG Zi-xuan;ZHANG Wen-zhu;CHENG Peng;XIE Shu-han(School of Information and Control Engineering,Xi′an University of Architecture and Technology,Xi′an 710055,China)

机构地区：[1]西安建筑科技大学信息与控制工程学院,西安710055

出　　处：《小型微型计算机系统》2023年第6期1290-1296,共7页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61473216)资助;陕西省重点研发计划项目(2021GY-066)资助;陕西省自然科学基础研究计划项目(2020JM-489)资助.

摘　　要：计算卸载决策是移动边缘计算的关键问题之一,与移动边缘计算系统中任务计算时延和计算能耗的优化密切相关.针对多用户多MEC服务器的移动边缘计算系统中的计算卸载决策问题,考虑MEC服务器计算资源有限的特点,以任务计算时延和计算能耗为优化目标,设计了一种基于混合果蝇算法的计算卸载决策方法HFOA.HFOA方法首先使用启发式方法和随机方法进行种群初始化;然后,使用基于概率选择的自适应方法进行嗅觉搜索;最后,基于模拟退火原理对视觉搜索后的最优解作进一步优化,以提升算法的局部搜索能力.通过仿真实验,将HFOA方法与其他计算卸载决策方法进行对比,结果表明HFOA方法在系统计算时延、计算能耗、收敛性等方面均有改进.Computation offloading decision is one of the crucial problems in Mobile Edge Computing,which related to the optimization of computation delay and energy consumption.In order to solve the computation offloading problem in MEC system with multiple mobile devices and MEC servers,considering the limited computation resource of MEC servers,a computation offloading decision strategy HFOA based on hybrid fruit fly optimization algorithm is proposed with the object to optimize the computation delay and energy consumption.First,HFOA improves the initial population with a heuristic method mixed with random method.Then,introduce an adaptive method with probability to improve the olfactory search.Finally,further optimize the solution after visual search based on simulated annealing method.Simulation results and statistical comparison with other strategies show HFOA has apparent improvement in computation delay,computation energy consumption and convergence.

关 键 词：移动边缘计算 计算卸载决策 计算时延 计算能耗 混合果蝇算法 

分 类 号：TN915[电子电信—通信与信息系统]