基于可行点追踪-连续凸逼近的移动边缘计算任务卸载  

Mobile Edge Computing Task Offloading Based on Feasible Point Tracking Continuous Convex Approximation

作　　者：陈刚 王志坚 徐胜超 CHEN Gang;WANG Zhi-jian;XU Sheng-chao(School of Data Science,Guangzhou Huashang College,Guangzhou 511300,China)

机构地区：[1]广州华商学院数据科学学院,广东广州511300

出　　处：《计算机与现代化》2023年第8期93-97,共5页Computer and Modernization

基　　金：国家自然科学基金面上项目(61772221);广州华商学院校内导师制科研项目资助(2023HSDS07)。

摘　　要：移动边缘计算任务卸载会受到邻近服务器的干扰,导致计算任务难以准确卸载到网络边缘服务器,因此设计基于可行点追踪-连续凸逼近的移动边缘计算任务卸载方法。该方法首先建立移动边缘计算任务的依赖模型,分析移动边缘计算任务的卸载需求。其次,考虑任务卸载时延和卸载能耗,以任务依赖模型为依据,建立任务卸载模型。最后,采用可行点追踪-连续凸逼近,将求解卸载模型的求解问题转变为线性松弛问题,引入迭代过滤函数追踪可行点,避免邻近服务器的干扰,对松弛变量进行连续凸逼近获取任务卸载模型的最优解,实现移动边缘计算任务的卸载。实验结果表明,本文方法负载低、卸载能耗低、卸载精度高。The task unloading of mobile edge computing will be interfered by adjacent servers,which makes it difficult to accu‐rately unload computing tasks to network edge servers.Therefore,a task unloading method of mobile edge computing based on feasible point tracking continuous approximation method is designed.This method firstly establishes the dependency model of mo‐bile edge computing tasks,and analyzes the unloading requirements of mobile edge computing tasks.Secondly,considering the task unloading delay and energy consumption,the task unloading model is established based on the task dependency model.Fi‐nally,the feasible point tracking continuous convex approximation method is used to transform the problem of solving the unload‐ing model into a linear relaxation problem.Iterative filter functions are introduced to track the feasible points to avoid interference from adjacent servers.The relaxation variables are continuously convex approximated to obtain the optimal solution of the task un‐loading model,so as to realize the unloading of mobile edge computing tasks.The experimental results show that the proposed method has low load,low energy consumption and high unloading accuracy.

关 键 词：可行点追踪-连续凸逼近法 移动边缘计算 任务依赖模型 线性松弛问题 任务卸载 

分 类 号：TP393.4[自动化与计算机技术—计算机应用技术]