基于次模优化的边云协同多用户计算任务迁移方法

Multiuser computation offloading for edge-cloud collaboration using submodular optimization

为了提升多用户计算任务卸载时的系统效用,提出了一种基于边云联合计算的多用户任务卸载方案。该方案在提升系统效用的同时,考虑了边云资源的协同优化问题。针对计算任务卸载模式的选择及边云资源分配的问题,设计了一种基于次模理论的贪心算法并充分利用了云端以及边缘端的计算和通信资源。仿真结果表明,所提方案能够有效降低计算任务执行的时延和能耗,且当多用户卸载计算任务时,所提方案在资源受限的条件下仍然能够保持稳定的系统性能。

A computation offloading scheme based on edge-cloud computing was proposed to improve the system utility of multiuser computation offloading.This scheme improved the system utility while considering the optimization of edge-cloud resources.In order to tackle the problems of computation offloading mode selection and edge-cloud resource allocation,a greedy algorithm based on submodular theory was developed by fully exploiting the computing and communication resources of cloud and edge.The simulation results demonstrate that the proposed scheme effectively reduces the delay and energy consumption of computing tasks.Additionally,when computing tasks are offloaded to edge and cloud from devices,the proposed scheme still maintains stable system utilities under ultra-limited resources.