摘要： 移动边缘计算（MEC）通过将计算和存储资源部署在无线网络边缘，使得用户终端可将计算任务卸载到边缘服务器进行处理，从而缓解终端设备资源受限与高性能任务处理需求之间的冲突。但随着任务卸载规模的不断增加，执行任务所产生的功耗急剧上升，严重影响了MEC系统的收益。建立任务队列动态调度模型，以队列上溢概率为约束构建最大化系统平均收益的资源优化模型。考虑到资源优化问题为不同时隙下的耦合问题，运用Lyapunov优化理论设计一种基于单时隙的资源分配算法，将优化问题转化为用户本地计算资源分配、功率和带宽资源分配以及MEC服务器计算资源分配3个子问题并分别进行求解。仿真结果表明，该算法在满足用户QoS需求的同时能够有效提高MEC系统的时间平均收益。

关键词: 移动边缘计算, 任务卸载, 资源分配, Lyapunov理论, 任务队列

Abstract: By deploying computing and storage resources at the edge of wireless network,Mobile Edge Computing(MEC) enables mobile devices to offload computing tasks to the MEC servers for processing,thereby effectively alleviating the contradiction between the limited resources of mobile devices and the high-performance task processing demands.However,with the increasing size of offloaded tasks,the power consumption generated by task execution rises dramatically,which significantly affects the profit of MEC systems.To solve the problem,this paper designs a dynamic scheduling model for task queue,and builds a resource optimization model to maximize the average revenue of the MEC system with the queue overflow probability as a constraint.Considering that the resource optimization problem is a coupling problem under different time slots,this paper uses the Lyapunov optimization theory to design a resource allocation algorithm based on a single time slot,and thus converts the optimization problem into three sub-problems:local computing resource allocation,power and bandwidth resource allocation,and MEC server computing resource allocation.Simulation results show that the proposed algorithm improves the time-average revenue of the MEC system while satisfying the QoS requirements of users.

Key words: Mobile Edge Computing(MEC), task offloading, resource allocation, Lyapunov theory, task queue

中图分类号: 

• TP391