移动边缘计算中分布式异构任务卸载算法

随着物联网(IoT)迅速发展，移动边缘计算(MEC)在提供高性能、低延迟计算服务方面的作用日益明显。然而，在面向IoT业务的MEC(MEC-IoT)时变环境中，不同边缘设备和应用业务在时延和能耗等方面具有显著的异构性，对高效的任务卸载及资源分配构成严峻挑战。针对上述问题，该文提出一种动态的分布式异构任务卸载算法(D2HM)，该算法利用分布式博弈机制并结合李雅普诺夫优化理论，设计了一种资源的动态报价机制，并实现了对不同业务类型差异化控制和计算资源的弹性按需分配，仿真结果表明，所提的算法可以满足异构任务的多样化计算需求，并在保证网络稳定性的前提下降低系统的平均时延。

A Distributed Heterogeneous Task Offloading Methodology for Mobile Edge Computing

With the rapid development of the Internet of Things (IoT), Mobile Edge Computing (MEC) becomes increasingly effective in improving processing capacity and providing low-latency computing services. However, in the time-varying MEC-IoT environment, heterogeneous devices and applications cause serious challenges on efficient task offloading and resource allocation. A Distributed Dynamic Heterogeneous task offloading Methodology (D2HM) algorithm is proposed in this paper by exploiting game theory and Lyapunov optimization, which can achieves heterogeneous control and allocation of computation resources by dynamic quote price mechanism. Simulation results show that the proposed algorithm can meet the diverse computing needs of heterogeneous tasks and reduce the average delay of the system while ensuring network stability.