By offloading computation-intensive and latency-sensitive tasks from mobile devices to the mobile edge server, mobile edge computing (MEC) has been considered as a promising technology in 5 G and beyond to reduce task delay and energy consumption. This paper considers an optimization framework of computation offloading in a MEC system with multiple devices. We aim to minimize the energy consumption of all devices with their task delay constraints by jointly optimize the communication and computation resource allocation at both the devices and the mobile edge server. Specifically, dynamic voltage scaling (DVS) technology is considered in each device to adjust the operating frequency according to its delay constraint. The formulated problem is a non-convex problem due to a couple among multiple variables. To tackle the non-convex problem, we first decompose the original problem by optimizing the offloading ratio and transmission power iteratively. Then, we proposed a joint communication and computation optimization algorithm based on the difference of convex function algorithms (DCA) to solve the optimization problem. Finally, simulation results show that the proposed joint communication and computation scheme significantly improves the energy efficiency of the devices comparing with the local computing scheme and server computing scheme.

中文翻译：

---

在移动边缘计算中使用动态电压调节实现节能任务卸载


通过将计算密集型和延迟敏感的任务从移动设备卸载到移动边缘服务器，移动边缘计算 (MEC) 被认为是 5G 及更高版本中一项有前景的技术，可减少任务延迟和能耗。本文考虑了具有多个设备的 MEC 系统中计算卸载的优化框架。我们的目标是通过联合优化设备和移动边缘服务器的通信和计算资源分配，最大限度地减少所有设备及其任务延迟约束的能耗。具体来说，每个器件都考虑动态电压调节（DVS）技术，根据其延迟约束来调整工作频率。由于多个变量之间存在耦合，所表述的问题是一个非凸问题。为了解决非凸问题，我们首先通过迭代优化卸载比和传输功率来分解原始问题。然后，我们提出了一种基于凸函数差分算法（DCA）的联合通信和计算优化算法来解决优化问题。最后仿真结果表明，与本地计算方案和服务器计算方案相比，所提出的联合通信计算方案显着提高了设备​​的能源效率。