A joint optimization scheme for task offloading and resource allocation based on edge computing in 5G communication networks
A joint optimization scheme for task offloading and resource allocation based on edge computing in 5G communication networks
复制标题
DOI：
10.1016/j.comcom.2020.07.008
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Shi Yang
中科院分区：
文献类型：
--
作者：
Shi Yang
随着5G通信网络的发展和智能终端的普及，各种新应用的计算资源密集型特点对智能终端的任务处理能力提出了严峻的挑战。为了提高任务处理效率，提出了一种5G通信网络中基于边缘计算的任务卸载和资源分配联合优化方案。首先，结合边缘计算和设备到设备通信技术，我们提出了三种基于5G边缘网络多用户网络系统模型的计算密集型任务处理模式，包括本地计算，雾节点计算，边缘节点计算。针对这三种计算模式分别建立了相应的时延模型、任务执行模型和卸载能耗计算模型。最后将计算任务卸载问题转化为时延和能耗的联合优化问题，包括CPU频率、卸载决策、传输带宽分配和卸载用户功率分配等优化问题。此外，利用内点法求解了该问题。仿真平台被用来证明我们所提出的方案的性能。实验结果表明，该方案能有效降低终端任务的时延和能耗，提高了任务处理效率和终端用户的体验质量。
With the development of 5G communication networks and the popularization of intelligent terminals, the computing resource intensive characteristic of various new applications poses a severe challenge to the task processing ability of intelligent terminals. In order to improve the efficiency of task processing, a joint optimization scheme for task offloading and resource allocation based on edge computing in 5G communication networks is proposed. Firstly, combining edge computing and Device-to-Device communication technologies, we propose three modes for processing computationally intensive tasks based on multi-user network system model for 5G edge networks, including local computing, fog node computing, edge node computing. Then, the corresponding time delay model, task execution model and offloading energy consumption computing model are constructed for these three computing modes. Finally, the problem of computing task offloading is transformed into a joint optimization problem of time delay and energy consumption, including optimization problems such as CPU frequency, offloading decision, transmission bandwidth allocation and power allocation of offloading users. Besides, the interior point method is utilized to solve this problem. Simulation platform is used to demonstrate the performance of our proposed scheme. The experimental results show that the scheme can effectively reduce the time delay and energy consumption of terminal tasks, which improves the efficiency of task processing and the experience quality of end users.