张海波¹,
荆昆仑¹,
刘开健^1,,,
贺晓帆²
1.重庆邮电大学通信与信息工程学院 重庆 400065
2.武汉大学电子信息学院 武汉 430000
基金项目:国家自然科学基金(61801065, 61601071)，****和创新团队发展计划基金(IRT16R72)，重庆市基础与前沿项目(cstc2018jcyjAX0463)

详细信息

作者简介:张海波：男，1979年生，副教授，研究方向为无线资源管理
荆昆仑：男，1995年生，硕士生，研究方向为移动边缘计算
刘开健：女，1981年生，讲师，研究方向为最优化算法
贺晓帆：男，1985年生，助理教授，研究方向为无线资源优化

通讯作者:刘开健　liukj@cqupt.edu.cn

中图分类号:TN929.5

计量

文章访问数:2816
HTML全文浏览量:1644
PDF下载量:188
被引次数:0

出版历程

收稿日期:2019-04-30
修回日期:2019-09-05
网络出版日期:2019-09-18
刊出日期:2020-03-19

An Offloading Mechanism Based on Software Defined Network and Mobile Edge Computing in Vehicular Networks

Haibo ZHANG¹,
Kunlun JING¹,
Kaijian LIU^1,,,
Xiaofan HE²
1. School of Communication and Information Engineering, Chongqing University of Posts andTelecommunications, Chongqing 400065, China
2. School of Electronic Information, Wuhan University, Wuhan 430000, China
Funds:The National Natural Science Foundation of China (61801065, 61601071), The Program for Changjiang Scholars and Innovative Research Team in University (IRT16R72), The General Project on Foundation and Cutting-edge Research Plan of Chongqing (cstc2018jcyjAX0463)


摘要
摘要:在新兴的车联网络中，汽车终端请求卸载的任务对网络带宽、卸载时延等有着更加严苛的需求，而新型通信网络研究中移动边缘计算(MEC)的提出更好地解决了这一挑战。该文着重解决的是汽车终端进行任务卸载时卸载对象的匹配问题。文中引入了软件定义车载网络(SDN-V)对全局变量统一调度，实现了资源控制管理、设备信息采集以及任务信息分析。基于用户任务的差异化性质，定义了重要度的模型，在此基础上，通过设计任务卸载优先级机制算法，实现任务优先级划分。针对多目标优化模型，采用乘子法对非凸优化模型进行求解。仿真结果表明，与其他卸载策略相比，该文所提卸载机制对时延和能耗优化效果明显，能够最大程度地保证用户的效益。
关键词:车联网/
软件定义网络/
移动边缘计算/
卸载机制
Abstract:In the emerging vehicular networks, the task of the car terminal requesting offloading has more stringent requirements for network bandwidth and offload delay, and the proposed Mobile Edge Computing (MEC) in the new communication network research solves better this challenge. This paper focuses on matching the offloaded objects when the car terminal performs the task offloading. By introducing the Software-Defined in-Vehicle Network (SDN-V) to schedule uniformly global variables, which realizes resource control management, device information collection and task information analysis. Based on the differentiated nature of user tasks, a model of importance is defined. On this basis, task priority is divided by designing the task to offload the priority mechanism. For the multi-objective optimization model, the non-convex optimization model is solved by the multiplier method. The simulation results show that compared with other offloading strategies, the proposed offloading mechanism has obvious effects on delay and energy consumption optimization, which can guarantee the benefit of users to the greatest extent.
Key words:Vehicular networks/
Software Defined Network (SDN)/
Mobile Edge Computing (MEC)/
Offloading mechanism



PDF全文下载地址:

https://jeit.ac.cn/article/exportPdf?id=fcf99421-1253-4e3d-8720-014a76969ae0