软件定义车联网中动态路由关键技术的研究

The number of VANET applications have been increasing a lot with the wide popularization of Intelligent Transportation System. However, many problems caused by the high speed of vehicles, like shorter lifecycle of links and more rapid changing of networks’ topologies, impede further developments of VAENT. To solve those problems, we involved VAENT with the technology of Software Defined Networks, which proved its advantage in dealing with dynamic topologies of networks. In this program we propose the concept of Software Defined Vehicular Networks, investigate its protocol architectures and reliable routing algorithms. Our works will ensure reliable data transmissions in complex road conditions and provide strong technical supports for Intelligent Transportation System. The main research topics in this program is described as follows: to adapt the highly dynamic topologies in VANET, we design new protocol architecture for SDVN; to overcome short lifecycle of links, we study the building techniques of reliable routing algorithm based on SDN controllers, investigate the multipath parallel routing strategies based on Digital Fountain Codes and research the congestion control mechanism based on self-adapted strategies; to diminish impacts of high speed and unbalanced distribution of vehicles, we study the actively optimizing strategies to control transmitting power and investigate green routing technology based on transmission power control. Our team has rich experience in the area of VANET and reliable routes, many papers written by us have been published on influential journals all over the world. It is our common desire to propose some innovative and leading solutions for SDVN research and raise international influence through executing this program.

随着智能交通的普及和发展，车联网的重要性与日俱增。路由是车联网的基础和关键。车速快及其导致的链路生命周期短、网络拓扑变化快给车联网路由带来了极大的挑战。融合软件定义网络在动态网络拓扑方面的优势，项目提出软件定义车联网的概念，研究其协议架构及可靠路由方案，旨在保证各种路况下的数据传输，为智能交通提供技术支撑。研究内容主要包括：针对拓扑变化快，设计软件定义车联网的新协议架构，实时匹配网络状态，跟踪网络拓扑变化；针对链路生命周期短，研究基于控制器的动态可靠路由建立技术，基于喷泉码的多路径并行路由策略，基于自适应路由的拥塞控制机制，面向拥塞避免的重路由选择；针对车速快，各路段车辆分布不均匀，研究基于功率控制的主动优化策略和绿色路由技术。团队在车联网及可靠路由方面具有良好的积累，多篇论文发表在国内外重要期刊上，期望通过本项目的执行能够提出一些新颖甚至国际上较为前沿的解决方案，提升国际影响力。

车联网中车辆高速移动导致网络拓扑变化快，链路生命周期短使得车辆间通信的可靠路由面临巨大挑战，且通信需求的爆发式增长使得网络性能下降，软件定义网络（SDN，Software Defined Network）中网络状态的实时获取和路由机制的动态可控这两大优势正好为应对车联网所面临的可靠路由和负载均衡两大挑战提供了明确思路。课题融合车联网与SDN技术，主要研究了软件定义车联网（SDVN，Software Defined Vehicular Network）的架构、机制及保障车辆高效可靠通信的路由及资源分配关键技术。主要研究成果包括：SDVN车联网系统架构的设计，并基于此架构研究了基于负载均衡和节能的多控制器部署策略，面向车联网安全业务的快速广播算法、基于道路连通性的动态可靠路由技术、基于喷泉码的多路径备份路由算法、基于自适应路由的拥塞控制机制和基于功率控制的绿色动态路由技术。在完成计划书研究内容的基础上，课题组进一步进行了车联网传输容量的理论分析，研究了支持5G车联网增强型业务的资源分配机制，并对车联网中的隐私保护技术进行了初步探索研究。在课题资助下，已提交国家发明专利申请13项，授权7项；在国内外发表学术论文28篇（其中SCI，EI分别已检索11篇和15篇）；分别完成博士、硕士学位论文3篇和14篇。本课题的研究成果将为安全、高效、绿色的交通出行提供理论与技术支撑，加速未来智能网联汽车的发展和应用。