车联网环境下基于5G和云平台的智能交通关键技术研究

As the representative product of the Internet of Things and mobile Internet, Vehicular Ad Hoc Network (VANET) has become one of the most important part of modem Intelligent Transportation System (ITS). With the emergence of 5G communication and the increasing maturity of cloud platform, VANET brings new opportunities for the development of intelligent transportation. A series of innovative research about key technologies of intelligent transportation based on 5G and cloud platform in VANET will be carried out in this project. Firstly, we research data transmission methods and optimization strategies, including opportunistic communication based on information entropy, multiple links and multi-channel data transmission methods, and optimization strategy based on incentive mechanism. Then, distributed and dynamic routing methods are studied, which involve a cost estimation model based on trajectory speculation, a cloud-assisted fusion method for routing information, and a path decision method based on dynamic overhead game theory. In addition, on the basis of a collaborative security application framework, emergency event processing methods are studied based on reliable communication. In particular, safety driving assist methods and emergency messages dissemination protocols are proposed with the help of 5G communication. Finally, we research the privacy-preserving and efficient signature verification scheme in VANETs, including the privacy-preserving during the dynamic path searching and cloud-assisted efficient signature verification scheme. The research results of this project will provide theoretical and technical support for the application and development of intelligent transportation in VANET.

车联网作为物联网和移动互联网发展的代表性产物，成为现代智能交通的重要组成部分。随着5G通信的出现和云平台的日趋成熟，给智能交通带来了新的发展机遇。本项目拟针对车联网环境下基于5G和云平台的智能交通关键技术进行创新性研究。首先研究数据传输方法和优化策略，包括基于信息熵和QoS感知的传输方法、多链路和多信道联合的传输方法，以及基于激励机制的优化策略；然后研究分布式动态寻径方法，包括基于轨迹推测的道路开销评估模型、基于云端辅助的寻路数据融合方法和基于动态开销博弈的路径决策方法；同时研究紧急事件处理方法，在基于云平台的车辆协作式安全应用框架上，分别研究基于5G协助的辅助安全驾驶方法和紧急消息广播机制；最后研究车联网中的隐私保护与消息高效签名认证方法，包括车辆动态寻径过程中的隐私保护和基于云辅助的消息高效签名认证方法。本项目的研究成果将为车联网环境下智能交通的应用和发展提供理论支撑和技术保障。

车联网作为5G和汽车领域跨界融合最具潜力的应用，已经成为我国战略性新兴产业的重要发展方向，也是备受学术界和产业界关注的研究热点。本项目针对5G车联网环境及基于云平台的数据分发、分布式寻径、紧急事件处理、隐私保护与签名认证等问题进行深入研究。主要研究成果包括：基于强化学习的分簇协同调度方法；面向云数据共享的安全可搜索加密算法；基于区块链的可搜索公钥加密并支持前向和后向隐私的云辅助车辆社交网络；基于区块链的匿名身份验证与密钥管理边缘计算框架；基于智能手机的一种多模态生物识别认证系统；使用不可信服务器发布具有隐私保护的众包统计信息方法；深度强化学习在智慧城市通信网络中的应用；基于区域感知和强化学习的多交叉路口信号灯控制方法；基于动态时空图注意力网络的信号灯控制方法；面向延迟容忍网络的自适应多重喷射和等待路由算法；以及基于边缘计算的交叉路口交通灯数据检测方案等等。本项目的系列研究成果为车联网环境下智能交通的应用和发展提供技术支撑。