基于车流波动理论保障安全类应用的车辆自组织网络多场景数据分发方法

As an efficient approach to improve the road safety, driving comfort and traffic efficiency, VANETs(Vehicular Ad Hoc Networks) have attracted a great many research interests in recent years. However, since most researchers still tend to apply the traditional data dissemination schemes designed in MANETs(Mobile Ad Hoc Networks) to VANETs, the safety related applications, which are taken as the major usecases in vehicular networks, are experiencing significant risks and threats. In this proposal, according to the traffic flow-density relationship from the famous traffic flow theory, we first classify the application scenarios of VANETs by the flow characteristics into 3 categories, say free flow, transition flow and congestion flow. Then, based on the above classification, different data dissemination concerns have been issued especially for the safety related usecases. Next, to increase the data delivery ratio in the free and transition flow state, the methods used to improve the connectivity between vehicles have been proposed and analyzed in details. In addition, the congestion control and avoidance strategies designed to balance the tradeoff between the safety guarantee and delay requirement have also been investigated when the traffic flow stays in the transition or congestion state. Finally, to solve the data dissemination issue in the popular urban environment, the packets routing schemes have been well studied under the congestion flow state. In summary, according to the aforementioned schemes designed to guarantee the success of data dissemination for safety related applications in VANETs, the possible loss of life and property due to accidents is expected to be reduced to a great extent. As a result, through the implementation of our project, the transformation process from the theoretical research to practical application of intelligent transportation systems especially VANETs, will be accelerated significantly in the future.

车辆自组织网络作为提升驾驶安全性、舒适性及行车效率的有效途径，近年来一直受到国内外学者的高度关注。然而，针对车辆自组织网络中最关键的道路交通安全类应用，其数据分发方法仍沿袭传统移动自组织网络研究中的思路，给行车安全及其相关应用带来极大威胁。本申请，首先基于车流波动理论，将车辆自组织网络应用场景按车流密度关系分为自由、过渡与拥塞流态场景，随后分场景提出了不同交通流态下适用的安全类应用数据分发方式；接下来研究了自由与过渡流下，制约数据分发成功率的连通能力提升与分析方法；之后讨论了过渡和拥塞流下，平衡安全性与时效性的数据分发拥塞避免方法；最后针对应用广泛的城市车辆网络，提出了拥塞流条件下有效保证安全相关紧急数据成功投递的路由方案。基于提出的数据分发策略，可大幅度提升车辆网络安全类应用的可靠性，避免行车过程中的事故，减少带来的经济损失和人员伤亡，加速智能交通系统从理论研究向实际应用转化的过程。

车辆自组织网络作为提升驾驶安全性、舒适性及行车效率的有效途径，近年来一直受到国内外学者的高度关注。然而，针对车辆自组织网络中最关键的道路交通安全类应用，其数据分发方法仍沿袭传统移动自组织网络研究中的思路，给行车安全及其相关应用带来极大威胁。本项目经过4年的研究与积累，首先基于车流波动理论，将车辆自组织网络应用场景按车流密度关系分为自由、过渡与拥塞流态场景，随后分场景提出了不同交通流态下适用的安全类应用数据分发方式，并进行了丰富的实验，发表了相应成果；接下来研究了自由与过渡流下，制约数据分发成功率的连通能力提升与分析方法，并进行了丰富的实验，发表了相应成果；之后研究了过渡和拥塞流下，平衡安全性与时效性的数据分发拥塞避免方法，并进行了丰富的实验，发表了相应成果；最后针对应用广泛的城市车辆网络，提出了拥塞流条件下有效保证安全相关紧急数据成功投递的路由方案，并进行了丰富的实验，发表了相应成果。项目所产生的部分成果已在高水平期刊和会议上进行了发表，项目负责人及其成员多次受邀在相关会议上进行了成果宣讲。部分专利及软著得到了进一步转化，产生了较好的经济和社会效益。