基于节点社会性行为分析的车辆网络优化技术研究

With the advancement of wireless technology, vehicular networks are emerging as a promising approach to addressing many important transportation problems such as road safety, efficiency, and convenience, which keep receiving research attention from both the academia and the industry. We have found that vehicles show obvious sociality in terms of their meetings in temporal and spacial dimensions, which would have significant impact on the network performance. Though there have been extensive studies on social network analysis in the Internet, it is very hard or not applicable to adopt existing schemes in vehicular networks.This project will develop a new set of solutions for efficient data communications and content dissemination in vehicular networks, via three following research thrusts. First, with the GPS (Global Positioning System) trace data of about tens of thousand experimental vehicles in Shanghai and Shenzhen, we will study and establish large-scale node sociality analysis methods. Second, we will mine the social behavior of vehicles and the sociality characteristics of urban vehicular networks. Third, based on the sociality characteristics, we will propose new data transmission and content dissemination schemes. The research outcomes will make a significant contribution to the development of future smart vehicles, and will provide the automotive industry with critical knowledge for driving the deployment of intelligent transportation systems.

)：无线车辆网络作为一种新型的无线移动自组织网络，因其独特的网络特性和潜在的重要社会经济应用价值，近年来受到国内外政府、企业和高等院校的高度重视。本课题前期研究发现车辆网络中车辆节点行为具有社会性，节点社会性行为具有环境感知性、多维时空性、动态随机性、天然分布性等特征，并且用户的社会性行为对网络性能产生了重要影响。传统互联网中用户社会关系分析及建模方法难以有效地应用在移动车辆网络环境中。本项目提出：1）构建大规模车辆网络的多尺度社会性行为研究平台；2）挖掘车辆网络中车辆节点的社会性行为特征；3）设计车辆节点社会性行为感知的网络优化技术。预期在车辆网络节点社会性行为挖掘和节点分类方法，以及基于节点社会性行为感知的车辆网络数据传输与分发的优化技术上取得较大的创新和突破。

无线车辆网络作为一种新型的无线移动自组织网络，因其独特的网络特性和潜在的重要社会经济应用价值，近年来受到国内外政府、企业和高等院校的高度重视。本课题着重无线车辆网络节点行为社会性和高效任务分发方案的研究。首先，构建了大规模车辆网络的多尺度社会性行为研究平台，该平台包括大规模、多维度、高可靠性的移动车辆数据集，以及相对应的社会性行为分析方法和衡量指标；其次，使用研究平台对车辆网络中车辆的社会性行为进行了相应的分析，发现网络中的车辆展现出较为动态的社会性，这种车辆社会性又具有很强的时间相关性；最后，基于对车辆网络中节点社会性的观察，提出了一种创新的车辆网络移动信息分发方案，使用马尔科夫链来对未来的车辆社会性进行推测，并采用贪心式启发方法来为移动信息分发服务选取出种子节点。我们基于包括出租车与公交车在内的三个真实车辆数据集进行了大量而充分的仿真实验。实验结果表明该方案能够有力地提高信息覆盖率以及覆盖强度。