Crowdsourcing application, deemed as a key evolution on the way to vehicular networking, has great potential to provide real-time services. However, existing cloud-based vehicular networking cannot support real-time data transmission with wasting massive bandwidth resources. This paper studies the crowdsourcing application in edge-assistant vehicular networking. To improve the real-time demand of data transmission, we propose the E-node of that owns the learning and semantic analysis abilities. Then we analyze two data transmission scenarios of crowdsourcing for collected data: road map uploading, traffic accident and traffic flow. On the other hand, to address the privacy leakages in the process of data aggregation and data distribution, we separately design time-tolerance anonymous privacy protection algorithm and k − 1 location-offset privacy protection algorithm. Finally, we conduct extensive experiments to verify the effectiveness of our proposed privacy protection algorithms, including time delay, offset probability, privacy leakage probability and accuracy.

众包应用程序被认为是通向车辆联网方法的关键演进，它具有提供实时服务的巨大潜力。但是，现有的基于云的车载网络不能支持浪费大量带宽资源的实时数据传输。本文研究了众包在边缘辅助车辆网络中的应用。为了提高数据传输的实时性要求，我们提出了ë -节点具有学习和语义分析能力。然后，我们针对收集的数据分析了众包的两种数据传输方案：路线图上传，交通事故和交通流量。另一方面，为了解决数据聚合和分发过程中的隐私泄露问题，我们分别设计了时间容忍匿名隐私保护算法和k  − 1位置偏移隐私保护算法。最后，我们进行了广泛的实验，以验证我们提出的隐私保护算法的有效性，包括时间延迟，偏移概率，隐私泄漏概率和准确性。