车联网轨迹数据隐私保护问题研究

发布时间：2018-01-01 18:37

本文关键词：车联网轨迹数据隐私保护问题研究　出处：《河南工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着车载定位技术和移动自组织网络技术的飞速发展,车联网(VANETs)相关技术已引起国内外政府、学术界和产业界的高度重视并成为热点话题。目前车联网与工业4.0、可穿戴设备成为当前物联网领域的三大研究热点。车联网的部署能够实现智能驾驶、交通管理以及车载娱乐等等。但是机遇与挑战并存,在车联网飞速发展的同时也面临着数据安全、隐私泄露的风险。未来车联网中每辆车上有超过80个传感器,每天向后台云服务器传输的数据量可达到100MB,这些数据覆盖了车辆和用户的个人信息、行为模式等敏感信息,攻击者可以通过窃听通信过程中的广播消息、分析数据、预测轨迹、跟踪车辆从而获取车辆的敏感数据。用户的个人隐私甚至人身安全都将受到威胁。因此想要大规模的部署车联网,首先要面临的问题就是数据的隐私安全问题,而对于车联网来说,其中最主要的数据就是车辆的轨迹数据。本文首先通过对车联网的潜在的轨迹隐私需求进行系统的分析。针对车联网不同应用场景下的轨迹隐私问题进行分析,针对如何保护车联网轨迹隐私数据和防止敏感信息泄露进行了阐述,分别介绍了目前主流的隐私保护技术并通过分析对现有的隐私保护方案进行对比,并说明其存在的问题。针对车联网轨迹隐私问题,分别在车联网网内数据传递与数据发布过程两种应用场景,引入差分隐私思想研究车载网络轨迹数据发布过程中的隐私保护问题,通过理论分析、建模和实验,探索适合车联网轨迹隐私保护机制。最后分别从数据质量和轨迹数据可用性两个角度,对本文提出的隐私保护算法进行性能测评并与现有的隐私保护方案进行对比。实验结果显示差分隐私在车联网网内轨迹隐私保护方面不仅有良好的隐私保护效果而且能够有效的缓解通信压力,对于面向位置服务的轨迹数据发布与现有的GNoise机制与PNoise机制进行对比显示在数据质量和隐私保护度均取得良好的实验结果。
[Abstract]:With the rapid development of mobile positioning technology and mobile self-organizing network technology, car Networking (VANETs) technology has attracted domestic and foreign government attaches great importance to the academia and industry and has become a hot topic. At present, car networking and industrial 4, wearable devices become the three hotspot in networking field. Car networking the deployment can realize intelligent driving, traffic management and vehicle entertainment and so on. But the opportunities and challenges, the rapid development of networking in the car while also facing the risk of data security, privacy leaks. The future car networking in every car has more than 80 sensors, the amount of data transmission of cloud servers back every day can reach 100MB these data, covering the vehicle and the user's personal information, behavior patterns and other sensitive information, an attacker can broadcast messages, eavesdropping on communication in the process of data analysis, trajectory, vehicle tracking A vehicle to obtain sensitive data. The user's personal privacy and personal safety will be threatened. Therefore, to large-scale deployment of car networking, privacy and security the first problem faced is the data, and for the car networking, the trajectory data of the primary data is the vehicle. This paper based on the Internet of vehicles the potential path privacy requirement analysis system. Aiming at the car networking trajectory privacy issues under different scenarios are analyzed, in order to protect the car networking track data privacy and prevent sensitive information leakage are described, respectively introduces the analysis comparison of the existing privacy protection scheme by the mainstream technology and privacy protection, and to illustrate the problem. Aiming at the car networking trajectory privacy issues, network network data transfer and data release process of two kinds of application respectively in the car The scene, introducing difference privacy protection privacy points thought of vehicle network trajectory data release process, through theoretical analysis, modeling and experiment, to explore suitable car networking trajectory privacy protection mechanism. Finally, from two aspects of data quality and track data availability, performance evaluation of privacy protection algorithm is proposed in this paper and compared with the existing privacy protection scheme. Experimental results show that the differential privacy and privacy protection is not only a good effect can effectively relieve the pressure in the car networking communication within the network path of privacy protection, to track data in location based services released comparison shown in the data quality and the privacy protection degree achieved good results with GNoise the mechanism and the existing PNoise mechanism.

【学位授予单位】：河南工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP309

【参考文献】