Privacy Protection Technology For Continuous Query In Location Based Services

Recently location based services(LBS) have been widely used, with the popularity of mobile smart devices and the development of wireless communications and positioning technology. Although LBS provides a lot of convenience to individuals, it poses the potential threat to their privacy. To enjoy interested services, an individual is supposed to submit his or her location and query, which can be used to breach the individual’s privacy. So this makes the research on LBS privacy protection very urgent and necessary.Firstly, the LBS privacy protection system architecture is studied, and a distributed privacy preserving system architecture based on the trusted third party(TTP) architecture has been proposed in this thesis. The system aims to protect query privacy and overcome the shortcomings in the TTP architecture, such as the single point of failure problem, the bottleneck of the system and so on. In this system, multiple semi-honest anonymizing servers are involved, and a user can divide his or her query into several shares according to the secret sharing scheme, and distribute them to these servers. Then, this thesis conducts an intensive study of LBS privacy protection algorithms, which shows that current works are almost about the snapshot query, neglecting the continuous query that may be more significant to study. Because the snapshot query can be treated as the special case of the continuous query, and the continuous query has a wide range of applications. However, most of the existing privacy protection algorithms for the continuous query have an unsatisfactory performance in terms of the privacy protection and quality of services(QoS). To solve this problem, a velocity based dynamic cloaking algorithm is proposed, which can be applied not only to the system architecture proposed in this thesis but also to the TTP architecture. The algorithm takes the moving status and trend into consideration, and introduces the velocity and acceleration similarity into the anonymizing process. Moreover, the consecutive generated cloaked sets are used to produce the new cloaked set. All of these contribute to the good performance of the algorithm with respect to privacy protection and QoS. In addition, users are allowed to define their own privacy and QoS parameter, and the algorithm can cloak for a single user and a batch of users respectively. Considering the privacy issue of the expired query, a concept of m-anonymity and a location predicting mechanism are proposed.An evaluation system for the continuous query privacy protection algorithm, including seven metrics with regard to privacy protection, QoS and performance, is presented at the end of this thesis. The simulation results show that the algorithm can protect mobile users’ privacy as well as provide high QoS.