Reasearch On The Location Anonymit Privacy Protection Technology

With the vigorous development of the wireless technology and mobile localization technology a new research field is opened up - location based service. When users needs this kind of service, they will have to send their precise location to service providers. In other words, they may face the risk of privacy being compromised. Therefore to protect the users' location information without being obtained by a potential attacker becomes a research hotspot of scholars, and they have already put forward a series of effective methods to protect the users' privacy. Position k-anonymous is one of the most common location privacy protection technology, which achieves the purpose of privacy protection by making the user's precise location generalized to an area with k-anonymous nature. A high performance, high efficiency of k-anonymous algorithm can not only protect users' privacy, but also improve the quality of service and reduce the communication expense.According to the users' different queries, this essay respectively and deeply researches on different attacks directed to static query and continuous queries, and the different measures which are needed to taken aiming at these attacks. In addition, this essay has also proposed two improved k-anonymous algorithm based on the original one, the main work and results are as follows:(1) Detailed location privacy protection methods and k-anonymous basic concepts, characteristics, and possible attack types, the analysis of k-anonymous technology research significance and the future development according to domestic and foreign research present situation, the further research on the common methods of implementation of k-anonymous technology and anonymity measure method.(2) The traditional k-anonymous scheme exists imprecise query results, especially in the scenarios of scarce users, it will produce larger anonymous area, thus increase the communication expense. In order to balance the contradiction between the quality of service and privacy protection, we propose a new scheme of dividing the sub division of anonymous region by dividing the anonymous region into a few scattered sub division of anonymous region. This method will not generate continuous cloaking region but directly divide into several Sub Clocking Regions, and randomly choose one cloaking region to take the place of the users'actual location and send query to the LBS Server. This method can more effectively protect the privacy of users, and can improve the quality of service, and reduce the communication expense.(3) In the scenario that the moving user keeps delivering queries to Location-Based Service, the attacker can infer the user's privacy according to the relationship of the customer's history requests. Terefore, a more optimized k-anonymous model is proposed. Under the condition of meeting the users' requirements of specified privacy degrees, this model can use the users' historical position in the active area to look for k-1 users which has largest probability density and the densest position distribution to make up common anonymity set. This method can get precise query results and largely decrease communication traffic.