Research On Continuous Nearest Neighbor Query In Location Privacy Protection

Along with the rapid development of wireless communication、spatial databasetechnology、 positioning technology,as well as the great popularity of location-basedwireless handheld devices, the Location-Based Service(LBS) has become more andmore popular。The LBS is increasingly appreciated by people for its so manycharacteristics, such as diversity、closely related to the people’s life、immediacy andso on. The LBS have penetrated into all aspects of people’s lives, which is affectingthe way we live、work and study. However, the LBS bring us tremendous convenience,it also imply the disclosure of location privacy. The purpose of location privacypreserving is not only to make people enjoy the high-quality services, but also notrevealing our location privacy, which needs to find out the trade-off between enjoyingservices and location privacy. In this paper, we have studied issues in both locationprivacy preserving in LBS services and privacy preserving-based continuousnearest-neighbor query. The main work can be described briefly as follows:Firstly, the framework of location privacy preserving system and locationK-anonymity model are established. The framework of location preserving system iscentralized architecture in this paper, which contains client-side、location anonymizerand location-based database server. The client-side sends a service request andprivacy request, the location anonymizer manages the users’ location informationdynamically and provides the location protection for the users, the location-baseddatabase responses the service request from the users. The k-anonymity model is usedto dynamically create the privacy requirements for mobile subscribers; locationanonymizer constructs the anonymous region for mobile subscribers to protect theirlocation privacy.Secondly, two new anonymity algorithms which are based on the centralizedarchitecture are proposed in this paper. The centralized architecture-based anonymityalgorithm requires timeliness and high efficiency. Existing algorithms incur highcloaking time and low accuracy. To address this issue, two cloaking algorithms,namely QFC and SWC, are proposed in this paper. In contrast to existing cloaking algorithm, QFC can reduce the CPU time significantly at the same cloaking accuracy;Another stepwise-based algorithms SWC,which can achieve a higher cloakingaccuracy at the cost of certain CPU time. The validity of the two proposed algorithmshave been proved by the experimental resultsLastly, an algorithm which can effectively deals with continuous nearest-neighborquery with privacy-preserving is proposed. Privacy-preserving continuousnearest-neighbor query is equivalent to anonymous region-based nearest-neighborquery in the framework of location privacy preserving. The location database serverwhich processes such query by snapshot will incur high CPU cost, existing algorithmhas a poor performance especially when the number of user is increasing. To copewith this challenge, an algorithm named RLPCNN which is based on reusingtechnique query updating is proposed in this paper. The algorithm can reduce thecomputational expense by using the similarity between the two adjacent time andmakes the answer set updated quickly, which can quicken the system response timemarkedly. The validity of the algorithm has been proved by the experimental results...