基于Voronoi图预划分的LBS位置隐私保护方法  

Location privacy-preserving method in LBS based on Voronoi division

作　　者：马春光[1] 周长利[1] 杨松涛[1] 赵蕴龙[1] 

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《通信学报》2015年第5期1-12,共12页Journal on Communications

基　　金：国家自然科学基金资助项目(61170241;61073042);中央高校基础科研业务费重大专项基金资助项目(HEUCFZ1105);高等学校博士学科点专项科研基金资助项目(20132304110017);黑龙江省自然科学基金资助项目(F201229)~~

摘　　要：为了解决服务器面临大量用户请求时匿名效率下降的问题,分别提出适用于静态用户和动态用户的协作匿名方法。首先基于Voronoi图划分全局区域,再由中心服务器组织本区域内用户实现协作匿名,由于服务器无需为每个用户单独构造匿名区,降低了服务端的负担;针对查询过程中用户提供真实位置信息带来位置隐私泄露的问题,提出了逆向增量近邻查询算法。用户以固定锚点代替真实位置,向位置服务器逐步获取兴趣点候选集并计算出想要的结果,避免位置隐私直接泄漏的同时获取精准查询结果。该算法同时解决了锚点与用户过近而带来的位置隐私被推断问题。实验表明本方法在有效保护用户位置隐私的同时,具有良好的工作效率。In view of low efficiency when the anonymous server faces a large number of users, a cooperative anonymous method is proposed for static and dynamic users respectively. Based on the Voronoi division of the global area, a central server organizes the users in its region to achieve cooperative anonymity, the server needn＇t construct anonymous region alone for each user, and the burden of the server is reduced. In view of providing actual location when users query from a LBS server, a decrease nearest neighbor query algorithm is proposed. A user＇s actual location is replaced with a stationary anchor location and gets the points of interest candidate set from the LBS server gradually. By running the algorithm, pre- cise results can be got and avoiding exposure to a user＇s location privacy. The algorithm can also help to reduce the possi- bility of location inference when the anchor chooses closely to the user. Experiments show that our method can guarantee the user＇s location privacy, and provide a good working efficiency.

关 键 词：位置隐私 协作匿名 锚点 逆向增量查询 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]