目前，大多数位置匿名算法会出现匿名区域较大、匿名时间较长、匿名不成功的可能性较高等问题，并且对包含更多隐私信息的查询隐私没有做到更好的保护。为解决这些问题，文章提出一种基于敏感度的个性化LBS（P,L,K）匿名模型，该模型在K匿名基础上，通过对查询内容设置不同的敏感度来满足P敏感约束和L覆盖性约束，达到保护查询隐私的目的，从而实现匿名隐私保护的个性化需求。同时，文章在该模型基础上提出基于网格和假用户匿名算法，该算法将整个匿名空间划分成m×n的网格，通过迭代寻找查询用户所在网格的邻域空间进而找到该用户的临时匿名空间，然后根据用户分布矩阵对临时匿名空间进行边缘剥离，直至满足用户面积约束条件。从对比实验结果可知，在满足用户个性化要求条件下，该方法匿名区域面积更小，从而提高了相对匿名度和用户的查询服务质量。%At present, most position anonymity algorithms exist larger anonymous region, longer anonymity time and higher possibility of unsuccessful anonymity, and inquiry details which may include more privacy information are not protected better. To solve these problems, this paper proposes an anonymous model called LBS(P,L,K),which is based on k-anonymous model .This model sets parameters P and L by sensitivity of the queries in order to protect privacy of user queries and personalized needs of users. At the same time, this paper proposes algorithm called grid-fake users anonymity algorithm, which can not only protect the location privacy, but also to protect the query privacy. The algorithm’s idea is as follows: ifrst the space is mapped to mxn grid, then iteration search neighborhood space of the grid of the user lies in until ifnds the Minimum contain space, then stripping the edges with smallest user distribution density one by one according to the density matrix, on purpose of ifnding the anonymous user set meeting the anonymity condition in a minimum range, and achieving a better balance between privacy and quality of service. By contrast experiment, the algorithm has a higher success rate of anonymity, a smaller anonymous area and a higher relative anonymity under meeting the requirements of individual users, so it increases the quality of the user's query service.
展开▼
