基于位置关联相似性的匿名算法

Location association similarity based anonymous algorithm

在基于位置的查询服务中,历史查询数据所包含的位置与兴趣点之间存在一定的关联差异,攻击者可以利用这种差异,通过使用匹配关联的方式,获取用户的位置隐私信息。针对这一问题,通过使用二分图刻画该服务中实体之间的匹配关系,分析了这种攻击方法的基本特点,并基于k-匿名思想提出了cor-k匿名算法。该算法有效地模糊了查询位置间的历史匹配关系,阻止了潜在的历史数据匹配关联。实验证明该算法能够在较短时间内完成处理过程,适合在实际环境下部署以保护用户的位置隐私。

In the location based query services,the historical query data contains the correlation probabilities between the location and the points of interest.With the difference of these probabilities,the attacker can launch a history matching correlation attack to obtain users' location privacy.In order to solve this problem,we first analyze the basic characteristics of this attack,and depicting the relationship between entities with binary chart.A cor-k algorithm is proposed on the basis of the k-anonymity algorithm.This algorithm blurs the matching correlation of query location efficiently,and prevents the attack of history matching correlation.Experiments show that the method can be finished in a relatively short period of time,and is suitable for deployment in a real environment.