基于位置关联相似性的匿名算法

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.