基于历史查询概率的K-匿名哑元位置选取算法

基于历史查询概率的哑元位置隐私保护机制存在匿名度低、隐匿区域小和位置分布不均匀的问题。提出K-匿名哑元位置选取（K-DLS）算法用于位置隐私保护。通过综合考虑匿名集的位置离散度和零查询用户，增强哑元匿名集的隐私性。利用熵度量选择哑元位置，使得哑元匿名集的熵值最优，并根据位置偏移距离优化匿名结果，增加匿名集的位置离散度。仿真结果表明，K-DLS算法的哑元匿名集离散度优于DLS、DLP、Enhanced_DLP等算法，能够有效提高用户位置的隐私保护效果。

The K-Anonymous Dummy Location Selection Algorithm Based on Historical Query Probability

The dummy-based location privacy mechanism using historical query probability suffers from low anonymity, small coverage area and imbalanced location distribution.To address the problem, a K-anonymous dummy-based locationselection algorithm is proposed for position privacy protection.The privacy of dummy anonymous set is enhanced by comprehensively considering the location dispersion of anonymous set and zero-query users.The algorithm selects the location of dummy through entropymeasure to make the entropy of the anonymous dummy set optimal.Then the anonymous result is optimized based on the offset distance of the location, and the location dispersion of the constructed anonymous set is improved.The simulation results show that the proposed algorithm displays a higher location dispersion degree of the dummy-based anonymous set than DLS, DLP, Enhanced_DLP and other algorithms.It significantly improves the performance of location privacy protection for users.