防止暴露位置攻击的轨迹隐私保护

为解决移动对象轨迹信息被大量收集所导致的轨迹隐私泄露问题，提出了基于假轨迹的轨迹隐私保护算法。在该算法中，考虑了用户的暴露位置，基于轨迹相似性和位置多样性的综合度量，设计了一种启发式规则来选择假轨迹，从而使得生成的假轨迹能有效隐匿真实轨迹和敏感位置。此外，还提出了轨迹有向图策略和基于网格划分的地图策略来优化算法的执行效率。基于真实的轨迹数据进行实验测试和分析，实验结果表明所提算法在保持数据可用性的情况下能有效保护真实轨迹。

Preventing location disclosure attacks through generating dummy trajectories

In order to solve the problem of trajectory privacy leakage caused by the collection of numerous trajectory information of moving objects， a dummy trajectory-based trajectory privacy protection algorithm was proposed. In this algorithm， considering the user’s locations under disclosure， a heuristic rule was designed based on the comprehensive measure of trajectory similarity and location diversity to select the dummy trajectories， so that the generated dummy trajectories were able to effectively hide the real trajectory and sensitive locations. Besides， the trajectory directed graph strategy and the grid-based map strategy were proposed to optimize the execution efficiency of the algorithm. Experimental results on real trajectory datasets demonstrate that the proposed algorithm can effectively protect the real trajectory with high data utility.