基于重力模型生成假轨迹的隐私保护方法

Generating Dummies Based on Gravity Model for User's Trajectory Privacy

针对连续查询场景中用户实时位置的隐私保护问题,设计了一种基于客户端的假轨迹生成方法.该方法使用网格划分地理空间,统计网格划分后每个网格内的历史查询数据.通过分析网格内的历史查询数据构建实时预测用户移动轨迹的重力模型.在重力模型基础上结合历史查询概率定义了轨迹熵度量轨迹隐私保护等级,并在最大运行速度限制下,提出了一种具有最大轨迹熵的基于k-匿名的假轨迹隐私保护算法.实验结果验证了所设计的假轨迹生成方法能够有效地保护真实轨迹的隐私.

The real-time location privacy preserving is a hotspot in continuous Location-Based Services （LBSs）.A client-based dummy trajectory generation method is proposed.Based on the spatial grid partition, the history data in each cell of this grid is analyzed.Using the gravity model, a prediction model for users movement pattern is built.Combined with the movement pattern model and the history query probability, the author defines a trajectory entropy to metric the trajectory privacy level.Based on k-anonymity principle, a limited velocity dummy trajectory generation algorithm with maximum trajectory entropy is proposed.Experiments from synthesis data and real-world data validate the effectiveness of our proposed method.