基于Fréchet距离函数的轨迹隐私保护方法

Trajectory Privacy Protection Method Based on Frechet Distance Function

针对传统的基于欧几里得距离函数计算轨迹相似性过程,要求轨迹等长且时间点对应,没有考虑轨迹的形状,一定程度影响了轨迹隐私保护的效果和数据的利用率问题,提出了一种基于Fréchet距离函数的轨迹隐私保护方法.该方法将轨迹运动方向相近且平均速度相近的轨迹匿名在一起,利用Fréchet距离方法计算轨迹间的距离,最后利用轨迹图之间的权值实现轨迹匿名集合.与其他隐私保护算法比较,在隐私保护和信息损失率上都有一定的改进,表明该方法在处理隐私保护数据方面是有效的.

For the traditional trajectory similarity calculation process based on Euclidean distance function,the trajectory is required to be equal in length and corresponding in time points,however,does not consider the shape of the trajectory,which affects the effect of trajectory privacy protection and the utilization of data to a certain extent.A research on trajectory privacy protection method based on Frechet distance function was conducted.In this method,trajectories with similar motion directions and similar average velocities were anonymous together,and the Frechet distance method was used to calculate the distance between trajectories.Finally,the weights between trajectory maps were used to realize the anonymous set of trajectories.Compared with other privacy protection algorithms,there are some improvements in privacy protection and information loss rate,which indicates that this method is effective in processing privacy protection data.