基于多边形构建的差分隐私位置保护方法

Differential Privacy Location Privacy Protection Method Based on Polygon Construction

笔者提出了一种基于多边形构建的差分隐私位置数据保护方案。在该方案中,算法首先从相关数据点集中选择邻接位置点,然后根据受保护的点选择的邻接点构成多边形,并求得多边形的质心,最后利用差分隐私算法加入多边形的质心,用加噪后的多边形质心代替被保护的点,然后重构并发布新的轨迹数据。实验结果表明,算法的隐私保护效果好,数据可用性高。

This paper proposes a differential privacy location data protection scheme based on polygon construction.In this scheme,the algorithm first selects the adjacent location points from the relevant data points,then constructs the polygon according to the adjacent points selected by the protected points,and obtains the center of the polygon.Finally,the center of the polygon is added by the differential privacy algorithm,and the protected points are replaced by the noisy polygon center,and then the new trajectory data is reconstructed and published.Experimental results show that the algorithm has good privacy protection effect and high data availability.