School of Electronic Information Engineering, Taiyuan University of Science and Technology, Taiyuan 030024, China.
College of Mathematics and Computer Science, Shanxi Normal University, Taiyuan 030039, China.
Sensors (Basel). 2023 May 31;23(11):5219. doi: 10.3390/s23115219.
Location-based services (LBS) are widely used due to the rapid development of mobile devices and location technology. Users usually provide precise location information to LBS to access the corresponding services. However, this convenience comes with the risk of location privacy disclosure, which can infringe upon personal privacy and security. In this paper, a location privacy protection method based on differential privacy is proposed, which efficiently protects users' locations, without degrading the performance of LBS. First, a location-clustering (L-clustering) algorithm is proposed to divide the continuous locations into different clusters based on the distance and density relationships among multiple groups. Then, a differential privacy-based location privacy protection algorithm (DPLPA) is proposed to protect users' location privacy, where Laplace noise is added to the resident points and centroids within the cluster. The experimental results show that the DPLPA achieves a high level of data utility, with minimal time consumption, while effectively protecting the privacy of location information.
基于位置的服务（LBS）由于移动设备和位置技术的快速发展而得到广泛应用。用户通常向 LBS 提供精确的位置信息以访问相应的服务。然而，这种便利性带来了位置隐私泄露的风险，可能会侵犯个人隐私和安全。在本文中，提出了一种基于差分隐私的位置隐私保护方法，该方法可以有效地保护用户的位置，而不会降低 LBS 的性能。首先，提出了一种位置聚类（L-clustering）算法，该算法根据多组之间的距离和密度关系将连续位置划分为不同的簇。然后，提出了一种基于差分隐私的位置隐私保护算法（DPLPA），该算法在簇内的驻留点和质心处添加拉普拉斯噪声以保护用户的位置隐私。实验结果表明，DPLPA 在最小化时间消耗的同时实现了高水平的数据效用，有效地保护了位置信息的隐私。
