Scholarlink

<p>To address the limitations of existing trajectory privacy protection approaches, particularly the low data utility and the difficulty of balancing privacy protection with analytical effectiveness in complex scenarios, this study proposes a trajectory data privacy protection method based on local differential privacy (TDPP-LDP). First, a sliding window algorithm is applied to extract representative stay points in order to identify high-risk sensitive regions within trajectories. Based on this, a spatial perturbation algorithm constrained by a maximum perturbation radius is designed to effectively protect sensitive locations while reducing the loss of data utility caused by trajectory direction mutation. Furthermore, to mitigate privacy threats arising from spatiotemporal reasoning, a temporal perturbation algorithm constrained by a time-safe region is introduced. Experimental results demonstrate that the proposed TDPP-LDP method achieves strong performance and significantly improves data utility while ensuring effective privacy protection for trajectory data.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Trajectory data privacy protection method based on local differential privacy

  • 期刊论文

  • 发表日期：

    2026年4月13日

查看全文
  • Zhang Lei,
  • Chen Jie,
  • Chen Yun,
  • Bai Yongbo,
  • Zhang Yunxiang

摘要

To address the limitations of existing trajectory privacy protection approaches, particularly the low data utility and the difficulty of balancing privacy protection with analytical effectiveness in complex scenarios, this study proposes a trajectory data privacy protection method based on local differential privacy (TDPP-LDP). First, a sliding window algorithm is applied to extract representative stay points in order to identify high-risk sensitive regions within trajectories. Based on this, a spatial perturbation algorithm constrained by a maximum perturbation radius is designed to effectively protect sensitive locations while reducing the loss of data utility caused by trajectory direction mutation. Furthermore, to mitigate privacy threats arising from spatiotemporal reasoning, a temporal perturbation algorithm constrained by a time-safe region is introduced. Experimental results demonstrate that the proposed TDPP-LDP method achieves strong performance and significantly improves data utility while ensuring effective privacy protection for trajectory data.