Privacy-Preserving Modeling of Trajectory Data: Secure Sharing Solutions for Trajectory Data Based on Granular Computing

Abstract

Trajectory data are embedded within driving paths, GPS positioning systems, and mobile signaling information. A vast amount of trajectory data play a crucial role in the development of smart cities. However, these trajectory data contain a significant amount of sensitive user information, which poses a substantial threat to personal privacy. In this work, we have constructed an internal secure information granule model based on differential privacy to ensure the secure sharing and analysis of trajectory data. This model deeply integrates granular computing with differential privacy, addressing the issue of privacy leakage during the sharing of trajectory data. We introduce the Laplace mechanism during the granulation of information granules to ensure data security, and the flexibility at the granularity level provides a solid foundation for subsequent data analysis. Meanwhile, this work demonstrates the practical applications of the solution for the secure sharing of trajectory data. It integrates trajectory data with economic data using the Takagi–Sugeno fuzzy rule model to fit and predict regional economies, thereby verifying the feasibility of the granular computing model based on differential privacy and ensuring the privacy and security of users’ trajectory information. The experimental results show that the information granule model based on differential privacy can more effectively enable data analysis.