A localized differential privacy-based location preservation method using a hidden Markov model in vehicular networks

In the rapid development of vehicular networks, the exchange of information between vehicles and between vehicles and the infrastructure is becoming increasingly frequent. Vehicle location data have become one of the core types of information. However, location data contain a large amount of user privacy, and once leaked, they can severely threaten user safety and freedom. Therefore, there is an urgent need for a method that can balance privacy protection, data availability, and processing efficiency. In this paper, we propose a localized differential privacy location protection method based on the hidden Markov model (HMM) in vehicular networks, which addresses the problem of vehicle location privacy protection. The method consists of an HMM-based continuous location privacy protection algorithm and a localized differential privacy perturbation algorithm. The algorithm introduces the HMM into the field and utilizes its ability to accurately predict the continuous changes in vehicle location, thereby providing a scientific basis for privacy protection. At the same time, it combines the spatial correlation of location distribution to construct a privacy-preserving security area, which effectively restricts the range of the localized differential privacy perturbation, reduces the error, and improves data availability while safeguarding privacy. Second, this method incorporates a two-stage localized differential privacy perturbation algorithm to achieve dynamic differential privacy protection of vehicle location, adapting to real-time changes in vehicle location data through collaboration between the client and server. Based on the experiments and analysis using an actual trajectory dataset, the results show that the method provides strong privacy protection, high data availability, and efficient processing, thereby verifying its feasibility and effectiveness.