Differential Privacy-Based Location Privacy Protection with Hilbert Curve in Vehicular Networks

The mishandled or inadequately of the location data can lead to significant privacy breaches. Existing studies employ differential privacy and Hilbert curve on a timestamp or multiple points, overlooking the correlation between a moving user’s location and its historical location, which can be susceptible to various attacks. To address this issue, we propose a complete trajectory data set protection mechanism that clusters the locations and generates anonymous sets of locations. The anonymous sets are utilized to generate fine-grained obfuscated results of the proposed differential privacy mechanism. We consider the correlation between the user’s current location and its location history in vehicular networks with the Hilbert curve, which provides fine-grained obfuscation and high-level privacy protection. By carefully crafting the weight of the proposed mechanism, we overcome the privacy breach caused by the DP in protecting the same location multiple times. The experimental results show that the proposed mechanism outperforms the existing studies in privacy protection.