FedVAE: Trajectory privacy preserving based on Federated Variational AutoEncoder
Yuchen Jiang, Ying Wu, Shiyao Zhang, James J. Q. Yu
TL;DR
FedVAE tackles trajectory privacy by integrating a variational autoencoder with federated learning to generate synthetic trajectories that preserve the original feature space while protecting user privacy. The method trains locally on user devices and aggregates models using FedAVG, producing a global VAE capable of generating new trajectories with similar distributions but low similarity to original data. Experiments on the Geolife dataset for travel mode identification show FedVAE achieves superior privacy protection (lower similarity) and higher downstream task accuracy compared to perturbation-based, MixZone, and k-anonymity baselines, with favorable distributional fidelity as indicated by KL divergence. The work demonstrates a scalable, privacy-utility-preserving trajectory generation framework with practical implications for ITS data sharing and privacy-aware analytics.
Abstract
The use of trajectory data with abundant spatial-temporal information is pivotal in Intelligent Transport Systems (ITS) and various traffic system tasks. Location-Based Services (LBS) capitalize on this trajectory data to offer users personalized services tailored to their location information. However, this trajectory data contains sensitive information about users' movement patterns and habits, necessitating confidentiality and protection from unknown collectors. To address this challenge, privacy-preserving methods like K-anonymity and Differential Privacy have been proposed to safeguard private information in the dataset. Despite their effectiveness, these methods can impact the original features by introducing perturbations or generating unrealistic trajectory data, leading to suboptimal performance in downstream tasks. To overcome these limitations, we propose a Federated Variational AutoEncoder (FedVAE) approach, which effectively generates a new trajectory dataset while preserving the confidentiality of private information and retaining the structure of the original features. In addition, FedVAE leverages Variational AutoEncoder (VAE) to maintain the original feature space and generate new trajectory data, and incorporates Federated Learning (FL) during the training stage, ensuring that users' data remains locally stored to protect their personal information. The results demonstrate its superior performance compared to other existing methods, affirming FedVAE as a promising solution for enhancing data privacy and utility in location-based applications.