FedTDP: A Privacy-Preserving and Unified Framework for Trajectory Data Preparation via Federated Learning
Zhihao Zeng, Ziquan Fang, Wei Shao, Lu Chen, Yunjun Gao
TL;DR
FedTDP addresses privacy and generalizability gaps in trajectory data preparation by orchestrating a privacy-preserving federated framework that leverages a server LLM and client-side SLMs. It introduces three core components: Trajectory Privacy AutoEncoder (TPA) for safe data transmission, Trajectory Knowledge Enhancer (TKE) to tailor TDP-oriented knowledge through prompt engineering and efficient tuning, and Federated Parallel Optimization (FPO) to boost training efficiency via split learning and parallel updates. Across six real datasets and ten TDP tasks, FedTDP outperforms 13 baselines, with improvements ranging from 4.84% to 45.22% relative to LLM-based competitors and at least 18.38% over non-LLM methods, while maintaining privacy. The approach advances practical federated trajectory data processing by combining privacy guarantees, multi-task capability, and efficiency, with code and data made available for reproducibility.
Abstract
Trajectory data, which capture the movement patterns of people and vehicles over time and space, are crucial for applications like traffic optimization and urban planning. However, issues such as noise and incompleteness often compromise data quality, leading to inaccurate trajectory analyses and limiting the potential of these applications. While Trajectory Data Preparation (TDP) can enhance data quality, existing methods suffer from two key limitations: (i) they do not address data privacy concerns, particularly in federated settings where trajectory data sharing is prohibited, and (ii) they typically design task-specific models that lack generalizability across diverse TDP scenarios. To overcome these challenges, we propose FedTDP, a privacy-preserving and unified framework that leverages the capabilities of Large Language Models (LLMs) for TDP in federated environments. Specifically, we: (i) design a trajectory privacy autoencoder to secure data transmission and protect privacy, (ii) introduce a trajectory knowledge enhancer to improve model learning of TDP-related knowledge, enabling the development of TDP-oriented LLMs, and (iii) propose federated parallel optimization to enhance training efficiency by reducing data transmission and enabling parallel model training. Experiments on 6 real datasets and 10 mainstream TDP tasks demonstrate that FedTDP consistently outperforms 13 state-of-the-art baselines.