CA-Gen: Trajectory Generation with Co-Movement Awareness

Synthetic trajectory generation is essential for addressing privacy concerns and data scarcity in mobility-related applications.However, existing methods in this area ignore co-movement patterns.In this paper, we propose CA-Gen, a Co-movement Aware trajectory Generation framework based on Generative Adversarial Networks (GANs).To enhance the realism of generated trajectories, CA-Gen integrates a waypoint-guided search strategy through frequent subsequence mining.Experiments on real-world datasets demonstrate that CA-Gen outperforms SOTA methods, generating trajectories that preserve both individual mobility behaviors and co-movement dynamics.Our implementation is available online 1 .