Bidirectional LSTM-Based Privacy Preserving Method for Trajectory Generation

DOI: https://doi.org/10.62517/jike.202404214

Author(s)

Xiangjie He, Tingting Gao, Yihan Yin, Wei Jiang

Affiliation(s)

Institute of Computer Science and Information Engineering, Harbin Normal University, Harbin, China

Abstract

To ensure the privacy of trajectory data while improving its usability upon release, machine learning algorithms can be applied to process trajectory data, thereby enhancing its usability. Addressing the issue of trajectory data release usability, we propose a trajectory privacy protection scheme that combines Bidirectional Long Short-Term Memory (BILSTM) networks and differential privacy (DP). The scheme prepossess the trajectory data using BILSTM to improve its usability. For the generated trajectory data, the Laplace mechanism in differential privacy is applied to add noise, thereby achieving privacy protection. The generalized trajectory data-set obtained is then released. This scheme ensures good data usability and offers certain efficiency advantages.

Keywords

Trajectory Privacy; Neural Networks; Trajectory Generation

References

[1] Cao Xiaoqian. Research on trajectory synthesis method based on GAN [D]. Zhejiang Normal University, 2023. [2] Gramaglia M, Fiore M, Furno A, et al. GLOVE: towards privacy-preserving publishing of record-level-truthful mobile phone trajectories[J]. ACM/IMS Transactions on Data Science (TDS), 2021, 2(3): 1-36. [3] Mahdavifar S, Deldar F, Mahdikhani H. Personalized privacy-preserving publication of trajectory data by general-ization and distortion of moving points[J]. Journal of Network and Systems Management, 2022, 30(1): 10. [4] Hou L, Yao N, Lu Z, et al. Tracking based mix-zone location privacy evaluation in VANET[J]. IEEE Transactions on Vehicular Technology, 2021, 70(10): 10957-10969. [5] Hua J, Gao Y, Zhong S. Differentially private publication of general time-serial trajectory data[C]. 2015 IEEE Conference on Computer Communications (INFOCOM), 2015, 549-557. [6] Dwork C, McSherry F, Nissim K, et al. Calibrating noise to sensitivity in private data analysis[C]. Theory of Cryp-tography: Third Theory of Cryptography Conference, 2006, 265-284. [7] Cai S, Lyu X, Li X, et al. A trajectory released scheme for the internet of vehicles based on differential privacy[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(9): 16534-16547. [8] Zhao X, Pi D, Chen J. Novel trajectory privacy-preserving method based on prefix tree using differential privacy[J]. Knowledge-Based Systems, 2020, 198: 105940. [9] Yan Yan, Cong Yiming, Adnan Mahmood, et al. A deep learning based approach for statistical release and privacy protection of location big data[J]. Journal of Communication,2022,43(01):203-216. [10] Chen S, Fu A, Shen J, et al. RNN-DP: A new differential privacy scheme base on Recurrent Neural Network for Dynamic trajectory privacy protection[J]. Journal of Network and Computer Applications, 2020, 168: 102736. [11] Shen Z, Zhang Y, Wang H, et al. BiGRU-DP: Improved differential privacy protection method for trajectory data publishing[J]. Expert Systems with Applications, 2024, 252: 124264. [12] QIN Chengyi, WU Lei, WEI Xiaochao, et al. Research and progress of location trajectory correlation differential privacy protection technology[J]. Journal of Cryptography, 2023, 10(06): 1118-1139.