TrjPrivacy

Source code of TKDE'23 paper: A Survey and Experimental Study on Privacy-Preserving Trajectory Data Publishing.

Compared Models

Formal models
- k-anonymity models: W4M [1], GLOVE [2], KLT [3]
- differential privacy models: DPT [4], AdaTrace [5]
Ad-hoc models
- mixzone [6]
- dummy [7]

Source Codes

We would like to thank the authors for kindly providing the publicly-available runable source code:

W4M
AdaTrace
DPT: we obtained it directly from the authors but don't have the right to make it public here.
GLOVE, KLT, mixzone, dummy: implemented by ourselves in Java and provided in this repository.

Environment

Tested in CentOS Linux and MacOS Monterey (jdk 17.0.1)

Datasets

T-Drive and Geolife are publicly available datasets provided by Microsoft. A small subset of T-Drive is provided in this project for testing.

Project Structure

Testing/                                  
├── BeijingRoadNetworkInfo/                -- the road network information, including road verties/edges and all Beijing POIs
├── formal_inputs/                         -- two small sets of T-Drive data have been uploaded here:
                                              one contains 100 objects' total data ("T-Drive_object/100.csv")
                                              the other contains 1000 objects' down-sampling data ("T-Drive_sample/3600.csv")
                                              some objects could be discarded due to too few data points
src/                                      
├── resources/config.properties            -- the program will read parameters here (modify it according to your requirements)
├── models/Main.java                       -- the entry of the program
├── models/                                -- four models: mixzone, dummy, GLOVE, KLT
├── shared/                                -- some useful functions/classes shared by privacy models
├── spatial/                               -- spatial structures used in this project
├── evaluation/                            -- the evaluation metrics (coming soon)

libs                                       -- some external library packages that are necessary to be included in the program
...

References

[1] O. Abul, F. Bonchi, and M. Nanni, “Anonymization of moving objects databases by clustering and perturbation,” Inf. Syst., vol. 35, no. 8, pp. 884–910, 2010.

[2] M. Gramaglia and M. Fiore, “Hiding mobile traffic fingerprints with GLOVE,” in Proc. CoNEXT. ACM, 2015, pp. 26:1–26:13.

[3] Z. Tu, K. Zhao, F. Xu, Y. Li, L. Su, and D. Jin, “Protecting trajectory from semantic attack considering k-anonymity, l-diversity, and t-closeness,” IEEE Trans. Netw. and Service Management, vol. 16, no. 1, pp. 264–278, 2019

[4] X. He, G. Cormode, A. Machanavajjhala, C. M. Procopiuc, and D. Srivastava, “DPT: differentially private trajectory synthesis using hierarchical reference systems,” Proc. VLDB, vol. 8, no. 11, pp. 1154–1165, 2015.

[5] M. E. Gursoy, L. Liu, S. Truex, L. Yu, and W. Wei, “Utility-aware synthesis of differentially private and attack-resilient location traces,” in Proc. SIGSAC. ACM, 2018, pp. 196–211

[6] X. Liu, H. Zhao, M. Pan, H. Yue, X. Li, and Y. Fang, “Traffic-aware multiple mix zone placement for protecting location privacy,” in Proc. INFOCOM. IEEE, 2012, pp. 972–980.

[7] X. Liu, J. Chen, X. Xia, C. Zong, R. Zhu, and J. Li, “Dummy-based trajectory privacy protection against exposure location attacks,” in Proc. WISA, ser. Lecture Notes in Computer Science, vol. 11817. Springer, 2019, pp. 368–381

Citation

If you find our library or the experimental results useful, please kindly cite the following paper:

@ARTICLE{jin2023,
  author={Jin, Fengmei and Hua, Wen and Francia, Matteo and Chao, Pingfu and Orlowska, Maria E and Zhou, Xiaofang},
  journal={IEEE Transactions on Knowledge and Data Engineering}, 
  title={A Survey and Experimental Study on Privacy-Preserving Trajectory Data Publishing}, 
  year={2023},
  volume={35},
  number={6},
  pages={5577-5596},
  doi={10.1109/TKDE.2022.3174204}
}

Please feel free to contact fengmei.jin@uq.edu.au if encountering any unexpected issues in this project.

CharlotteE67/TrjPrivacy