This is the source code used for the research paper Automated Website Fingerprinting through Deep Learning:
https://tor-wf-dl.distrinet-research.be/
The code is currently not maintained, email us for updates: vera.rimmer@cs.kuleuven.be
If you found our framework or data useful for your research, please cite:
@inproceedings{Rimmer2018,
title={Automated Website Fingerprinting through Deep Learning},
author={Rimmer, Vera and Preuveneers, Davy and Juarez, Marc and Van Goethem, Tom and Joosen, Wouter},
booktitle={Proceedings of the 25nd Network and Distributed System Security Symposium (NDSS 2018)},
year={2018},
organization={Internet Society}
}
The datasets used for the evaluation are available to download under these links:
Closed World --
https://tor-wf-dl.distrinet-research.be/files/tor_100w_2500tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_200w_2500tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_500w_2500tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_900w_2500tr.npz
Open World --
https://tor-wf-dl.distrinet-research.be/files/tor_open_200w_2000tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_open_400000w.npz
Concept drift --
https://tor-wf-dl.distrinet-research.be/files/tor_time_test3d_200w_100tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_time_test10d_200w_100tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_time_test2w_200w_100tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_time_test4w_200w_100tr.npz
https://tor-wf-dl.distrinet-research.be/files/tor_time_test6w_200w_100tr.npz