M. Dijkslag, M. Damie, F.W. Hahn, A. Peter, "Passive query-recovery attack against secure conjunctive keyword search schemes" (ACNS 2022)
This repo is based on a copy of the original Score Attack publicly available at: Refined-score-atk-SSE, the original score attack is presented in: M. Damie, F. Hahn, A. Peter, "A Highly Accurate Query-Recovery Attack against Searchable Encryption using Non-Indexed Documents" (USENIX 2021)
This code aims at simulating the CKWS-adapted-Refined-score-attack against SSE as PoC.
Download the repo and launch setup.sh.
Then, run the dataset preprocessor to speed up experiments.
python dataset_preprocessor.py
To run an attack scenario:
python attack_scenario.py
NOTE: The co-occurrence matrices generated by the code can be very big, therefore you can gain significant speed-up if you run large matrix operations on a GPU. The code is written to run in such fashion using the Tensorflow library.
All the procedures we used to produce our results are in result_procedures.py.
In this repo you can find bash scripts that were used to run experiments on a cluster running with SLURM workload scheduler.