Source codes and data for thesis

Master's thesis: Algebraic Cryptanalysis of Small Scale Variants of the AES via Reduction Theory

Author: Tomáš Faikl

FNSPE in Prague, FIT

Installation:

Create conda environment and install newest SageMath (tested with 10.4, 2024-07-19) via conda. Optionally install also galois for linear algebra over Numpy matrices in finite fields.

Compile CodeRed reduction library in packaged format in packages/. Then, install all supplied python packages via packages/install-pkgs.sh. If needed, adjust the number of cores in packages/aesalgae/config.py.

Finally, run

# run SR(2,4,2,4), PC=32, r=1, preprocessing: HME -> LLL; Groebner basis using F4 with 24 threads and GPU
sage scripts/aes.py -n 2 -r 4 -c 2 -e 4 --pc-pairs 32 -T24 --preprocessing monomelim,lll --pre-reduce-mult 1

# run SR(3,2,2,4), PC=1, no preprocessing; Groebner basis using F4 with 24 threads and GPU
sage scripts/aes.py -n 3 -r 2 -c 2 -e 4 --pc-pairs 1 -T24

Alternatively, execute batches via scripts/aes-batch.json and supply json file describing the batch. Used examples are in scripts/batches/*.json.