Differential-and-Linear-Cryptanalysis

Issues

  • Still need to manually simplify POS when using MILP to model S-Box

    • possible solution: solve with pyeda.boolalg.espresso
  • Equations is too long for large (8-bit) S-Box

    • possible solution: ignore small bias

TODO

  • simplify logical condition with python

  • modeling Feistel Network

Reference