Rust implementation of the algebraic immunity computation of Boolean functions, based on the paper
"Efficient Computation of Algebraic Immunity for Algebraic and Fast Algebraic Attacks"
(Armknecht et al., 2006, DOI: https://doi.org/10.1007/11761679_8). However, the algorithm presented in the paper relies on implicit assumptions that make direct implementation infeasible. Our implementation uses two new propositions from our paper "Computing the Restricted Algebraic Immunity, and Application to WPB Functions"
Luca Bonamino and Pierrick Méaux. IACR ePrint 2025/1779, that resolve this issue: they drastically reduce complexity and make the non-restricted algorithm implementable in practice — while also enabling our efficient extension to the restricted case.
This Rust implementation is wrapped as a Python package using PyO3 (https://pyo3.rs) and maturin (https://github.com/PyO3/maturin).
This library was developed to provide a robust and reliable implementation of algebraic immunity computation. It addresses a correctness issue in SageMath’s BooleanFunction.algebraic_immunity() method, which can raise internal errors or produce incorrect results for certain Boolean functions — particularly those with two variables, such as [1, 0, 0, 1].
This Rust implementation has been tested extensively and is suitable for both small and large truth tables, with a focus on correctness and Python accessibility.
You can install the package via PyPI or use the pre-built wheels provided in the GitHub releases.
To install the package directly from PyPI:
pip install algebraic_immunityIf PyPI installation doesn't work for your platform, you can use the pre-built wheels:
- Run the following script to determine the correct wheel for your platform:
python construct_wheel_url.py- Then install the wheel using:
pip install <output of the previous command>from algebraic_immunity import AlgebraicImmunity
truth_table = [0, 1, 1, 1, 1, 0, 0, 1]
n = 3
ai = AlgebraicImmunity.algebraic_immunity(truth_table, n)
print(f"Algebraic immunity: {ai}")- Parts of this repository are more broadly relevant to operations over GF(2) matrices. These components could be extracted into a standalone Rust crate in the future.
- The current implementation sometimes uses the
clonemethod unnecessarily to bypass Rust’s borrowing rules. These instances will be reviewed and optimized in future updates.