RUS  ENG
Full version
JOURNALS // Informatika i Ee Primeneniya [Informatics and its Applications] // Archive

Inform. Primen., 2019 Volume 13, Issue 1, Pages 67–74 (Mi ia580)

This article is cited in 1 paper

Polynomial algorithms for constructing local affinities of quadratic Boolean functions

O. A. Logachevab, A. A. Sukayeva, S. N. Fedorova

a Information Security Institute, M. V. Lomonosov Moscow State University, 1 Michurinskiy Prosp., Moscow 119192, Russian Federation
b Institute of Informatics Problems, Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: Due to the affine normal form, one can consider a Boolean function as affine on certain flats in its domain — so-called local affinities. This Boolean function representation — affine approximation — could be useful for solving systems of nonlinear equations over two-element field. The problem of solving these systems (of a special sort) arises, in particular, in some methods of the information security tools design and analysis. The paper describes an approach to finding local affinities for quadratic Boolean functions which is based on Dickson's*theorem. By this, one obtains affine normal forms for such functions. Besides, the paper concerns the efficiency of corresponding algorithms. This approach can be profitable for constructing efficient methods of solving systems of quadratic Boolean equations via “approximation” of corresponding Boolean functions by their affine normal forms.

Keywords: Boolean function, system of quadratic Boolean equations, vector space partition, flat, local affinity, Dickson's theorem, affine normal form (ANF) of Boolean function, algebraic cryptanalysis.

Received: 11.01.2019

DOI: 10.14357/19922264190110



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2025