RUS  ENG
Full version
JOURNALS // Diskretnyi Analiz i Issledovanie Operatsii // Archive

Diskretn. Anal. Issled. Oper., 2020 Volume 27, Issue 2, Pages 117–135 (Mi da953)

This article is cited in 3 papers

Exact formula for exponents of mixing digraphs for register transformations

V. M. Fomichevabcd, Ya. E. Avezovab

a Financial University under the Government of Russian Federation, 49 Leningradskii Avenue, 125993 Moscow, Russia
b National Research Nuclear University MEPhI, 31 Kashirskoe Highway, 115409 Moscow, Russia
c Institute of Informatics Problems of FRC CSC RAS, 44 Bld. 2 Vavilov Street, 119333 Moscow, Russia
d Security Code LLC, 10 Bld. 1 Pervyi Nagatinskii Driveway, 115230 Moscow, Russia

Abstract: A digraph is primitive if some positive degree of it is a complete digraph, i. e. has all possible edges. The least degree of this kind is called the exponent of the digraph. Given a primitive digraph, the elementary local exponent for some vertices $u$ and $v$ is the least positive integer $\gamma$ such that there exists a path from $u$ to $v$ of every length at least $\gamma$. For transformation on the binary $n$-dimensional vector space that is given by a set of $n$ coordinate functions, the $n$ vertex digraph corresponds such that a pair $(u,v)$ is an edge if the $v$th coordinate component of transformation essentially depends on $u$th variable. Such a digraph we call a mixing digraph of transformation.
We study the mixing digraphs of widely used in cryptography $n$-bit shift registers with nonlinear Boolean feedback function (NFSR), $n>1$. We find the exact formulas for the exponent and elementary local exponents for $n$-vertex primitive mixing digraph associated to NFSR. For pseudo-random sequences generators based on the NFSRs, our results can be applied to evaluate the length of blank run. Bibliogr. 20.

Keywords: mixing digraph, primitive digraph, locally primitive digraph, feedback shift register, exponent of a digraph.

UDC: 519.17

Received: 06.09.2019
Revised: 27.09.2019
Accepted: 19.02.2020

DOI: 10.33048/daio.2020.27.670


 English version:
Journal of Applied and Industrial Mathematics, 2020, 14:2, 308–319

Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024