A new cryptanalytic attack for PN-generators filtered by a Boolean function

Sabine Leveiller, Gilles Zémor, Philippe Guillot, Joseph Boutros

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

We present a new cryptanalytic attack on PN-generators filtered by a Boolean function. The key-idea is to jointly combine the knowledge of the Boolean function and the LFSR structure so as to introduce a new iterative decoding technique. The results we obtained prove to be very good: indeed, even in the case of systems for which classical iterative algorithms appear powerless, our scheme enables us to recover the LFSR initial state with a very low error probability. The latter is such that ending the algorithm by an exhaustive search among sequences of limited length is hardly needed. Therefore, the overall complexity of our scheme remains very small and the algorithm appears as an excellent candidate for cryptanalysis of filtered PN-generators.

Original languageEnglish
Pages (from-to)232-249
Number of pages18
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2595
Publication statusPublished - 1 Dec 2003
Externally publishedYes

Fingerprint

Linear Feedback Shift Register
Boolean functions
Boolean Functions
Attack
Generator
Iterative Decoding
Exhaustive Search
Cryptanalysis
Error Probability
Iterative Algorithm
Iterative decoding
Knowledge

Keywords

  • APP decoding
  • Boolean functions
  • Fourier transform
  • Stream ciphers

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{59a128410ac74cb5b81c7c75bbadff46,
title = "A new cryptanalytic attack for PN-generators filtered by a Boolean function",
abstract = "We present a new cryptanalytic attack on PN-generators filtered by a Boolean function. The key-idea is to jointly combine the knowledge of the Boolean function and the LFSR structure so as to introduce a new iterative decoding technique. The results we obtained prove to be very good: indeed, even in the case of systems for which classical iterative algorithms appear powerless, our scheme enables us to recover the LFSR initial state with a very low error probability. The latter is such that ending the algorithm by an exhaustive search among sequences of limited length is hardly needed. Therefore, the overall complexity of our scheme remains very small and the algorithm appears as an excellent candidate for cryptanalysis of filtered PN-generators.",
keywords = "APP decoding, Boolean functions, Fourier transform, Stream ciphers",
author = "Sabine Leveiller and Gilles Z{\'e}mor and Philippe Guillot and Joseph Boutros",
year = "2003",
month = "12",
day = "1",
language = "English",
volume = "2595",
pages = "232--249",
journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - A new cryptanalytic attack for PN-generators filtered by a Boolean function

AU - Leveiller, Sabine

AU - Zémor, Gilles

AU - Guillot, Philippe

AU - Boutros, Joseph

PY - 2003/12/1

Y1 - 2003/12/1

N2 - We present a new cryptanalytic attack on PN-generators filtered by a Boolean function. The key-idea is to jointly combine the knowledge of the Boolean function and the LFSR structure so as to introduce a new iterative decoding technique. The results we obtained prove to be very good: indeed, even in the case of systems for which classical iterative algorithms appear powerless, our scheme enables us to recover the LFSR initial state with a very low error probability. The latter is such that ending the algorithm by an exhaustive search among sequences of limited length is hardly needed. Therefore, the overall complexity of our scheme remains very small and the algorithm appears as an excellent candidate for cryptanalysis of filtered PN-generators.

AB - We present a new cryptanalytic attack on PN-generators filtered by a Boolean function. The key-idea is to jointly combine the knowledge of the Boolean function and the LFSR structure so as to introduce a new iterative decoding technique. The results we obtained prove to be very good: indeed, even in the case of systems for which classical iterative algorithms appear powerless, our scheme enables us to recover the LFSR initial state with a very low error probability. The latter is such that ending the algorithm by an exhaustive search among sequences of limited length is hardly needed. Therefore, the overall complexity of our scheme remains very small and the algorithm appears as an excellent candidate for cryptanalysis of filtered PN-generators.

KW - APP decoding

KW - Boolean functions

KW - Fourier transform

KW - Stream ciphers

UR - http://www.scopus.com/inward/record.url?scp=24144441508&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24144441508&partnerID=8YFLogxK

M3 - Article

VL - 2595

SP - 232

EP - 249

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SN - 0302-9743

ER -