FERROALUMINIUM

FERROALUMINIUM ( Fast Dated Encipherment Algorithm ) is a algorithm of Chiffrement per block proposed like a faster and sure alternative of OF the. Published in 1987 by Akihiro Shimizu and Shoji Miyaguchi of NTT, this coding was passably studied and made it possible to refine the methods of Cryptanalyse. It has from this mainly contributed point of view, at the sides of, the birth of the linear Cryptanalyse and differential .

FEAL-4

FERROALUMINIUM was updated on several occasions but the diagram remains similar and works on a block of 64 bits. The oldest architecture, FEAL-4, called upon 4 turns and a key of 64 bits. One finds however in the literature of the mentions of FEAL-1 and FEAL-2 but there exists little of information on this subject. FEAL-4 quickly will be cryptanalysé and broken. Bert den Boer showed a vulnerability as of the publication of coding. One year later, den Boer describes an attack which requires a reasonable number of selected clear texts (of 100 with 10 000 messages) and Sean Murphy publishes in 1990 an improvement with only 20 preset clear texts by the Cryptanalyste. These methods contain already elements which will appear later in the differential cryptanalyse.

FEAL-8

By doubling the number of rounds, the FERROALUMINIUM originators thought of eliminating the attacks but as of 1989, Eli Biham and Adi Shamir give the broad outlines to the conference Securicom , of a differential attack. Henri Gilbert and Chassé publishes an attack similar to a differential cryptanalyse in 1990, the result required 10 000 pairs of selected clear texts.

FEAL-N and FEAL-NX

To answer these attacks, the originators paramétrisent the number of revolutions: FEAL-N (1990). The parameter NR was chosen by the user and a named alternative FEAL-NX, uses a key of 128 bits. In 1991, the differential cryptanalyse of Biham and Shamir shows that FEAL-N just like FEAL-NX can be broken, with however a relatively large complexity. Under the assumption of an attack with text clearly chosen, other attacks, anticipating the linear cryptanalyse, could break FEAL-4 with 5 clear texts (Matsui and Yamagishi in 1992). FEAL-6 could be broken with 100 texts and FEAL-8 with 2¹⁵ messages.

See too

N-Hash

References

Eli Biham, Adi Shamir: Differential Cryptanalysis off Ferroaluminium and N-Hash. EUROCRYPT 1991:1 - 16
Bert den Boer, Cryptanalysis off F.E.A.L., EUROCRYPT 1988:293 - 299
Henri Gilbert, Guy Driven out: In Statistical Attack off the FEAL-8 Cryptosystem. CRYPTO 1990:22 - 33.
Shoji Miyaguchi: The FERROALUMINIUM Cipher Family. CRYPTO 1990:627 - 638
Shoji Miyaguchi: The FEAL-8 Cryptosystem and has Call for Attack. CRYPTO 1989:624 - 627
Mitsuru Matsui, Atsuhiro Yamagishi: In New Method for Known Plaintext Attack off Cipher FERROALUMINIUM. EUROCRYPT 1992:81 - 91
Sean Murphy, The Cryptanalysis off FEAL-4 with 20 Chosen Plaintexts. J. Cryptology 2 (3): 145-154 (1990)
A. Shimizu and S. Miyaguchi, Fast encipherment dated algorithm FERROALUMINIUM, Advances in Cryptology - Eurocrypt '87, Springer-Verlag (1988), 267-280.
Anne Tardy-Corfdir, Henri Gilbert: In Known Plaintext Attack off FEAL-4 and FEAL-6. CRYPTO 1991:172 - 181

External bonds

the page of FERROALUMINIUM
an article on sci.crypt of Peter Gutmann about FERROALUMINIUM
US License n°4850019

Random links: