1. Introduction
I
nformation security is generally accepted to be essential to
modern business and technology, both for privacy of
transactions and communications, as well as for defense
against malicious intruders. Cryptography is the study of
information security and the feasibility of communication over
an insecure channel while preserving the secrecy of the infor-
mation transmitted [1], [2]. Cryptographic techniques should
offer at least the following three security features concerning
data transmission: confidentiality, authentication and integrity.
Confidentiality is fundamental—third parties are expected to
see the encrypted data but should not be able to decipher it.
Authentication methods allow the receiver to verify that the
sender is legitimate. Lack of authentication makes systems vul-
nerable to fraudulent transactions and denial of service attacks.
Integrity of the transmitted data must be verifiable, i.e., the
receiver should be able to check that no part of the message
M. Arvandi, S. Wu,
and A. Sadeghian
Ryerson University, CANADA
© ARTVILLE
On the Use of Recurrent
Neural Networks to Design
Symmetric Ciphers
Abstract: In this article, we
describe an innovative form of
cipher design based on the use
of recurrent neural networks.
The well-known characteristics
of neural networks, such as par-
allel distributed structure, high
computational power, ability to
learn and represent knowledge
as a black box, are successfully
applied to cryptography. The
proposed cipher has a relatively
simple architecture and, by
incorporating neural networks,
it releases the constraint on the
length of the secret key. The
design of the symmetric cipher
is described in detail and its
security is analyzed. The cipher
is robust in resisting different
cryptanalysis attacks and pro-
vides efficient data integrity and
authentication services. Simula-
tion results are presented to val-
idate the effectiveness of the
proposed cipher design.
Digital Object Identifier 10.1109/MCI.2008.919075
42 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | MAY 2008 1556-603X/08/$25.00©2008IEEE