Cryptography Reference
In-Depth Information
from the source to the destination) is at least to some extent independent of the
noise on the channel from the sender to the adversary. This system demonstrates
that a mere difference in the signals received by the legitimate receiver and by
the adversary, but not necessarily with an advantage to the legitimate receiver, is
sufficient for achieving security. From Maurer's results, one may also conclude
that, for cryptographic purposes, a given communication channel that is noisy is not
necessarily bad. In addition, such a channel should not be turned into an error-free
channel by means of error-correcting codes, but rather that cryptographic coding and
error-control coding should be combined.
10.5
FINAL REMARKS
In this chapter, we elaborated on symmetric encryption systems and had a closer
look at some exemplary systems (i.e., the block ciphers DES and AES, and the
stream cipher ARCFOUR). These systems are arbitrarily chosen and only reflect
their wide deployment. In fact, the DES has been widely deployed for financial
applications, whereas the AES is slowly but steadily replacing DES. RC4 is widely
deployed because it is built into many commercial off-the-shelf (COTS) products
(e.g., Web browsers and mail clients). There are many other symmetric encryption
systems developed and proposed in the literature that are not addressed in this topic.
Examples include the other AES finalists (i.e., the competitors of Rijndael), MISTY1
[28, 29], Camellia [30], and SHACAL-2 from the New European Schemes for
Signatures, Integrity and Encryption (NESSIE) project, as well as some more recent
proposals (e.g., [31]). It goes without saying that all of these systems represent good
alternatives.
All symmetric encryption systems in use today look somehow similar in the
sense that they all employ a mixture of more or less complex functions that are
iterated multiple times (i.e., in multiple rounds) to come up with something that
is inherently hard to understand and analyze. There are also many details in a
cryptographic design that may look mysterious or arbitrary to some extent. For
example, the S-boxes of DES look arbitrary, but they are not and are well chosen to
protect against differential cryptanalysis (which was published almost two decades
after the DES specification). Against this background, one may get the impression
that it is simple to design and come up with a new symmetric encryption system.
Unfortunately, this is not the case, and the design of a symmetric encryption system
that is secure and efficient is still a very tricky business. Legions of systems had
been proposed, implemented, and deployed before a formerly unknown successful
attack was discovered and applied to break them. Such a discovery then often
brings the end to several symmetric encryption systems. For example, the discovery
