Cryptography Reference
In-Depth Information
In this chapter we have examined a number of historical cryptosystems. Although
none of these cryptosystems are suitable for modern use, they provide several
important design lessons for modern cryptosystems. These include:
• A large keyspace is necessary in a practical cryptosystem, but a large keyspace
alone does not guarantee security.
• It is possible to break a cryptosystem without first determining the key.
Designers of cryptosystems thus need to worry about much more than just
protection of the key.
• The ciphertext produced by a cryptosystem should disguise the statistics of the
underlying plaintext alphabet.
• Effective techniques for disguising plaintext statistics include increasing alpha-
bet size, ensuring that plaintext letters encrypt to a variety of different ciphertext
letters, and introducing positional dependency, but these properties alone do
not guarantee security.
• Efficiency and security are often traded off against one another when designing
a cryptosystem.
• It is unlikely that anyone will use a cryptosystemwhere the balance of efficiency
versus security is inappropriate. In particular, secure cryptosystems that are
inefficient to use in practice are not attractive for most applications.
• We can design cryptosystems to be secure against attacks that we know
and understand, but unknown attacks could be discovered at any time in the
future.
Most of the modern cryptosystems that we will later examine have been designed
with these lessons in mind.
There are many topics that explain historical cryptosystems and an interested
reader should have no trouble finding them. These include general texts such
as Stinson [185] and topics more dedicated to classical cryptography such as
Spillman [180]. Many of these will providemore details on how to break the Vigenère
Cipher (see also [176] for this).
One of the most compelling publicists of the links between historical and modern
cryptography has been Simon Singh. His influential and very readable The Code Book
[176] has some very interesting background information on historical cryptography
(andmuchmore), andwasmade into a five-part television series calledThe Science of
Secrecy. He has produced a version of this topic for younger audiences [177], as well
as maintaining a website [175] from which a related CD-ROM can be downloaded.
