Cryptography Reference
In-Depth Information
replaces the letters of the alphabet with short sequences of dots and dashes. Note
that Morse Code has nothing to do with secrecy, since the codebook in this case
is well known. Morse Code was designed to efficiently transmit messages over
telegraph wires. Another example of a code is ASCII, which provides a means of
converting keyboard symbols into data suitable for processing on a computer (see
the Mathematics Appendix).
If a codebook is kept secret, and is only known by the sender and the receiver of
some data, then the resulting code can be regarded as a type of cryptosystem. In this
case the encryption algorithm is simply to replace the plaintext with its matching
ciphertext entry in the codebook. The decryption algorithm is the reverse process.
The encryption (and decryption) key is the codebook specification itself. For
example, Morse Code is not a cryptosystem because there is only one way of
replacing letters by dots and dashes. However, if the rule for replacing letters
by dots and dashes was kept secret from everyone except a chosen sender and
receiver, then we could regard this as a cryptosystem.
In general, cryptosystems based on codebooks only tend to be referred to as
'codes' when the codebook describes ways of replacing dictionary words by other
words. Thus the term 'code' is most likely to be encountered in reference to
historical cryptosystems or recreational puzzles. The types of cryptosystem that
we will be most interested in do not convert words into words, but rather convert
sequences of ones and zeros into other sequences of ones and zeros. While we
could produce 'codebooks' for these modern cryptosystems, the codebooks would
have to be so large that they would be impractical to use.
The term 'code' is also often used as an abbreviated form of
error-correcting
code
. This is a technique that can be deployed in order to enable the recovery of
correct data from 'noisy' data containing accidental errors that are introduced in
an unreliable channel. Error-correcting codes have nothing to do with preventing
data from being seen by unauthorised users. While they are related to data
integrity, error-correcting codes do not protect data from being
deliberately
manipulated by an attacker. Therefore, we cannot really regard them as
cryptographic primitives.
1.4.5
Steganography
Another concept often confused with cryptography is
steganography
, which is
also concerned with preventing unauthorised users from accessing plaintext data.
However, the basic assumptions behind the use of steganography are rather
different from those of cryptography. Steganography is essentially the study of
information hiding
. The main aim of steganography is for a sender to transfer a
plaintext to a receiver in such a way that only the receiver can extract the plaintext
because only the receiver knows that a hidden plaintext exists in the first place, and
how to look for it (for example, by extracting information from a digital image). In
steganography an 'interceptor' may well be unaware that observed data contains
