Cryptology
Visit our simulations on cryptology for more information.

SOME AUTHORITATIVE REFERENCES:

SOME WEB RESOURCES:

REPRESENTING TEXT
All of these methods are probably still in use today

  1. Text in coded messages is often represented by the lower-case alphabet alone, with no spaces, tabs, paragraphs or punctuation of any kind. If this is done, then it also can be represented by the decimal numbers 1-26 or their binary equivalent. In programming applications this can be easily accomplished by referencing a small subset of the ASCII code.
  2. There is also no reason why the entire ASCII character set cannot be used. In this case each character may be represented by 7 bits, or simply by one 8-bit byte.
  3. Numerical codes pose a different problem. Every pair of decimal digits will represent the numbers from 00 to 99. If we reduce the alphabet to 25 characters, dropping the "Z,"we can represent the alphabet by the ranges 00-24, 25-49, 40-64 and 65-99, selecting one of the four ranges at random. However, this is not particularly efficient.
  4. We can increase efficiency if we choose an extended "alphabet" of 100 single letters as well as common digraphs and trigraphs. Such a table is shown among the images below. The first number of the pair may reference the row, the second number the column. Of course, an efficient table for each language will be different.
  5. Another method is to have 4 or 5-digit numbers reference an entry for a word or phrase in a code book. This practice had its beginning in telegraphy in the 19th Century.

ABSOLUTE SECURITY with a TRULY RANDOM KEY
One-Time-Systems (Pads, Silk, Tape or Disk) The Vernam Cipher

A system in which a different scrambled alphabet is used for each character in a message. The set of references to the scrambled alphabets used in a message is referred to as its key and is supplied to the sender and recipient on pads of paper, silk, tape or disk. Typically, the cleartext is added to the keytext to produce a cryptotext. If one knows the key to the ciphertext the cleartext may be recovered by subtracting the keytext from the cryptotext. (All addition and subtraction is done modulo 26; the alphabet wraps around.) The key could be the text from a page of a widely published novel, lead article in an international newspaper or the headline story on cnn.com. This method is unbreakable if, and only if:

  1. The keytext is of the same length as the cleartext.
  2. The keytext is truly random. (The randomness of noise from electronic components and radioactive decay are good candidates.)
  3. The keytext is only used once, and is then destroyed.

The problem with the one-time-pad is that both the sender and recipient have to have copies of a number of keys. The keys must be delivered by courier and every precaution must be made to insure that the keys are not intercepted and copied.

Some One-Time-Systems:

Second World War
Second World War
East German
Russian
Russian
A Windows application for PCs.
A numeric representation of text.

PRETTY GOOD SECURITY with a PSEUDO-RANDOM KEY
Early Beginnings

Below is illustrated one excessively complex device for generating pseudo-random keys by hand. From a few initial settings it was possible to generate a long string of seemingly random numbers. The device was too complicated for use in the field and tended to fall apart if not assembled on a desk. It was abandoned soon after it was introduced to the West German Army during the cold war.

Most of the cipher equipment developed during the Second World War and the Cold War relied on mechanical and electromechanical methods of generating a long pseudo-random number key from a much shorter list of settings. These were mechanical and electromechanical dedicated computers, precursors of today's silicon based machines. The settings, which could change by the hour, the day, the week or month, would be distributed to the sender and receiver long before they were needed. Once the settings were input into the machine, the machine would generate a long key which was otherwise extremely difficult to predict by an outside party. Perusing the various Web links listed above will show you examples of many such devices, the best known being the German Enigma machine.

Another way to look at pretty good security is to say it's not secure enough. If it's not totally secure, then the codes (the ciphers) can be cracked. And so the cryptanalysts set out to search for solutions using guesses (cribs) and advanced statistical methods, frequently aided by cryptographer's mistakes. Inspired by the work of the Poles in early successes in breaking the German Enigma codes, the British and the Americans built huge electromechanical devices dedicated to narrowing the search for plaintexts. These were early computers, programmed in hardware to solve a narrow range of critically important problems. It was during this period of intense research and development that Alan Turing, and others, devised the general architecture of the modern computer.

SOME EXCELLENT VIDEO RESOURCES ON CRACKING THE ENIGMA
Kept secret for half a century.