Difference between revisions of "Random numbers, Encryption and Hashing"

From Organic Design wiki
m
m (fner fner)
Line 11: Line 11:
 
∞-distributed is an [[W:Independent and identically-distributed random variables|iid]] process chacterized as a sequence of numbers
 
∞-distributed is an [[W:Independent and identically-distributed random variables|iid]] process chacterized as a sequence of numbers
  
A [[W:Pseudorandom number generator|pseudo random number generator]] is a generator that produces a series of numbers that pass a reasonable set of statistical testes.
+
A [[W:Pseudorandom number generator|pseudo random number generator]] is a generator that produces a series of numbers that pass a reasonable set of statistical tests.
  
 
In the 1950's Linear congruential random number generators (LCG's) were used to generate pseudo-random numbersbased on the computation;
 
In the 1950's Linear congruential random number generators (LCG's) were used to generate pseudo-random numbersbased on the computation;

Revision as of 03:19, 20 January 2008

Random numbers

1. Definition: A random sequence must be generated as if they were independent draws from a well mixed urn where each number is represented once in the urn.

Mathematicians prefer abstract specifications not descriptions of processes.

2. Definition: A random sequence is one in which each number in the sequence is independent and equally probable.

3. Definition: A random sequence is one in which all finite sub-sequences of numbers are ∞-distributed.

∞-distributed is an iid process chacterized as a sequence of numbers

A pseudo random number generator is a generator that produces a series of numbers that pass a reasonable set of statistical tests.

In the 1950's Linear congruential random number generators (LCG's) were used to generate pseudo-random numbersbased on the computation;

[math] r_i = (c_1 \times r_{i-1} + c_2 ) \mbox{MOD} c_3[/math]

where [math]c_1, c_2, c_3[/math] are constants, and [math]r_{i-1}[/math] is the last pseudo-randomly generated number. All LCG's eventually repeat, and have many non random properties based on prior observations.

Encryption

Encryption is the idea of locaking away information so that the only a person with a key can unock and read it. The simplest example of a code is the alphabetic substitution cipher. This is easily crackable due to inherent structure in languages, for example spaces between words, and repeated letters.

See also;

Encryption is used in User Authentication. The idea is rather than encrypting an entire file of passwords once, encrypt each users password via an encryption process used to generate a standard message phrase. Then only the result of the encryption is required to be stored for validation, not the actual password that generates the result. e.g.

buz:wYdB]eaWLpD[tz[b]z: Robert Uzgalis

Here, buz is the username, but the password is never stored only the result of the encryption process for validation - wYdB]eaWLpD[tz[b]z.