Cryptography using evolutionary computing

We present a method of generating encryptors, in particular, Pseudo Random Number Generators (PRNG), using evolutionary computing. Working with a system called Eureqa, designed by the Cornell Creative Machines Lab, we seed the system with natural noise sources obtained from data that can include atmospheric noise generated by radio emissions due to lightening, for example, radioactive decay, electronic noise and so on. The purpose of this is to `force´ the system to output a result (a nonlinear function) that is an approximation to the input noise. This output is then treated as an iterated function which is subjected to a range of tests to check for potential cryptographic strength in terms of a positive Lyapunov exponent, maximum entropy, high cycle length, key diffusion characteristics etc. This approach provides the potential for generating an unlimited number of unique PRNG that can be used on a 1-to-1 basis. Typical applications include the encryption of data before it is uploaded onto the Cloud by a user that is provided with a personalised encryption algorithm rather than just a personal key using a `known algorithm´ that may be subject to a `known algorithm attack´ and/or is `open´ to the very authorities who are promoting its use.