Utilizing Hopfield Neural Network for Pseudo-Random Number Generator

Abstract

Pseudo-Random Number Generators (PRNGs) play a vital role in many cryptography functions such as encryption, authentication, and identification. Producing a Pseudo Random Number (PRN) including high randomness is a big challenge for researchers. This paper introduces a model for PRNG via employing Hopfield Neural Network (HNN) that has produced unpredictable output under specific circumstances. The random numbers generated by means of HNN are evaluated through the National Institute of Standards and Technology (NIST) statistical test and ENT test. Effectiveness of the proposed model has been revealed based on the results recorded over the evaluation metrics.