Researches of algorithm of PRNG on the basis of bilinear pairing on points of an elliptic curve with use of a neural network

Abstract

In this paper pseudorandom number generator based on elliptic curve bilinear pairing is developed. Residue number system and approximate method are used for effictive realization of modular operations over finite field that allows to increase the speed of pseudorandom number generator for −256 by 2,15 times compared to similar PRNG that uses positional notation. The developed pseudorandom number generator based on neural network has as good statistical properties as random sequences from site random.org and passes Diehard tests.