Pseudo Random Numbers Generator Based On One-Dimensional Cellular Automata

Pseudo-random numbers generator based on cellular automata is a method to determine pseudo-random numbers that use of their time of space and discrete state,it is different from traditional congruence and linear feedback shift register which has composition of simple unit, the role of partial between units and highly parallel information processing.at the same time, it shows complex features of global properties.Since 1985 Wolfram first time proposed the concept of ranom sequences based on one-dimensional cellular automata, it was widely utilized in cipher, random numbers generator and private key password. Since the simplicity of component unit in cellular automata, it has a unique advantage in the password field.This paper based on the In-depth study of pseudo-random numbers generator based on one-dimensional extended cellular automata.First of all, from the three variables, four variables and five variables, it achieved output pseudo-random numbers research in the most basic pseudo-random numbers generator based on cellular automata, its cellular automata use unified cellular rules and cycle boundary, randomly selects a cellular outputs its state (0 or 1), uses NIST testing package tests and gets random cellular automata rules.Secondly, in the output mode of pseudo-random numbers generator based on one-dimensional extended cellular automata, this paper uses five parallel output mode and NIST test package tests it with export, to improve the efficiency of generating random numbers.Finally, this paper introduces the Genetic Algorithm(GA), To the amendment entropy of cellular state sequences as the fitness of genetic evolution, then obtained the optimum pseudo-random numbers generator in the initial conditions, and uses NIST test package tests its export, confirms it's the optimum random.