A new pseudo random number generator based on Chebyshev maps and parameter switching

Abstract

Digital implementations of chaotic systems are periodic and they are degraded versions of the original chaotic system. Hence, all digital pseudo random number generators based on chaotic systems inherently contain several weaknesses. These weaknesses are eliminated using various methods like cascading multiple chaotic systems and applying small perturbation to the calculations. Another method is switching parameter values of a dynamical system to obtain different chaotic behaviors. There is an example of such a method in the literature which demonstrates the switching between continuous-time ODE based Chen-like and Lorenz-like systems. In this study, a similar approach is followed and four different discrete-time Chebyshev maps are utilized for pseudo random number generation by switching a single parameter value. Just like in the other study about parameter switching method, the resulting attractor is a combination of the individual attractors of the switched systems. A direct approach is used for bit extraction and the resulting random sequences successfully pass the NIST SP 800-22 randomness tests. © 2018 IEEE.