A Digital Random Number Generator Based on Regular Sampling of Double Scroll Chaos

Abstract

This paper presents the design of a novel digital random number generator (RNG) based on the distribution of regular samples from a double scroll chaotic system. In this study, a double scroll chaotic system is designed by using the Euler approximation and the nonlinearity feature of sawtooth wave in a 3^rd order ordinary differential equation (ODE). By examining the distribution of the values of chaotic state variables, it is demonstrated that the distribution has at least two regions and a binary sequence can be obtained by exploiting these two regions. The XOR is used to combine the resulting binary sequences obtained from these two regions, and thus, the RNG is designed. The RNG output satisfies all statistical randomness tests in the NIST 800–22 package without post-processing. The proposed RNG is implemented using Verilog hardware description language and prototyped on a KCU105 Evaluation Board Xilinx Ultrascale Field Programmable Gate Array (FPGA) with less than 1% LUT and Flip-Flop utilization and data throughput of up to 1.3125 Mbps.