Efficient Post-Processing Module for a Chaos-based Random Bit Generator

In this paper we propose an efficient low-hardware complexity post-processing module for a feedback-corrected chaos-based True Random Bit Generator (TRBG) with high Average Shannon Entropy (ASE). The design of the postprocessing module was based both on the theoretical analysis of the information generation mechanism of the considered source and on numerical simulations. A prototype of the module was tested on an implementation of the TRBG with an actual ASE slightly lower than 1 bit/time step, obtaining random sequences that fully pass all the well-known statistical tests of the Diehard suite.