A TRNG using chaotic entropy pool as a post-processing technique: analysis, design and FPGA implementation

Abstract

In this study, the hardware implementation of a TRNG using an alternative post-processing technique is presented. ROs (ROs) are used as the noise source of TRNG and true randomness is obtained from the jitter. The use of an entropy pool composed of discrete-time chaotic systems (quadratic map, cubic map, Bernoulli shift map and tent map) is proposed as the post-processing technique for TRNG. In the system, the statistical quality of true random numbers with low statistical quality, obtained from jitter, is improved to meet cryptographic qualifications with the proposed post-processing technique. Unlike other known post-processing techniques in the literature, the post-processing technique based on the effective use of chaos is user-controlled, and is dynamically adaptable to the chosen chaotic systems. Thus, optional post-processing inputs fed from a single chaotic system or different combinations of those systems can be obtained for the TRNG. The performance of the proposed TRNG with this post-processing is high. In addition to this, the fact that any random number sequence can be generated with the contribution of more than one chaotic input in the post-processing stage makes both the post-processing and TRNG safer and more secure. In addition to the chaos-based comprehensive security analysis of the post-processing, the statistical validation of the TRNG is made by using test-based tools. As a result, it has been found that the proposed post-processing technique can be used for cryptographic purposes.