Self‐tunable chaotic true random bit generator in current‐mode CMOS circuit with nonlinear distortion analysis

Abstract

We discuss the low-complexity integrated circuit design of a true random bit generator with parametric self-tuning capabilities. The design targets a mixedsignal complementary metal-oxide-semiconductor (CMOS) current-mode circuit implementing the Sawtooth chaotic map. Starting from the analysis of the proposed circuit topologies, we investigate how the dynamical system behavior is affected by nonlinear distortions, discussing theoretical simplified models and resorting to qualified numerical simulations. The circuit solutions proposed in this paper and validated with experiments represent an improvement of the designs published in literature, proposing technical solutions to solve stability issues and introducing parametric adjustment capabilities while maintaining a strong link with the studied theoretical models.

A Stochastic Algorithm to Design Min-Entropy Tuning Controllers for True Random Number Generators

“…10. The cumulative distribution function (19) assuming p H = 1 16 •130%, for different values of m. The grayed area represents the levels for which…”

“…where η H is the cumulative probability distribution of a negative binomial random variable with mean value m(1 − p H )/ p H . Using (19), proper values of T 0 can be chosen such to have P(m +min i η i ≤ T 0 ) < α. For example, in Fig.…”

“…For example, in Fig. 10 we reported the numerical computation of (19) assuming p H = 1 16 • 130%. The grayed area represents the levels for which P(min i η i ≤ z 0 ) < α = 2 −20 .…”