Noise-Based Simulation Approach For Statistical Analysis of Parameter Fluctuations in Circuits

A. Kloes (THM Univ. of Applied Sciences, Germany), A. Nikolaou (THM Univ. of Applied Sciences, Germany and Univ. Rovira i Virgili, Spain), N. Dersch (THM Univ. of Applied Sciences, Germany), C. Roemer (THM Univ. of Applied Sciences, Germany and Univ. Rovira i Virgili, Spain), G. Darbandy, M. Schwarz (THM Univ. of Applied Sciences, Germany)

Circuit simulations including statistical variability are usually carried out following the Monte-Carlo (MC) approach, which is very time consuming. As an alternative and numerically very efficient method, the Noise-Based Variability Approach (NOVA) has been proposed for the consideration of fluctuating device parameters with a Gaussian statistical distribution function. In this presentation, NOVA reviewed in comparison to MC simulations for the case of circuits with organic thin-film transistors including parameter fluctuations. Furthermore, results for the application of NOVA for the simulation of the inference process in crossbar arrays of neuromorphic computing structures is shown.