Comparative study on deconvolution function dependencies of RTN/RDF effect estimation errors in analyzing sub-nm-scaled SRAM margins

Comparative study between the proposed technique and MATLAB-built-in deconvolution-functions with regard to deconvolution errors is discussed, which have a crucial impact in reversing the effects of convolution with Random Telegraph Noise (RTN) and Random Dopant Fluctuation (RDF) on overall SRAM margin variations. The proposed algorithm successfully avoids noise amplification thanks to eliminating the need for any operations of differential, division, and maximum-likelihood gradient sequence. This advantage over the MATLAB-built-in deconvolution-functions has been demonstrated for the first time with applying it to a real analysis for the effects of the RTN/RDF on the overall SRAM margin variations. It has been shown that the proposed technique can reduce fail-bit-count estimation error based on the convolution of the deconvoluted-RTN with the RDF by 10¹⁴-fold compared with the MATLAB-built-in deconvolution-functions.