A model-based technique for efficient evaluation of noise robustness

In this paper, we present an analytical technique for deriving noise rejection curves (NRCs) and the associated noise susceptibility metric under parameter variations (L_eff, V_T, V_DD and channel width, W). The method involves modeling of the pull-up and pull-down resistances using approximated BSIM4 device equations. Compared to circuit simulation results, the analytical model provides more than live orders of magnitude speedup while maintaining an average (maximum) error of 1.3% (5%) over the entire range of parameter variations, which makes it suitable for design optimization for noise robustness.