An enhanced macromodeling approach for differential output drivers

Abstract

This paper presents an approach for building new compact macromodels of differential output drivers. Composed of enhanced physical-based elements, the new models are capable of capturing the important intrinsic nonlinear and dynamic characteristics of the drivers. We demonstrate the approach with two typical digital drivers, low-voltage differential signaling (LVDS) driver and pre-emphasis driver. The obtained macromodels achieve excellent accuracy in capturing behaviors at various input patterns, loading conditions and supply voltages.