Macromodeling based variability analysis of an inverted embedded microstrip line

Abstract

A multivariate macromodel of the per unit length parameters of an inverted embedded microstrip line is built starting from a limited number of highly accurate but computationally expensive electromagnetic simulations. Besides the frequency dependency, the macromodel also encompasses the influence of a geometrical parameter that determines the shape of the cross-section of the metallic interconnect. This allows to assess the influence of the etching process through a variability analysis of the interconnect's behavior in both frequency and time domain. The analysis is carried out via a robust Monte Carlo approach, which has been made computationally feasible thanks to the macromodeling step.