An Objective Function Formulation for Circuit Parameter Extraction Based on the Kullback-Leibler Distance

Numerical circuit parameter extraction is widely used for modeling many kinds of high-frequency structures. In particular, parameter extraction can be used to find the optimal parameter values of an equivalent circuit model to match as close as possible a given target response. This paper presents a novel objective function formulation for parameter extraction based on the Kullback-Leiber distance. A rigorous graphical and numerical comparison of the proposed new formulation against classical l-th norm formulations is included. One- and two-dimensional synthetic examples are used to illustrate the advantages of the proposed Kullback-Leiber distance formulation. Our results indicate that the proposed new formulation yields similar or better-behaved parameter extraction objective functions than those obtained from l-th norm formulations.