Parametric modeling of millimeter-wave passive components using combined neural networks and transfer functions

Abstract

This paper propose to develop the combined neural networks and transfer functions (neuro-TF) for parametric modeling of millimeter-wave passive components. Artificial neural networks (ANN) techniques are recognized as a powerful tool for modeling the EM behavior of microwave components. In this paper, we train the ANN to map geometrical variables onto coefficients of transfer functions. The model obtained using our proposed technique can achieve good accuracy, and can be further used in the high-level design. Two millimeter-wave examples are used to demonstrate the validity of this technique.