Premium
Recent advances in parametric modeling of microwave components using combined neural network and transfer function
Author(s) -
Feng Feng,
Zhang Jianan,
Zhang Wei,
Zhao Zhihao,
Jin Jing,
Zhang QiJun
Publication year - 2020
Publication title -
international journal of numerical modelling: electronic networks, devices and fields
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.249
H-Index - 30
eISSN - 1099-1204
pISSN - 0894-3370
DOI - 10.1002/jnm.2733
Subject(s) - transfer function , parametric statistics , artificial neural network , computer science , parametric model , sensitivity (control systems) , microwave , control theory (sociology) , algorithm , electronic engineering , artificial intelligence , engineering , mathematics , telecommunications , statistics , electrical engineering , control (management)
Abstract Parametric modeling of electromagnetic (EM) behaviors has become important for EM design optimizations of microwave components. This paper provides an overview of recent advances in parametric modeling of microwave components using combined neural network and transfer function (neuro‐TF). Transfer functions are used to represent the EM responses of passive components vs frequency. With the help of the transfer function, the nonlinearity of the neural network structure can be significantly decreased. We first introduce the neuro‐TF modeling approach in rational format. We also review the pole‐residue‐based neuro‐TF modeling technique. The orders of the pole‐residue transfer functions may vary over different regions of geometrical parameters. A pole‐residue tracking technique can be used to solve this order‐changing problem. As a further advancement, we discuss the sensitivity analysis‐based neuro‐TF modeling technique. The purpose is to increase the model accuracy by utilizing EM sensitivity information and to speed up the model development process by reducing the number of training data required for developing the model. After the modeling process, the trained model can be used to provide accurate and fast prediction of the EM responses w.r.t. the geometrical variables and can be subsequently used in the high‐level circuit and system design.