Microwave and millimeter‐wave device and circuit design based on physical modeling

Snowden, C.M.

doi:10.1002/mmce.4570010103

Cited by 6 publications

(4 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Device modeling and parameter extraction are being studied in various ways by research groups from all around the world. Based upon device characterization, model classification can be made into three main categories: physical models, 1,2 behavioral models, 3,4 and equivalent‐circuit based models 5,6 . Many researchers are now getting creative with a hybrid strategy that combines these classified model techniques to achieve the desired outcomes.…”

Section: Introductionmentioning

confidence: 99%

Auto‐encoder based hybrid machine learning model for microwave scaled GaAs pHEMT devices

Bhargava

Vadalà

Majumdar

et al. 2022

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

In this article, a study of performing machine learning (ML) based modeling for semiconductor devices has been developed using experimental microwave data. Characterization of gallium arsenide (GaAs) pseudomorphic high electron mobility transistors (pHEMTs) with different gate widths is used as the illustrative example to demonstrate the accuracy and effectiveness of the presented modeling procedure. The tested devices are based on the multifinger layout, in which the total gate width (W) is obtained by multiplying the number of fingers (N f ) and their length (W 0 ). Machines are trained with scattering (S-)parameter measurements up to 65 GHz by using the extreme gradient boosting (XGBoost) algorithm with K-fold cross-validation. Then, the output of the trained machine is utilized by the parameters such as N f and W 0 inside the Auto-encoder (AE) model. In particular, the ML model with AE has a maximum of 99.88% prediction accuracy, despite the uncertainty inherent in the microwave measurements and the unavoidable deviations from the ideal behavior of the analyzed devices.

show abstract

Section: Introductionmentioning

confidence: 99%

Auto‐encoder based hybrid machine learning model for microwave scaled GaAs pHEMT devices

Bhargava

Vadalà

Majumdar

et al. 2022

Int J RF Mic Comp-Aid Eng

View full text Add to dashboard Cite

show abstract

“…7,8,11,13,18,[40][41][42][43][44][45][46][47][48][49] The main advantage of an ANN-based FET model is that, because of the "black-box" nature of ANN models, the S-parameters can be straightforwardly and accurately reproduced without requiring the extraction of an equivalent-circuit model [50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68] or a detailed knowledge of the FET physics. [69][70][71][72][73] As a matter of fact, by exploiting ANNs, it is possible mathematically describe the observable inputoutput relationships. Nevertheless, it can be quite challenging to determine the most appropriate ANN model for the specific case study, depending on the FET technology and operating conditions.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, ANNs are widely used also for modeling and predicting the scattering ( S‐ ) parameters of microwave field‐effect transistors (FETs) . The main advantage of an ANN‐based FET model is that, because of the “black‐box” nature of ANN models, the S‐ parameters can be straightforwardly and accurately reproduced without requiring the extraction of an equivalent‐circuit model or a detailed knowledge of the FET physics . As a matter of fact, by exploiting ANNs, it is possible mathematically describe the observable input‐output relationships.…”

Section: Introductionmentioning

confidence: 99%

A review on the artificial neural network applications for small‐signal modeling of microwave FETs

Marinković

Crupi

Caddemi

et al. 2019

Int J Numerical Modelling

View full text Add to dashboard Cite

The purpose of this paper is to provide a comprehensive overview of the fieldeffect transistor (FET) small-signal modeling using artificial neural networks (ANNs). To gain an in-depth insight into how to effectively develop an ANN model, we present a comparative study on the application of the ANNs for modeling the scattering (S-) parameters of a variety of FET technologies versus bias point, ambient temperature, and geometrical dimensions. As will be shown, the main challenge consists of identifying the most appropriate ANN model for the specific case under study. This is because the performance of an ANN-based model can vary significantly, depending especially on the choice of the model structure and the size and parameters of the chosen ANN. In addition, the choice of the model is related directly to the behavior of the FET characteristics, which might greatly depend on the selected device technology and operating conditions. The analysis of the present comparative study allows understanding how to properly construct ANN models to perform at their best for a successful FET modeling.

show abstract

“…The driving force behind the stream of FETs from fabrication to design is their wide field of applications, which can be thought of as the channel's electric field forcing the electrons to drift from source to drain. Typically, FET models are classified into three main groups: physical models based on device physics , behavioral models based on a mathematical description of the observable input output relationships , and equivalent circuit models representing a good compromise between the previous two . As a matter of fact, equivalent circuit models are determined from experimental measurements but keeping the connection with the physical phenomena occurring in the device.…”

Section: Introductionmentioning

confidence: 99%

The large world of FET small-signal equivalent circuits (invited paper)

Crupi

Caddemi

Schreurs

et al. 2016

Int J RF and Microwave Comp Aid Eng

View full text Add to dashboard Cite

The small-signal equivalent circuit modeling of microwave field-effect transistors (FETs) is an evergreen and ever flourishing research field that has to be up-to-date with technological developments. Hence, modeling techniques must be continuously adapted and extended to suit best evolving technologies. The extraction of a FET high-frequency small-signal equivalent circuit is a very active and broad research area of significant interest, owing to its use as a prerequisite for noise and large-signal modeling. The aim of this invited article is to provide in-depth knowledge, critical understanding, and new insights into how to extract a FET small-signal equivalent circuit from both theoretical and practical perspectives. To illustrate potential solutions to the key challenges faced by researchers, experimental results for different semiconductor technologies are reported and discussed. The study is focused on the hot research topic of the cold approach that has been, and still is, the most widely used technique for extracting FET small-signal models and on the active role of the transconductance for successful modeling. V C 2016 Wiley Periodicals, Inc. Int J RF and Microwave CAE 26:749-762, 2016.

show abstract

Microwave and millimeter‐wave device and circuit design based on physical modeling

Cited by 6 publications

References 70 publications

Auto‐encoder based hybrid machine learning model for microwave scaled GaAs pHEMT devices

Auto‐encoder based hybrid machine learning model for microwave scaled GaAs pHEMT devices

A review on the artificial neural network applications for small‐signal modeling of microwave FETs

The large world of FET small-signal equivalent circuits (invited paper)

Contact Info

Product

Resources

About