Combined empirical and look-up table approach for non-quasi-static modelling of GaN HEMTs

This paper describes a demonstrator with a small size, incorporating the concepts developed in the ESA-funded project NATALIA. This demonstrator is a smaller version (16 patches) of the final prototype (156 patches). The small prototype is described in detail and measurement results of the farfield, nearfields, power consumption and thermal behaviour are presented. It is shown that the performance is according to expectation and that the design is very robust. The outcome of all experiments verifies the NATALIA concepts.

show abstract

“…Its precision is a lot higher than one of the model with look-up table approach for non-quasi-static modeling of GaN HEMT in Ref. [11].…”

Section: Resultsmentioning

confidence: 97%

“…In the MLP neural network, the neurons are classified into layers. For example, in a three-layer perceptron as depicted follows, the first layer is the input layer, the second layer is the hidden layer, and the third layer is the output layer [11].…”

Section: Structure Of Neural Networkmentioning

confidence: 99%

Small-scale prototype of a Ku-band phased array for mobile satellite communications

Baggen

Vaccaro

Rio

et al. 2011

2011 IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications

View full text Add to dashboard Cite

show abstract

“…The second test set consisting of 3000 samples was meant for extrapolation assessment. Therefore, the input variables ranges were extended as follows: 15,8] dBm. This test set was constrained by the maximum dissipated power P diss < 0.3 W, maximum instantaneous gate voltage V gmax (t) < 0.4 V, and maximum instantaneous drain voltage V dmax (t) < 10.5 V. The region common with the interpolation test set was excluded.…”

Section: Methodsmentioning

confidence: 99%

“…The other way of constructing a hybrid model is to propagate or add the response of one type of model through the response of a second model. This solution is particularly useful when used to combine a behavioral model for the nonlinear transistor core and an empirical model for the parasitic components [3], [13]- [15].…”

Section: Introductionmentioning

confidence: 99%

Hybrid Nonlinear Modeling Using Adaptive Sampling

Barmuta

Avolio

Ferranti

et al. 2015

IEEE Trans. Microwave Theory Techn.

View full text Add to dashboard Cite

Abstract-This paper proposes a direct method for the extraction of empirical-behavioral hybrid models using adaptive sampling. The empirical base is responsible for the functionality over a wide range of variables, especially in the extrapolation range. The behavioral part corrects for the errors of the empirical part in the region of particular interest, thus, it improves the accuracy in the desired region. Employment of response surface methodology and adaptive sampling allows full automation of the hybrid model extraction and assures its compactness. We used this approach to build a hybrid model composed of a robust empirical model available in CAD tools and a Radial Basis Functions interpolation model with Gaussian basis function. We extracted the hybrid model from measurements of a 0.15 µm GaAs HEMT and compared it to the pure behavioral and pure empirical models. The hybrid model yields higher accuracy while maintaining extrapolation capabilities. Additionally, the extraction time of the hybrid model is relatively low and we show that a good accuracy level can be achieved with a small number of measurements.

show abstract

“…In the MLP neural network, the neurons are classified into layers. For example, in a three‐layer perceptron as depicted follows, the first layer is the input layer, the second layer is the hidden layer, and the third layer is the output layer .…”

Section: Structure Of Neural Networkmentioning

confidence: 99%

A Novel Modeling of Millimeter‐Wave Al_0.27Ga_0.73N/AlN/GaN Hemt Based on Artificial Neural Network

Cheng

Zhang³

2013

Micro & Optical Tech Letters

View full text Add to dashboard Cite

An artificial neural network (ANN) was used for modeling millimeter‐wave Al0.27Ga0.73N/AlN/GaN high electron mobility transistor (HEMT) with multi‐biases in this article. Millimeter‐wave Al0.27Ga0.73N/AlN/GaN HEMT with gate width of 2 × 75 µm and gate length of 0.3 µm was designed and fabricated at first, and then its performance was measured for modeling. NeuroModelerPlus_V2.1E software was trained to learn the input‐output relationship from the data about DC and AC performances of proposed device. ANN DC model and AC model were set up and embedded in software of ADS together to form a model of millimeter‐wave Al0.27Ga0.73N/AlN/GaN HEMT, which can be used in circuit design. The results of the simulation of the model showed that it corresponded with the measurement results. © 2013 Wiley Periodicals, Inc. Microwave Opt Technol Lett 55:2124–2127, 2013

show abstract

Combined empirical and look-up table approach for non-quasi-static modelling of GaN HEMTs

Cited by 6 publications

References 16 publications

Small-scale prototype of a Ku-band phased array for mobile satellite communications

Small-scale prototype of a Ku-band phased array for mobile satellite communications

Hybrid Nonlinear Modeling Using Adaptive Sampling

A Novel Modeling of Millimeter‐Wave Al_0.27Ga_0.73N/AlN/GaN Hemt Based on Artificial Neural Network

Contact Info

Product

Resources

About

Combined empirical and look-up table approach for non-quasi-static modelling of GaN HEMTs

Cited by 6 publications

References 16 publications

Small-scale prototype of a Ku-band phased array for mobile satellite communications

Small-scale prototype of a Ku-band phased array for mobile satellite communications

Hybrid Nonlinear Modeling Using Adaptive Sampling

A Novel Modeling of Millimeter‐Wave Al0.27Ga0.73N/AlN/GaN Hemt Based on Artificial Neural Network

Contact Info

Product

Resources

About

A Novel Modeling of Millimeter‐Wave Al_0.27Ga_0.73N/AlN/GaN Hemt Based on Artificial Neural Network