Modeling Of Microwave Devices Research Articles

Advances in hybrid format‐based neuro‐transfer function techniques for parametric modeling of microwave components

AbstractElectromagnetic (EM) parametric modeling has become significant for EM designs of microwave devices. This paper outlines recent advances in hybrid format‐based neuro‐transfer function (TF) techniques for EM parametric modeling of microwave devices. To solve the problem of high‐sensitivity, a novel decomposition approach is discussed to develop a rational‐based neuro‐TF model of EM behavior of microwave devices. To handle the issue of non‐smoothness and discontinuity, a parametric modeling technique incorporating pole‐residue/rational and neural network hybrid transfer function (short for rational/pole‐residue hybrid neuro‐TF) of EM behavior is reviewed. This technique effectively combines rational and residue‐pole formats of the transfer functions. Compared with the residue‐pole‐based neuro‐TF modeling approach and the rational‐based neuro‐TF modeling approach, the rational/pole‐residue hybrid neuro‐TF technique allows for better accuracy in large geometrical changes and high order applications. A parametric modeling method combining neural network and polynomial‐transfer function (neuro‐PTF) is further presented as an advanced version of the rational/pole‐residue hybrid neuro‐TF method. In this approach, the pole–residue‐based transfer function and the polynomial function are used together to describe the EM responses, and can produce more accurate models, especially with large geometrical variables. Following the modeling process, trained models can be used to provide fast and accurate EM response predictions and can subsequently be used for advanced circuit and system design.

An Adaptive Rational Fitting Technique of Sommerfeld Integrals for the Efficient MoM Analysis of Planar Structures

The integral equation method is one of the most successful computational models for microwave devices or integrated circuits in planar layered media. However, the efficient and accurate evaluation of the associated Green’s function consisting of Sommerfeld integrals (SIs) is still a remaining challenge. To mitigate this difficulty, this work proposes a spatial domain rational function fitting technique (RFFT) for SIs so that the approximation accuracy is controllable. In conjunction with an adaptive sampling strategy, the proposed RFFT minimizes the orders of rational functions, and the resultant SI evaluation efficiency is optimized. In addition, we investigate the semi-analytical singularity treatment for the rational expression of SIs in method of moment (MoM) implementation. Extensive simulation of representative planar devices validates the correctness of the proposed method and demonstrates its superior performance over conventional SI approximation methods.

Meet the Editors

Meet the Editors

Применение нейронных сетей при построении нелинейных моделей полевых транзисторов

The nonlinear model of a field-effect transistor based on the mathematical apparatus of the theory of artificial neural networks was developed. The main feature of this model is the possibility of training neural networks used to approximate the current-voltage characteristic and the gate-drain and gate-source capacitances of a nonlinear transistor model by using optimization algorithms built into popular microwave CAD systems. This makes it possible to use the well-known advantages of neural networks in the problems of function approximation to increase the reliability of the results of non-linear modeling of microwave devices based on field-effect transistors.

A Time Delay Neural Network Based Technique for Nonlinear Microwave Device Modeling.

This paper presents a nonlinear microwave device modeling technique that is based on time delay neural network (TDNN). The proposed technique can accurately model the nonlinear microwave devices when compared to static neural network modeling method. A new formulation is developed to allow for the proposed TDNN model to be trained with DC, small-signal, and large signal data, which can enhance the generalization of the device model. An algorithm is formulated to train the proposed TDNN model efficiently. This proposed technique is verified by GaAs metal-semiconductor-field-effect transistor (MESFET), and GaAs high-electron mobility transistor (HEMT) examples. These two examples demonstrate that the proposed TDNN is an efficient and valid approach for modeling various types of nonlinear microwave devices.

Spectral models of microwave devices with longitudinal interaction

An attempt is made to expand the spectral approach, implemented formerly for the M-type devices, onto devices with longitudinal interaction like one-beam floating-drift klystrons and travelling-wave tubes. The models are intended for engineering analysis and optimization of wide class of the O-type microwave devices. The general features of the spectral models and applied algorithms of the numerical experiment are described. Preliminary results of the simulations are quoted.

Automated Neural-Based Modeling of Microwave Devices Using Parallel Computation and Interpolation Approaches

Automated model generation (AMG) is an automated artificial neural network (ANN) modeling algorithm, which integrates all the subtasks (including adaptive sampling/data generation, model structure adaptation, training, and testing) in neural model development into one unified framework. In existing AMG, most of the time is spent on data sampling and model structure adaptation due to the iterative neural network training and the sequential computation mechanism. In this paper, we propose an advanced AMG algorithm using parallel computation and interpolation approaches to speed up the neural modeling of microwave devices. Efficient interpolation approaches are incorporated to avoid repetitive training of the intermediate neural networks during adaptive sampling process in AMG. Parallel computation formulation based on a multi-processor environment is proposed to further save time during interpolation calculation, data generation, and model structure adaptation process. Examples of automated modeling of two microwave filters are presented to show the advantage of this paper.

A general neuro-space mapping technique for microwave device modeling

Accurate modeling of nonlinear microwave devices is critical for reliable design of microwave circuit and system. In this paper, a more general neuro-space mapping (Neuro-SM) method is proposed to fulfill the needs of the increased modeling complexity. The proposed technique retains the capability of the existing dynamic Neuro-SM in modifying the dynamic voltage relationship between the coarse model and the desired model. The proposed Neuro-SM also considers dynamic current mapping besides voltage mappings. In this way, the proposed Neuro-SM generalizes the previously published Neuro-SM methods and has the potential to produce a more accurate model of microwave devices with more dynamics and nonlinearity. A new formulation and new sensitivity analysis technique are derived to train the general Neuro-SM with dc, small-, and large-signal data. A new gradient-based training algorithm is also proposed to speed up the training. The validity and efficiency of the general Neuro-SM method are demonstrated through a real 2 × 50 μm GaAs pseudomorphic high-electron mobility transistor (pHEMT) modeling example. The proposed general Neuro-SM model can be implemented into circuit simulators conveniently.

Modeling Instabilities in Relativistic Electronic Beams in the CST Particle Studio Environment

The paper is devoted to the development of the classical vircator model in the CST Particle Studio for the investigation of the physical mechanisms of the development, coexistence, and interaction of Pierce/Bursian and diocotron instabilities in relativistic electronic streams. Such a model is numerically analyzed by the CST Particle Studio. We find that a numerical instability develops, which leads to an additional small-scale space density modulation in the electron beam during its motion in the system. We find the character time scale of the instability and determine its relation with a spatial mesh step. We study the instability in detail and propose a method to suppress it by the CST Particle Studio. Also, important notes concerning the development of models of microwave devices in the CST Particle Studio are presented.

A Wiener-Type Dynamic Neural Network Approach to the Modeling of Nonlinear Microwave Devices

A new Wiener-type dynamic neural network (DNN) approach for nonlinear device modeling is proposed in this paper. The proposed analytical formulation of Wiener-type DNN structure consists of a cascade of a simplified linear dynamic part and a nonlinear static part. The simplified linear dynamic equations are obtained by vector fitting enhancing the efficiency of the model. A new formulation and new sensitivity analysis technique is derived to train the proposed Wiener-type DNN with dc, small- and large-signal data. A new gradient-based training algorithm is also formulated to speed up the training of proposed Wiener-type DNN model. The proposed Wiener-type DNN model can be trained to be accurate relative to device data. Furthermore, the proposed Wiener-type DNN provides enhanced convergence properties over existing neural network approaches such as time delay neural network (TDNN) and TDNN with extrapolation. Application examples on modeling GaAs metal-semiconductor-field-effect transistor, GaAs high-electron mobility transistor (HEMT), and a real <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2 \times 50$ </tex-math></inline-formula> gatewidth GaAs pseudomorphic HEMT are presented. The use of Wiener-type DNN model in harmonic balance simulations demonstrates that the Wiener-type DNN is a robust method for developing models for different types of microwave devices. These models could be implemented into circuit simulators conveniently.

Modeling of magnetoelectric microwave devices

The possibility of computer modeling implementation of electrically controlled magnetoelectric (ME) microwave devices is considered. The computer modeling results of different structures of ME microwave devices based on layered ferrite-piezoelectric structure formed on the slot line, microstrip line and coplanar waveguide are offered. Results are reported as frequency dependencies of insertion losses of ME devices.

A Novel Dynamic Neuro-Space Mapping Approach for Nonlinear Microwave Device Modeling

This letter presents a novel dynamic Neuro-space mapping (Neuro-SM) technique for nonlinear device modeling. This is an advance over the existing static Neuro-SM which aims to map a given approximate device model towards an accurate model. The proposed technique retains the ability of static Neuro-SM in modifying the effects of nonlinear resistors and current sources. The proposed technique can also make up for any capacitive effects and non-quasi-static effects that maybe missing in the given model, which is not achievable by the existing static Neuro-SM. In this way, the dynamic Neuro-SM model can exceed the accuracy limit of the static Neuro-SM. The validity and efficiency of the proposed approach are verified through two high-electron mobility transistor (HEMT) modeling examples.

Accurate Modelling of Liquid Crystal-Based Microwave Devices

Liquid crystal substrates have been shown to provide the means to develop low-cost, reconfigurable, adaptive and tuneable microwave devices for mobile and wireless communication systems. In order to take maximum advantage of the possibilities that these materials offer and to design LC-based devices appropriately, techniques for the characterisation of the liquid crystal dielectric properties are needed. Similarly, appropriate modelling methods are required to simulate accurately the switching behaviour of the liquid crystal and the characteristics of the wave propagation through the devices, taking full consideration of the point-by-point variation of the material tensor permittivity.

A Goal-Oriented Error Estimator for Reduced Basis Method Modeling of Microwave Devices

This letter proposes a novel a-posteriori error estimator suitable for the reduced order modeling of microwave circuits. Unlike the existing error estimators based on impedance function residuals, the new one exploits the residual error associated with the computation of the scattering matrix. The estimator can be effectively used in the Reduced Basis Method (RBM) to automatically generate reduced-order models. The results of numerical experiments show that the new error estimator predicts the errors in the computed S-parameters more accurately than the other existing approaches, and that it can be used to automate the generation of wideband reduced models.

Compact Behavioral Models of Nonlinear Active Devices Using Response Surface Methodology

This paper presents the response surface methodology in modeling of nonlinear microwave devices. First, different combinations of sampling techniques and types of radial basis functions are evaluated in simulations of the drain current of a 0.15- μm GaAs HEMT transistor described by the Chalmers model. It allows to determine the best settings of the response surface methodology for the modeling of active microwave devices. It is shown that the best sampling strategy is a combination of space-exploration (Voronoi), problem-exploitation (LOLA), and model-error-driven sample rankers. From the various radial basis function models, the fastest convergence is achieved with exponential functions. This knowledge is then used in behavioral modeling of a low-power amplifier AG303 measured in the load-pull setup. It is shown that the response surface methodology outperforms commonly used factorial design of experiments. Moreover, it gives accurate models within just a few tens of samples. However, attention has to be paid at the noisy regions, which might be oversampled by the sampling techniques.

Hybrid modeling of microwave devices using multi-kernel support vector regression with prior knowledge

This article proposes a support-vector hybrid modeling method of microwave devices when only a small number of measurements are available. In this method, a hybrid model of microwave device has been obtained by combining a coarse model and a support-vector model, where the coarse model is complemented by a support-vector model capable of correcting the difference between the measurements and the coarse model. The support-vector model was developed using a novel algorithm. In the algorithm, multi-kernel and prior knowledge from a calibrated simulator were incorporated into the framework of the linear programming support vector regression by utilizing multiple feature spaces and modifying the optimization formulation. The experimental results from two microwave devices show that the hybrid modeling can enhance the physical meaning of the support-vector model and improve the modeling accuracy for a small dataset, and that the proposed algorithm shows great potential in some applications where sufficient experimental data is difficult and costly to obtain, but the prior knowledge from a simulation model is available. The hybrid modeling is suited to a microwave computer-aided design tool or an automatic tuning robot. © 2014 Wiley Periodicals, Inc. Int J RF and Microwave CAE 25:219–228, 2015.

A Microwave Modeling Oxymoron?: Low-Frequency Measurements for Microwave Device Modeling

For a number of decades, the modeling of microwave transistors relied on specific well-known characterization methods. Those methods include S-parameters measurement through vector network analyzers (VNAs) ranging from the lower end of the RF spectrum to the millimeter-wave (mm-wave) region and load pull measurements of transistors used for the design of power amplifiers (PAs). Later, the availability of more powerful computer-aided design (CAD) software enabled the need for models of active microwave devices, thus driving a huge amount of research activity on microwave transistor modeling. Simultaneously, new transistor technologies were invented, combining working concepts such as heterojunction bipolar transistors (HBTs), metal semiconductor field effect transistors (MESFETs) or high electron mobility transistors (HEMTs) and new materials such as gallium arsenide (GaAs), gallium nitride (GaN), indium phosphide (InP), and silicon germanium (SiGe), to cite only the main ones.

Overview of Microwave Device Modeling Techniques Based on Machine Learning

This paper overviews the microwave device modeling techniques based on machine learning. The proposed methods are divided into two categories, active and passive components modeling. These modeling techniques aim to characterize the microwave device efficiently and accurately. The active components consist of MESFET, HEMT, pHEMT, and so on. The passive devices include inductor, capacitor, Lange coupler, coplanar waveguide, filter, etc. A variety of modeling techniques has been successfully used, e.g., artificial intelligence, support vector regression, genetic algorithm, etc. Increased investigations published in the literature are a true testament of the appeal based on machine learning.

Extracting Volterra series representation from X-parameters for the modeling of microwave device

As a classic mathematical means, Volterra series are good representations for the description of nonlinear microwave devices and systems. However, the difficulty in extracting accurate kernel functions is the most important reason that limits wider application of Volterra series. In this research, the X-parameters proposed in recent years are used to resolve this problem for the first time. The extraction method is detailed, and the results show that it is very convenient to obtain Volterra series representation by using X-parameters. A power amplifier was measured to verify the proposed approach, and a good agreement has been achieved between the measured results and the predicted data based on the Volterra series. The method proposed in this paper is useful for accurate and fast modeling of modern microwave devices.

Modeling and simulations of high-power microwave devices using the CHIPIC code

AbstractThe CHIPIC code, a fully electromagnetic particle-in-cell (PIC) code for modeling and simulations of high-power microwave (HPM) devices, is introduced in this paper. It consists of a two-dimensional (2D) code and a three-dimensional (3D) code. The 2D code can model and simulate HPM devices with symmetric structure on 2D Cartesian, cylindrical and polar grids, while the 3D code can model and simulate HPM devices on 3D Cartesian and cylindrical grids. The fields are calculated using the finite-difference time-domain scheme, and the particles are described by the PIC scheme. Various types of boundary conditions have also been implemented for different kinds of applications. In addition, the 3D code is specifically designed for high-performance modeling and computing. It uses the message passing interface and the open specifications for multiprocessing (OpenMP) for parallelization. Its parallel design ensures that it is capable of efficiently executing on a variety of architectures. In order to allow efficient use of parallel architectures, it provides automated partitioning and dynamic load balancing. Even though this code is still in development, it has successfully simulated various real-world HPM experimental devices. Simulation results on some typical HPM devices by using the CHIPIC code are given, which agree well with those obtained from some well-known PIC codes. Direction for future work is also presented.