J. R. Loo-Yau

Center for Research and Advanced Studies of the National Polytechnic Institute | Cinvestav · Departamento de Diseño Electrónico (Guadalajara)

In this work, a procedure to obtain two-port admittance models of buried cables illuminated by incident electromagnetic fields for the analysis of transients in the frequency domain is presented. The procedure is based on cascading two-port admittance models of various cable segments, avoiding the numerical problems that chain matrix representation...

This letter presents the digital predistortion for power amplifiers (PAs) using a new behavioral model that combines a nonlinear autoregressive exogenous architecture with a support vector regression (NARX-SVR) algorithm. Unlike other models based on SVR, the NARX-SVR requires lower support vector numbers (SVN) and floating-point operations (FLOPs)...

In this work, saturation and hysteresis loops are included in a four-port transformer model for electromagnetic transients computation in the frequency domain. Approximations with linear segments of the hysteresis characteristic and saturation curve are used. To obtain time domain results, the numerical Laplace transform is employed and the results...

This work demonstrates a straightforward analytical helpful technique to determine the transistor’s input and output reflection coefficients to achieve the maximum conversion gain (CG) of RF frequency multiplier (FM) circuits using conventional $X$ -parameters ( $Z_{L}$ = $50 \Omega $ ) of a transistor. Two frequency doubler (FD) circuits, one...

This work presents a theoretical analysis of the frequency dependence of the load impedance seen from the transistor’s intrinsic current source and the package plane for designing class-J*/B/J power amplifiers (PAs). The presented results are possible thanks to the new analytical expression for calculating the output current. Moreover, the proposed...

In this letter, the intrinsic current and voltage waveforms of a power transistor are measured by using a time-domain low-frequency (LF) active harmonic load-pull (LP) system. The measured waveforms are used to study the device's behavior under loading conditions corresponding to the impedance design space of the resistive-reactive (R-R) class-J mo...

In order to develop equivalent circuit models of microwave transistors, electrical measurements performed on the device to be modeled are mapped to a network of circuit components that mimic the device's physical structure. To accomplish this, S‐parameters measurements using a calibrated vector network analyzer are required. This article presents a...

This article presents an improved nonlinear empirical I/V model suitable for GaAs and GaN FETs. The new drain‐to‐source current formulation accurately represents the symmetric and the asymmetric bell‐shaped transconductance (gm) for all VDS values. Besides modeling with high accuracy the I/V characteristic, the proposed model can fit the first and...

This letter demonstrates for the first time the usefulness of the X-parameters under 50-Ω load condition (conventional X-parameters) for simulating and designing RF feedback oscillator circuits. The measurement procedure to characterize a GaAs FET ATF34143 with conventional X-parameters is presented. The measured X-parameters' data are subsequently...

For determining the line characteristic impedance ${Z}_{1}$ of uniform transmission lines, it is typically necessary to use a calibrated vector network analyzer (VNA), whose calibration reference impedance ${Z}_{2}$ is known. Using the eight-error term model for modeling a VNA and based on the fact that the values of the error terms are unique...

In this paper, a model of aerial multiconductor transmission lines with frequency-dependent electrical parameters for the simulation of time-domain electromagnetic transients due to incident electromagnetic fields is presented. The frequency dependency of the electrical parameters is taken into account using the penetration impedance; this allows a...

In this paper a model for Nonuniform Transmission Lines for electromagnetic transient analysis that incorporates frequency dependency of electrical parameters, variation of line electrical parameters with respect to distance, and distributed excitations due to incident electromagnetic fields is presented. The model is developed using the Method of...

In this work, a compact microstrip lowpass‐bandpass diplexer using radial stubs (RSs) is proposed. By controlling the input impedance of the RS resonators, it is possible to obtain a selective lowpass filter (LPF) and a narrow bandpass filter (BPF) with a fractional bandwidth of 5%, without increasing the insertion losses (ILs) of both filters. The...

In this article, the effect of the second harmonic on the microwave performance of RF feedback oscillators is investigated. A microstrip bandpass filter, which combines the characteristics of a 1st order coupled line and a hairpin filter allowing to reject the second harmonic, is used as feedback circuit. It is demonstrated that by mitigating the e...

This paper deals with modeling of drain current IDS(VGS,VDS) and intrinsic nonlinear capacitances Cgs(VGS,VDS) and Cgd(VGS,VDS) of the CMOS FET by means of empirical analytical expressions. The proposed models are based on exponential series allowing modeling of the CMOS FET from the linear to the breakdown region. Several High Voltage CMOS (HVMOS)...

This paper presents a novel band-pass microstrip filter based on a 1 st order parallel coupled microstrip filter with minimum fractional bandwidth. One of the main characteristic of the proposed microstrip filter is its high rejection level at the 2 nd harmonic of the resonance frequency (f 0 ). Electromagnetic simulations based on the method of Mo...

In this study, a model of one-phase transformers with saturation effects, suitable for low- and mid-frequency electromagnetic transient simulations using the numerical Laplace transform is presented. Substitution and superposition theorems are invoked to convert the saturation non-linear condition of the transformer into a series of time-sequential...

In this study, a new model for aerial multiconductor transmission lines with frequency dependent electrical parameters is presented. The new model is called here A-line model and is a two-port model based on the method of characteristics and finite differences that does not require segmenting the transmission line. The frequency dependence of elect...

Using a field-programmable gate array (FPGA) development board, a digital signal processor (DSP) builder, and the phase-to-amplitude conversion principle, a low-cost system for measuring the amplitude-to-amplitude (AM/AM) and amplitude-to-phase (AM/PM) distortion curves of radio frequency (RF) power amplifiers (Pas) is presented. The state of the a...

This work presents a digital predistortion (DPD) scheme to linearize power amplifiers (PAs) using a recurrent neural network called Nonlinear AutoRegressive with eXogenous input model (NARX) neural network (NARXNN). The architecture of the NARXNN is based on a class of discrete‐time nonlinear system named NARX. Its topology has embedded memory at t...

The thru-reflect-match (TRM) calibration technique uses as standards a through connection of ports 1 and 2 of the vector network analyzer, two symmetrical highly reflecting loads, and two broadband loads (match) of impedance close to the measuring system impedance (Z0). The classical theory of the TRM calibration refers the calibration to the imped...

This work presents a method to determine the parasitic capacitances of GaN FETs, especially the Cpd. The proposed method is based on an idea used for GaAs FET that assumes that Cpd is directly proportional to Cds by a factor of 0.25. The contribution of this work lies in the adaptation of this technique to be suitable for GaN FETs by proposing a me...

In this paper an effective method for modeling parasitic capacitances of Gallium Nitride (GaN) FET transistors is presented. The proposed method is a modification of a method that was proposed for Gallium Arsenide (GaAs) transistors. The modification consists of the computation of a constant relating to the asymmetry of the transistor when it is pi...

This paper presents the FPGA implementation of a NARX neural network for the modeling nonlinear systems. The complete neural architecture was implemented with Verilog language in Xilinx ISE Tool with the Virtex-6 FPGA ML605 Evaluation Kit. All operations, such as data processing, weight connections, multipliers, adders and activation function were...

This work presents a novel Digital Predistortion (DPD) scheme based on a NARX network, suitable for linearizing power amplifiers (PAs). The NARX network is a Recurrent Neural Network (RNN) with embedded memory that allows efficient modeling of nonlinear systems. Its neural architecture is very effective to model long term dependencies, such as the...

In this paper the hardware implementation of a NARX neural network algorithm using a Field Programmable Gate Array (FPGA) is presented. A NARX network is a Recurrent Neural Network (RNN) suitable for modeling nonlinear systems with promising results for the modeling of the inverse characteristics (AM/AM and AM/PM) of Power Amplifiers (PAs). The imp...

The last seven years, Cinvestav-GDL and CICESE, two of the most important research centers in Mexico, have been working together in several microwave research topics. This paper describes the relationship of these two academic institutions regarding to the research activities in this field. A brief history and the research interests of these instit...

In this work a technique to heighten the breakdown voltage and the transition frequency (fT)in standard MOS technology is presented. By using an optimized extended drift region at the drain, a CMOS FET can achieve higher breakdown voltage. To enhance the operation frequency, the standard analog/digital pads were modified to decrease coupling effect...

This letter deals with the de-embedding of on-wafer CMOS FETs embedded in symmetrical and reciprocals pads. A de-embedding method, that uses a calibrated vector network analyzer and two microstrip lines fabricated on a lossy SiO2-Si substrate, is introduced. The proposed method not only allows the characterization on the interconnection lines but a...

This article presents a method to extract the elements of the electrical equivalent circuit model CMOS pads using two uniform transmission lines fabricated on a lossy Si substrate. The lengths of the uniform transmission lines are different, while its characteristic impedances are equal and can be of an arbitrary value. The proposed method is usefu...

By using a neural network approach that takes into account input/output relationship data along with derivative information in the training process, a fast and straightforward methodology to obtain the quasi-static model of GaN FETs is introduced. This method uses data obtained from pulsed I/V and S-parameter measurements to train three different n...

In this paper, the line-offset offset-open offset-short (LZZ) calibration technique for vector network analyzers (VNA) is introduced. The LZZ uses as calibration standards a fully known transmission line and two offset reflecting loads. The mathematical formulation of the LZZ is based on the use of -parameters for modeling the imperfect VNA as well...

This paper deals with an analytical methodology to synthesize microstrip input and output (I/O) matching networks for class F-1 PAs up to the third harmonic using short stubs. Physical dimensions of the microstrip are determined by means of T-parameters. To verify the proposed method, a GaN class F-1 PA was designed at 2.5 GHz. Experimental results...

Using the wave cascading matrix formalism and the 8-term error model for modeling the imperfect measurement system vector network analyzer a new approach of the Thru-Reflect-Line (TRL) calibration technique is introduced and compared with the already existing TRL calibration technique. This new approach allows the utilization of arbitrary character...

A reliable calibration of the vector network analyzer is needed in order to characterize microwave devices. For some VNAs, such as HP8510, solving the 8 error model is not enough to accurately compute the device under test (DUT) S-parameters, but also, a correction of the switching errors inherent to the measurement must be performed. In this paper...

Artificial neural networks are used as a tool in the development of reliable models for field effect transistors. This letter presents an improvement of the classical backpropagation algorithm allowing for the training process information contained in the first derivatives of Ids with respect to the control voltages Vgs and Vds. Excellent agreement...

The aim of this contribution is to show why sources of non-idealities are actually a concept of reason in order to define tradeoffs in the design of analog circuits. A tradeoff is commonly picked up from an analytical design-model, which tries to explain a given phenomenon under study by using physical theories underlying the role of non-idealities...

Using ABCD parameters and the 8-term error model to represent a non-ideal vector network analyzer (VNA), a new procedure for implementing the Thru-Reflect-Line calibration technique (TRL) is introduced. The novelty of this TRL is the use of ABCD-parameters, instead of transmission parameters, to represent a non-ideal vector network analyzer (VNA)....

By using transmission line measurements of arbitrary characteristic impedance, this paper introduces an alternative method to de-embed parasitic structures such as pads and interconnection lines. The proposed method uses two uniform transmission lines of arbitrary characteristic impedance as calibration standard (L-L method). The two transmission l...

A new bandstop-coupled line filter using spurline structures has been presented in this work. The new bandstop filter enhances the rejection bandwidth of bandstop coupled line filters in a 50%. To demonstrate the idea, classical bandstop-coupled line filters with and without simple spurlines structure were designed using Rogers material, and their...

An improved two-tier L-L method for characterizing symmetrical microwave test fixtures is introduced in this paper. The improved two-tier L-L method is implemented with two uniform transmission lines having the same characteristic impedance and propagation constants, but different lengths. The ABCD parameters of the test fixture and the embedded de...

This article presents an improved nonlinear IDS(VGS, VDS) model useful for modeling the experimental pulsed I-V measurement data of GaN FETs operated at large drain-source voltages. This improved IDS(VGS, VDS) model is based on the Chalmers model. The main features of the improved nonlinear IDS(VGS, VDS) model are the two analytical expressions tha...

A novel procedure to characterize coaxial line- stretcher phase shifter based on the two-line method is presented in this work. Intrinsic losses and differential phase shift are determined using S-parameters data measured with an uncalibrated vector network analyzer. To validate the method, different trombone-type line-stretcher phase shifters are...

We propose and demonstrate experimentally that spurline structures enhance the rejection bandwidth of microwave bandstop coupled line filters. We have investigated the influence of spurlines structures on the rejection bandwidth of a typical microwave coupled line filters (with a notch frequency at 3.0 GHz). Momentum simulations and experimental re...

This paper introduces a method for improving the bandwidth of GaN class-F PAs. The key idea behind this broadband class-F PA lies in the bandpass filter (BPF) used at the input matching network of a conventional class-F PA. Harmonic impedance optimization at the input of the FET is performed in order to design the BPF. Theoretical and experimental...

In this paper a straightforward method to determine the parasitic gate resistance (R_g) of GaN FET is introduced. The method uses a simple linear regression to directly determine the value of R_g without prior knowledge of the Schottky diode resistance R₀ and capacitance C₀. Furthermore, the method requires...

By using an indirect method for determining the characteristic impedance of uniform transmission lines embedded in coaxial connectors, a novel two tier L-L de-embedding method is presented. The proposed de-embedding method is suitable for S parameters characterization of GaN HEMTs packaged transistors mounted on a symmetrical and reciprocal test fi...

Heterojunction field effect transistors (HFET) based on gallium nitride (AlGaN/GaN) and metal semiconductor field effect transistors (MESFETs) based on silicon carbide (SiC) are the preferred transistors for high-power amplifier circuit designs rather than MESFETs, high electron mobility transistors (HEMTs) and pseudomorphic HEMTs based on gallium...

Using measurement techniques and appropriate models for III-V semiconductor compound transistors we have implemented a large signal model for a commercial LDMOS transistor. Based on this model a 4 Watt UHF class AB power amplifier was simulated and evaluated. The power amplifier was characterized using an aluminum test bench and microstrips. Experi...

Recently, a new method for calculating gate resistance and inductance of on-wafer AlGaN/GaN HEMTs was presented. In this paper the performance of the method mentioned before is investigated on commercially available packaged power GaN HEMTs (CHG35015) and power SiC MESFETs (CRF24010f for the determination of Rg and Lg of the equivalent circuit mode...

This letter presents an extension of the analytical method proposed in (Mader and Popovic, 1995) to determine the impedance of the load network of a class E power amplifier (PA). The extended method lies in a new analytical expression for calculating the output load network, which allows for the parasitic elements and the time that the input signal...

In FET I(V) nonlinear modeling, the I(V) experimental data are important because they allow one to determine the initial values of the empirical nonlinear model. Pulsed I(V) are preferred over DC measurements because of dispersion phenomena. However, the simulation of a class E PA using I(V) empirical nonlinear model constructed from DC or pulsed m...

In this paper, a simple and effective method of extracting the values of the parameters of the Angelov nonlinear I(V) model of GaAs HJ-FETs is presented. In the proposed method, the coefficients of the function ψ are determined from I(V) pulse measurements, using the least-squares optimization method. The utility of the proposed method is demonstra...

GaAs FET's are widely used in system such LNA, PA, mixers, frequency multipliers, oscillators and attenuators. Recent works have shown the usefulness of these transistors in the optoelectronics field, as photodetectors or optoelectric mixers. In this paper an experimental study explores the idea of unifying the detection and mixing of two optical m...

This work deals with in-fixture calibration using line-attenuator-reflect, LAR, and line-attenuator-line, LAL, as calibration standards. A novel procedure for in-fixture calibration using line-attenuator-line is presented. To demonstrate the usefulness of the proposed in-fixture calibration, the measured S parameters of a medium power GaAs HJ-FET N...

Not Available

~This paper presents a theoretical study focus on the impact of the parasitic resistances of a GaAs FET, R. and Rs, on the voltage and current waveform of a transmission line class E PA at microwave frequencies. In order to consider the parasitic resistances, the classical equivalent circuit of a transmission line class E PA was modified. This feat...

This paper presents an IDS(VGS,VDS)model to represent PHEMT behavior in the reverse )0 (< DS V , and forward zone )0 (> DS V .The model predicts with high accuracy the measured data as well as higher orders derivatives of the transconductance over a large range of VDS bias.These characteristics are important for the analysis of intermodulation dist...

This work deals with a nonlinear model for the gate-source capacitance CGS (VGS, VDS) and gate-drain capacitance CGD (VGS, VDS) of GaAs MESFET, HEMT and PHEMT transistors. An analytical bias dependent expression for modeling the CGS (VGS, VDS) and CGD (VGS, VDS) capacitances is developed. The CGS (VGS, VDS) and CGD (VGS, VDS) experimental values ar...