2024 IEEE MTT-S Young Professionals Workshop on Modeling, Optimization, and Measurement Techniques for Active Devices (MOMA)

Joint-chairs:
Gian Piero Gibiino (University of Bologna, Italy), Justin King (Trinity College Dublin, Ireland)

Abstract:
The third edition of this virtual workshop aims to discuss recent and prospective trends in the broad field of RF active device modeling and characterization plus measurement techniques, spanning from the component/device level up to the system level. There are six presentations from our Young Professional (YP) speakers where they will discuss their ongoing research and latest results. The first presentation will be an invited industry talk from AMCAD.

Event Date / Time:
Tue October 29,  2024
13.00 UTC to 17.00 UTC
14.00 CET to 18.00 CET
09:00 EDT to 13.00 EDT

This workshop will be conducted virtually via Zoom. It is open to everyone free of charge.
Please register here in advance to receive the Zoom link.

Workshop Agenda:

Session 1: 13.00 UTC – 15.00 UTC

[S1-1] 13:00 – 13:50 UTC – Wafae El Fennouri and Frédérique Simbelie (AMCAD Engineering, France)

Industry Talk: “From component modeling for circuit design to circuit modeling for system”
Abstract – This presentation discusses innovative approaches in transistor and circuit modeling for 5G and 6G telecommunications. To reach a first-pass design success, the modeling process must adapt from transistor to system levels to predict and enhance RF performance. This presentation introduces a novel methodology for characterizing GaN HEMT technologies that involves measuring the current-voltage (I-V) characteristic in a non-50 Ω environment to accurately model dynamic trapping states. The trapping state of charge is dynamically controlled using calibrated RF pre-pulses derived from preliminary large-signal measurements. Furthermore, using such transistor models, the presentation highlights a new Power Amplifier modeling flow at the circuit design stage, to capture PA behavior under various conditions (instantaneous input power, carrier frequency and load impedance variations), offering valuable insights for circuit performance optimization, system design and integration.

[S1-2] 13:50 – 14:25 UTC – Miles Lindquist (Ohio State University, US)

 “Large-Signal Pulsed RF Measurements to Investigate Trapping Effects in GaN HEMTs Using a Real-Time NVNA”
Abstract – This talk will detail the construction and calibration of a real-time NVNA and present pulsed RF large-signal measurements obtained with it. The NVNA, which is comprised of two real-time oscilloscopes, is able to measure full loadlines one RF period at a time making it ideal to track slow-moving transient effects. This testbed has been used primarily to track drain-lag effects in GaN HEMTs that result from large-signal RF operation of the device, where the instantaneous drain voltage of the device will fill additional traps when above the set drain bias. In one application, a set of pulsed class-B active loadpull measurements are performed at different bias conditions and a model based on Shockley-Read-Hall statistics is developed to reproduce the pulsed quiescent drain currents measured before and after each application of the pulsed class-B excitations. Additionally, a method of calibration for modulated measurements will be presented.

[S1-3] 14:25 – 15:00 UTC – Felix Heinz (Fraunhofer IAF, Germany)

 “Characterization and modeling of InGaAs HEMTs at room temperature and cryogenic temperature for ultra-low-noise applications”
Abstract – High-electron-mobility transistors (HEMTs) based on an InGaAs channel offer best noise performance at RF-frequencies among today’s various transistor technologies. In order to achieve best noise in circuits, very accurate simulations are necessary. This is especially true for microwave monolithic integrated circuit (MMIC) design, where design iterations are costly and time consuming. This demands for precise characterization techniques and scalable device models to achieve best circuit performance. This talk deals with the characterization and modeling of metamorphic HEMTs with sub-100-nm gate length for low-noise applications at room temperature and under cryogenic conditions. Current measurement methods, models and RF-circuits are presented.

10-minute break

Session 2: 15.10 UTC – 16:55 UTC

[S2-1] 15:10 – 15:45 UTC – Simone Cangini (University of Bologna, Italy)

“Non-Linear Modeling and Simulation of mm-Wave GaN MMIC Varactors by Network Synthesis”
Abstract – Varactor devices are promising building blocks for a variety of tunable microwave circuits, requiring accurate models for their simulation in CAD environments. This presentation introduces a novel technique for automatically deriving equivalent-circuit models for varactors targeting high-frequency applications. The approach, based on the Vector Fitting (VF) algorithm and the Bott-Duffin (BD) network synthesis, allows to produce a behavioral lumped-component topology directly from measured multi-bias S-parameters data, without requiring prior information about the physical structure of the device. Multiple Gallium Nitride (GaN) varactor device models are reported as long as the implementation of the proposed nonlinear circuit in a commercial CAD tool.

[S2-2] 15:45– 16:20 UTC – Dries Peumans (Vrij Universiteit Brussels, Belgium)

“Combining local modelling and Volterra series for enhanced nonlinear behavioural modelling and power amplifier linearization”
Abstract – Nonlinear behavioural modelling is essential for the effective implementation of digital predistortion, significantly enhancing performance, efficiency, and signal quality in communication systems. This presentation introduces an innovative nonlinear modelling technique that employs local polynomial approximations of Volterra series kernels in the frequency domain under wideband excitations. This method results in a bandwidth-limited and parameter-efficient digital predistortion technique.

[S2-3] 16:20 – 16:55 UTC – Raúl Criado (Universitat Politècnica de Catalunya, Spain)

“Cascaded DPD models for wideband PA linearization”
Abstract – This workshop presentation will explore cascaded digital predistortion (DPD) models for the linearization of wideband power amplifiers (PAs). We will discuss the challenges of linearizing wideband signals and present a novel approach that utilizes cascaded DPD models to enhance linearity while maintaining efficiency. We will highlight the significant improvements offered by cascaded DPD models over traditional single-stage polynomial-based models. Attendees will gain insights into the underlying principles, implementation strategies, and performance improvements achieved with this proposed methodology.

5-min closing