

Contributions
LSNA Technology Library
On this page
General
 October 2009  17th EuMW
Rome, Italy
OnWafer LSNA Measurements Including DynamicBias
Gustavo Avolio, Guillaume Pailloncy, Dominique Schreurs, Marc Vanden Bossche, and Bart Nauwelaers
A novel setup extending Large Signal Network Analyzer (LSNA) capabilities is described in this work. The new
setup allows the simultaneous onwafer measurement of highfrequency response (600 MHz50 GHz) and currents/voltages
induced at lowfrequency scale (10 kHz24 MHz) when the nonlinear DUT is excited by a periodic modulated signal.
Experiments carried out on FinFET devices are reported. It will be shown that, in certain conditions, the contribution of lowfrequency
information cannot be disregarded as it causes a significant discrepancy to appear in the current/voltage at the DUT terminals.
Presentation: (pdf  600 KB)
 December 2008  72nd ARFTG
Portland, Oregon, USA
Simultaneous measurement of high and low frequency response of nonlinear microwave circuits
Gustavo Avolio, Guillaume Pailloncy, Dominique Schreurs, Marc Vanden Bossche, and Bart Nauwelaers
In this work, measurements of the high frequency as well as the low frequency response of a nonlinear
microwave circuit are reported. The developed setup is based on an extension of the LSNA and it enables
the simultaneous measurement of baseband response and RF behaviour. Thanks to this capability a direct
correlation between the variation of baseband impedance and asymmetry of distortion components around
the fundamental carrier is possible. Experimental results of twotone measurements, carried out on a
hybrid GaAs microwave circuit, are shown.
Poster: (pdf  638 KB)
 November 2008  InMMIC Workshop
Málaga, Spain
Complete characterisation of LF and RF dynamics at device terminals within microwave circuits
Gustavo Avolio, Dominique Schreurs, Bart Nauwelaers, Guillaume Pailloncy, and Marc Vanden Bossche
This work presents a way to determine the complete response, encompassing both the low and highfrequency
components, at the device terminals within a microwave circuit. The measurement setup is based on an
extension of the largesignal network analyser. Experimental results on a GaAs power amplifier are analysed.
Presentation: (pdf  980 KB)
 December 2008  MOSAK Meeting
San Francisco, USA
Using LargeSignal Measurements for Transistor Characterization and Model
Verification in a Device Modeling Program.
Maciej Myslinski (K.U.Leuven), Giovanni Crupi (Univ. of Messina), Marc Vanden Bossche (NMDG),
Dominique Schreurs (K.U.Leuven), Bart Nauwelaers (K.U.Leuven)
The trend in microwave industry to deliver more for less leads to new semiconductor technologies, increasing level of
integration and complexity of the circuits, as well as shorter timetomarket of the designed components. To support
these evolutions, there has been a growing need not only for for more advanced characterization techniques but also
for more accurate and reliable device models.
LargeSignal Network Analyzer (LSNA), also known as Nonlinear Vector Network Analyzer (NVNA), measurements provide
complete information about the behavior of the measured device under realistic microwave signal conditions. On the
other hand, the available semiconductor device model extraction software tools offer an integral platform to process,
visualize, analyze various measurement data, and to apply them for the modeling purposes. Unfortunately, these programs,
despite their constant development, do not benefit from the largesignal measurement capability.
In this work we extend the functionality of one of such software tools, namely ICCAP, to work also with the largesignal
measurements. The LSNAmeasured data are incorporated into the software environment and after simple processing
are provided to the user in frequency, time and power domains.
The application of these data to the transistor characterization and the model verification is demonstrated on the
example of a MOSFET device and its Angelov model, respectively.
Presentation: slides+notes (pdf  2.3MB)
 slides only (pdf  2.2MB)

October 2007  TC33 workshop on high–frequency phase calibration techniques – EuMW
Munich, Germany
Frans Verbeyst organized the workshop with contributions from leading National Metrology Institutes like NIST, PTB and NPL
and universities like VUB (ELEC).
The workshop was organized and chaired by IEEE T&M TC33 and sponsored by TARGET and IEEE.
Marc Vanden Bossche is technical chair of IEEE T&M TC33.
Phase Calibration, an Overview from a LargeSignal Network Analysis Point of View.
Frans Verbeyst
First, it is explained why (accurate) phase is important and what LargeSignal Network Analysis is all about. Next, the
calibration process of a LargeSignal Network Analyzer is explained with a focus on phase calibration. Phase calibration
of a LargeSignal Network Analyzer requires a calibrated Harmonic Phase Reference. This, in its turn, requires an ideal sampling
oscilloscope. Because "real" sampling oscilloscopes suffer from both time base and vertical errors, each of them either need to
be measured, estimated and compensated or must be avoided. Also, one must properly deal with the effect of adapters and
mismatches.
The characterization of the dynamics of the sampling scope (nonideal amplitude and phase characteristic) is based on either
the nosetonose or EOSbased calibration procedure. Both are explained shortly. Next, the former is explained in more detail,
while the latter will be explained during the other presentations. Two applications of a nosetonose calibrated sampling
oscilloscope are explained, one of them being the HPR calibration, the other being the characterization of an optoelectric
component. It is explained that EOSbased calibration of a sampling oscilloscope is the "inverse" of the second application.
For each application, the equivalent model is shown. Discrepancies found between EOSbased and nosetonose based sample scope
calibration are highlighted. Finally it is explained that time base related errors must be dealt with, independent of the
choice of EOS or nosetonose.
Presentation: (pdf  1MB)

September 2007  RFMTC
Gävle, Sweden
Extending vector network analyzers for the nonlinear characterization of diodes, transistors and amplifiers in frequency and time domain.
Frans Verbeyst
After explaining the capibilities of existing characterization techniques, this workshop gives a short introduction to LargeSignal
Network Analysis theory.
Turning the theory into practice, it is explained how vector network analyzers can be extended to characterize the harmonic nonlinear
behavior of components, even under non50 Ohm circumstances.
This measurement capability results into different new applications. Some of them will be discussed briefly. To conclude the symbiose
between largesignal network analyzers and new type of tuners will be illustrated.
Presentation: (pdf  4.3MB)

May 2007  IMTC
Warsaw, Poland
Enhanced Time Base Jitter Compensation of Sine Waves.
Frans Verbeyst (NMDG/VUB), Yves Rolain (VUB), Rik Pintelon (VUB), Johan Schoukens (VUB)
The goal of this paper is to estimate the amplitude of a sine wave in the presence of time base jitter, time base drift, time
base distortion and additive noise. This work is motivated by a comparative study of the amplitude distortion estimated using a
nosetonose and electrooptic sampling based calibration of a highfrequency sampling oscilloscope. It uses the exact expression
of the variance of a sine wave in the presence of normally distributed additive and jitter noise, instead of a Taylor approximation
of this expression.
Article: (pdf  400kB) –
Presentation: (ppt  1MB)

May 2007  IMTC
Warsaw, Poland
System Identification Approach Applied to Drift Estimation.
Frans Verbeyst (NMDG/VUB), Rik Pintelon (VUB), Yves Rolain (VUB), Johan
Schoukens (VUB), Tracy S. Clement (NIST)
A system identification approach is applied to estimate the time base drift introduced by a highfrequency sampling
oscilloscope. First, a new least squares estimator is proposed to estimate the delay of a set of repeated measurements in the presence
of additive and jitter noise. Next, the effect of both additive and jitter noise is studied in the frequency domain using simulations.
Special attention is devoted to the covariance matrix of the experiments, which is used to construct a weighted least squares estimator that
minimizes the uncertainty of the estimated delays. Comparative results with respect to other stateoftheart methods are shown.
Finally, the enhanced method is applied to estimate the drift observed in repeated impulse response measurements of an optoelectrical
converter using an Agilent 83480A sampling oscilloscope in combination with a 83484A 50 GHz electrical plugin.
Article: (pdf  700kB) –
Presentation: (ppt  1.5MB)

September 2006  PhD dissertation (Vrije Universiteit Brussel)
Brussels, Belgium
Contributions to LargeSignal Network Analysis.
Frans Verbeyst
This PhD contains contributions to LargeSignal Network Analysis at the level of raw LSNA measurements, accurate phase,
Volterrabased behavioural modelling and both measurements and modelling under mismatch conditions.
(pdf  5.6 MB)

April 2006  IMTC
Sorrento, Italy
System Identification Approach Applied to Jitter Estimation.
Frans Verbeyst (NMDG/VUB), Yves Rolain (VUB), Johan Schoukens (VUB),
Rik Pintelon (VUB)
A system identification approach is applied to estimate the jitter introduced by a highfrequency sampling oscilloscope. An
extended model is proposed to describe the sample variance of a set of repeated (impulse response) measurements in the presence
of additive and jitter noise. Then, the (weighted) leastsquares and maximum likelihood estimator are introduced to estimate the
standard deviation of this additive and jitter noise.
First, results are shown based on simulations. These allow to test both the correctness of the implementations, to verify the
ability to detect model errors and to study the effect of uncertainties on the input signal.
Next, the jitter and additive noise standard deviation are estimated on real measurements by performing impulse response measurements
using an Agilent 83480A sampling oscilloscope in combination with 83484A 50 GHz electrical plugins.
Additional challenges, such as the conjugated effect of time base drift and time base distortion, are described and correctly taken
care of, demonstrating the real power of a solid stochastical framework.
Article: (pdf  1.4MB) –
Presentation: (pdf  1.5MB)
Award: Honorable mention
recognized by the Award Commission of Agilent Technologies

June 2005  EEEfCOM
Ulm, Germany
A Network Analyzer for Active Components.
M. Vanden Bossche (NMDG), R. Tuijtelaars (BSW)
This presentation demonstrates how an active component can be characterized from small  signal behavior (sparameters) to
largesignal behavior with one connection. The component is analyzed with a one tone and modulation signals in a 50 Ohm environment
and non50 Ohm environment..
(pdf  3 MB)
NMDG Contributions to NVNA Users' Forum

June 2010  ARFTG
IMS  Anaheim, USA
Switching amplifiers and LargeSignal Deembedding.
Guillaume Pailloncy
This presentation demonstrates how the R&S/NMDG ZVxPlus simplifies the design of amplifiers.
Using largesignal deembedding to bring measurement reference planes close to the intrinsic nonlinearities, one can indeed
optimize an amplifier in different classes of operation using directly the theoretical design techniques available in
textbooks.
(pdf  748 KB)

June 2009  ARFTG
IMS  Boston, USA
Technology update: Pulse and modulation capabilities with the ZVxPlus.
Marc Vanden Bossche
This presentation demonstrates how active components, being in a nonlinear mode of operation, can be properly characterized
with the ZVxPlus under pulse and modulation conditions.
(pdf  1.6 MB)

June 2006  ARFTG
IMS  San Francisco, USA
Metrology @ NMDG.
Frans Verbeyst
A short overview is given of the metrology applied at NMDG with respect to phase calibration of largesignal network analyzers.
(pdf  0.8 MB)

June 2006  ARFTG
IMS  San Francisco, USA
Avoiding common pitfalls when going beyond Sparameters.
Frans Verbeyst
Lesson #1: Nonlinear behaviour shows up in Sparameter measurements as "deterministic noise".
Lesson #2: b2(f) / a1(f) = S21(f) if and only if a2(f) = 0
(pdf  0.4 MB)

October 2004  EUMC
Rai, Amsterdam
Realtime PA characterization and behavioral modeling "based on LSNA technology".
F. Verbeyst
Combination of RealTime PA Analysis and Behavioral Modeling.
(pps  0.7 MB)

June 2004  ARFTG
IMS – Forth Worth, Texas
An improved Harmonic Phase Reference for LSNA capabilities up to 50 GHz.
M. Vanden Bossche
Some results are shown to the NVNA community of the phase calibration efforts up to 50 GHz going on at NMDG in support of
the MTT463 product.
(pdf  6.3 MB)

June 12th 2003  ARFTG
IMS  Philadelphia
Calibration methods for LSNAbased loadpull measurements .
M. Vanden Bossche
It is demonstrated how the ATS tuners and software from Maury Microwave work together with a largesignal network
analyzer and how the combination gives unprecedented amount of information what is going on at the device level.
(pdf  0.7 MB)

December 4, 2002  ARFTG
Monarch Hotel, Washington, D.C.
1st ARFTG NVNA Users’ Forum.
NMDG Engineering in Sp(ac)e
The situation, mission and vision explained.
Highlight on a specific topic: NARROWBAND and BROADBAND Modulation
(link)
NMDG Contributions to IMS MicroApps

June 2010
IMS  Anaheim, USA
Sfunctions, the "Sparameters" for nonlinear devices.
Guillaume Pailloncy
This article describes the NMDG "Sfunctions" measurementbased behavioral model.
When imported in simulators, Sfunctions, the "Sparameters" for nonlinear components, allow to
accurately predict both harmonic and modulation behavior of cascaded components.
Article (pdf  83 KB)
Presentation (pdf  506 KB)

June 2009
IMS  Boston, USA
Fast component characterization using modern network analysis techniques.
Guillaume Pailloncy
Presently source and loadpull techniques are used to find the optimal source and load matching
circuitry of active components. While the present tuning process is cumbersome, this paper presents
fast analysis techniques, based on largesignal network analyzer, that overcome the drawbacks of the
existing state of the art solutions.
(pdf  2.6 MB)


