Showing papers by "Johan Schoukens published in 2009"

Wiener-Hammerstein Benchmark

Abstract: This paper describes a benchmark for nonlinear system identification A Wiener-Hammerstein system is selected as test object In such a structure there is no direct access to the static nonlinearity starting from the measured input/output, because it is sandwiched between two unknown dynamic systems The signal-to-noise ratio of the measurements is quite high, which puts the focus of the benchmark on the ability to identify the nonlinear behaviour, and not so much on the noise rejection properties The benchmark is not intended as a competition, but as a tool to compare the possibilities of different methods to deal with this specific nonlinear structure

Time-variant harmonic signal modeling by using polynomial approximation and fully automated spectral analysis

Abstract: We present a novel approach to modelling time-variant harmonic content in audio signals. We show that both amplitude and fundamental frequency time variations can be compactly captured in a single time polynomial which modulates the fundamental harmonic component. A correct estimation of the fundamental frequency is assured through the fully automated spectral analysis method (ASA).The best-fit is easily obtained by linear least-squares, given the fact that the set of equations is linear-in-parameters. In contrast to the existing methods, the proposed approach is designed to properly describe harmonic structures in audio signals under conditions of both AM and FM modulations and low signal-to noise ratios.

Blind Maximum-Likelihood Identification of

Abstract: This paper is about the identification of discrete-time Wiener systems from output measurements only (blind identifi- cation). Assuming that the unobserved input is white Gaussian noise, that the static nonlinearity is invertible, and that the output is observed without errors, a Gaussian maximum-likelihood esti- mator is constructed. Its asymptotic properties are analyzed and the Cramer-Rao lower bound is calculated. A two-step procedure for generating high-quality initial estimates is presented as well. The paper includes the illustration of the method on a simulation example.

Modeling the baseband output envelope of a Microwave detector

Abstract: The current state-of-the-art for the calibration of wave spectra containing spectral lines that lie on a harmonic grid (f 0 , 2f 0 …) relies on the well-established step recovery diode method [1]. The challenge in narrow band modulated measurements resides in the calibration of the instrument's phase distortion for such signals. A crystal detector (HP 420C, [2,3]) is studied as a candidate reference element for the phase calibration of the Large-Signal network analyzer (LSNA) for modulated excitations as the detector translates the envelope of the RF-signal to IF frequencies. A realistic crystal detector itself will introduce phase distortion, hence the identification and validation of a parametric black-box model of the detector was undertaken. In this work a nonlinear feedback model for a crystal detector is constructed: the model contains a low-pass filter in the feedforward path and a Wiener system in the feedback loop. In the first part of the paper the model extraction is discussed. The model is estimated from baseband data and needs to be validated for use with high frequency signals: the model needs to be extrapolated to RF frequencies. Therefore the physical representation of the model structure is translated into its differential equation in section 4. By means of this equation the low frequency output envelope is computed for RF input signals and compared to the measured envelope in section 5.

Wave distorsion in multiplying, switching or sampling mixers

Abstract: Conceptually, engineers tend to replace mixers by multipliers, controlled switches or samplers. It is clear that these models lead to very different signal processing operations and therefore are in general incompatible. In the special case that the mixers are used as frequency translation devices and one is interested in the first lower or upper conversion band only, the equivalence between these models is shown for general multiple tone input signals. Distortion of the shape of the converted temporal waveforms is shown to occur. Fortunately, this distortion does not jeopardize the measurement results obtained by mixing or sampling waveform measurement front ends as the envelope of the modulated signal is shown to be insensitive to the type of distortion at hand.