Showing papers by "Johan Schoukens published in 2011"

Optimal multisine excitation design for broadband electrical impedance spectroscopy

Abstract: Electrical impedance spectroscopy (EIS) can be used to characterize biological materials in applications ranging from cell culture to body composition, including tissue and organ state. The emergence of cell therapy and tissue engineering opens up a new and promising field of application. While in most cases classical measurement techniques based on a frequency sweep can be used, EIS based on broadband excitations enables dynamic biological systems to be characterized when the measuring time and injected energy are a constraint. Myocardial regeneration, cell characterization in micro-fluidic systems and dynamic electrical impedance tomography are all examples of such applications. The weakness of such types of fast EIS measuring techniques resides in their intrinsic loss of accuracy. However, since most of the practical applications have no restriction over the excitation used, the input power spectrum can be appropriately designed to maximize the accuracy obtained from the measurements. This paper deals with the problem of designing the optimal multisine excitation for electrical bioimpedance measurements. The optimal multisine is obtained by the minimization of the Cramer–Rao lower bound, or what is the same, by maximizing the accuracy obtained from the measurements. Furthermore, because no analytical solution exists for global optimization involving time and frequency domains jointly, this paper presents the multisine optimization approach partially in both domains and then combines the results. As regards the frequency domain approach, a novel contribution is made for the multisine amplitude power spectrum. In the time domain, multisine is optimized by reducing its crest factor. Moreover, the impact on the information and accuracy of the impedance spectrum obtained from using different multisine amplitude power spectra is discussed, as well as the number of frequencies and frequency distributions. The theory is supported by a set of validation measurements when exciting with the optimal and flat multisine signals and compared to a single frequency ac impedance analyzer when characterizing an RC circuit. In vivo healthy myocardium tissue electrical impedance measurements show that broadband EIS based on multisine excitations enable the characterization of dynamic biological systems.

Detecting and analyzing non-linear effects in respiratory impedance measurements

Abstract: This article describes the nonlinear effects in the respiratory impedance and in the related measurement instrumentation during the forced oscillation technique (FOT) measurements. First, the principle of FOT-measurements and nonlinear variance analysis is explained. Two methods are considered to detect nonlinear effects: a robust method and a fast method. These methods are employed to compare the nonlinear distortions in a prototype device and a commercial device, respectively. The identification signal for the respiratory impedance is optimized to reduce the nonlinear distortions. Finally, the nonlinear effects are measured in the respiratory impedance. A set of lung function tests are performed in several groups of patients with various lung conditions (healthy, asthma, cystic fibrosis, smoking). Based on the measured data from the patients, the corresponding nonlinear distortions can be quantified and evaluated for classification purposes.

Identifying a Wiener system using a variant of the Wiener G-Functionals

Abstract: This paper concerns the identification of nonlinear systems using a variant of the Wiener G-Functionals. The system is modeled by a cascade of a single input multiple output (SIMO) linear dynamic system, followed by a multiple input single output (MISO) static nonlinear system. The dynamic system is described using orthonormal basis functions. The original ideas date back to the Wiener G-functionals of Lee and Schetzen. Whereas the Wiener G-Functionals use Laguerre orthonormal basis functions, in this work Takenaka-Malmquist orthonormal basis functions are used. The poles that these basis functions contain, are estimated using the best linear approximation of the system. The approach is illustrated on the identification of a Wiener system.

Method and device for determining non-linear effects in the respiratory impedance

Abstract: The present invention is directed to a method and device for measuring the respiratory impedance over low frequencies, e.g. in a range between 0.1 and 50 Hz. The present invention provides a method for determining non-linear effects in respiratory impedance of a subject having a respiratory system. The method comprises constructing an excitation signal comprising a plurality of excitation frequencies, applying the constructed excitation signal to an actuator, e.g. an FOT device, for applying pressure oscillations to the respiratory system of the subject, measuring the frequency response function of the respiratory impedance and performing a frequency analysis, and determining presence of non-excited harmonics in the frequency response function.

A state-space view on locally-stable, globally-unstable nonlinear models driven by Gaussian burst inputs

Abstract: In this paper, the behaviour of nonlinear dynamic systems driven by stationary random excitations is studied from a model-based perspective - i.e. starting from a perfect knowledge of the system under study and its driving random input - over a finite time interval (a burst excitation is assumed). For a given discrete-time nonlinear state-space model operating in the neighbourhood of a stable equilibrium, a “blow-up” is seen as the event of escaping out of a region of attraction. Based on Laplace integration, a method is outlined to approximate a future state's probability density function (pdf) at low excitation amplitudes. Inspection of this pdf can reveal additional insights into the complex behaviour of an abstract state-space model, compared with the simulation approach. The probability of staying inside the region of attraction (viz. obtaining a bounded operation subject to an input active in a finite time interval) can be obtained by integration of this pdf. The state pdf estimation is illustrated with numerical Monte-Carlo simulation experiments.

Identification of Stiffness and Damping in Nonlinear Systems

Abstract: Resonance nonlinearities in vibrating mechanical structures are either due to stiffness, damping, or a combination of both. A method is presented to detect the nonlinear distortions, determine and quantify the distortion levels. This is achieved by configuring a nonlinear device as a second-order, single-input-single-output (SISO) closed-loop feedback system such that static nonlinearities are confined to the feedback path, and the dynamic linear part is modeled as the forward gain. The closed-loop system is then subjected to random phase multisine excitations. This makes it possible to model the linear part by its frequency response function, thus facilitating the characterization of the nonlinear part. There is a good agreement between the estimated and experimental data. The results indicate distortion nonlinearities due to stiffness and damping with distinguishable levels.

Constrained time-variant signal modeling for identifying colliding harmonics in sound mixtures

Abstract: We present a general approach to identifying time-variant colliding harmonics in pitched steady-state monophonic sound mixtures. Each sound is described by a linear-in-parameter quasi-harmonic model which captures properly instantaneous time variations of the non-stochastic sound energy. The model parameters corresponding to colliding harmonics are estimated on the basis of multiple-solution cost function L2 minimization with regularization. The major advantage against the state-of-the art methods is that no additional information about the underlying sounds in the mixture is needed. A comparative study shows that the proposed method performs significantly better than an existing algorithm for separation of monaural pitched sounds from a mixture.

On the efficiency loss of the local polynomial method for single experiment MIMO frequency response matrix extraction

Abstract: The estimation the Frequency Response Matrix (FRM) of multiple-input multiple-output (MIMO) Linear Time-Invariant (LTI) systems is often the first step in analyzing, modeling and controlling the system at hand. Recently, the so-called local-polynomial method (LPM), was developed to extract a nonparametric FRM at all frequency bins out of a single MIMO experiment. This LPM has an efficiency loss when it comes to the variability of the estimated FRM. This paper studies the sources of the efficiency losses in the LPM when using a single experiment to determine the FRM for MEMO systems. A theoretical study of the efficiency shows that a zippered multisine experiment has a low efficiency loss in the LPM when using only a single experiment This is confirmed using simulation results.