Menu
EN

Vyhledávání

Lidé

prof. Ing. Pavel Sovka, CSc.

Všechny publikace

An Embedded Implementation of Discrete Zolotarev Transform Using Hardware-Software Codesign

  • Autoři: Kubák, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioengineering. 2021, 30(2), 364-371. ISSN 1210-2512.
  • Rok: 2021
  • DOI: 10.13164/re.2021.0364
  • Odkaz: https://doi.org/10.13164/re.2021.0364
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The Discrete Zolotarev Transform (DZT) brings an improvement in the field of spectral analysis of non-stationary signals. However, the transformation algorithm called Approximated Discrete Zolotarev Transform (ADZT) suffers from high computational complexity. The Short Time ADZT (STADZT) requires high segment length, 512 samples, and more, while high segment overlap to prevent information loss, 75 % at least. The STADZT requirements along with the ADZT algorithm computational complexity result in a rather high computational load. The algorithm computational complexity, behavior, and quantization error impacts are analyzed. We present a solution which deals with high computational load employing co-design methods targeting Field Programmable Gate Array (FPGA). The system is able to compute one-shot DZT spectrum 2 048 samples long in ≈22ms. Real-time STADZT spectrum of a mono audio signal of 16 kHz sampling frequency can be computed with overlap of 91 %.

Validation of freely-available pitch detection algorithms across various noise levels in assessing speech captured by smartphone in Parkinson's disease

  • DOI: 10.1016/j.bspc.2019.101831
  • Odkaz: https://doi.org/10.1016/j.bspc.2019.101831
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Measuring the fundamental frequency of the vocal folds F-0 is recognized as an important parameter in the assessment of speech impairments in Parkinson's disease (PD). Although a number of F-0 trackers currently exist, their performance in smartphone-based evaluation and robustness against background noise have never been tested. Monologues from 30 newly-diagnosed, untreated PD patients and 30 matched healthy control participants were collected. Additive non-stationary urban and household noise at different SNR levels was added to the recordings, which were subsequently assessed by 10 freely-available and widely-used pitch-tracking algorithms. According to the comparison of all investigated pitch detectors, sawtooth inspired pitch estimator (SWIPE) was the most robust and accurate method in estimating mean F-0 and its standard deviation. However, at a low 6 dB SNR level, a combination of more algorithms may be needed to achieve the desired precision. Monopitch, calculated as F-0 standard deviation and estimated by SWIPE, proved to be robust in distinguishing between the PD and healthy control groups (p < 0.001). We anticipate that monopitch may serve as a quick and inexpensive biomarker of disease progression based on longitudinal data collected via smartphone, without any logistical or time constraints for patients and physicians. (C) 2020 Elsevier Ltd. All rights reserved.

Short-time fractal analysis of biological autoluminescence

  • Autoři: Dlask, M., Kukal, J., Poplová, M., prof. Ing. Pavel Sovka, CSc., Cifra, C.M.
  • Publikace: PLoS ONE. 2019, 14(7), ISSN 1932-6203.
  • Rok: 2019
  • DOI: 10.1371/journal.pone.0214427
  • Odkaz: https://doi.org/10.1371/journal.pone.0214427
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Biological systems manifest continuous weak autoluminescence, which is present even in the absence of external stimuli. Since this autoluminescence arises from internal metabolic and physiological processes, several works suggested that it could carry information in the time series of the detected photon counts. However, there is little experimental work which would show any difference of this signal from random Poisson noise and some works were prone to artifacts due to lacking or improper reference signals. Here we apply rigorous statistical methods and advanced reference signals to test the hypothesis whether time series of autoluminescence from germinating mung beans display any intrinsic correlations. Utilizing the fractional Brownian bridge that employs short samples of time series in the method kernel, we suggest that the detected autoluminescence signal from mung beans is not totally random, but it seems to involve a process with a negative memory. Our results contribute to the development of the rigorous methodology of signal analysis of photonic biosignals.

Demand flow system with two controlled valves for support of spontaneous breathing during high-frequency oscillatory ventilation

  • Autoři: Ráfl, J., prof. Ing. Pavel Sovka, CSc., Roubík, K.
  • Publikace: The Eighth Young Biomedical Engineers and Researchers Conference YBERC 2018. Košice: Technical University of Kosice, 2018. ISBN 978-80-8086-271-8.
  • Rok: 2018
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    High-frequency oscillatory ventilation (HFOV) is a debated mode of mechanical ventilation of adult patients suffering from Acute Respiratory Distress Syndrome (ARDS). The Demand Flow System (DFS) has been developed for support of spontaneous breathing during HFOV. The DFS compensates for the pressure changes in a ventilator circuit of a high-frequency oscillatory ventilator 3100B caused by spontaneous breathing and reduces imposed breathing effort of a patient. The aim of this work was to develop the DFS for support of spontaneous breathing during HFOV with two controlled valves, inspiratory and expiratory, to optimize the DFS with respect to the reduction of a patient’s breathing effort, and to evaluate the effectiveness of versions of the DFS for support of spontaneous breathing during HFOV. A system was designed with the control valves placed independently of the HFO ventilator bias flow, allowing to separate HFOV from the action of the DFS. For the new version of the DFS, a control algorithm was implemented, based on a combination of a proportional-summation-difference controller and a filter of high-frequency (HF) oscillations. Results of animal experiments show that complete filtration of HF oscillations from measured pressure signal is essential for proper functioning of the DFS with two control valves. Filtration of HF oscillations limits the ability of a controller to compensate for the pressure changes when compared to conventional ventilators; however, it prevents vibrations of the DFS valves and assures that the DFS does not interfere with HFOV and related gas exchange. The new version of DFS reduces imposed breathing effort during HFOV better by 11% on average than a previous version of the DFS with one controlled valve and a linear-quadratic Gaussian controller. Results of this work might help to allow successful weaning from HFOV to extubation.

Fractional Brownian Bridge Model for Alzheimer Disease Detection from EEG Signal

  • Autoři: Dlask, M., Kukal, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: ICSPIS 2018 Int. Conf. on Signal Processing and Informationam Security. Dubai: UAE University, 2018. 1. vol. 1. ISBN 978-1-7281-0257-3.
  • Rok: 2018
  • DOI: 10.1109/CSPIS.2018.8642720
  • Odkaz: https://doi.org/10.1109/CSPIS.2018.8642720
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    A number of biomedical data can be investigated using methods of fractal geometry. A measurement of their nonlinear character and chaoticity can be used for subsequent data classification or irregularity detection. In this paper, we introduce the method of the fractional Brownian bridge for the Hurst exponent estimation from a signal and apply it to the electroencephalogram (EEG) data. The technique is used to detect the early stages of Alzheimer’s disease, exhibiting significant performance when compared with control patients. The measures of variability where the most significant changes occur together with the recommended EEG channels are presented in the paper.

Performance optimization methods for switched-capacitor biquadratic filters

  • DOI: 10.2478/jee-2018-0050
  • Odkaz: https://doi.org/10.2478/jee-2018-0050
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    A class of computer-aided optimization methods based on Differential Evolution (DE), Particle Swarm Optimization (PSO) and Nelder Mead algorithms applied to a switched-capacitor (SC) filter circuit design are investigated. Comparisons of these algorithms applied to a 4th order biquadratic two-channel filter bank CMOS design on 0. 35 μm technology are made. The frequency responses of the biquadratic filters must match ideal responses in a finite number of iterations with a limited number of ”particles”. The original and derived methods are evaluated on the base of their convergence progress and their reliability over different starting populations. An optimal design approach based on combining algorithms is derived as a more suitable and more reliable method for SC circuit optimization.

Suppression of overlearning in independent component analysis used for removal of muscular artifacts from electroencephalographic records

  • DOI: 10.1371/journal.pone.0201900
  • Odkaz: https://doi.org/10.1371/journal.pone.0201900
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper addresses the overlearning problem in the independent component analysis (ICA) used for the removal of muscular artifacts from electroencephalographic (EEG) records. We note that for short EEG records with high number of channels the ICA fails to separate artifact-free EEG and muscular artifacts, which has been previously attributed to the phenomenon called overlearning. We address this problem by projecting an EEG record into several subspaces with a lower dimension, and perform the ICA on each subspace separately. Due to a reduced dimension of the subspaces, the overlearning is suppressed, and muscular artifacts are better separated. Once the muscular artifacts are removed, the signals in the individual subspaces are combined to provide an artifact free EEG record. We show that for short signals and high number of EEG channels our approach outperforms the currently available ICA based algorithms for muscular artifact removal. The proposed technique can efficiently suppress ICA overlearning for short signal segments of high density EEG signals.

Evaluation of Computing Symmetrical Zolotarev Polynomials of the First Kind

  • Autoři: Kubák, J., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Radioengineering. 2017, 26(3), 903-913. ISSN 1210-2512.
  • Rok: 2017
  • DOI: 10.13164/re.2017.0903
  • Odkaz: https://doi.org/10.13164/re.2017.0903
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This report summarize and compares various methods for computing the symmetrical Zolotarev Polynomial of the first kind and its spectrum with each other. Suitable criteria are suggested for the comparison. The best numerical stability shows the method employing Chebyshev polynomial recurrence. In case of the polynomial spectrum computation the best method is the one using the difference backward recursion introduced by M. Vlček. Both methods are able to generate the polynomial of high degree up to, at least, 2000, using 32-bit IEEE floating point arithmetics.

Poisson pre-processing of nonstationary photonic signals: Signals with equality between mean and variance

  • DOI: 10.1371/journal.pone.0188622
  • Odkaz: https://doi.org/10.1371/journal.pone.0188622
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Photonic signals are broadly exploited in communication and sensing and they typically exhibit Poisson-like statistics. In a common scenario where the intensity of the photonic signals is low and one needs to remove a nonstationary trend of the signals for any further analysis, one faces an obstacle: due to the dependence between the mean and variance typical for a Poisson-like process, information about the trend remains in the variance even after the trend has been subtracted, possibly yielding artifactual results in further analyses. Commonly available detrending or normalizing methods cannot cope with this issue. To alleviate this issue we developed a suitable pre-processing method for the signals that originate from a Poisson-like process. In this paper, a Poisson pre-processing method for nonstationary time series with Poisson distribution is developed and tested on computer-generated model data and experimental data of chemiluminescence from human neutrophils and mung seeds. The presented method transforms a nonstationary Poisson signal into a stationary signal with a Poisson distribution while preserving the type of photocount distribution and phase-space structure of the signal. The importance of the suggested pre-processing method is shown in Fano factor and Hurst exponent analysis of both computer-generated model signals and experimental photonic signals. It is demonstrated that our pre-processing method is superior to standard detrending-based methods whenever further signal analysis is sensitive to variance of the signal.

Differences in mean arterial pressure of young and elderly people measured by oscilometry during inflation and deflation of the arm cuff

  • DOI: 10.1515/bmt-2015-0098
  • Odkaz: https://doi.org/10.1515/bmt-2015-0098
  • Pracoviště: Katedra fyziky, Katedra teorie obvodů
  • Anotace:
    Systemic arterial blood pressure (BP) is one of the most important parameters of the cardiovascular system. An oscillometric NIBP monitor was specifically designed to measure oscillometric pulsations and mean arterial pressure (MAP) during inflation and deflation of the cuff. Nineteen healthy young (age 23.1±1.7 years; mean±SD) and 35 elderly (83.9±7.9 years; mean±SD) subjects were studied. Differential analysis of MAP during inflation and deflation show mean |ΔMAP|=2.9±2.6 mm Hg in the young group (mean±SD) and |ΔMAP|=6.3±5.2 mm Hg for seniors (mean±SD). There was a significant difference (p<0.05) in means of |ΔMAP| measured during cuff inflation and cuff deflation between both groups. In about 50% of elderly subjects |ΔMAP| was higher than 5 mm Hg. Potential clinical relevance of the method needs to be further evaluated.

MONITORING OF TREMOR: DESIGN AND REALIZATION OF MEASURING DEVICE

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The paper deals with an initial study of monitoring and classification of a tremor. The different types of tremor are currently classified according to clinical scales, unfortunately, the objective approach is still missing. The principle design of a measuring device is presented in the paper together with the first examples of obtained and processed signals. The device is based on the 3-axis accellerometer and 3-axis gyroscope chip (MotionTracking sensor MPU-6050).

DESIGN AND REALIZATION OF MEASURING DEVICE FOR TREMOR EVALUATION

  • Autoři: Ing. Jan Havlík, Ph.D., Szentpétery, T., Němečková, M., Vávrová, D., Řasová, K., Zeman, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: INTERNATIONAL WORKSHOP ON COMPUTATIONAL INTELLIGENCE FOR MULTIMEDIA UNDERSTANDING. Praha: Czech Technical University in Prague, 2015. ISBN 978-1-4673-8457-5.
  • Rok: 2015
  • DOI: 10.1109/IWCIM.2015.7347079
  • Odkaz: https://doi.org/10.1109/IWCIM.2015.7347079
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Tremor as an involuntary rhythmic oscillatory movement of a body part belongs to one of the most disabling features of multiple sclerosis. Our research presented in this paper deals with the developing of an evaluation system of tremor. The system which was designed and ealized consists of the portable telemetry units with accelerometers and the base station. The system is able to store 3 axes accelerometry data in the dynamic range of ± 8 g and the sample frequency of 50 Hz. The system allows storing raw data for further processing in PC. The device was verified in the experiment with the gravitational acceleration. The illustration of results from pilot study are presented in the result section. These results show the system ability to trace various types of tremors and distinguish between them.

APPLICATION OF MOTION DETECTION METHODS IN HUNTINGTON’S DISEASE ANIMAL TRACKING SYSTEM

  • Autoři: Machek, J., Parák, J., Pokorný, M., Baxa, M., Juhas, Š., Klempíř, J., Motlík, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: YBERC 2014. Bratislava: STU v Bratislave, FEI, 2014. pp. 177-180. ISBN 978-80-971697-0-1.
  • Rok: 2014
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The presented paper describes implementation of motion detection methods in the automatic Huntington’s disease animal model movement tracking system. Animals used in this study were transmutagen pigs. All of the animal models were recorded by a static videocamera placed above the piggery. Animal movements were recorded during all day. The aim of our study is to accurately measure overall movement as well as changes in certain parts of the day. The several image segmentation methods and motion analysis were designed, implemented and tested. The best ones were selected and used for further recorded videos batch processing.

Approximation of the null distribution of the multiple coherence estimated with segment overlapping

  • DOI: 10.1016/j.sigpro.2013.10.003
  • Odkaz: https://doi.org/10.1016/j.sigpro.2013.10.003
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this fast communication we suggest an approximation of the null distribution of the multiple coherence (MC) estimated with segment overlapping. The approximation is based on the formulas known for the non-overlapped segmentation, but the parameter corresponding to the number of segments is altered. The suggested approximation is statistically tested through a Monte Carlo simulation, and it is shown that its precision is quite high for a considerable range of MC parameters.

Invariance of the Null Distribution of the Multiple Coherence

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this paper we investigate the invariance of the null distribution of the multiple coherence (MC) to the statistics of the examined signals. We show that when the MC is computed between a group of signals x_i[n], i=1,...,K and a signal y[n], the null distribution of the MC is independent of the distribution of x_i[n] and y[n] if at a given frequency the joint distribution of the spectra of the segments of x_i[n] and y[n] is rotationally symmetric with respect to the rotation of the spectra of the segments of x_i[n] or y[n]. The significance of this result lies in the improvement of the multiple coherence analysis. Hitherto, the null distribution of the MC was known only for signals with the multivariate Gaussian distribution; therefore, an MC estimate could be evaluated for its statistical significance only in this limited case. With the results presented in this paper, it will be possible to evaluate the statistical significance of MC estimates for much wider class of signals.

Optimized switched capacitor biquads for two-channel quadrature-mirror filter bank

  • DOI: 10.1049/cp.2014.0724
  • Odkaz: https://doi.org/10.1049/cp.2014.0724
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper deals with the design and optimization of biquadratic switched capacitor filters for 4th order quadrature-mirror filter banks allowing for real switch effects (e.g. on-state resistance, charge injection...) and the folded cascode OpAmp designed in 0.35 um CMOS technology. The filter is designed as a cascade of two biquads and each filter is separately optimized using the Differential Evolution Algorithm to obtain the required transfer function with deviation of less than ±0.1 dB from the designed transfer function. The filters process a baseband signals (0 - 8 kHz) using a clock frequency of 16 kHz. The resulting capacitor values are presented.

Overcoming Inter-Subject Variability in BCI Using EEG-Based Identification

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Kostílek, M.
  • Publikace: Radioengineering. 2014, 23(1), 266-273. ISSN 1210-2512.
  • Rok: 2014
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The high dependency of the Brain Computer Interface (BCI) system performance on the BCI user is a well-known issue of many BCI devices. This contribution presents a new way to overcome this problem using a synergy between a BCI device and an EEG-based biometric algorithm. Using the biometric algorithm, the BCI device automatically identifies its current user and adapts parameters of the classification process and of the BCI protocol to maximize the BCI performance. In addition to this we present an algorithm for EEG-based identification designed to be resistant to variations in EEG recordings between sessions, which is also demonstrated by an experiment with an EEG database containing two sessions recorded one year apart. Further, our algorithm is designed to be compatible with our movement-related BCI device and the evaluation of the algorithm performance took place under conditions of a standard BCI experiment. Estimation of the mu rhythm fundamental frequency using the Frequency Zooming AR modeling is used for EEG feature extraction followed by a classifier based on the regularized Mahalanobis distance. An average subject identification score of 96 % is achieved.

Autoregressive causal relation: Digital filtering approach to causality measures in frequency domain

  • Autoři: Bořil, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Signal Processing. 2013, 23(5), 1756-1766. ISSN 1051-2004.
  • Rok: 2013
  • DOI: 10.1016/j.dsp.2013.04.006
  • Odkaz: https://doi.org/10.1016/j.dsp.2013.04.006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    A novel measure of the Autoregressive Causal Relation based on a multivariate autoregressive model is proposed. It reveals the strength of the connections among a simultaneous time series and also the direction of the information flow. It is defined in the frequency domain, similar to the formerly published methods such as: Directed Transfer Function, Direct Directed Transfer Function, Partial Directed Coherence, and Generalized Partial Directed Coherence. Compared to the Granger causality concept, frequency decomposition extends the possibility to reveal the frequency rhythms participating on the information flow in causal relations. The Autoregressive Causal Relation decomposes diagonal elements of a spectral matrix and enables a user to distinguish between direct and indirect causal relations. The main advantage lies in its definition using power spectral densities, thus allowing for a clear interpretation of strength of causal relation in meaningful physical terms. The causal measures can be used in neuroscience applications like the analysis of underlying structures of brain connectivity in neural multichannel time series during different tasks measured via electroencephalography or functional magnetic resonance imaging, or other areas using the multivariate autoregressive models for causality modeling like econometrics or atmospheric physics but this paper is focused on theoretical aspects and model data examples in order to illustrate a behavior of methods in known situations.

Implementation of a Two-Channel Maximally Decimated Filter Bank using Switched Capacitor Circuits

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The aim of this paper is to describe the implementation of a two-channel filter bank (FB) using the switched capacitor (SC) technique considering real properties of operational amplifiers (OpAmps). The design procedure is presented and key recommendations for the implementation are given. The implementation procedure describes the design of two-channel filter bank using an IIR Cauer filter, conversion of IIR into the SC filters and the final implementation of the SC filters. The whole design and an SC circuit implementation is performed by a PraCAn package in Maple. To verify the whole filter bank, resulting real property circuit structures are completely simulated by WinSpice and ELDO simulators. The results confirm that perfect reconstruction conditions can be almost accepted for the filter bank implemented by the SC circuits. The phase response of the SC filter bank is not strictly linear due to the IIR filters. However, the final ripple of a magnitude frequency response in the passband is almost constant, app. 0.5 dB for a real circuit analysis.

Potential approximation in realistic Laplacian computation

  • DOI: 10.1016/j.clinph.2012.08.020
  • Odkaz: https://doi.org/10.1016/j.clinph.2012.08.020
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Objective: This paper aims to improve the shortcomings of the extant methodologies for realistic Laplacian (RL) computation, and correct the erroneous claims published in the past. Methods: We implemented several variants of RL computation methods, using various potential approximation techniques and different regularization approaches. The individual variants of the RL computation were tested using simulations based on a realistic head model computed with the boundary element method (BEM). The results which disagreed with previously published works were further analyzed, and the reasons for the disagreement were identified. Results: We identified the best regularization techniques for the surface potential approximation, and we showed that once these techniques are used there is often little difference between various potential approximations, which is in contrast with previous claims that promoted the radial basis function (RBF) approximation. Further, our analysis shows that the RBF approximation suffers from Runge phenomenon, which cannot be mitigated simultaneously for both deep and shallow sources; therefore, its good performance is guarantied only if a priori knowledge about the source depth is available. Conclusions: The previously published methodology for RL computation was not optimal. Improvements are possible if the newly suggested approach is used. Significance: The methodology presented in our paper allows more efficient utilization of the RL, providing a useful tool for processing of high density EEG recordings. Presented techniques allow to achieve high EEG spatial resolution, and avoid unnecessary spatial blurring caused by the problems in the previously published RL methodology.

Statistical evaluation of coherence estimated from optimally beamformed signals.

  • DOI: 10.1016/j.compbiomed.2013.06.010
  • Odkaz: https://doi.org/10.1016/j.compbiomed.2013.06.010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this paper we investigate a situation where we want to perform a coherence analysis of two signal sources, one of which is measured directly, and the other is measured through a sensor array affected by noise. To extract the latter signal, we suggest the use of the optimal beamforming with reference. We note, however, that this approach results in a coherence estimate that is noticeably biased, and cannot be evaluated by the known statistical tests. We therefore derive a new statistical test, that allows the evaluation of the biased coherence estimate. We illustrate the applicability of our methodology on the coherence analysis of EEG and EMG signals. We note that the suggested approach has several advantages over the surface Laplacian, which is a spatial filter commonly used in the EEG-EMG coherence analysis.

Switched Capacitor Circuit Implementation of Two-Channel Filter Bank

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The aim of this paper is to describe the implementation of a two-channel filter bank (FB) using the switched capacitor (SC) technique. The implementation of two-channel filter bank using an IIR Cauer filter, conversion of IIR into the SC filters and the final implementation of the SC filters is described. The whole filter bank was realized on a printed board and the magnitude frequency response was measured for different timing of clock signals. Resulting magnitude frequency response is given. Measured responses are compared with simulated ones.

Using ADZT for Signal Reconstruction

  • Autoři: Ing. Pavel Máša, Ph.D., prof. Ing. Pavel Sovka, CSc., Vlček, M., Špetík, R.
  • Publikace: Circuit Theory and Design (ECCTD), 2013 European Conference on. Dresden: Technische Universität, 2013. pp. 1-4. http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6653328. ISBN 978-3-00-043430-3.
  • Rok: 2013
  • DOI: 10.1109/ECCTD.2013.6662335
  • Odkaz: https://doi.org/10.1109/ECCTD.2013.6662335
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The Approximate Discrete Zolotarev Transform (ADZT) has been introduced recently. The aim of this paper is to describe the possibilities and limitations of ADZT when used for a signal reconstruction. The signal reconstruction is shown for three types of signals. Only one signal segment is used, that is why there is no need of an overlap-add (or overlap-save) technique. ADZT behaviour is compared to DFT behaviour. Unlike DFT synthesis vectors, the ADZT synthesis vectors are signal dependent. There is only one constraint: these vectors are given by the inversion of a linear combination of the DFT base vectors. This fact forms the main difference between DFT and ADZT syntheses. The signal-dependent base of ADZT brings some advantages but also some new problems with the interpretation of the results.

Zernike Polynomials and their Spectral Representation

  • Autoři: Vlček, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of the 2013 International Conference on Electronics, Signal Processing and Communication Systems (ESPCO 2013). Venice: EUROPMENT, European Society for Applied Sciences and Development, 2013. pp. 21-25. http://www.europment.net/library/2013/venice/ESPCO.pdf. ISBN 978-1-61804-207-1.
  • Rok: 2013
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The Zernike polynomials are known in optical physics, and they are used for the various diffractions and aberrations problems of lenses. They are defined on a circle, so that their representation decouples radial and axial coordinates. It is know that the Zernike radial polynomials are represented through Jacobi polynomials. This paper deals with Chebyshev expansions for Jacobi polynomials. We have developed the recursive evaluation for spectral coefficients used in these expansions. These consequently provide a straightforward interpretation of Fourier transform of Zernike polynomials.

Cesta k Aproximovanej Diskrétnej Zolotarev Kosínusovej Transformácii

  • Autoři: Janík, J., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: LETNÍ DOKTORANDSKÉ DNY 2012. Praha: České vysoké učení technické v Praze, 2012. pp. 14-19. ISBN 978-80-01-05050-7.
  • Rok: 2012
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    V tomto článku predstavujeme novú Aproximovanú Diskrétnu Zolotarev Kosínusovú Transformáciu (ADZCT), ktorej zmyslom je vylepšiť vlastnosti Diskrétnej Kosínusovej Transformácie (DCT). Zaoberáme sa vývojom transformácie a problémami, na ktoré sme počas vývoja narazili.

Comparison of selected methods for spectral analysis of non-stationary signals

  • Autoři: Turoň, V., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: 2012 International Conference on Applied Electronics. Pilsen: University of West Bohemia, 2012. pp. 317-321. ISSN 1803-7232. ISBN 978-80-261-0038-6.
  • Rok: 2012
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this paper we present a comparison of selected methods that are suitable for the spectral analysis of non-stationary signals. The first selected method is the Wavelet transform (WT), often used in practical applications for analysing and pre-processing electrocardiograph signals (ECG). The second method is an adaptive modification of the widely-used Short-Time Fourier Transform (STFT). It achieves satisfactory results for multispectral analysis of non-stationary signals. The last selected method is our Short-Time Approximated Discrete Zolotarev Transform (STADZT), which exploits the unique spectral properties of selective Zolotarev polynomials and achieves better time and frequency resolution than STFT.

How we came up with ADZCT

  • Autoři: Janík, J., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: 2012 International Conference on Applied Electronics. Pilsen: University of West Bohemia, 2012. pp. 119-122. ISSN 1803-7232. ISBN 978-80-261-0038-6.
  • Rok: 2012
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    We introduce here the new Approximated Discrete Zolotarev Cosine Transform (ADZCT) which purpose is to improve the properties of the Discrete Cosine Transform in the field of non-stationary signals. The paper presents the development of the ADZCT from the Discrete Zolotarev Cosine Transform step by step and deals with the issues that were met during the development. It is shown that the ADZCT achieves remarkable results in detection of nonstationary signals.

How we came up with ADZCT The way from DZCT to ADZCT

  • Autoři: Janík, J., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: 2012 International Conference on Applied Electronics. Pilsen: University of West Bohemia, 2012. pp. 119-122. ISSN 1803-7232. ISBN 978-80-261-0038-6.
  • Rok: 2012
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    We introduce here the new Approximated Discrete Zolotarev Cosine Transform (ADZCT) which purpose is to improve the properties of the Discrete Cosine Transform (DCT) in the field of non-stationary signals. The paper presents the development of the ADZCT from the Discrete Zolotarev Cosine Transform (DZCT) step by step and deals with the issues that were met during the development. It is shown that the ADZCT achieves remarkable results in detection of non-stationary signals.

Real-Time Robust Heart Rate Estimation Based on Bayesian Framework and Grid Filters

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this chapter we discuss derivation, implementation and testing of a robust real-time algorithm for the estimation of heart rate (HR) from electrocardiograms recorded on subjects performing vigorous physical activity. We formulate the problem of HR estimation as a problem of inference in a Bayesian network, which utilizes prior information about the probability distribution of HR changes. From this formulation we derive an inference procedure, which can be implemented as a grid filter. The resulting algorithm can then follow even a rapidly changing HR, whilst withstanding a series of missed or false QRS detections. Also, the HR estimate is complete with confidence intervals to allow the identification of the moments, where the precision of HR estimation is lowered. Additionally, the computational complexity of this algorithm is acceptable for battery powered portable devices.

Vybrané adaptivní metody pro spektrální analýzu signálů

  • Autoři: Turoň, V., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: LETNÍ DOKTORANDSKÉ DNY 2012. Praha: České vysoké učení technické v Praze, 2012. pp. 20-23. ISBN 978-80-01-05050-7.
  • Rok: 2012
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Tento příspěvek popisuje vybrané spektrální metody, které se v praxi používají k analýze signálů. Uvedené metody jsou rozděleny do dvou základních skupin: i) klasické metody (Krátkodobá Fourierova transformace, Vlnková transformace, Wigner-Villova transformace a Hilbert-Huangova transformace); ii) adaptivní metody, které jsou založeny na adaptivním nastavení délky okna Krátkodobé Fourierovi transformace podle různých kriterií (špičatosti spektra, velikosti spektrálního prosakování a okamžité frekvence).

A way to a new multi-spectral transform

  • Autoři: Janík, J., Turoň, V., prof. Ing. Pavel Sovka, CSc., Špetík, R., Vlček, M.
  • Publikace: Proceedings of the 11th WSEAS International Conference on Signal Processing, Computational Geometry and Artificial Vision. Athens: WSEAS Press, 2011. pp. 177-182. ISBN 978-1-61804-027-5.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    We introduce fundamentals of a new selective transform which is purposely called the Zolotarev transform. The paper contains innovative and original methods and algorithms for multi-spectral analysis of nonstationary signals using Zolotarev piolynomials.

Design and Realization of a Filter Bank by Switched Capacitor Technique

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The aim of this paper is to describe a method of filter bank realization using the switched capacitor (SC) technique. The analysis of steps needed for this realization is performed. The correct solution requires the IIR filter bank synthesis followed by the conversion to SC circuits. Due to simplicity only the two-channel filter bank is used. The main points of the whole design procedure and key recommendations for the realization are given. The complete design of SC circuits was performed by PraCAn package in Maple program and resulted circuit structures were simulated in WinSpice. The obtained results are in agreement with theoretical derivations.

Exploiting Temporal Context in High-Resolution Movement-Related EEG Classification

  • Autoři: Doležal, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioengineering. 2011, 2011(3), 666-676. ISSN 1210-2512.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution presents an application of a movement-related EEG temporal development classification which improves the classification score of voluntary movements controlled by closely localized regions of the brain. A dynamic Hidden Markov Model-based (HMM) classifier specifically designed to capture EEG temporal behavior was used. Surprisingly, HMM classifiers are rarely used for BCI design despite of their advantages. Because of this we also experimented with Learning Vector Quantization, Perceptron, and Support Vector Machine classifiers using a feature space which captures the temporal dynamics of the data. The results presented in this work show that HMM achieves the best performance due to an a priori information on physiological behavior of EEG inserted to the HMM classifier. Feature extraction process and problems with classification were analyzed as well. Classification scores of 66.7% - 94.7% were achieved in our experiments.

Porovnání dvou spektrálních metod pro analýzu akustických signálů

  • Autoři: Turoň, V., Janík, J., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Akustické listy. 2011, 17(4), 26-30. ISSN 1212-4702.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Spektrální analýza akustických signálů je komplexní obor digitálního zpracování signálů. V našem příspěvku se zaměřujeme na porovnání dvou metod spektrální analýzy. První z nich byla speciálně navržena pro analýzu audio signálů. Tato metoda využívá Diskrétní Fourierovu transformaci s adaptivním nastavováním délky okna. Výběr optimální délky okna je založen na principu minimální velikosti spektráního prosakování. Druhá metoda je nová Aproximovaná Diskrétní Zolotarevova transformace, která se zdá být vhodným nástrojem pro analýzu nestacionárních signálů.

Signals and Images of Nanoscale World for Teaching of Digital Signal Processing

  • Autoři: Kučera, O., prof. Ing. Pavel Sovka, CSc., Nejepsová, M.
  • Publikace: Proceedings of the 8th International Conference on Digital Technologies 2011. Žilina: Slovenská elektrotechnická společnost, 2011. pp. 74-78. ISBN 978-80-554-0437-0.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Subject of this project was to create four lessons for the teaching of digital signal processing using data from nanoscale material characterization methods. The database, which contains random quan-tum signals and topographic pictures of atomic and molecular lattices, was taken by Atomic Force Micros-copy and Scanning Tunneling Microscopy. The aim is to make introductory courses of signal theory and sig-nal processing more attractive and to motivate stu-dents with tasks inspired by contemporary nanosci-ence. The database is extensive enough for modifica-tion of current tasks and for future development of new assignments.

Study of ADZT properties for spectral analysis

  • Autoři: Turoň, V., Janík, J., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Proceedings of the 11th WSEAS International Conference on Signal Processing, Computational Geometry and Artificial Vision. Athens: WSEAS Press, 2011. pp. 171-176. ISBN 978-1-61804-027-5.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this paper we present a new time-frequency transform called the Approximate Discrete Zolotarev Transform (ADZT), and demonstrate its efficacy in analysing non-stationary signals. Several experiments have confirmed two main properties of ADZT: i) in contrast to the Short-Time Fourier Transform (STFT) or the Wavelet Transform (WT), the choice of window type and length are not critical; and ii) the similarity between ADZT and the Fourier transform (FT) implies a simple physical interpretation of results in contrast to Wigner-Ville time-frequency distribution (WVD) or Choi-Williams time-frequency distribution (CWD).

System Interpretation of Causality Measures in Frequency Domain Used in EEG Analysis

  • Autoři: Bořil, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: 19th European Signal Processing Conference (EUSIPCO 2011). Herent: EURASIP, 2011. p. 1539-1543. ISSN 2076-1465.
  • Rok: 2011
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper suggests new measures for the evaluation of an absolute, ratio and relative causal relation in frequency domain in the context of multivariate autoregressive models. The idea is explained on four synoptic artificial data experiments. In each example, a model without causal connection is presented and then, a causal connection is added. The influence of this modification is analysed and interpreted in the scope of LTI digital filters. Using the proposed measures, a comparison of results of state-of-the-art frequency domain methods Generalized Partial Directed Coherence (GPDC) and Direct Directed Transfer Function (dDTF) is performed to evaluate their behavior. The concept is demonstrated on real EEG data of awake resting state of human brain.

Introducing the Discrete Zolotarev Cosine Transform

  • Autoři: Janík, J., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Technical Computing Bratislava 2010. Bratislava: RT systems, s.r.o, 2010. 18th Annual Conference Proceedings. ISBN 978-80-970519-0-7.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The discrete Zolotarev cosine tranform represents a new time-frequency tranform related to the discrete cosine transform. The paper deals with the DCT transform and describes possibilities of the basis functions substitution.

Introduction to Selective Zolotarev -Cosines

  • Autoři: Janík, J., Vlček, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analýza a zpracování řečových a biologických signálů - sborník prací 2010. Praha: České vysoké učení technické v Praze, 2010. pp. 40-43. ISBN 978-80-01-04680-7.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Paper deals with introduction to zolotarev cosines, describe standard discrete cosine transform and an idea to substitute standard cosines for selective cosines. The paper points out issues of setting of the selectivity.

Metody pro analýzu kauzálních vztahů v EEG

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Analýza dynamických vztahů v mozku je významnou součástí neurofyziologie. Umožňuje vyhodnocovat nejen sílu vazeb, ale také směr jejich působení. Grangerova kauzalita a příbuzné metody založené na vícerozměrných autoregresních modelech jsou již po delší dobu používány pro analýzu směrů ovlivňování ve vícekanálových datech, nedávno byla navržena nová metoda index sklonu fáze (PSI) pracující na zcela odlišném principu. Tato práce shrnuje základní rozdíly mezi vybranými metodami a ukazuje jejich použití na modelových datech.

PERFORMANCE STUDY OF CAUSALITY MEASURES

  • Autoři: Bořil, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Technical Computing Bratislava 2010. Bratislava: RT systems, s.r.o, 2010, pp. 1-5. 18th Annual Conference Proceedings. ISBN 978-80-970519-0-7. Available from: http://www.posterus.sk/tcb2010/
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Analysis of dynamic relations in human brain has became an important task in neurophysiology in recent time. Unlike a coherence analysis used for revealing cooperating parts in the brain, causality measures make possible not only to evaluate the strength of relations but also a direction of their influence. Well-accepted Granger causality and other derived causality measures based on multivariate autoregressive models have been used to analyse directions of information flow in multichannel time series in econometrics and neurophysiology for a long time. Recently, a new method Phase Slope Index (PSI) based on completely different approach has been proposed. This paper compares conditional Granger causality with PSI, shows some important issues with PSI and proposes an idea how to interpret results of both methods used on the same data set.

Role of the feedback in accuracy of SPM measurement

  • Autoři: Kučera, O., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Seeing at the Nanoscale VIII. Basel: University of Basel, 2010, pp. 217.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Besides proper selection of probe stiffness, SPM users must correctly set parameters of the feedback loop. In this contribution we quantitatively analyze the influence of the feedback setting on the accuracy of SPM measurement.

Selective Zolotarev-Cosines for Selective Spectral Transforms

  • Autoři: Janík, J., Vlček, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Technologies 2010. Žilina: TU v Žilině, 2010. ISBN 978-80-554-0304-5.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Paper deals with the odd and even selective cosine. There are introduced mathematical definitions and possibilities of the discrete cosine trasnform improvement.

The Zolotarev transform

  • Autoři: Turoň, V., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Digital Technologies 2010. Žilina: TU v Žilině, 2010. pp. 1-2. ISBN 978-80-554-0304-5.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The paper introduces the new time-frequency Zolotarev transform, which improves several disadvantages of the common used Fourier transform, such as spectral leakage suppression.

The Zolotarev transform and selected application

  • Autoři: Turoň, V., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Technical Computing Bratislava 2010. Bratislava: RT systems, s.r.o, 2010. pp. 1-4. 18th Annual Conference Proceedings. ISBN 978-80-970519-0-7.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The paper is focused on possibilities of application of the Zolotarev transform in lossy signal compression and analysing of non-statioary signals. The results are compared with widely used Fourier transform.

Zolotarevova transformace a spektrální analýza

  • Autoři: Turoň, V., Špetík, R., prof. Ing. Pavel Sovka, CSc., Vlček, M.
  • Publikace: Analýza a zpracování řečových a biologických signálů - sborník prací 2010. Praha: České vysoké učení technické v Praze, 2010. pp. 104-108. ISBN 978-80-01-04680-7.
  • Rok: 2010
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Tento článek je věnován možnosti využití Zolotarevovy transformace při kratkodobé spektrální analýze nestacionárních signálu. Výsledky této analýzy jsou porovnány s běžně používanou Fourierovou transformací.

Active Brain Centres Selection for Function Connection Analysis

  • Autoři: Bořil, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Technical Computing Prague 2009. Praha: Humusoft, 2009, pp. 18-21. ISBN 978-80-7080-733-0. Available from: http://dsp.vscht.cz/konference_matlab/MATLAB09/prispevky/018_boril.pdf
  • Rok: 2009
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In recent years, the possibility of tomographic processing of EEG has been becoming more common, so the surface electrode records are projected into the 3D space. This allows a localization of more deeply located sources, whose activity on the scalp is very low. For subsequent processing of the tomographic EEG, like an analysis of relations among the active centres, there is a need of convenient algorithm of region of interest (ROI) selection. Several methods of ROI selection are proposed in this paper. We have developed a new viewer in Matlab environment of output files of sLoreta system. It allows manual color legend settings common to all tomographic images. The viewer is useful for direct connection to the algorithms of selection of active centres presented in the paper and implemented in Matlab.

Detekce očních artefaktů v 19-ti svodovém EEG s využitím analýzy nezávislých components

  • Autoři: Kužílek, J., prof. Ing. Pavel Sovka, CSc., Kopřivová, J., Lhotská, L.
  • Publikace: Proceedings of the 8th Czech-Slovak Conference Trends in Biomedical Engineering. Bratislava: Slovak University of Technology in Bratislava, 2009, pp. 80-83. Slovak University of Technology in Bratislava. ISBN 978-80-227-3105-8.
  • Rok: 2009
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Cílem práce je prozkoumat a ověřit možnosti využití analýzy nezávislých komponent (ICA) při detekci očních artefaktů v 19-ti svodovém EEG záznamu. Pro vlastní detekci je nutné nalézt vhodné příznaky počítané na komponentách vzniklých po ICA a provést výběr vhodných klasifikátorů. Výběr vhodných příznaků se provádí na základě shlukové analýzy algoritmem K-means. K detekci očního artefaktů se využívá klasifikátorů vzniklých pomocí algoritmů K-nearest neigbours, Support Vector Machines (SVM) a AdaBoost.

Recording and recognition of movement related EEG signal

  • Autoři: Doležal, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2009. Plzeň: Západočeská univerzita v Plzni, 2009. p. 95-98. ISSN 1803-7232. ISBN 978-80-7043-781-0.
  • Rok: 2009
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This article describes method of recording movement EEG database and experimental results of movement classification.

Robust Heart Rate Estimation Based on Imperfect QRS Detection

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper presents an algorithm for the estimation of heart rate (HR) based on imperfect detection of QRS complexes.It aims to be applicable in real world HR measurement on individuals working in highly stressful and harsh environ- ments (fireman, soldiers, etc...). The described algorithm proved to be a robust method for the estimation of HR based on intensely corrupt ECG that can be recorded on individuals performing dangerous, physically demanding and stressful tasks.

ACE Strategy Using Virtual Electrodes

  • Autoři: Vondrášek, M., Tichý, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images; Biosignal 2008 proceedings. Brno: VUTIUM Press, 2008. p. 1-7. ISSN 1211-412X. ISBN 978-80-214-3613-8.
  • Rok: 2008
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The hearing impairment is one of the frequent diseases. The profound hearing loss patients with senzorineural disorder can use cochlear implant. The cochlear implant is electronic device that bypasses a non-functional inner ear by stimulating hearing nerve by patterns of electrical pulses, so that sounds and speech cam be medicated to profound hearing loss patients. Cochlear implant can substitute hearing only partially. Patients with cochlear implants have problems with speech understanding; some of them could not recognize speakers. Also effect of using cochlear implant in noisy environment is very limited. Our research is focused on better speech understanding and music perception enhancement using better frequency discrimination for Nucleus(R) 24 cochlear implants. That was achieved by introducing virtual channels in ACE speech processing algorithm.

Avoidance of Speckle Noise in Laser Vibrometry by the Use of Kurtosis Ratio: Application to Mechanical Fault Diagnostics

  • DOI: 10.1016/j.ymssp.2007.08.008
  • Odkaz: https://doi.org/10.1016/j.ymssp.2007.08.008
  • Pracoviště: Katedra teorie obvodů, Katedra měření
  • Anotace:
    This paper presents a statistical technique to enhance vibration signals measured by Laser Doppler Vibrometry (LDV). The method has been optimised for LDV signals measured on bearings of universal electric motors and applied to quality control of washing machines. Inherent problems of LDV are addressed, particularly the speckle noise occurring when rough surfaces are measured. The presence of speckle noise is detected using a new scalar indicator Kurtosis Ratio (KR), specifically designed to quantify the amount of random impulses generated by this noise. The KR is a ratio of the standard kurtosis and a robust estimate of kurtosis, thus indicating the outliers in the data. Since it is inefficient to reject the signals affected by the speckle noise, an algorithm for selecting an undistorted portion of a signal is proposed. The algorithm includes band-pass filtering and segmentation of the signal, as well as thresholding of the KR computed for each filtered signal segment.

Classification Algorithms Comparison for EEG Movement Recognition

  • Autoři: Doležal, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images; Biosignal 2008 proceedings. Brno: VUTIUM Press, 2008. ISSN 1211-412X. ISBN 978-80-214-3613-8.
  • Rok: 2008
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This article provides a comparison of algorithms for single-trial EEG classification. EEG accompanying fast extension (E) and flexion (F) movement of right index finger is classified. The aim of our study was to verify the necessity of using the EEG time context information for classification of very closely localized and similar movements. Comparison of our system based on Hidden Markov Models (HMM) with Support Vector Machine (SVM) and Perceptron is presented. The application field of our research results is in the BCI domain to improve the movement classification resolution of the existing systems.

Electrode Position Scalling in Realistic Laplacian Computation

  • DOI: 10.1109/TBME.2008.921168
  • Odkaz: https://doi.org/10.1109/TBME.2008.921168
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This note discusses the effects of the electrode position scaling on the realistic Laplacian (RL) computation. It is shown that when the RL is estimated with the help of Tikhonov regularization and the generalized cross-validation (GCV) criterion, improper electrode position scaling may influence the GCV criterion, which results in the decrease of RL precision. We identify what the proper scaling should be, and we provide a closer examination of how the GCV criterion is affected by the electrode position scaling.

High-Resolution Movement EEG Classification

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Computational Intelligence and Neuroscience. 2008, 2008(15.01.08), 1-10. ISSN 1687-5265.
  • Rok: 2008
  • DOI: 10.1155/2007/54925
  • Odkaz: https://doi.org/10.1155/2007/54925
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The aim of the contribution is to analyze possibilities of high-resolution movement classification using human EEG. For this purpose, a database of the EEG recorded during right-thumb and little-finger fast flexion movements of the experimental subjects was created. The statistical analysis of the EEG was done on the subject's basis instead of the commonly used grand averaging. Statistically significant differences between the EEG accompanying movements of both fingers were found, extending the results of other so far published works. The classifier based on hidden Markov models was able to distinguish between movement and resting states (classification score of 94-100%), but it was unable to recognize the type of the movement. This is caused by the large fraction of other (nonmovement related) EEG activities in the recorded signals. A classification method based on advanced EEG signal denoising is being currently developed to overcome this problem.

Independent Component Analysis of Movement-Related EEG and Classification of Selected Components

  • Autoři: Ručkay, L., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2008. Pilsen: University of West Bohemia, 2008. pp. 175-178. ISBN 978-80-7043-654-7.
  • Rok: 2008
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This contribution describes a method for EEG decomposition with the help of independent component analysis (ICA), and classification of EEG reconstructed from movement-related independent components (ICs). The application is targeted to the brain-computer interface (BCI). Our previous works showed that it is possible to decompose EEG into movement-related and non-movement-related ICs, and classification of closely localized movements. The selection of only movement-related ICs for EEG reconstruction leads to BCI EEG classification score increasing. We reached classification score increasing by 2% on average by method described here.

Performance Study of Bivariate Granger Causality

  • Autoři: Bořil, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Technologies 2008. Žilina: Žilinská universita, Elektrotechnická fakulta, 2008. ISBN 978-80-8070-953-2.
  • Rok: 2008
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Identification of statistic causal relations among simultaneously recorded signals is an important problem in the area of multidimensional time series analysis with applications in many domains spreading from biology to economics. Simultaneous records of many of electrodes in neurophysiology (electroencephalograph - EEG) produce large amount of data. Multidimensional analysis of time series provides an instrument for evaluation of interactions of brain neurons in these data. Granger causality is a tool for revealing and examining such relations.

Robust and Computationally Efficient QRS Detector for HR estimation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper introduces a QRS detection algorithm aimed for the use in portable ECG devices carried by intensely moving subjects. The algorithm features high robustness against several types of artifacts while it keeps its computational requirements low enough to be implemented on an 8051 type microcontroller.

Selection and Classification of EEG Movement-Related Independent Components

  • Autoři: Ručkay, L., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images; Biosignal 2008 proceedings. Brno: VUTIUM Press, 2008. p. 1-5. ISSN 1211-412X. ISBN 978-80-214-3613-8.
  • Rok: 2008
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This contribution describes a method for EEG decomposition with the help of independent component analysis (ICA), and classification of movement-related independent components (ICs). The selection of movement-related ICs for EEG classification leads to BCI EEG classification score increasing. We reached classification score increasing by 5-30% depending on subject and movement type by method described here.

Approximation of statistical distribution of magnitude squared coherence estimated with segment overlapping

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper suggests an approximation to the statistical distribution of a magnitude squared coherence (MSC) function estimated with segment overlapping. So far, the statistical distribution is known for an MSC computed without segment overlapping only. However, as the overlapped segmentation provides a more accurate MSC estimate, its statistical distribution is desired to allow its evaluation. This paper provides an approximation of the cumulative density function, probability density function, confidence limit formula, confidence interval computation and a procedure for the comparison of two MSC estimates, all for an MSC accurately estimated with the segment overlapping. Additionally, the benefits of the knowledge of the approximated statistics are illustrated on EEG-EMG (i.e. cortico-muscular) MSC estimates.

Detection of Cortical Oscillations Induced by SCS Using Power Spectral Density

  • Autoři: Svoboda, L., Stančák, A., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioengineering. 2007, 16(4), 38-45. ISSN 1210-2512.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Chronic, intractable pain of lower back and lower extremity might develop as the result of unsuccessful surgery of back. This state called failed-back surgery syndrome cannot be effectively treated by pharmacotherapy. Electric stimulation of the dorsal spinal cord (SCS) is applied to relieve the pain. According to the medical hypothesis, oscillatory activity, which might be related to the analgesic effects, may occur in the cortex during the stimulation. To confirm the presence of the SCS induced oscillations, a new method of detection was designed. The analysis of EEG data was performed using power spectral density, confidence intervals, visualization and group statistic for its verification. Parameters of the method were experimentally optimized to maximize its reliability. During ongoing SCS, statistically significant changes were detected and localized at the stimulation frequency and/or its subharmonic or upper harmonic over central midline electrodes in eight patients.

ICA Model Order Estimation Using Clustering Method

  • Autoři: Ručkay, L., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioengineering. 2007, 16(4), 51-57. ISSN 1210-2512.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    In this paper a novel approach for independent component analysis (ICA) model order estimation of movement electroencephalogram (EEG) signals is described. The application is targeted to the brain-computer interface (BCI) EEG preprocessing. The selection of only movement related ICs might lead to BCI EEG classification score increasing. The real number of the independent sources in the brain is an important parameter of the preprocessing step. Previously, we used principal component analysis (PCA) for estimation of the number of the independent sources. However, PCA stimates only the number of uncorrelated and not independent components ignoring the higher-order signal statistics. In this work, we use another approach - selection of highly correlated ICs from several ICA runs. The ICA model order estimation is done at significance level alpha = 0.05 and the model order is less or more dependent on ICA algorithm and its parameters.

Induced Cortical Activity Detection using Spectral Analysis of Electroencephalogram

  • Autoři: Svoboda, L., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2007. Plzeň: Západočeská universita, 2007. pp. 211-214. ISBN 978-80-7043-537-3.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Failed-back surgery syndrome (FBSS) develops after unsuccessful discotomy in some patients. It causes a chronic, neuropathic pain of lower back and legs that cannot be effectively suppressed by chemical analgesics. To alleviate the pain, an electric stimulation of the dorsal spinal cord (SCS) can be applied. Its analgesic effects might be connected with induced rhythmical activity that occurs in pain matrix during the stimulation. Hence, we analyzed EEG recordings of ten patients, whose neuropathic pain was reduced using SCS, to find out whether the SCS induced the cortical oscillations at the same frequency. The analysis was performed using power spectral density (PSD). During SCS, statistically significant increase of PSD was found at the stimulation frequency over the central midline electrodes in eight patients. The results suggest that SCS induces the oscillatory activity in the medial motor and premotor cortical regions, which can be successfully detected by spectral analysis.

Localization of Cortical Oscillations Induced by SCS Using Coherence

  • Autoři: Svoboda, L., Stančák, A., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioengineering. 2007, 16(4), 46-50. ISSN 1210-2512.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper suggests a method based on coherence analysis and scalp mapping of coherence suitable for more accurate localization of cortical oscillations induced by electric stimulation of the dorsal spinal cord (SCS), which were previously detected using spectral analysis. While power spectral density shows the increase of power during SCS only at small number of electrodes, coherence extends this area and sharpens its boundary simultaneously. Parameters of the method were experimentally optimized to maximize its reliability. SCS is applied to suppress chronic, intractable pain by patients, whom pharmacotherapy does not relieve. In our study, the pain developed in lower back and lower extremity as the result of unsuccessful vertebral discotomy, which is called failed-back surgery syndrome (FBSS). Our method replicated the results of previous analysis using PSD and extended them with more accurate localization of the area influenced by SCS.

Regularization Techniques in Realistic Laplacian Computation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper explores regularization options for the ill-posed spline coefficient equations in the realistic Laplacian computation. We investigate the use of the Tikhonov regularization, truncated singular value decomposition, and the so-called lambda-correction with the regularization parameter chosen by the L-curve, generalized cross-validation, quasi-optimality, and the discrepancy principle criteria. The provided range of regularization techniques is much wider than in the previous works. The improvement of the realistic Laplacian is investigated by simulations on the three-shell spherical head model. The conclusion is that the best performance is provided by the combination of the Tikhonov regularization and the generalized cross-validation criterion-a combination that has never been suggested for this task before.

Seminars Focused on Spatial Filtering of Acoustic Signals

  • Autoři: Bolom, V., prof. Ing. Pavel Sovka, CSc., Bergl, P.
  • Publikace: Proceedings CD-ROM of Digital Technologies 2007. Žilina: Slovenská elektrotechnická společnost, 2007. ISBN 978-80-8070-786-6.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper presents two seminars focused on spatial filtering of acoustic signals. This task is a subset of multichannel signal processing and can be applied for speech enhancement in noisy environment. The seminars are suitable for students of master study program who are interested in digital signal processing. The objective of the seminars is to provide a brief and intelligible introduction to spatial filtering. The education is supplemented by practical tasks.

Spectrum Enhancement of Laser Vibrometer Signals Using a New Speckle Noise Reduction Technique and Angular Resampling

  • Autoři: Vass, J., Randall, R.B., Cristalli, C., Torcianti, B., prof. Ing. Pavel Sovka, CSc., prof. Ing. Radislav Šmíd, Ph.D.,
  • Publikace: 14th International Congress on Sound and Vibration (ICSV14) - Proceedings. Cairns: International Institute of Acoustics and Vibration (IIAV), 2007. ISBN 978-0-7334-2516-5.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů, Katedra měření
  • Anotace:
    In this paper, a new method to enhance vibration signals measured by Laser Doppler Vibrometry (LDV) is proposed. The method has been applied to LDV signals measured on washing machines for the purpose of quality control in production lines.

Technologie hlasových komunikací

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Monografie přináší ucelenou informaci o metodách číslicového zpracování signálů lidské řeči, a to v oblastech přenosu, syntézy a rozpoznávání. Poslouží jako studijní literatura k předmětu Digitální zpracování signálu řeči, který je zahrnut do magisterského programu strukturovaného studia i předmětu Fonetické signály a jejich kódování, který je ve studijním plánu doktorského programu. Jejím posláním je nejen seznámit širší veřejnost s touto vědní disciplínou všeobecně, ale publikovat i některé vědecké výsledky získané při výzkumu v laboratoři katedry teorie obvodů na Fakultě elektrotechnické ČVUT.

The 3D Surface Laplacian Filtration with Integrated Sampling Error Compensation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This work analyses the influence of the EEG recording sampling instants systematic time shift on the subsequent surface filtering. The theoretical model of the common mode (CM) signal rejection with a general surface filter is presented. Interpolation is proposed to align sampling instants. Promising results for two configurations commonly used in brain-computer interface (BCI) domain are shown. Suggested interpolation techniques might be generalized and used in any case when there is a sequentially-sampled sensor array followed by any postprocessing.

Výzkum a návrh algoritmů pro klasifikaci EEG

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Ručkay, L.
  • Publikace: Trendy v biomedicínském inženýrství, Sborník 7. česko-slovenské konference. Praha: ČVUT v Praze, Fakulta biomedicínského inženýrství, 2007. pp. 174-177. ISBN 978-80-01-03777-5.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Příspěvek shrnuje současný stav výzkumu, vývoje a návrhu algoritmů pro číslicové zpracování EEG signálů na Katedře teorive obvodů FEL ČVUT Praha.

Výzkum a návrh algoritmů pro klasifikaci EEG

  • Autoři: Šťastný, J., Ručkay, L., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Trendy v biomedicínském inženýrství, Sborník 7. česko-slovenské konference. Praha: ČVUT v Praze, Fakulta biomedicínského inženýrství, 2007, pp. 174-177. ISBN 978-80-01-03777-5. Available from: http://amber.feld.cvut.cz/fpga
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Práce ve stručnosti shrnuje dosavadní výsledky výzkumu v oblasti návrhu algoritmů pro klasifi-kaci EEG signálu v naší laboratoři. Cílem našeho výzkumu je návrh algoritmů pro rozpoznávání EEG signálu s vysokým rozlišením prostřednictvím neinvazivního snímání. V rámci naší práce byl navržen klasifikační systém pro single-trial, off-line rozpoznávání EEG postavený na skrytých markovských modelech a další existujících části řetězce zpracovávajícího EEG signál byly optimalizovány . Klasifi-kátor byl testován na několika databázích EEG signálů obsahujících EEG doprovázející volní pohyb pokusné osoby. Výsledky ukazují, že je z EEG signálu možné odlišit pohyby pravého ramene a ukazo-váčku pokusné osoby i vlastní směr pohybu ukazováčku (extenze/flexe). V současné době pracujeme na dalším vylepšení systému, které by mělo umožnit odlišit pohyby jednotlivých prstů.

Zvýrazňování řeči pomocí vícekanálového zpracování

  • Autoři: Bolom, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Technical Computing Prague 2007. Praha: Humusoft, 2007. ISBN 978-80-7080-658-6.
  • Rok: 2007
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Úlohu zvýraznování řeci přítomné ve smesi akustických signálu lze rešit pomocí vícekanálového zpracování. Příspevek se zabývá použitím adaptivního algoritmu Generalised Sidelobe Canceller. Vlivem užitecného řecového signálu dochází pri adaptaci vah k jejich rozvažování. V tomto príspevku je diskutována možnost, jak rozvažování zabránit. Je navržena modifikace algoritmu s použitím detektoru řečové aktivity.

Detekce elektrokortikálních rytmů vnucených míšním neurostimulátorem u pacientů trpících chronickou bolestí

  • Autoři: Svoboda, L., Stančák, A., Kozák, J., Vrba, I., prof. Ing. Pavel Sovka, CSc., prof. Ing. Roman Čmejla, CSc., Poláček, H., Vrána, J., Tintěra, J., Bukovský, S.
  • Publikace: 53. společný sjezd české a slovenské společnosti klinické neurofyziologie. Praha: MH Consulting, 2006. pp. 32. ISBN 80-7254-916-2.
  • Rok: 2006

Detekce korové aktivity vyvolané míšním neurostimulátorem u pacientů trpících nezvladatelnou neuropatickou bolestí

  • Autoři: Svoboda, L., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Analýza a zpracování řečových a biologických signálů - Sborník prací 2006. Praha: České vysoké učení technické v Praze, 2006. pp. 57-65. ISBN 80-01-03621-9.
  • Rok: 2006

EEG-based Biometric Person Identification

  • Autoři: Šťastný, J., Vrchota, P., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006. pp. 76-77. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This work deals with the basic research in the field of the person identification by means of EEG signals. This topic has been nearly neglected in the publications so far. The only exception known to us were works [3-5]. Owning to the small number of citations on this topic we wanted to analyze the feasibility of the EEG-based person identification as the first step of our work. We started with the verification of results published in work [3] and their further extension

EEG-EMG Coherence Enhancement

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The paper introduces a new approach to the estimation of the EEG-EMG coherence, which is used to examine the functional connection between a human brain and muscles. A typical EEG-EMG coherence estimation, with a magnitude squared coherence (MSC) barely exceeding 0.15, is enhanced so that MSC reaches or even goes above 0.5. The proposed method is mathematically analyzed, and its properties are discussed. Additionally, the paper includes several EEG-EMG coherence analysis results, with MSC exceeding 0.5.

Modelling and Recognition of Movement Related EEG Signal

  • Autoři: Doležal, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2006. Plzeň: ZČU Plzeň, 2006. pp. 27-30. ISBN 978-80-7043-442-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This article describes method and process of nonmovement related (resting) EEG signal generation along with the reached new classification results including the false alarm rate. Artificial EEG signal was generated by AR modelling. The AR model parameters were estimated from short segments (3 sec) of resting EEG already present in the database using the autocorrelation method. Owing to large intra- and inter-personal variability one set of parameters had to be estimated for each person and electrode. The artificial EEG was used for the classification with satisfactory classification results.

Movement-Related EEG Decomposition Using Independent Component Analysis

  • Autoři: Ručkay, L., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2006. Plzeň: ZČU Plzeň, 2006. pp. 149-152. ISBN 978-80-7043-442-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    A novel approach to the movement-related EEG decomposition by means of ICA is presented in this contribution.

Movement-Related EEG Decomposition Using Independent Component Analysis

  • Autoři: Ručkay, L., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006. pp. 78-80. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    A novel approach to the movement-related EEG decomposition by means of ICA is presented in this contribution.

Reducing Speckle Noise in Vibration Signals Measured with Laser Doppler Vibrometry

  • Pracoviště: Katedra teorie obvodů, Katedra měření
  • Anotace:
    The article describes a new method of processing vibration data for detection of mechanical faults in electric motors on the production line.

Virtual Channels in NucleusR 24 Cochlear Implants

  • Autoři: Vondrášek, M., prof. Ing. Pavel Sovka, CSc., Tichý, T.
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006. pp. 146-148. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Cochlear Implant is an electronic device, which substitute hearing to profoundly deaf people. Non-functioning Organ of Corti is substituted by direct stimulation of fibres of the Cochlear nerve, which are tonotopically arranged along the cochlear duct. Stimulation of apical fibres results in perception of low-pitched sounds; the closer to the base stimulation occurs, the higher is the pitch of the percept. As the number of electrodes is limited, the number of tones cochlear implant users are able to differentiate is limited, too. This paper describes a method of implementing virtual channels in Nucleus 24 cochlear implants by sequential stimulation of adjacent electrodes.

Virtual Electrodes Discrimination in Nucleus 24 Cochlear Implant

  • Autoři: Vondrášek, M., Tichý, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Applied Electronics 2006. Plzeň: ZČU Plzeň, 2006. p. 229-232. ISBN 978-80-7043-442-0.
  • Rok: 2006
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Cochlear Implant is an electronic device, which substitute hearing to profoundly deaf people. Non-functioning Organ of Corti is substituted by direct stimulation of fibres of the Cochlear nerve, which are tonotopically arranged along the cochlear duct. Stimulation of apical fibres results in perception of low-pitched sounds; the closer to the base stimulation occurs, the higher is the pitch of the percept. As the number of electrodes is limited, the number of tones cochlear implant users are able to differentiate is limited, too. This paper describes a method of implementing virtual channels in Nucleus 24 cochlear implants by sequential stimulation of adjacent electrodes. Finnaly the virtual electrode discrimination test is proceeded.

Zdokonalenie odhadu povrchového Laplasiánu využívajúceho aproximáciu realistického tvaru hlavy spline funkciami

Zvýrazňování řeči pomocí vícekanálového zpracování

  • Autoři: Bolom, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Technical Computing Prague 2006 - Sborník příspěvků 14. ročníku konference. Praha: Humusoft, 2006. pp. 22. ISBN 80-7080-616-8.
  • Rok: 2006

Zvýšenie presnosti koherenčnej analýzy pomocou novo odvodených štatistík pre odhad koherencie s prekryvom segmentov

Automatic classification of extra- and intracerebral points in source dipole analysis based on triangular model of realistic cerebral surface

  • Autoři: Mlynář, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Neuroscience 05 - Abstract Viewer/Itinerary planer. Washington, DC: Society for Neuroscience, 2005. pp. 689.4.
  • Rok: 2005

Číslicové filtry

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Článek je věnován úvodu do číslicové filtrace a vizualizaci frkevnčních charakteristik filtrů.

Číslicové filtry II

Číslicové filtry III

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Článek popisuje často používané typy filtrů a ilustruje jejich chování.

Comparison of Different Methods of Time Shift Measurement in EEG

  • Autoři: Jiruška, P., Prokš, J., Drbal, O., prof. Ing. Pavel Sovka, CSc., Marusič, P., Mareš, P.
  • Publikace: Physiological Research. 2005, 54(54), 459-465. ISSN 0862-8408.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution describes of the comparison of different methods of time shift measurement in EEG for epilepsy.

Confidence Limit Computation for EEG-EMG Coherence Estimated with Segment Overlapping

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper suggests a formula for the computation of a confidence limit for a magnitude squared coherence (MSC) estimated with segment overlapping. So far, the confidence limit could be computed exactly only if MSC was estimated without segment overlapping. In case of overlapped segmentation the confidence limit could be only estimated through a quantile estimation, a procedure, which is shown to be imprecise. The suggested formula allows a solid evaluation of MSC estimated with segment overlapping, which provides more precise results. The confidence limit computation is intended for the cortico-muscular coherence analysis, which investigate coupling between the electroencephalogram (EEG) and the electromyogram (EMG). The formula is shown to be valid for signals with length typically used for EEG-EMG MSC computation - in particular the signal length assuring at least 100 degrees of freedom of an MSC estimate.

Impact of Variance Prior to Sequential DOA Estimate

  • Autoři: Štrupl, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Technologies 2005. Žilina: Slovenská elektrotechnická společnost, 2005. pp. 84-89. ISBN 80-8070-486-4.
  • Rok: 2005

Movement-Related EEG Separation Using Independent Component Analysis

  • Autoři: Konopka, O., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005. pp. 1-4. ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This work analyses the possibility of the decomposition of the EEG recorded during right shoulder and index finger (SF) movements to movement and non-movement components using Independent Component Analysis (ICA) [1]. Purpose of this work is to enhance the EEG preprocessing for brain computer interface which can reduce EEG activity that is not related to performed movements and thus further improve EEG classification results [2][3].

On the Use of the Constant Phase Term in the EEG-EMG Time Delay Estimation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper discusses the treatment of the constant phase term that is used to fit the phase of a cross-spectrum in the estimation of the time delay between the record of electrical activity of a brain (EEG), and the record of electrical activity of a contracted muscle (EMG). Commonly, when the EEG-EMG time delay is estimated as a slope of the phase of an EEG-EMG cross-spectrum the phase spectrum is fitted with a line given by a slope and an additive constant phase term. We point out that if analyzed signals are short the permanent use of the constant phase term may be unwise. First, based on the analysis of EEG-EMG data sets, we show that the constant phase term is often not significantly different from zero. Second, we illustrate that the use of the constant phase term provides results with much greater variance than without it. As a consequence, we suggest to limit the use of the constant phase term only to the cases where it is showed to be significantly different from zero.

Optimal Parametrization Selection for the Brain-Computer Interface

  • Autoři: Šťastný, J., Zejbrdlich, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: AEE´05 - Proceedings of the 4th WSEAS International Conference on: Applications of Electrical Engineering. Athens: WSEAS Press, 2005. pp. 300-304. ISBN 960-8457-13-0.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution deals with the optimization of the EEG off-line, single-trial movement classification by means of parameterization tuning. The data we classify represent manifestations of the simple movements performed by the right shoulder (proximal movement) and right index finger (distal movement) of experimental subjects. We implemented several approaches to the EEG parameterization and compared results in order to increase the recognition score. The results are compared with the results from our earlier works and will form a strong basis for the coming experiments with a new EEG database. The target of our experiments is the implementation of the Brain Computer Interface machine recognizing movements performed on one side of the body using the non-invasive EEG scanning.

Recognition of Direction of Finger Movement from EEG Signal Using Markov Models

  • Autoři: Doležal, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005. pp. 1-4. ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This contribution deals with the application of Hidden Markov Models (HMM) on E/F movements off-line single-trial classification. The aim of our study is to verify possibilities of the very closely localized and similar movements classification. The application of our results lies in construction of the Brain - Computer Interface (BCI) based on recognition of these movements.

Speech Reconstruction and Objective Measurement for Cochlear Implants

  • Autoři: Vondrášek, M., prof. Ing. Pavel Sovka, CSc., Tichý, T.
  • Publikace: Digital Technologies 2005. Žilina: Slovenská elektrotechnická společnost, 2005. pp. 102-107. ISBN 80-8070-486-4.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Cochlear Implant is an electronic device, which substitute hearing to profoundly deaf people. Non-functioning Organ of Corti is substituted by direct stimulation of fibres of the Cochlear nerve, which are tonotopically arranged along the cochlear duct. New strategies had to be tested using simulations before the changes are verified oon patints. The speech reconstruction is algorithm which creates speech from curent pusles normaly used for stimulation. The priginal and reconstructed speech is finally compared using cepstral distance.

Speech Reconstruction for Cochlear Implants

  • Autoři: Vondrášek, M., Tichý, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005. ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution comparises two methods of speech reconstruction for speech conding strategies for cochlear implants.

Study of Speech Defects Caused by Landau-Kleffner Syndrome for Small Children

  • Autoři: Čadan, L., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Technologies 2005. Žilina: Slovenská elektrotechnická společnost, 2005. pp. 17-22. ISBN 80-8070-486-4.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution analyses speech attributes suitable for the analysis of speech defects caused by epilepsy.

The Influence of the EEG Recording Machine Time Multiplex on the Laplacian Filter: The Simulation with the Real Shaped Head Model

  • Autoři: Mlynář, J., Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005. pp. 1-4. ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The work analyzes the influence of the EEG recording machine time channel multiplex (TCX) on the Laplacian filter output signal-to-noise ratio. The real shaped head model is used for the analysis. The target of the analysis was to obtain a reliable answer to a question if the channel time multiplex should be compensated or not. The final answer is 'yes'.

Analysis and Optimization of Telephone Speech Command Recognition System Performance in Noisy Environment

Application of Zolotarev Polynomials for Non-Stationary Signals Spectral Analysis

  • Autoři: Špetík, R., Vlček, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Digital Technologies 2004. Žilina: Žilinská univerzita, 2004. pp. 49-54. ISBN 80-8070-334-5.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    An alternative method for analysis of non-stationary signals is proposed. In the core of this method Zolotarev polynomials and functions derived from Zolotarev polynomials are used. Zolotarev polynomials can be considered as a generalisation of the Chebyshev polynomials. Basic principles and some experimental results of this method are compared to short time Fourier transformation as a representative of the most challenging competitor.

Audio Signal Segmentation Using Recursive Bayesian Change-point Detectors

Audio Signal Segmentation Using Recursive Bayesian Change-point Detectors

Bayesian and Monte Carlo Change-Point Detection

Bayesian and Monte Carlo Change-Point Detection

Bayesian Detection and Localization of Signal Changes

Beamforming Applied on EEG-EMG Coherence Computation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    We present a beamforming application on electroencephalogram (EEG) records with an aim to improve coherence between the EEG and an electromyogram (EMG) record of an iso- metrically contracted muscle. The EEG beamforming is shown to be superior to the currently used EEG spatial filtration, so called surface Laplacian. Moreover, the beamforming has demonstrated the ability to reveal EEG-EMG binding in 6-12Hz band, which has hitherto been observed rarely.

Beamforming EEG odhaluje EEG-EMG koherence v pásmu 6-12Hz

Comparison of PMC and MMSE Methods for Robust Speech Recognition

Comparison of Selected PMC Based Robust Speech Recognition Approaches

  • Autoři: Novotný, J., prof. Ing. Pavel Sovka, CSc., prof. Ing. Jan Uhlíř, CSc.,
  • Publikace: Proceedings of the Polish-Czech-Hungarian Workshop on Circuit Theory, Signal Processing and Telecommunications. Budapest: Budapest University of Technology and Economics, 2004. pp. 191-197. ISBN 96-34215-70-X.
  • Rok: 2004

Design of Feedforward ANC System

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: The Eleventh Internation Congress on Sound and Vibration. St.Petersburg: The International Institute of Acoustics and Vibration (IIAV), 2004. pp. 315-322. ISBN 5-7325-0816-3.
  • Rok: 2004

Design of Feedforward ANC Systems

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Inter-Noise 2004, Proceedings. Praha: Česká akustická společnost, 2004.
  • Rok: 2004

Effects of Muscle Contraction on Somatosensory Event-Related EEG Power and Coherence Changes

  • Autoři: Svoboda, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Neurophysiologie Clinique. 2004, 34(1), 62-73. ISSN 0987-7053.
  • Rok: 2004

Error Analysis of a Surface Potential in the Sphere-Shaped Forward EEG Problem

  • Autoři: Mlynář, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Digital Technologies 2004. Žilina: Žilinská univerzita, 2004. pp. 125-129. ISBN 80-8070-334-5.
  • Rok: 2004

Family of Recursive Bayesian Detectors and Applications for Signal Segmentation

Family of Recursive Bayesian Detectors and Applications for Signal Segmentation

Fractional Delay of EEG Signal

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: MATLAB 2004 - Sborník příspěvků 12. ročníku konference. Praha: VŠCHT, 2004. pp. 569-574. ISBN 80-7080-550-1.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution deals with the design of the fractional delay filters for EEG signals.

Increasing Robustness of Telephone Speech Recognition System Using Silence Model Adaptation Technique

MEG-EMG Coherence Enhancements

Model Order Selection Using Penalty Criteria and Bayesian Evidence

Reconstruction of Scalp and Brain from MR Images for Cortical Mapping of Electroencephalographic Potentials

  • Autoři: Mlynář, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2004. pp. 324-326. ISSN 1211-412X. ISBN 80-214-2633-0.
  • Rok: 2004

Recursive Bayesian Autoregressive Changepoint Detector for Sequential Signal Segmentation

Rozhraní mezi mozkem a počítačem

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A., Bočan, P.
  • Publikace: Česká hlava - Svět vědy. 2004, 2(5/2004), 18-20. ISSN 1214-0902.
  • Rok: 2004

Rozhraní mozek - počítač: výzkum na ČVUT FEl

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: 51. Neurofyziologický kongres. České Budějovice: Neurofyziologická laboratoř Nemocnice České Budějovice, 2004. pp. 60.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Náš příspěvek mapuje minulost, současnost a budoucnost výzkumu v oblasti brain-computer interface v laboratoři zpracování biologických signálů na katedře teorie obvodů elektrotechnické fakulty Českého vysokého učení technického v Praze.

Spectral Analysis of Non-Stationary Signals using Zolotarev Polynomials

  • Autoři: Špetík, R., Vlček, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: 7th COST 276 Workshop. Ankara: Middle East Technical University, 2004. pp. 6.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    An alternative method for analysis of non-stationary signals is proposed. In the core of this method Zolotarev polynomials and functions derived from Zolotarev polynomials are used. Zolotarev polynomials can be considered as a generalisation of the Chebyshev polynomials. Basic principles and some experimental results of this method are compared to short time Fourier transformation as a representative of the most challenging competitor.

Study and Application of Silence Model Adaptation for Use in Telephone Speech Recognition System

Towards Brain-Computer Interface: Analysis of EEG Accompanying Thumb and Little Finger Flexions

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Proceedings of the Polish-Czech-Hungarian Workshop on Circuit Theory, Signal Processing and Telecommunications. Budapest: Budapest University of Technology and Economics, 2004. pp. 106-111. ISBN 96-34215-70-X.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution presents the short-time spectra analysis of EEG signals accompanying brisk right index and little finger flexions. Analysed data were processed with the aim of brain computer interface (BCI) construction. Regular patterns of μ and event related desynchronization (ERD) as well as synchronization (ERS) were revealed with the help of the sign test. The characteristic parameters of these patterns were measured. Basic differences between both movements were confirmed.

Towards Brain-Computer Interface: Analysis of EEG Accompanying Thumb and Little Finger Flexions

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Applied Electronics 2004. Plzeň: Západočeská univerzita v Plzni, 2004. pp. 204-207. ISBN 80-7043-274-8.
  • Rok: 2004
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The contribution presents the short-time spectra analysis of EEG signals accompanying brisk right index and little finger flexions. Analysed data were processed with the aim of brain computer interface (BCI) construction. Regular patterns of μ and event related desynchronization (ERD) as well as synchronization (ERS) were revealed with the help of the sign test. The characteristic parameters of these patterns were measured. Basic differences between both movements were confirmed.

A Family of Coherence-Based Multi-Microphone Speech Enhancement Systems

Analysis and Simulation of Frost's Beamformer

Cross-Language Experiment

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Markov model theory is already older more than half a century but to the broader engineering awareness it has entered not long time ago with the application oriented articles. The intense grow of computational power and broad availability of powerful computers in the last two decades allowed implementation and usage of voice classification systems based on Hidden Markov Models (HMM) in real life. In the text below we would like to briefly make the reader familiar with the fundamentals of HMM and to present one - not so typical - example of application of HMM usage. In the continuation of our contribution we will deal with the application of HMM for the biosignal classification - to be more precise, to the human simple movements using EEG classification.

Dvouvstupový implementačně efektivní systém zvýrazňování řeči

EEG Signal Classification: Introduction to the Problem

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Radioengineering. 2003, 12(3), 51-55. ISSN 1210-2512.
  • Rok: 2003
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Markov model theory is already older more than half a century but to the broader engineering awareness it has entered not long time ago with the application oriented articles. The intense grow of computational power and broad availability of powerful computers in the last two decades allowed implementation and usage of voice classification systems based on Hidden Markov Models (HMM) in real life. In the text below we would like to briefly make the reader familiar with the fundamentals of HMM and to present one - not so typical - example of application of HMM usage. In the continuation of our contribution we will deal with the application of HMM for the biosignal classification - to be more precise, to the human simple movements using EEG classification.

Parallel Model Combination Evaluation od SpeechDat(E) Telephone Database

  • Autoři: Novotný, J., prof. Ing. Pavel Sovka, CSc., prof. Ing. Jan Uhlíř, CSc.,
  • Publikace: Proceedings of the Polish-Hungarian-Czech Workshop on Circuit Theory, Signal Processing, and Applications. Praha: ČVUT FEL, Katedra teorie obvodů, 2003. pp. 53-58. ISBN 80-01-02825-9.
  • Rok: 2003

Počítač, který dokáže číst myšlenky

  • Autoři: prof. Ing. Pavel Sovka, CSc., Králík, J., Šťastný, J., Stančák, A.
  • Publikace: ComputerWorld. 2003, 14(2), 23-24. ISSN 1210-9924.
  • Rok: 2003

Analysis and Simulation of Frost Beamformer

  • Autoři: Štrupl, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002. p. 129-131. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002

Číslicové zpracování signálů

EEG Coherence during Paintful Cutaneous Stimulation

  • Autoři: Svoboda, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Proceedings of Joint Session with 38th Interdisciplinary Conference on Clinical and Experimental Study of Higher Nervous Functions - Electrophysiology of CNS. Olomouc: Společnost pro psychosomatické integrace, 2002. pp. 50-51. ISBN 80-86358-04-6.
  • Rok: 2002

EEG Signal Classification and Segmentation by Means of Hidden Markov Models

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002. pp. 415-417. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002
  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    The article describes the optimization of off-line classification of simple movements using a system based on Hidden Markov Models (HMM). Movements of the proximal arm (shoulder) and distal arm (finger) were classified using scalp EEG signals and reached classification score was typically about 80%. The classification of movement-related EEG data based on HMM yielded higher recognition scores than previously reported classification scores based on artificial neural networks (NaN). Next we tested the stability of the training algorithm and tried to utilize the trained models for the automatic segmentation of EEG realizations. We also tested the classification of brisk right index finger extensions and flexions.

Head Reconstruction from Magnetic Resonance Images and Forward Problem in EEG Solution via Multiple Technique

  • Autoři: Králík, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Proceedings of Joint Session with 38th Interdisciplinary Conference on Clinical and Experimental Study of Higher Nervous Functions - Electrophysiology of CNS. Olomouc: Společnost pro psychosomatické integrace, 2002. p. 40-41. ISBN 80-86358-04-6.
  • Rok: 2002

Head Reconstruction from Magnetic Resonance Images and GMT Solution of the Forward Problem in EEG

  • Autoři: Králík, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002. pp. 323-325. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002

Intra- and Inter-hemispheric Coupling of Electroencephalographic 8-13 Hz Rhythm in Humans and Force of Static Finger Extension

  • Autoři: Svoboda, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Neuroscience Letters. 2002,(334), 191-195. ISSN 0304-3940.
  • Rok: 2002

Study of Detection of Changes in EEG Signal

Using of Bayesian Detectors for EEG Signal Changes

Úvod do bayesovské klasifikace dat

Vícevstupové metody redukce šumů v řeči

Active Hearing Protection

  • Autoři: Bašta, I., Píč, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Poster 2001. Praha: České vysoké učení technické v Praze, Fakulta elektrotechnická, 2001. pp. E2.
  • Rok: 2001

Active Hearing Protection - Realisation

  • Autoři: Píč, V., Bašta, I., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of Workshop 2001. Praha: České vysoké učení technické v Praze, 2001. pp. 396-397. ISBN 80-01-02335-4.
  • Rok: 2001

Blind Location of Phonetic Boundaries

EEG signal classification

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Proceeding of the 23rd Annual International Conference of the IEEE,Engineering in Medicine and Biology Society. New York: IEEE Press, 2001. pp. 132. ISBN 0-7803-7211-5.
  • Rok: 2001

EEG Signal Classification

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Proceeding of the 23rd Annual International Conference of the IEEE,Engineering in Medicine and Biology Society. New York: IEEE Press, 2001. pp. 642-644. ISBN 0-7803-7211-5.
  • Rok: 2001

Estimation of Boundaries between Speech Units Using Bayesian Changepoint Detectors

Head Reconstruction and Localization of Brain Activity Using Bayesian Evidence

Implementation Effective One-Channel Noise Reduction System

  • Autoři: Tihelka, J., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Eurospeech 2001 Scandinavia. Aalborg: Aalborg University, 2001. p. 1863-1866. ISBN 87-90834-09-7.
  • Rok: 2001

Neural Network Implementation for Blind Signal Separation

  • Autoři: Zeman, T., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Polish-Czech-Hungarian Workshop on Circuit Theory, Signal Processing and Telecommunications Network. Budapest: Technical University, 2001. pp. 152-156.
  • Rok: 2001

Study of Segmental Fast ICA

  • Autoři: prof. Ing. Pavel Sovka, CSc., Zeman, T.
  • Publikace: Proceedings of the 17th International Congress on Acoustics. Rome: 17th International Congress on Acoustics, 2001. pp. 150-151.
  • Rok: 2001

Active Hearing Protector - Simulation and Realisation

  • Autoři: Bašta, I., Píč, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of 7th International Congress on Sound and Vibration. Auburn: International Institute of Acoustics and Vibration, 2000. pp. 313-318.
  • Rok: 2000

Adaptivní tlumení hluku ve sluchátkách

  • Autoři: Píč, V., Bašta, I., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Sborník 60. akustického semináře & 36. akustické konference. Praha: České vysoké učení technické v Praze, 2000. pp. 137-140. ISBN 80-01-02179-3.
  • Rok: 2000

Distal-proximal motion classification from EEG signal

  • Autoři: Šťastný, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Biosignal 2000. Brno: VUTIUM Press, 2000. pp. 112-114. ISBN 80-214-1610-6.
  • Rok: 2000

Distal-proximal Motion Classification from EEG Signals

  • Autoři: Šťastný, J., Stančák, A., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2000. pp. 112-114. ISBN 80-214-1610-6.
  • Rok: 2000

Echo Cancellation for Car

  • Autoři: Brodský, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Speech Processing. Praha: AV ČR, Ústav radiotechniky a elektroniky, 2000. pp. 32-33. ISBN 80-86269-05-1.
  • Rok: 2000

Echo Cancellation I: Algorithms Simulation

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Contribution summarizes results of echo cancellation with frequency -domain algorithms.

Echo Cancellation II: Double Talk Detection and Environmental Noise Influence

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Contribution summarizes results of echo cancellation with frequency -domain algorithms under the simultaneous activity of both speakers: double-talk. A proper speech activity detector is suggested and optimized.

Echo Cancellation of Hands-Free Telephony in the Car

  • Autoři: Brodský, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Radioelektronika 2000 - Conference Proceedings. Bratislava: STU v Bratislave, FEI, 2000. pp. 89-91. ISBN 80-227-1389-9.
  • Rok: 2000

Mapping of EEG Signal on the Head Surface

  • Autoři: Králík, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Poster 2000. Praha: České vysoké učení technické v Praze, Fakulta elektrotechnická, 2000. pp. NS14.
  • Rok: 2000

Mapping of EEG Signals on the Head Surface

  • Autoři: Králík, J., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2000. pp. 267-268. ISBN 80-214-1610-6.
  • Rok: 2000

Respiratory Oscillations in the Time-varying EEG-EMG Coherence during Maintained Finger

  • Autoři: Stančák, A., prof. Ing. Pavel Sovka, CSc., Hloušková, B.
  • Publikace: Brain Topography. 2000, 12(4), 295. ISSN 0896-0267.
  • Rok: 2000

Simulation of ANC System in Headset Driven by Various Types of Adaptive Algorithms

  • Autoři: Píč, V., Bašta, I., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of Workshop 2000. Praha: České vysoké učení technické v Praze, 2000. pp. 198. ISBN 80-01-02229-3.
  • Rok: 2000

Application of Bayesian Detector in Speech Processing

ASR with Noisy Speech Pre-processing and Phoneme Model Re-estimation

CAR2 - Czech Database of Car Speech

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper presents new Czech language twochannel (stereo) speech database recorded in car environment. The created database was designed for experiments with speech enhancement for communication purposes and for the study and the design of a robust speech recognition systems. Tools for automated phoneme labelling based on Baum-Welch reestimation were realised. The noise analysis of the car background environment was done.

Combined Noise Suppresioon System for Monaural Cochlear Implants

Czech Language Database of Car Speech and Environmental Noise

Influence of Parameter Estimation in Kalman Filtering of Speech Signals

Methods and Means for Limiting the Impact of Unwanted Acoustic Fields Measurement of Noise Penetration into Headset for Use in ANC

  • Autoři: Píč, V., Bašta, I., prof. Ing. Pavel Sovka, CSc., Daněk, R.
  • Publikace: Proceedings of Workshop 99. Praha: České vysoké učení technické v Praze, 1999. pp. 16.
  • Rok: 1999

Modified Spectral Subtraction Based Speech Enhancement

Notes on Bayesian Detector Use

Phoneme Model Based ASR of Words in Car Environment

Real-Time Fixed-Point DSP-Implementation of Spectral Substraction Algorithm for Speech Enhancement in Noisy Environment

Simulace systému aktivní protihlukové ochrany ve sluchátku

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc., Bašta, I.
  • Publikace: Sborník 58. akustického semináře. Praha: České vysoké učení technické v Praze, 1999. pp. 46-49. ISBN 80-01-01988-8.
  • Rok: 1999

Simulation of ANC System Driven by Various Types of Adaptive Algorithms

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc., Bašta, I.
  • Publikace: Proceedings of the Matlab DSP Conference 1999. Helsinki: COMSOL OY, 1999. pp. 112-115. ISBN 952-91-1375-7.
  • Rok: 1999

Study of Blind Deconvolution Based on Information Maximatisation

  • Autoři: prof. Ing. Pavel Sovka, CSc., Machuca, E.
  • Publikace: Proceedings of the 1999 IEEE International Workshop on Acoustic Echo and Noise Control. Murray Hill, NJ: Bell Laboratories, Lucent Technologies, 1999. pp. 176-179.
  • Rok: 1999

System for Echo Cancellation of Hands-free Telefones during Double Talk

  • Autoři: Brodský, M., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: 9th International Czech-Slovak Scientific Conference Radioelektronika 99. Brno: VUT v Brně, 1999. pp. 178-181. ISBN 80-214-1327-1.
  • Rok: 1999

The Simulation of Active Ear Defender with Different Algorithms

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc., Bašta, I.
  • Publikace: Poster 1999. Praha: České vysoké učení technické v Praze, Fakulta elektrotechnická, 1999. pp. 120-121.
  • Rok: 1999

The Study of Algorithms for Active Ear Defender

  • Autoři: Píč, V., prof. Ing. Pavel Sovka, CSc., Bašta, I.
  • Publikace: 9th International Czech-Slovak Scientific Conference Radioelektronika 99. Brno: VUT v Brně, 1999. pp. 174-177. ISBN 80-214-1327-1.
  • Rok: 1999

The Using of Bayesian Autoregressive Detector for Speech Segmentation

Two-Chanel Modified Coherence Noise Reduction on TMS320C31

Two-Chanel Modified Coherence Noise Reduction on TMS320C31

Vliv dechového rytmu na kortikomotoneuronální spojení při izometrické extenzi distální končetiny

  • Autoři: Hloušková, B., prof. Ing. Pavel Sovka, CSc., Stančák, A.
  • Publikace: Sborník z 35. konference Společnosti o experimentálním a klinickém výzkumu vyšších nervových funkcí. Olomouc: Lékařská fakulta Univerzity Palackého, 1999. pp. 25.
  • Rok: 1999

Adaptive Filtering for Vowel Description

Adaptive Formant Tracking and Classification of Czech Vowels

Czech Car Noisy Speech Database

Database of Car Speech, Analysis of Collected Data, Tools for Automated Labeling

Experimental Study of Speech Recognition in Noisy Environment

French-Czech Cross-Language Experiment

Speech Segmentation Using Bayesian Autoregressive Changepoint Detector

Study of 2-input 2-output Blind Signal Separation by Output Decorrelation

  • Autoři: Machuca, E., prof. Ing. Pavel Sovka, CSc., Davídek, V.
  • Publikace: Radioengineering. 1998, 7(4), 38-41. ISSN 1210-2512.
  • Rok: 1998

Study of Blind Signal Separation Using LMS Algorithm

  • Autoři: Machuca, E., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of the 8-th Czech-German Workshop on Speech Processing. Praha: AV ČR, Ústav radiotechniky a elektroniky, 1998. pp. 49-50. ISBN 80-86269-00-0.
  • Rok: 1998

Suppression of Acoustic Noise in Speech Using Kalman Filtering

Study of Speech Recognition in Noisy Environment

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    Achieving reliable performance in speech recognition is the car for mobile telephony application has been studying intensively for more than one decade. This paper addresses effects of mismatched conditions and their minimization with respect to the performance of speaker-independent isolated. word recognition in the car-noise environment without consideration of Lombard effect. This study is primarily intended to study the dependence of the recognition rate on the SNR of an input signal without and with noise enhancement preprocessing, especially to find conditions under that the modified spectral subtraction call be effectively used for the speech recognition ill a real non-stationary car-noise environment. If as the worst recognition rate. is admitted e.g. 80%, then tile use of spectral subtraction methods enables to use wider interval of input SNRs: for the trainig: made oil a clean speech this interval is (40,6)(1) dB; for the training made on a noisy speech this interval is (40,-2) dB; for tile training performed oil an enhanced speech this interval is (40,-8) dB. The third case gives tile widest interval of SNRs in which a recogniser (with tile final recognition rate ill tile interval of (100,80)%) can be used.

The Problems of Robust LPC Parametrization

Extended Spectral Subtraction

Extended Spectral Subtraction

Noise Cancellation Systems

Cepstral Speech/Pause Detectors

Číslicové zpracování signálů

Implementation of Spectral Subtraction

Neural Network Based Enhancement of Coherence Detector Speech-Pause

  • Autoři: Kříž, B., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Proceedings of the Polish-Czech-Hungarian Workshop on Circuit Theory and Applications. Warsaw: University of Technology, 1995. pp. 41-46.
  • Rok: 1995

Odhady časového zpoždění u vícekanálových signílů řeči

Real-Time Implementation of Spectral Subtraction Algorithm for Suppression of Acoustic Noise Speech

  • Autoři: Davídek, V., prof. Ing. Pavel Sovka, CSc., Šika, J.
  • Publikace: Eurospeech 95 Proceedings. Berlin: ESCA, 1995. pp. 141-144. ISSN 1018-4074.
  • Rok: 1995

Speech Detection in the Real Car Environment

Speech/Pause Detection for Real-Time Implementation of Spectral Subtraction Algorithm

Sprachpausendetektion zur realisierung von spektralsubtraktionsverfahren

  • Autoři: Davídek, V., prof. Ing. Pavel Sovka, CSc.,
  • Publikace: Elektronische Sprach Signalverarbeitun. Dresden: TU Dresden, 1995. pp. 14-20. ISSN 0940-6832.
  • Rok: 1995

Studie koherenčního detektoru řeč/pauza

The Effects of Respiratory Frequency on Noninvasive Parameters of Baroreceptor-Cardiac Reflex Sensitivity

The Study of Speech/Pause Detectors for Speech Enhancement Methods

University Courses in DSP

Speech Recognition Systems

The Study of Adaptive Two Microphone Methods for Noise Suppression in a Car Noisy Environment

The Study of Two Channel Methods for Noise Suppression

Electroencephalographic Correlates of Paced Breathing

  • Autoři: Stančák, A., Pfeffer, D., Hrudová, L., prof. Ing. Pavel Sovka, CSc., Dostálek, C.
  • Publikace: Neuroreport. 1993, 4(6), 723-726. ISSN 0959-4965.
  • Rok: 1993

Noise Cancellation Methods

Noise Suppression System for a Car

Noise Suppression System for Speech Degraded in Running Car

Non-musical Tone Spectral Subtraction

One Channel Suppression System for a Car

Seminars Focused on Spatial Filtering of Acoustic Signals

  • Pracoviště: Katedra teorie obvodů
  • Anotace:
    This paper presents two seminars focused on spatial filtering of acoustic signals. This task is a subset of multichannel signal processing and can be applied for speech enhancement in noisy environment. The seminars are suitable for students of master study program who are interested in digital signal processing. The objective of the seminars is to provide a brief and intelligible introduction to spatial filtering. The education is supplemented by practical tasks.

Za stránku zodpovídá: Ing. Mgr. Radovan Suk