Spectral Estimation of Nonstationary EEG Using Particle Filtering With Application to Event-Related Desynchronization (ERD)

This paper proposes non-Gaussian models for parametric spectral estimation with application to event-related desynchronization (ERD) estimation of nonstationary EEG. Existing approaches for time-varying spectral estimation use time-varying autoregressive (TVAR) state-space models with Gaussian state noise. The parameter estimation is solved by a conventional Kalman filtering. This study uses non-Gaussian state noise to model autoregressive (AR) parameter variation with estimation by a Monte Carlo particle filter (PF). Use of non-Gaussian noise such as heavy-tailed distribution is motivated by its ability to track abrupt and smooth AR parameter changes, which are inadequately modeled by Gaussian models. Thus, more accurate spectral estimates and better ERD tracking can be obtained. This study further proposes a non-Gaussian state space formulation of time-varying autoregressive moving average (TVARMA) models to improve the spectral estimation. Simulation on TVAR process with abrupt parameter variation shows superior tracking performance of non-Gaussian models. Evaluation on motor-imagery EEG data shows that the non-Gaussian models provide more accurate detection of abrupt changes in alpha rhythm ERD. Among the proposed non-Gaussian models, TVARMA shows better spectral representations while maintaining reasonable good ERD tracking performance.

[1]  Simon J. Godsill,et al.  Particle methods for Bayesian modeling and enhancement of speech signals , 2002, IEEE Trans. Speech Audio Process..

[2]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[3]  P. Moral.  Nonlinear filtering : Interacting particle resolution , 1997 .

[4]  H. Sorenson,et al.  Nonlinear Bayesian estimation using Gaussian sum approximations , 1972 .

[5]  Aysin Ertüzün,et al.  Modeling non-Gaussian time-varying vector autoregressive processes by particle filtering , 2010, Multidimens. Syst. Signal Process..

[6]  Tomoyuki Higuchi,et al.  Self-Organizing Time Series Model , 2001, Sequential Monte Carlo Methods in Practice.

[7]  Simon J. Godsill,et al.  On sequential Monte Carlo sampling methods for Bayesian filtering , 2000, Stat. Comput..

[8]  Braham Barkat,et al.  A high-resolution quadratic time-frequency distribution for multicomponent signals analysis , 2001, IEEE Trans. Signal Process..

[9]  G. Kitagawa.  A self-organizing state-space model , 1998 .

[10]  A. Doucet,et al.  Monte Carlo Smoothing for Nonlinear Time Series , 2004, Journal of the American Statistical Association.

[11]  C. Braun,et al.  Adaptive AR modeling of nonstationary time series by means of Kalman filtering , 1998, IEEE Transactions on Biomedical Engineering.

[12]  S. Tseng,et al.  Evaluation of parametric methods in EEG signal analysis. , 1995, Medical engineering & physics.

[13]  M. S. Woolfson,et al.  Study of cardiac arrhythmia using the Kalman filter , 1991, Medical and Biological Engineering and Computing.

[14]  Mika P. Tarvainen,et al.  Estimation of nonstationary EEG with Kalman smoother approach: an application to event-related synchronization (ERS) , 2004, IEEE Transactions on Biomedical Engineering.

[15]  William D. Penny,et al.  Bayesian nonstationary autoregressive models for biomedical signal analysis , 2002, IEEE Transactions on Biomedical Engineering.

[16]  G Pfurtscheller,et al.  Adaptive Autoregressive Modeling used for Single-trial EEG Classification - Verwendung eines Adaptiven Autoregressiven Modells für die Klassifikation von Einzeltrial-EEG-Daten , 1997, Biomedizinische Technik. Biomedical engineering.

[17]  Simon J. Godsill,et al.  An Overview of Existing Methods and Recent Advances in Sequential Monte Carlo , 2007, Proceedings of the IEEE.

[18]  Saeid Sanei,et al.  The application of particle filters in single trial event-related potential estimation , 2009, Physiological measurement.

[19]  Klaus-Robert Müller,et al.  The BCI competition 2003: progress and perspectives in detection and discrimination of EEG single trials , 2004, IEEE Transactions on Biomedical Engineering.

[20]  Nando de Freitas,et al.  Sequential Monte Carlo Methods in Practice , 2001, Statistics for Engineering and Information Science.

[21]  Etienne Perret,et al.  Sequential Parameter Estimation of Time-Varying Non-Gaussian Autoregressive Processes , 2002, EURASIP J. Adv. Signal Process..

[22]  Neil J. Gordon,et al.  Editors: Sequential Monte Carlo Methods in Practice , 2001 .

[23]  G. Kitagawa.  Non-Gaussian State—Space Modeling of Nonstationary Time Series , 1987 .

[24]  A. Doucet,et al.  A Tutorial on Particle Filtering and Smoothing: Fifteen years later , 2008 .

[25]  Mohammad Emtiyaz Khan,et al.  An Expectation-maximization Algorithm Based Kalman Smoother Approach for Event-related Desynchronization (erd) Estimation from Eeg , 2022 .

[26]  Roberto Hornero,et al.  Adaptive modeling and spectral estimation of nonstationary biomedical signals based on Kalman filtering , 2005, IEEE Transactions on Biomedical Engineering.

[27]  F. L. D. Silva,et al.  Event-Related Desynchronization , 1999 .

[28]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[29]  G. Kitagawa.  Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear State Space Models , 1996 .

[30]  Michael A. West,et al.  Combined Parameter and State Estimation in Simulation-Based Filtering , 2001, Sequential Monte Carlo Methods in Practice.

[31]  M. Tarvainen,et al.  Time-varying analysis of heart rate variability signals with a Kalman smoother algorithm , 2006, Physiological measurement.

[32]  Ercan E. Kuruo,et al.  Modeling non-Gaussian time-varying vector autoregressive processes by particle filtering , 2009 .

[33]  Arnaud Doucet,et al.  An overview of sequential Monte Carlo methods for parameter estimation in general state-space models , 2009 .