Universiti Teknologi Malaysia Institutional Repository

Spectral Estimation of Non-Stationary EEG using Particle Filtering with Application to Event-Related Desynchronization (ERD)

Ting, Chee-Ming and Salleh, Sh-Hussain and M. Zainuddin, Zaitul and Bahar, Arifah (2010) Spectral Estimation of Non-Stationary EEG using Particle Filtering with Application to Event-Related Desynchronization (ERD). IEEE Trans. On Biomedical Engineering, 58 (2). 321-331 . ISSN 0018-9294

[img] PDF
68Kb

Official URL: http://dx.doi.org/10.1109/TBME.2010.2088396

Abstract

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.

Item Type:Article
Uncontrolled Keywords:event-related desynchronization (ERD), particle filters (PF), time-varying autoregressive (TVAR) models
Subjects:Q Science > Q Science (General)
Divisions:?? FBSK ??
ID Code:25989
Deposited By: Liza Porijo
Deposited On:21 Jun 2012 07:33
Last Modified:21 Jun 2012 07:33

Repository Staff Only: item control page