Limits...
Time-frequency analysis of band-limited EEG with BMFLC and Kalman filter for BCI applications.

Wang Y, Veluvolu KC, Lee M - J Neuroeng Rehabil (2013)

Bottom Line: Time-Frequency analysis of electroencephalogram (EEG) during different mental tasks received significant attention.The proposed method is applied to BCI Competition data IV for ERD detection in comparison with existing methods.For ERD detection, BMFLC-KF outperforms STFT and BMFLC-KS in real-time applicability with low computational requirement.

View Article: PubMed Central - HTML - PubMed

Affiliation: College of IT Engineering, Kyungpook National University, 1370 Sanyuk-dong, Daegu, 702-701, South Korea. veluvolu@ee.knu.ac.kr.

ABSTRACT

Background: Time-Frequency analysis of electroencephalogram (EEG) during different mental tasks received significant attention. As EEG is non-stationary, time-frequency analysis is essential to analyze brain states during different mental tasks. Further, the time-frequency information of EEG signal can be used as a feature for classification in brain-computer interface (BCI) applications.

Methods: To accurately model the EEG, band-limited multiple Fourier linear combiner (BMFLC), a linear combination of truncated multiple Fourier series models is employed. A state-space model for BMFLC in combination with Kalman filter/smoother is developed to obtain accurate adaptive estimation. By virtue of construction, BMFLC with Kalman filter/smoother provides accurate time-frequency decomposition of the bandlimited signal.

Results: The proposed method is computationally fast and is suitable for real-time BCI applications. To evaluate the proposed algorithm, a comparison with short-time Fourier transform (STFT) and continuous wavelet transform (CWT) for both synthesized and real EEG data is performed in this paper. The proposed method is applied to BCI Competition data IV for ERD detection in comparison with existing methods.

Conclusions: Results show that the proposed algorithm can provide optimal time-frequency resolution as compared to STFT and CWT. For ERD detection, BMFLC-KF outperforms STFT and BMFLC-KS in real-time applicability with low computational requirement.

Show MeSH

Related in: MedlinePlus

Parameter tuning for Kalman filter. (a) Parameter q selection based on S1(t); (b) Parameter q selection based on S2(t); (c) Parameter R selection with S1(t) for fixed q; (d) Subject #1 C3(all trials); (e) Subject #1 C4(all trials).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4222486&req=5

Figure 4: Parameter tuning for Kalman filter. (a) Parameter q selection based on S1(t); (b) Parameter q selection based on S2(t); (c) Parameter R selection with S1(t) for fixed q; (d) Subject #1 C3(all trials); (e) Subject #1 C4(all trials).

Mentions: For implementation of BMFLC-KF/KS, the two parameters, the state noise covariance Q and measure noise covariance R should be properly tuned. In the following, several experiments are conducted for identification of parameters to achieve better accuracy. To start with, we assume that the state noise covariance is a diagonal matrix in the form of Q=q∗I. Then the parameter q is selected such that the root-mean-square (RMS) error is minimized. In [34], the measurement noise covariance R was estimated online by using the innovation process of the Kalman filter. Then an optimal value for q is selected to minimize the RMS error. Further, the selection of q is also performed for pre-fixed R. Experiments are first performed with synthetic signal S1 and S2 and the corresponding results are shown in Figure 4(a) and 4(b). It shows that when q>0, the RMS error is below 3% when R is estimated online and 1% for pre-fixed R. Based on the result of earlier experiment, we initialize q=0.05 and then optimize R. The results obtained for signal S1(t) are shown in Figure 4(c). As we vary the value of R, the RMS error remains constant. This further shows that the error performance of BMFLC-KF/KS is highly dependent on the selection of q.


Time-frequency analysis of band-limited EEG with BMFLC and Kalman filter for BCI applications.

Wang Y, Veluvolu KC, Lee M - J Neuroeng Rehabil (2013)

Parameter tuning for Kalman filter. (a) Parameter q selection based on S1(t); (b) Parameter q selection based on S2(t); (c) Parameter R selection with S1(t) for fixed q; (d) Subject #1 C3(all trials); (e) Subject #1 C4(all trials).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4222486&req=5

Figure 4: Parameter tuning for Kalman filter. (a) Parameter q selection based on S1(t); (b) Parameter q selection based on S2(t); (c) Parameter R selection with S1(t) for fixed q; (d) Subject #1 C3(all trials); (e) Subject #1 C4(all trials).
Mentions: For implementation of BMFLC-KF/KS, the two parameters, the state noise covariance Q and measure noise covariance R should be properly tuned. In the following, several experiments are conducted for identification of parameters to achieve better accuracy. To start with, we assume that the state noise covariance is a diagonal matrix in the form of Q=q∗I. Then the parameter q is selected such that the root-mean-square (RMS) error is minimized. In [34], the measurement noise covariance R was estimated online by using the innovation process of the Kalman filter. Then an optimal value for q is selected to minimize the RMS error. Further, the selection of q is also performed for pre-fixed R. Experiments are first performed with synthetic signal S1 and S2 and the corresponding results are shown in Figure 4(a) and 4(b). It shows that when q>0, the RMS error is below 3% when R is estimated online and 1% for pre-fixed R. Based on the result of earlier experiment, we initialize q=0.05 and then optimize R. The results obtained for signal S1(t) are shown in Figure 4(c). As we vary the value of R, the RMS error remains constant. This further shows that the error performance of BMFLC-KF/KS is highly dependent on the selection of q.

Bottom Line: Time-Frequency analysis of electroencephalogram (EEG) during different mental tasks received significant attention.The proposed method is applied to BCI Competition data IV for ERD detection in comparison with existing methods.For ERD detection, BMFLC-KF outperforms STFT and BMFLC-KS in real-time applicability with low computational requirement.

View Article: PubMed Central - HTML - PubMed

Affiliation: College of IT Engineering, Kyungpook National University, 1370 Sanyuk-dong, Daegu, 702-701, South Korea. veluvolu@ee.knu.ac.kr.

ABSTRACT

Background: Time-Frequency analysis of electroencephalogram (EEG) during different mental tasks received significant attention. As EEG is non-stationary, time-frequency analysis is essential to analyze brain states during different mental tasks. Further, the time-frequency information of EEG signal can be used as a feature for classification in brain-computer interface (BCI) applications.

Methods: To accurately model the EEG, band-limited multiple Fourier linear combiner (BMFLC), a linear combination of truncated multiple Fourier series models is employed. A state-space model for BMFLC in combination with Kalman filter/smoother is developed to obtain accurate adaptive estimation. By virtue of construction, BMFLC with Kalman filter/smoother provides accurate time-frequency decomposition of the bandlimited signal.

Results: The proposed method is computationally fast and is suitable for real-time BCI applications. To evaluate the proposed algorithm, a comparison with short-time Fourier transform (STFT) and continuous wavelet transform (CWT) for both synthesized and real EEG data is performed in this paper. The proposed method is applied to BCI Competition data IV for ERD detection in comparison with existing methods.

Conclusions: Results show that the proposed algorithm can provide optimal time-frequency resolution as compared to STFT and CWT. For ERD detection, BMFLC-KF outperforms STFT and BMFLC-KS in real-time applicability with low computational requirement.

Show MeSH
Related in: MedlinePlus