| Presentation | 2011-06-30 Optimal Design of a Spatio-Temporal Filter Bank for EEG Signal Classification Hiroshi HIGASHI, Toshihisa TANAKA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose a method to design multiple FIR filters and the associated spatial weights associated for feature extraction in EEG signal classification for motor imagery based brain computer interface (MI-BCI). The spatial weights for electrodes called common spatial pattern (CSP) are known to be effective for feature extraction of MI-BCI. However, to achieve accurate classification in CSP, the frequency filter should be properly designed. To this end, several methods for designing the filter have been proposed. However, the existing methods cannot consider plural brain activities corresponding to different frequency bands and different spatial patterns such as activities of mu and beta rhythms. To address this problem, we design a set of filters that can successfully classify a learning datasets. To efficiently extract these brain activities, we design a cost function that has optimal parameters of spatial weights and filter coefficients. Moreover, by introducing a constraint of orthogonality between vectors of filter coefficients, the proposed method can design multiple FIR filters and the associated spatial weights. In the classification experiment of MI-BCI, the proposed method outperforms existing methods. Moreover, the experiment result suggests that the proposed method can automatically detect and extract the frequency bands that contain brain activities related to motor imagery. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | brain computer interface / common spatial pattern / finite impulse response filter / generalized eigenvalue problem |
| Paper # | CAS2011-15,VLD2011-22,SIP2011-44,MSS2011-15 |
| Date of Issue |
| Conference Information | |
| Committee | MSS |
|---|---|
| Conference Date | 2011/6/23(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Mathematical Systems Science and its applications(MSS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Optimal Design of a Spatio-Temporal Filter Bank for EEG Signal Classification |
| Sub Title (in English) | |
| Keyword(1) | brain computer interface |
| Keyword(2) | common spatial pattern |
| Keyword(3) | finite impulse response filter |
| Keyword(4) | generalized eigenvalue problem |
| 1st Author's Name | Hiroshi HIGASHI |
| 1st Author's Affiliation | Department of Electronic and Information Engineering, Tokyo University of Agriculture and Technology() |
| 2nd Author's Name | Toshihisa TANAKA |
| 2nd Author's Affiliation | Department of Electronic and Information Engineering, Tokyo University of Agriculture and Technology |
| Date | 2011-06-30 |
| Paper # | CAS2011-15,VLD2011-22,SIP2011-44,MSS2011-15 |
| Volume (vol) | vol.111 |
| Number (no) | 105 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |