Name
Title | ||
|---|---|---|
Frequency Recognition Based On Optimized Power Spectral Density Analysis For Sssep-Based Bcis |
Abstract | ||
|---|---|---|
This paper presents a novel three-class Steady-state somatosensory evoked potentials (SSSEPs)-based BCI paradigm for target identification. The improved stimulation pattern accompanied by rhythmic pulses (i.e., 'Tic-Tic-Toc') was provided to the arm, waist and thigh of three healthy subjects by tactile tactors vibrating at different frequencies. The subjects were asked to selectively focus their attention on flutter sensation derived from stimulation of one site among three. To improve classification accuracy, we added a posterior processing after the power spectral density (PSD) analysis to reduce the inter-frequency variation and named the new method D-PSD. Experimental results for three subjects suggested that D-PSD method and the 'Tic-Tic-Toc' pattern increased accuracy (60.41%) compared with traditional PSD (42.56%) and absence of 'Tic-Tic-Toc' pattern (54.72%), respectively. These indicate that our BCI paradigm deserves being further explored as an alternative to SSVEP-based BCI applications. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1007/978-3-319-67777-4_7 | INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING, ISCIDE 2017 |
Keywords | Field | DocType |
Brain-computer interface, Electroencephalogram, Steady-state somatosensory evoked potentials, Power spectral density | Pattern recognition,Computer science,Brain–computer interface,Somatosensory evoked potential,Flutter,Speech recognition,Spectral density,Artificial intelligence,Rhythm,Sensation | Conference |
Volume | ISSN | Citations |
10559 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 5 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xing Han | 1 | 0 | 3.38 |
| Yadong Liu | 2 | 105 | 14.04 |
| Yang Yu | 3 | 38 | 5.71 |
| Zongtan Zhou | 4 | 412 | 33.89 |