A tensorial approach to single trial recognition for Brain Computer Interface

Author

Phan, Anh Huy ; Cichocki, Andrzej ; Vu-Dinh, Thanh

Author_Institution

Brain Sci. Inst., Lab. for Adv. Brain Signal Process., RIKEN, Wako, Japan

fYear

2010

Firstpage

138

Lastpage

141

Abstract

In this paper, we propose a tensorial approach to single trial recognition in a EEG-based BCI system related to movement related potentials. In this approach input data are considered as tensors instead of more conventional vector or matrix representations. Feature extraction for multiway EEG spectral tensors is solved by using tensor (multi-array) decompositions. For the same EEG motor imagery dataset, the developed algorithms improved the accuracy of classification by almost 10% compared with the common spatial pattern (CSP) method.

Keywords

brain-computer interfaces; electroencephalography; feature extraction; matrix algebra; medical signal processing; pattern classification; tensors; EEG motor imagery dataset; EEG spectral tensor; EEG-based BCI system; brain computer interface; common spatial pattern; conventional vector; feature extraction; matrix representation; movement related potential; single trial recognition; tensorial approach; Accuracy; Electroencephalography; Feature extraction; Signal processing algorithms; Tensile stress; Training; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Advanced Technologies for Communications (ATC), 2010 International Conference on

Conference_Location

Ho Chi Minh City

Print_ISBN

978-1-4244-8875-9

Type

conf

DOI

10.1109/ATC.2010.5672697

Filename

5672697