Common Tensor Discriminant Analysis for human brainwave recognition accelerated by massive parallelism

Author

Gajdos, Petr ; Dohnalek, Pavel ; Bobrov, Pavel

Author_Institution

Dept. of Comput. Sci., VrB - Tech. Univ. of Ostrava, Ostrava, Czech Republic

fYear

2013

fDate

12-14 Aug. 2013

Firstpage

189

Lastpage

193

Abstract

In this paper, a massively parallel implementation of Common Tensor Discriminant Analysis is presented with applications to human brainwave pattern recognition. The implementation, accelerated by the NVIDIA Compute Unified Device Architecture technology, is shown to be 11.49x faster than the original MATLAB version. Before processing by the discriminant analysis, the data is segmented by a sliding window and converted into the time-frequency domain by the continuous wavelet transform.

Keywords

brain-computer interfaces; electroencephalography; medical signal processing; parallel architectures; pattern recognition; tensors; time-frequency analysis; wavelet transforms; MATLAB version; NVIDIA compute unified device architecture technology; common tensor discriminant analysis; continuous wavelet transform; human brainwave pattern recognition; human brainwave recognition; massive parallelism; massively parallel implementation; sliding window; time-frequency domain; Algebra; Europe; Graphics processing units; Libraries; MATLAB; Mesons; BCI; CTDA; parallelism; pattern matching; tensor;

fLanguage

English

Publisher

ieee

Conference_Titel

Nature and Biologically Inspired Computing (NaBIC), 2013 World Congress on

Conference_Location

Fargo, ND

Print_ISBN

978-1-4799-1414-2

Type

conf

DOI

10.1109/NaBIC.2013.6617860

Filename

6617860