Title :
Achieve single-channel BCI: The choice of electrode position and signal characteristics
Author :
Guixian, Xiao ; Wei, Yan
Author_Institution :
Huangshi Inst. of Technol., Huangshi, China
Abstract :
Adopting the "Imitating-natural-reading(INR)" elicited N2_P3 as carriers between brain and computer. The feature selection and classification of the evoked potentials with wavelet transform and Back Propagation(BP) neural network was used to obtain brain-computer interface(BCI) control signal. Followed by using 20 single-channel electroen-cephalogram(EEG) to single-trial estimation, the results showed that channel P8 recognition rate is above 95% in each segment as 200 ms, 250 ms and 300 ms. thus proved the feasibility of single-channel BCI.
Keywords :
backpropagation; brain-computer interfaces; electroencephalography; medical signal processing; neural nets; wavelet transforms; back propagation neural network; brain-computer interface control signal; channel P8 recognition rate; electrode position; electroen-cephalogram; evoked potential classification; feature selection; imitating-natural-reading; signal characteristics; single-channel brain-computer interface; single-trial estimation; wavelet transform; Biological neural networks; Brain computer interfaces; Computer interfaces; Computer networks; Data mining; Electrodes; Electroencephalography; Feature extraction; Frequency; Neural networks; BP neural network; Brain computer interface; hnitating-natural-reading(INR); wavelet transform;
Conference_Titel :
Information Management and Engineering (ICIME), 2010 The 2nd IEEE International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-5263-7
Electronic_ISBN :
978-1-4244-5265-1
DOI :
10.1109/ICIME.2010.5477481
