TY - JOUR
T1 - Efficient implementation and design of a new single-channel electrooculography-based human-machine interface system
AU - Wu, J. F.
AU - Ang, A. M.S.
AU - Tsui, K. M.
AU - Wu, H. C.
AU - Hung, Y. S.
AU - Hu, Y.
AU - Mak, J. N.F.
AU - Chan, S. C.
AU - Zhang, Z. G.
N1 - Publisher Copyright:
© 2004-2012 IEEE.
PY - 2015/2
Y1 - 2015/2
N2 - This brief introduces a new and practical human-machine interface (HMI) system based on single-channel electrooculography (EOG) signals. The proposed system uses a consumer wireless recording device to collect EOG and employs new encoding/decoding paradigms to convey users' intentions with EOG from eye movements including blinking and looking up. The simplicity and mobility of the system provides a comfortable and practical solution to HMI. Furthermore, to reduce the hardware complexity and power consumption of the signal processing modules of the EOG-based HMI system, a novel multiplierless implementation is developed, where all the algorithms involved, such as bandpass filtering, wavelet filtering, and support vector machine, can be realized using a limited number of adders and shifters only. Experimental results show that the proposed system offers a simple, practical, and yet reliable EOG-based HMI with low complexity and power consumption.
AB - This brief introduces a new and practical human-machine interface (HMI) system based on single-channel electrooculography (EOG) signals. The proposed system uses a consumer wireless recording device to collect EOG and employs new encoding/decoding paradigms to convey users' intentions with EOG from eye movements including blinking and looking up. The simplicity and mobility of the system provides a comfortable and practical solution to HMI. Furthermore, to reduce the hardware complexity and power consumption of the signal processing modules of the EOG-based HMI system, a novel multiplierless implementation is developed, where all the algorithms involved, such as bandpass filtering, wavelet filtering, and support vector machine, can be realized using a limited number of adders and shifters only. Experimental results show that the proposed system offers a simple, practical, and yet reliable EOG-based HMI with low complexity and power consumption.
KW - electrooculography
KW - human-machine interface
KW - multiplier-less implementation
KW - support vector machine
KW - wavelet
UR - http://www.scopus.com/inward/record.url?scp=84923083892&partnerID=8YFLogxK
U2 - 10.1109/TCSII.2014.2368617
DO - 10.1109/TCSII.2014.2368617
M3 - Article
AN - SCOPUS:84923083892
SN - 1549-7747
VL - 62
SP - 179
EP - 183
JO - IEEE Transactions on Circuits and Systems II: Express Briefs
JF - IEEE Transactions on Circuits and Systems II: Express Briefs
IS - 2
M1 - 6949633
ER -