전자공학회논문지 (Journal of the Institute of Electronics and Information Engineers)

  • 제52권10호
  • /
  • Pages.139-147
  • /
  • 2015
  • /
  • 2287-5026(pISSN)
  • /
  • 2288-159X(eISSN)
대한전자공학회 (The Institute of Electronics and Information Engineers)
권호 표지
DOI QR코드

DOI QR Code

CCA와 PSDA를 결합한 SSVEP 기반 BCI 시스템의 주파수 인식 기법

Frequency Recognition in SSVEP-based BCI systems With a Combination of CCA and PSDA

  • Lee, Ju-Yeong (Dept. of Electronics Engineering, Pusan National University) ;
  • Lee, Yu-Ri (Dept. of Electronics Engineering, Pusan National University) ;
  • Kim, Hyoung-Nam (Dept. of Electronics Engineering, Pusan National University)
  • 투고 : 2015.06.08
  • 심사 : 2015.10.01
  • 발행 : 2015.10.25
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Steady state visual evoked potential (SSVEP)는 뇌파의 종류 중 하나로서 다른 뇌파에 비해 훈련 시간이 짧고, 비교적 높은 신호대잡음비 (signal-to-noise ratio)와 높은 정보전달량 (information transfer rate)을 가지고 있어서 최근에 뇌-컴퓨터 접속 장치 (brain-computer interface; BCI)에 많이 사용되고 있다. SSVEP 신호를 분석하는 기존 기법에는 전력 스펙트럼 밀도 분석 (power spectral density analysis; PSDA)과 정준상관분석 (canonical correlation analysis; CCA)이 있다. 그러나 PSDA는 단일 전극만을 사용하기 때문에 잡음에 취약한 단점이 있고, CCA는 PSDA보다 높은 정확도를 가지지만 사인-코사인을 기준 신호로 가지므로 짧은 시간 윈도우 길이를 가질 경우 실시간 BCI 시스템에 적용되기 어렵다. 따라서 본 논문에서는 기존의 기법들의 한계점을 보완하기 위해 CCA의 결과로 얻을 수 있는 정준변수 간의 전력차이를 이용하는 CCA와 PSDA를 결합한 기법을 제안한다. 실험 결과를 통해, SSVEP 기반 BCI 시스템이 짧은 시간 윈도우 길이를 가질 때 제안된 기법이 기존의 CCA 기법에 비해 더욱 높은 주파수 인식 정확도를 가짐을 보여준다.

Steady state visual evoked potential (SSVEP) has been actively studied because of its short training time, relatively higher signal-to-noise ratio, and higher information transfer rate. There are two popular analysis methods for SSVEP signals: power spectral density analysis (PSDA) and canonical correlation analysis (CCA). However, the PSDA is known to be vulnerable to noise due to the use of a single channel. Although conventional CCA is more accurate than PSDA, it may not be appropriate for the real-time SSVEP-based BCI system when it has short time window length because it uses sinusoidal signals as references. Therefore, the two methods are not efficient for the real-time BCI system that requires a short TW and a high recognition accuracy. To overcome this limitation of the conventional methods, this paper proposes a frequency recognition method with a combination of CCA and PSDA using the difference between powers of canonical variables obtained from the results of CCA. Experimental results show that the performance of the combination of CCA and PSDA is better than that of CCA for the case of a short TW.

키워드

참고문헌

  1. Chang-Jin Kong. "Problem Analysis & Directions for Improvement of Neuroethics in Brain-Computer Interface." KOREA INFORMATION SCIENCE SOCIETY Korea computer congress 2009, Vol. 36, no. 1, pp. 376-381, 2009.
  2. R. Srinivasan, "Methods to improve the spatial resolution of EEG," International Journal of Bioelectromagnetism, Vol. 1, no. 3, pp. 102-111, 1999.
  3. Kab-Mun Cha and Hyun-Chool Shin, "Brain-Machin Interface Using P300 Brain Wave." IEIE, Vol. 47, no. 5, pp. 18-23, Sep. 2010.
  4. Lin, Zhonglin, et al., "Frequency recognition based on canonical correlation analysis for SSVEP-based BCIs." Biomedical Engineering, IEEE Transactions, Vol. 53, no. 12, pp. 2610-2614, December 2006. https://doi.org/10.1109/TBME.2006.886577
  5. Muller-Putz, Gernot R., et al., "Steady-state visual evoked potential (SSVEP)-based communication: impact of harmonic frequency components." Journal of neural engineering, Vol. 2, pp. 123-130, October 2005. https://doi.org/10.1088/1741-2560/2/4/008
  6. Bin, Guangyu, et al., "An online multi-channel SSVEP-based brain-computer interface using a canonical correlation analysis method." Journal of neural engineering, Vol. 6, June 2009.
  7. Li, Yun, et al., "Analysis of phase coding SSVEP based on canonical correlation analysis (CCA)." Neural Engineering (NER), 2011 5th International IEEE/EMBS Conference on. pp. 368-371, IEEE, 2011.
  8. Pan, Jie, et al., "Enhancing the classification accuracy of steady-state visual evoked potential-based brain-computer interfaces using phase constrained canonical correlation analysis." Journal of neural engineering, Vol. 8, 2011.
  9. Zhang, Y. U., et al., "Frequency recognition in SSVEP-based BCI using multiset canonical correlation analysis." International journal of neural systems, Vol. 24, no. 2, January 2014.
  10. Zhang, Yu, et al., "SSVEP recognition using common feature analysis in brain-computer interface." Journal of neuroscience methods, Vol. 244, pp. 8-15, 2015. https://doi.org/10.1016/j.jneumeth.2014.03.012
  11. da Cruz, Janir Nuno, et al., "Adaptive time-window length based on online performance measurement in SSVEP-based BCIs." Neurocomputing, Vol. 149, pp. 93-99, February 2015. https://doi.org/10.1016/j.neucom.2014.01.062
  12. Wei, Qingguo, Meixia Xiao, and Zongwu Lu., "A comparative study of canonical correlation analysis and power spectral density analysis for SSVEP detection." Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2011 International Conference on. Vol. 2. IEEE, 2011.
  13. Hotelling, Harold, "Relations between two sets of variates." Biometrika, Vol. 28, pp. 321-377, 1936. https://doi.org/10.1093/biomet/28.3-4.321
  14. Zhang, Yangsong, et al., "Multivariate synchronization index for frequency recognition of SSVEP-based brain-computer interface." Journal of neuroscience methods, Vol. 221, pp. 32-40, 2014. https://doi.org/10.1016/j.jneumeth.2013.07.018
  15. Borga, Magnus, "Canonical correlation: a tutorial." On line tutorial http://people.imt.liu.se/magnus/cca 4, 2001.