Skip to main content
. Author manuscript; available in PMC: 2009 Jun 30.
Published in final edited form as: J Neurosci Methods. 2008 Apr 8;171(2):331–339. doi: 10.1016/j.jneumeth.2008.03.015

Fig. 4.

Fig. 4

Ratio of evoked response power to variance of evoked response (based on 200 repetitions of a bootstrap resampling), for various noise reduction schemes. A: selection of single channel with best SNR. B: average of 20 channels with best SNR. C: matched filter (coefficients as in Fig. 4, top, center). D: matched filter with coefficients weighted by SNR. E: first PCA component. F: best PCA component (greatest evoked-to-total power ratio). G: best ICA component (RUNICA). H: best ICA component (AMUSE). I: our method.