Open
Description
Paper
Link: https://iopscience.iop.org/article/10.1088/1741-2552/ab2373/meta
Year: 2020
Summary
- to utilize the training data corresponding to not only the target stimulus but also the neighboring stimuli for learning and consequently better performance in learning
Learnings
- extended CCA (eCCA) and the ensemble TRCA (eTRCA) [8] methods equipped with learning from subject’s training data are representative in the midst and have demonstrated excellent performance in target recognition in SSVEP-based BCIs
- eCCA is proposed to compute the spatial filters using both the subject-specific reference signals and the sine–cosine reference signals, thus it performs much better than the other CCA-based methods. in which an additional offline calibration stage is conducted for each subject to generate individual data for learning
- TRCA is a spatial filtering approach for the time-locked near-infrared spectroscopy (NIRS) signal, which finds the spatial filter by maximizing the covariance of the time-locked signals across different trials
- to guarantee good performance, the number of calibration trials for each visual stimulus cannot be small for the eCCA method and the eTRCA method. become unreliable in the case of small training data so that the resulting spatial filters may not be accurate
- a subject’s SSVEP scalp topography varies not too much across different stimuli in an SSVEP-based BCI without cognitive tasks
- SSVEPs share a common pattern in spatial distribution across different stimulus frequencies and moreover the spatial filters for the SSVEPs are similar to each other
- frequencies within the range of [8 Hz, 15.8 Hz] should be similar to each other in an ideal case