Abstract
Various improved canonical correlation analysis (CCA) methods were developed for enhancing the performance of steady-state visual evoked potential (SSVEP)-based brain–computer interfaces (BCIs). Among them, the method combining CCA spatial filters from sine-cosine references and individual templates yielded the highest performance. However, the CCA aims to optimize the correlation between two sets of variables rather than the signal-to-noise ratio (SNR) of the SSVEP signals, upon which the performance of an SSVEP-based BCI depends mainly. In this paper, a novel algorithm, namely, maximum signal fraction analysis (MSFA), is proposed for creating spatial filters based on individual training data. The spatial filter for a specific stimulus target is estimated by directly maximizing the averaged SNR of the observed signals across multiple trials. An individual template is calculated for each target by averaging training signals of multiple trials. Target recognition is based on template matching between filtered template signals and a single-trial testing signal. Classification performance of the MSFA-based method was evaluated on a benchmark dataset and compared with that of the CCA-based methods. The results suggest that the proposed MSFA method significantly outperforms the CCA-based methods in terms of classification accuracy, and thus, it has great potential to be applied in the real-life SSVEP-based BCI systems.
Highlights
A brain-computer interface (BCI) is a non-muscular communication channel between the brain and an external device [1]
We proposed an maximum signal fraction analysis (MSFA)-based algorithm for target recognition in order to enhance the performance of steady-state visual evoked potentials (VEP) (SSVEP)-based BCIs
Paired t-tests showed that the r-squares achieved by EN MSFA and EN FBMSFA were significantly higher than those achieved by EX canonical correlation analysis (CCA) and EX FBCCA respectively. These results suggested that EN MSFA and EN FBMSFA could increase the distance of feature values between targets and non-targets compared to extended CCA (EX CCA) and EX FBCCA respectively
Summary
A brain-computer interface (BCI) is a non-muscular communication channel between the brain and an external device [1]. Such a channel can help people with severe motor disabilities communicate with the external world and improve their quality of lives. The steady-state VEP (SSVEP)-based BCIs have become popular due to the advantages of high. In the EEG signals recorded on the scalp, the useful signal for target recognition is generated from task-related brain activities, while the noise is derived from spontaneous brain activities and other task-unrelated sources, which include power line interference, eye movements and eye blinks, etc. The low SNR of EEG signals severely limits the practical applications of SSVEP-based BCIs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
