VMD based wavelet hybrid denoising and improved FBCCA algorithm: A new technique for wearable SSVEP recognit ion

Abstract The brain-computer interface (BCI) based on steady-state visual evoked potential (SSVEP) has gained increasing attention due to its non-invasiveness, low user training requirement, and high information transfer rate. In order to enhance the performance of SSVEP detection, we propose an improved joint model that combines variational mode decomposition (VMD) and wavelet fusion with filter bank canonical correlation analysis (FBCCA). The model is validated on awearable SSVEP-BCI dataset. By integrating decomposition and denoising techniques, the model employs DFA thresholding and applies deep filtering using discrete wavelet transform (DWT) and wavelet packet transform (WPT) to denoise the wearable EEGs. The filtered components are reconstructed along with the components that do not require filtering. Subsequently, identification is conducted using FBCCA, which employs a combination of filters to delineate frequency bands. Ultimately, the classification accuracy for dry and wet electrodes reaches 72.46% and 88.29% respectively. Compared to existing research results on this dataset, dry and wet electrodes show improvements of around 13% and10% respectively. This hybrid model provides a new perspective for wearable SSVEP recognition research and holds high potential for widespread application.