Literature Review of Signal Processing Methods for SSVEP-Based BCI

A Brain–Computer Interface (BCI) is a communication system based on neural activity. Brain Computer Interfaces (BCIs) can provide severely impaired users with alternative communication paths, by means of interpretation of the user’s brain activity. Among BCI operating paradigms, SSVEP is largely exploited for its potentially high throughput and reliability. Steady State Visually Evoked Potentials (SSVEP) are signals that are natural responses to visual stimulation at specific frequencies.

The signal processing algorithm is of key importance to the performance of BCI systems, and therefore plays a significant role in practical applications. The aim of this paper is to suggest a robust algorithm to implement SSVEP-based BCI. Techniques employed for signal preprocessing, feature extraction, and feature classification are discussed. Selected methods for pre-processing are: band-pass filtering, notch filtering, MEC and MCC. These methods are compared on the basis of performance. The spectral representation of EEG signals is required as visually evoked potential affects the fundamental as well as harmonic frequencies of the particular frequency component. Thus, Fourier transform is used. The value comparison and classifier methods are used for feature classification.