SSVEP Based Correlation Analysis for Efficiency Enhancement of Brain-Computer Interfaces

Recent progress in technologies and medical science have given momentum to Brain-computer interfaces (BCIs) system in generalizing human-computer communication, specifically for the people struggling with neuro-muscular disabilities. Currently, BCI systems heavily rely upon efficient electroencephalograms (EEG) signal detection from the current available solutions due to their noninvasiveness. However, the existing BCI systems works efficiently for less number of user instructions. Therefore, in this study, Adaptive Correlation based Component Analysis (ACCA) method is proposed to establish a proper Brain-computer interfaces (BCIs) system between user and computer. Here, correlation analysis is performed to identify spatial probability for a particular frequency using Adaptive Correlation based Component Analysis (ACCA) method. Maximum correlation between EEG signal phases triggered by both single and multiple flickers are generated using ACCA method to efficiently extract SSVEP component. The performance of proposed ACCA method is analysed against traditional SSVEP extraction methods in terms of classification accuracy, time consumption and Information Transfer Rate (ITR) for different subjects.