Computational Intelligence and Neuroscience / 2016 / Article / Tab 3

Research Article

Time-Shift Correlation Algorithm for P300 Event Related Potential Brain-Computer Interface Implementation

Table 3

Performance comparison of the proposed approach and conventional CCA SVM.

MethodsClassification accuracy/illegal trial rates (%)
Subject 1Subject 2Subject 3
Proposed BCICCA SVMProposed BCICCA SVMProposed BCICCA SVM

Average 1 epoch90.0/30.087.92/40.084.81/21.083.52/46.075.00/28.074.00/50.0
Data transfer rate (bit/min.)49.4138.4943.8228.1724.4516.09

Average 2 epochs95.35/14.096.22/15.5088.64/12.091.44/18.5083.72/14.087.1/19.0
Data transfer rate (bit/min.)38.7739.6829.1830.7222.6525.01

Average 3 epochs100.0/3.03100.0/7.096.67/9.0997.78/9.2590.0/9.0990.48/12.0
Data transfer rate (bit/min.)36.5735.0728.7830.3021.1820.95

Accuracy rate = correct/(correct + incorrect).
Illegal trial rate = illegal/total trials.