Classification of Motor Imagery EEG Signals Based on Energy Entropy

Feature extraction and classification of EEG signals is core issues on EEG-based brain computer interface (BCI). Typically, such classification has been performed using signals from a set of selected EEG sensors. Because EEG sensor signals are mixtures of effective signals and noise, which has low signal-to-noise ratio, motor imagery EEG signals can be difficult to classification. Energy entropy was used to preprocess motor imagery EEG data, and the Fisher class separability criterion was used to extract features. Finally, classification of four types motor imagery EEG was performed by a method based on the statistical theory. An average of 85% classification accuracy of the six type combination and the three subjects was achieved. The results showed that motor imagery EEG signals can be extracted using energy entropy and that these extracted features offered clear advantages for classification.