<![CDATA[The analysis of the feature’s uniqueness of the electroencephalograph (EEG) signal extracted by continuous wavelet transform (CWT) method against the five BCI wheelchair control indicators has been done. The usage of Friedman method as measuring the uniqueness level of EEG signal features as well as their significance is used in this research. The EEG signals from three subjects that sitting on a regular chair were recorded when they were performing mental commands as seem as controlling a wheelchair with five control indicators. The recorded signals are decomposed by CWT. The absolute mean (|µ|) and the deviation standard (σ) of the CWT decomposition results are used as feature. Then, the uniqueness of |µ| and σ features are analyzed using Friedman Method. Based on the experiment results, it is known that the proposed method is able to map features according to their uniqueness level. The experiment result shows that the highest uniqueness value of |µ| feature from three subjects are 400 (“forward – backward†indicators), 437 (“neutral – turn left†indicators), and 597 (“neutral – turn left†indicators) respectively. While the highest uniqueness value of σ feature from each subjects are 380, 419, and 568 respectively in the same indicator pairs as |µ| feature.]]>
Copyrights © 2018
