Abstract— Brain-computer interface (BCI) is a hardware and software communication system that enables humans to interact with their surrounding without the involvement of peripheral nerves and muscles by using control signals generated from electroencephalographic activity. In this paper, we report on results of developing motor imagery feature extraction method for BCI. The wavelet coefficients were used to extract the features from the motor imagery EEG and the Bayes Net, SVM and RBFN were utilized to classify the pattern of left, right hand movement and forward imagery. The performance was tested using dataset from BCI competition III and satisfactory results are obtained with accuracy rate as high as 99.0674%.
Index Terms— Brain computer interface, EEG, wavelet transform, bayes net, SVM, and RBFN.
The authors are with the Universiti Teknologi PETRONAS, Department of Electrical & Electronic Engineering, Center of Intelligent Signal & Imaging Research (CISIR) 31750 Tronoh, Malaysia (e-mail: eltaf55@yahool.com, mzuki_yusoff@petronas.com.my, nidalkamel@petronas.com.my, aamir_saeed@petronas.com.my).
[PDF]
Cite: Eltaf Abdalsalam Mohamed, Mohd Zuki B. Yusoff, Nidal Kamel Selman, and Aamir Saeed Malik, " Enhancing EEG Signals in Brain Computer Interface Using Wavelet Transform," International Journal of Information and Electronics Engineering vol. 4, no. 3, pp. 234-238, 2014.
