• Priyanka Chaturvedi
• Rajesh Parasar

Brain Computer Interface establishes a communication channel between human brain and external world. Brain Computer Interface is capable of controlling numerous applications and assistive devices such as controlling a cursor on computer screen, movement of a robotic arm or a wheel chair. The efficiency of a Brain Computer Interface system completely rely on efficient preprocessing and classification algorithms. In present work, a methodology of feature extraction and classification of Electroencephalogram signals is proposed for implementation of Brain Computer Interface for physically disabled. The EEG signal under consideration has been recorded from seven different subjects performing five different mental tasks. Time Frequency Energy Distribution spectrum is computed from the coefficient obtained from Hilbert Huang transform of Electroencephalogram signals. Four statistical parameters are calculated from the TFED of signals as features. For classification, Support Vector Machine classifier model is employed. The results of the classification show efficacy of present methodology of feature extraction and classification for implementation of Brain Computer Interface.

Brain Computer Interface (BCI), Electroencephalogram (EEG), Hilbert Huang Transform (HHT), Support Vector Machine (SVM), Classification

"Feature Extraction and Classification of Electroencephalogram signals for implementation of Brain Computer Communication interface", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.4, Issue 1, page no.12-21, January-2017, Available :http://www.jetir.org/papers/JETIR1701003.pdf

"Feature Extraction and Classification of Electroencephalogram signals for implementation of Brain Computer Communication interface", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.4, Issue 1, page no. pp12-21, January-2017, Available at : http://www.jetir.org/papers/JETIR1701003.pdf