Wavelet Energy based Neural Fuzzy Model for Automatic Motor Imagery Classification

Brain-computer interface (BCI) is a communication system by which a person can send messages without any use of peripheral nerves and muscles. BCI systems might help to restore abilities to patients who have lost sensory or motor function because of the damaged region, such as amyotrophic lateral sclerosis (ALS), spinal cord injury, brainstem stroke, or quadriplegic patients. Brain computer interfacing can be effectively implemented by analyzing EEG signals generated in the brain. This paper presents a method for accurately classifying EEG signals generated by imagery left and right hand movements. Firstly, wavelet transform and energy of the decomposed signal is used to obtain the final feature vector matrix. Secondly, the feature data is classified using ANFIS. . The Mutual Information value calculated is 1.2942 bit. The classification accuracy achieved 93.5% in the course of testing on the data from subject. Support Vector Machine is also used to compare the performance with ANFIS. includes the wavelet transfo