Comparative study of brain waves classification using fast fourier transform and feed forward neural network
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] Brain is one of the complex network known. There are several techniques available to acquire signals originating inside brain. If we can classify those signals based on frequency, it will be easy to understand how brain works. But the signals recorded have very low amplitudes and it is often added with nose. Different parts of the brain emit different kind of frequency, which are broadly classified as Delta, Theta, Alpha, Beta and Gamma. Our first task is to classify these different signals based on their frequencies. Two of such classification techniques are presented here. First, calculation of frequency using Fast Fourier Transform, a linear process and second, Artificial Neural Network based on Feed Forward Model (FFNN), non-linear process. Artificially simulated data show that with no noise added to the signal, FFT provides 100% accurate classification. When external White or Pink Noise was added to the signal, FFT accuracy got plummeted, but accuracy by FFNN surpassed FFT results. A comparative study between both these methods has been presented, and it showed that, not only signal amplitude but also number of cycles affect signal classification accuracy.
Access is limited to the campuses of the University of Missouri.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.