Classification of EEG signals using empirical mode decomposition and lifting wavelet transforms

The electroencephalogram (EEG) signals are a sturdy tool for tracing brain variations during different periods of life, also it plays a prominent and considerable role in the diagnosis of various diseases. In our previously published papers [1-8] we have worked on diverse problems that can be analyzed by neural networks. In this paper, we have chosen EEG signals due to its increase its application in Motor Learning predicaments. EEG recordings complied and manifold over the duration of an elongated time frame encompasses an enormous quantum of EEG data. The study of signals and decomposition of these signals activity contribute a way to diminish the computational cost and emend the enforcement of the classifiers. We have proposed a unique classification of signals in which we have used empirical mode decomposition and variety of lifting wavelet transform schemes for the compression of signals. The procedure for making a resolution contains four stages: (a) extraction of the signals, (b) signal preprocessing and filtering,(c) compression using Empirical mode decomposition or lifting wavelet Transform schemes and (d) classification using artificial neural network enforcement. The outcomes contributed the fact that there exists ability in the proposed algorithm for the classification of EEG signals.