Curvelet Transform Based EEG Signal Analysis Using Pca

Abstract

The knowledge of Brain-Computer Interface (BCI) provides a direct exchange of information from the human brain and external devices. In BCI design structure, electroencephalography (EEG) identifies to be the major deliberately calculate the recordings of brain activity. Our proposed method is used to extract and analyze the characteristics of the EEG signal. They organize signal for BCI can be discriminate against and serve up human emotions. The projected method recognizes EEG information retrieving and computing feature extraction and classification. These signals have dissimilar frequency stages for Data waves, theta, alpha and beta. The combination of curvelet transforms (CT) and the principal component analysis (PCA) compute the dimensionality minimize and optimal characteristic extraction. The categorization of EEG signals, ANN (Artificial Neural Network) impact on this process of classification. This paper also provides a similarity between the projected two tools PCA and CT, with a combination of ANN.