EEG signal with feature extraction using SVM and ICA classifiers

Identifying artifacts in EEG data produced by the neurons in brain is an important task in EEG signal processing research. These artifacts are corrected before further analyzing. In this work, fast fixed point algorithm for Independent Component Analysis (ICA) is used for removing artifacts in EEG signals and principal component analysis (PCA) tool is used for reducing high dimensional data and spatial redundancy can be removed. Support vector machine (SVM) tool is used for pattern recognition of EEG signals and the extracted parameters used to implement testing phase of the SVM on the hardware.