Artifact Removal using Elliptic Filter and Classification using 1D-CNN for EEG signals

Convolution plays major role in deep learning. CNN is the basic architecture and most of the other deep learning architectures ae built based on CNN. However, 1D-CNN is suitable for EEG signals as it is time-series one dimensional characteristic signal. In this paper, we have proposed a 6-layer 1D-CNN to analyze the classification of EEG. Elliptic filter helps to filter Multivariate EEG signal of sleep apnea patient and then classify sleep stages using CNN. Simulations are carried out using Mat Lab.