An Optimized Approach for Noise Signal Removal in Electroencephalogram

Abstract

Electroencephalogram (EEG) addresses the electrical action of the mind recorded by setting a few anodes on the scalp. EEG signals are mind boggling in nature and comprise of different antiques like: visual, solid, cardiovascular and so forth. The curios evacuation in EEG signs can be significantly displayed by thinking about it of type Additive White Gaussian Noise (AWGN) in nature. Autonomous part Analysis (ICA) is known for its capacity to sift through the curios from the sign and subsequently it is utilized to adjust the source signal into two combinations such that the mind cues and the ancient rarities get isolated. In spite of the fact that, there is a limitation that ICA must be performed on multi-channel signal info. In the current case as the information EEG is single-channel; subsequently ICA is applied in blend with Stationary Wavelet Transform (SWT) in this paper for commotion separating of EEG signals. The quantitative assessment of the proposed approach has been made utilizing Signal-to-Noise Ratio (SNR) boundary which portrays agreeable sifting at different force levels of AWGN.