Comparison of Deep Learning & Adaptive Algorithm Performance for De-Noising EEG

Abstract Various forms of artifacts can readily contaminate an electroencephalogram recorded using surface electrodes. A comparison several (EEG) de-noising methods is shown here. Five distinct noise are reduced three different strategies, and the results compared. These procedures Recursive Least Squares (RLS) adaptive algorithm, Mean (LMS) method, Fully Connected Neural Network (FCNN). The time-domain plots real EEG signal, noisy forecasted signal. For comparing performance techniques here relative-root-mean-square-error (RRMSE) signal-to-noise-ratio were used. Here, exploring values parameters, we find that FCNN predicts a better result than other two algorithms.