Advanced Techniques in Eeg Signal Analysis for Electronic Engineering Applications

This study provides a novel method that uses Stationary Wavelet Transform (SWT) with different threshold functions to adjust for ocular artifact (OA) in single-channel EEG signals. Metrics including power spectrum, ΔSNR (signal-to-noise ratio improvement), ARR (artifact rejection ratio), CC (correlation coefficient), and RMSE (root mean square error) are used to compare the efficacy of the SWT method to the current Discrete Wavelet Transform (DWT) approach. The findings demonstrate that the UT (Universal Threshold) and MT (Minimax Threshold) functions better preserve the neural information in non-artifact regions, as demonstrated by better CC and RMSE scores, while the NT (NeighShrink Threshold) function achieves superior artifact rejection as indicated by higher ΔSNR and ARR values. Because of expanded overt repetitiveness, SWT is more computationally requesting than DWT, notwithstanding its predominant execution in artifact removal. Moreover, neither SWT nor DWT are versatile, and the wavelet determination influences how well they work. Ensuing examination should focus on improving the NT edge for better sign uprightness and making more versatile wavelet techniques for higher artifact removal effectiveness.