A TKF Based Adaptive Filtering for ECG Signal

Most of the time, various kinds of noises corrupt an ECG signal. A portion of these commotions are power line interface, benchmark float, muscle withdrawal, movement relics, electrosurgical noise, instrumentation commotion and electromyography noises. A method capable of significantly filtering ECG signal noises must be developed urgently. An EMD, a three-iteration Kalman adaptive filter (TKF), and a brand-new Customized Otsu Thresholding (COT)-based method for de-noising the ECG signal are all presented in this work. This work is capable of denoising lower and higher order IMFs (Intrinsic mode functions) in place of old EMD-based de-noising methods, which only denoise IMFs of lower orders. In addition to EMD and TKF operation, this work has been utilized to further enhance signal quality. After taking into consideration the first iteration, the efficiency of the TKF measure, and applying two additional adaptive iterations of the Kalman filter to further improve the signal quality on the standard MIT-BIH arrhythmia database, the described technique's validity is evaluated. In order to validate the proposed method in a noisy real-time environment, Gaussian noise of various SNR levels is added to the initial signals.