Performance Analysis of Wavelet Transform in the Removal of Baseline Wandering from ECG Signals in Children with Autism Spectrum Disorder (ASD)

Abstract The electrocardiogram (ECG) signals are used for prediction of various
cardiovascular diseases and also as a prominent signal for developing intelligent
healthcare and wearable systems. The ECG signal is mostly corrupted with many

external noises due to baseline wandering (BW), poor contact of electrodes, move-
ment of the subject, and power line interferences. In order to retain the morpholog-
ical information of the ECG signal, these noises are removed by various methods

suggested by many researchers. In this work, we remove the BW in the ECG signal
recorded from one male child with autism spectrum disorder (ASD) using discrete
wavelet transform (DWT) such as Daubechies, Symlet, and Coiflet wavelet func-tions to denoise thecorrupted ECG signal. The mean square error (MSE), peaksignal-to-noise ratio (PSNR), noise power (Pn), and percentage root mean squaredifference (PRD) were analyzed which proves that ‘db4’ wavelet