Estimation of noise in ECG signals using wavelets

ECG signal and generally biomedical signals change their statistical properties over time. The most powerful tool for analyzing non-stationary signals is the wavelet transform. We developed and evaluated a noise perturbing ECG detection system based on the multi-resolution analysis. We considered four types of ECG signal, without additional noise and corrupted by three types of noise and we realized the signals classification using wavelet filtering and statistical analyses. The ECG data is taken from the standard MIT-BIH Arrhythmia database, while the signals of noise belong to MIT-BIH Noise Stress Test Database.