ECG Signal Preprocessing: High-pass and Band-stop Filter Implementation

This document provides a detailed explanation of ECG signal preprocessing methodology. First, we implement a high-pass filter to eliminate respiratory-induced baseline wander, typically using Butterworth or Chebyshev filters with cutoff frequencies around 0.5-1.0 Hz. The implementation involves designing filter coefficients using MATLAB's signal processing toolbox functions like butter() or cheby1(), followed by applying the filter using filtfilt() for zero-phase distortion. Next, we employ a band-stop filter (notch filter) centered at 50/60 Hz to suppress power line interference, which can be implemented using iirnotch() function with specific quality factor parameters. Finally, we conduct performance benchmarking between these filter implementations, comparing computational efficiency through processing time measurements using tic/toc commands and analyzing frequency response characteristics using fvtool(). These preprocessing steps significantly enhance signal quality by removing artifacts while preserving diagnostically relevant ECG components.