Ordinal Pattern-Based Mode Decomposition for Phonocardiogram Signal Analysis

The Ordinal Pattern-Based Mode Decomposition (OPMD) is a novel data-driven method recently introduced for processing weakly stationary signals embedded in noise. It exploits the filtering potential of ordinal patterns (OP) to decompose time series into multiple intrinsic oscillatory mode functions. We demonstrate its effectiveness in simulated and real recordingsphonocardiogram (PCG) signal decomposition by comparing it with well-established data-driven methods, including Empirical Mode Decomposition (EMD), Variational Mode Decomposition (VMD), and Empirical Wavelet Transform (EWT). OPMD shows strong potential as a competitive alternative, offering enhanced component separation and reducing mode mixing.