Fourier Analysis Based Respiration Rate Estimation Using Corrupted Photoplethysmogram Signal

Abstract

This work proposes a Fourier series-based Fourier decomposition method (FDM) technique to analyze the photoplethysmography (PPG) signals to estimate respiration rate (RR). The proposed method consists of data acquisition, preprocessing, signal decomposition into Fourier intrinsic band function (FIBFs), Superposition of clean FIBFs, fast Fourier transform, spectral peak identification, and estimation of RR. The efficacy of the proposed FDM-based RR estimation method is evaluated using the CapnoBase database, a benchmark RR estimation database. The proposed method improves estimation accuracy with a mean absolute error (MAE) value of 1.02 breaths per minute (BPM) and a root mean square error value (RMSE) of 1.22. The proposed method can accurately estimate vital physiological parameter RR from the PPG signal.