DSP removal of respiratory trend in photoplethysmographic blood volume pulse measurements

Abstract

Three different digital signal processing algorithms for removing the respiratory trend from photoplethysmographic blood volume pulse (BVP) signals are investigated. The goal of this processing is to avoid the distortion introduced by the low-frequency trend contaminating the BVP signals recorded for the purpose of exercise evaluation. The three approaches are compared in terms of the improvement in signal-to-noise ratio (SNR) achieved and the preservation of the morphology of the BVP pulses, as both are important for the determination of indices derived from this signal.