Pulse wave signal modelling and feature extraction based on Lognormal function from photoplethysmography in wireless body area networks

Dazhou Li, Yuanlu Xu, Wei Gao

Abstract: Highlights•We found and demonstrated that the Lognormal functions model has a stronger physiological characterization than the Gaussian functions model.•From a physiological view, we give the physiological significance of the parameters of the Lognormal functions.•To avoid nonlinear increase in the computational complexity of the basis functions model, we used multiple iterations of local curve fitting instead of the traditional one-shot global curve fitting.•We propose a conformation criterion by calculating the first-order and second-order derivatives of the fitted waveform, starting from the physiological formation principle and propagation characteristics of the PPG signal.