PPG signal characterization using windkessel model

Abstract

The cardiovascular system is one of the most important systems in the life process of humans and animals. At present, cardiovascular diseases have become a common problem affecting human life and health. Therefore, safe and reliable detection of non-invasive vital signs has received more and more attention recently. A common method is to use pulse oximeter (PPG) signals collected through wearable sensors such as smart watches. Through the establishment of a model to study the characteristics of PPG signal to analyze the subject's cardiovascular health, it is possible to prevent diseases and assist in the purpose of detecting the disease. In this thesis, the most widely used dual elastic cavity model is used to characterize and study the PPG signal. The corresponding electrical network model is established according to the cardiovascular system characteristics, and four parameters C_1, C_2, R, and L are determined to investigate the cardiovascular properties. The PPG signal is used to predict the parameter values through curve fitting and error analysis has been performed using the parameters.