(A) real-time photoplethysmography motion artifact reduction system robust against distance variation

Abstract

Recently, Photoplethysmography has been in the spotlight as a good method for obtaining biosignals conveniently since it is a non-invasive method and does not require any complicated signal line. However, despite these advantages, due to the nature of the optical sensor, it is vulnerable to body movements that often change the distance between the sensor and the skin. Therefore, in order to solve this problem, various compensation algorithms have been proposed based on the independent/dependent relationship between the Photoplethysmogram and the motion artifact. This thesis is based on the dependent relationship between them, and an algorithm to restore the distortion due to motion is introduced. The proposed algorithm partially obtains the AC component of the Photoplethysmogram using the DC component of the Photoplethysmogram, which is generally discarded and then attempts to finally restore the signal with a relocation algorithm using quasi-periodicity.