Photoplethysmography and accelerometer sensor signals for the detection of physical activity

Abstract

Wearable devices for monitoring human physiological parameters have become popular, and due to their low cost, the most common method of monitoring human information in such devices is the use of photoplethysmography (PPG) signals. Nevertheless, the accurate estimation of the PPG signal recorded from the subject's wrist during various physical exercises is often a challenging problem because the initial PPG signal is heavily corrupted by motion artifacts. Long Short Time Memory (LSTM) is built to recognize activities.