Sleep performance and physical activity estimation from multisensor time series sleep environment data

Abstract

Sleep is an essential physiological function, needed for the proper functioning of the brain and therefore for the general well-being and for a good quality of life. Currently, more and more people use mobile and wearable devices to monitor their sleep and physical activity. But while some of those recent devices may even provide reliable measurements of sleep structure, they do not provide a consolidated and definite answer for what has contributed to that situation. With the present study, we intend to verify and establish relationships between some environmental factors, such as temperature, humidity, luminosity, noise or air quality parameters and between sleep performance and physical activity and assert legal issues with GRPD law. A multisensor monitoring system was used to obtain a real dataset consisting of 55 night sessions of time series sleep environment data. We have also explored the feasibility of using time series machine learning models to predict sleep stages and to estimate the level of physical activity of a person.