Sleep Posture Recognition With a Dual-Frequency Cardiopulmonary Doppler Radar

John E Kiriazi,Olga Boric-Lubecke,Victor M Lubecke,Shekh M M Islam

doi:10.1109/access.2021.3062385

John E Kiriazi, Olga Boric-Lubecke + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3062385

Copy DOI

Abstract

While Doppler radar can be used to measure cardiopulmonary vital signs during sleep, meaningful diagnostic assessments are often subject to knowledge of a subject's changing sleep posture. The torso Effective Radar Cross Section (ERCS) and displacement magnitude were studied for 20 human subjects in three imitated sleep posture categories using a dual-frequency Doppler radar system in an exploratory examination of the feasibility of using radar to recognize body orientation. Box plot statistical analyses were performed for comparative assessment of ratio variations in ERCS and respiration depth for three different imitated sleep postures. The observed statistical trends and correlations were applied to a physical model to develop posture decision algorithms with initial supine posture data used as a reference. A single-frequency algorithm tracked postures without error for 90% of the subjects using 2.4 GHz data, and 80% using 5.8 GHz data. As accuracy limitations were complementary, a dual-frequency algorithm was developed which recognized postures without error for 100% of the subjects.

Highlights

Sleep is a biological imperative for humans, and quality of sleep plays an important role in ensuring people stay active, healthy, and energetic [1]
This paper examines an approach for recognizing three categories of sleep posture while simultaneously measuring diagnostic cardiopulmonary patterns using an unobtrusive and non-contact Doppler radar system
A dual-frequency Doppler radar system that integrates sleep posture recognition with cardiopulmonary monitoring has been successfully demonstrated, with the measurement results for a 20-subject human study analyzed for the recognition of three critical categories of imitated sleep postures

Summary

INTRODUCTION

Sleep is a biological imperative for humans, and quality of sleep plays an important role in ensuring people stay active, healthy, and energetic [1]. Frequency modulated continuous wave (FMCW) radar was applied with a multipath analysis of reflections used to distinguish sleep postures using a neural network model [31] This method requires calibration data with subjects wearing accelerometers in different sleep postures in a fixed home environment (accuracy 94.1%). Measurements carried out on twenty human subjects are presented, and data statistics are used to show the trend by which the ERCS and torso displacement magnitude are correlated to changes in posture Based on these results, a decision algorithm was developed to investigate the ability to automatically recognize the sleep-type posture of a subject. Proper thresholds and using a dual-frequency radar, this research demonstrates that such a system can differentiate whether a subject maintains a supine posture or switches to a prone or side posture

SYSTEM ACCURACY ASSESSMENT

RESULTS AND STATISTICS

POSTURE RECOGNITION

VALIDATION OF THE ALGORITHM IN ORDINARY ENVIRONMENT

CONCLUSION