Contactless Respiration Variability Detection and Accuracy Test Using UWB Radar

This paper investigates the potential of radar technology for precise and non-intrusive detection of respiration rate variability. UWB radar, with its ultra-short pulses and extensive bandwidth, offers significant advantages in capturing subtle chest wall movements associated with respiration. It possesses the unique ability to penetrate clothing and physical barriers, making it an excellent candidate for remote physiological monitoring. This ultra-wideband radar system ensures the extraction of accurate respiration waveforms, and deep learning models, including VGG16, Inception V3, and ResNet50, are employed to evaluate respiration rate variability. Remarkably, VGG16 attains outstanding accuracy in results. This study advances the field of radar-based respiration monitoring, emphasizing the importance of robust signal processing and deep learning techniques. It showcases the potential of UWB radar for non-contact respiration monitoring, with applications spanning healthcare and in-home environments, promising to revolutionize the assessment of well-being and health.