A Study of Detection of Drowsiness and Awakeness using Non-contact Radar Sensors

Biometric information is used in a variety of industrial fields. Heart rate and respiration rate, in particular, are widely applied not only in medical institutions but also in life safety. However, a sensor must be worn or directly attached to the human body to obtain a bio signal, which is inconvenient and limits its application. In this study, a 24 GHz radar sensor module is developed, and an algorithm is implemented by analyzing the frequency and peak values of a human participant’s heartbeat and respiration signals in an unconstrained state. In the experiment, the existing ECG equipment (MP150) and radar sensor module are compared. The results indicate that the average value of MP150 is higher than that of the Doppler sensor in terms of all parameters; however, the deviation of the Doppler sensor is small, and the bias is low. Furthermore, it is confirmed that the HRV decreases in the drowsy state compared to that in the wakeful state in both devices. These results confirm that bio-signals change during drowsiness, and conversely, drowsiness can be detected through changes in bio-signals, which is a significant finding.