Doppler Radar Sensors for Non-contact Human Vital Signal Detection

Abstract

In this thesis, Doppler radar sensors for Non-contact Human vital signal detection have been discussed. Doppler radar sensor for non-contact human vital signal detection has three main problems known as ‘null point problem’, ‘moving body problem’ and ‘external interferer problem’. This paper investigates about solution for solving these problems. A single chip 24 GHz Doppler radar sensor whose sensitivity is high enough to detect small skin motion produced by activity of heart and lung. However, the sensor is also sensitive to unwanted motion. It is hard to measure the heartbeat signal of moving human body because the unwanted motion produces noise signal whose bandwidth is similar to heartbeat signal. Therefore a self-mixing radar structure and a Doppler radar sensor using two-frequency is proposed to detect the heart more effectively. The Doppler radar sensor using two-frequency is proposed to solve ‘moving body problem’. The proposed two-frequency radar sensor uses frequency dependent characteristics of human body tissue and heartbeat signal can be measured using the proposed radar sensor structure. However, the heartbeat signal power was smaller (10 dB) than that of general Doppler radar sensor. But we found the heartbeat signal has high frequency components and if the operating frequency of Doppler radar sensor is low, the noise signal produced by unwanted motion can be in different frequency domain to heartbeat signal. However SNR of the radar sensor should be improved to detect the high frequency components of heartbeat signal. Therefore a coded quadrature Doppler radar sensor is proposed to increase the SNR of the sensor system. Comparing to general Doppler radar sensor, the coded quadrature Doppler radar sensor is strong to external interferer (interferer signal is suppressed by 30 dB comparing to general Doppler radar sensor) and has low noise characteristic (noise suppression is over 10 dB) in frequency range of 10~20 Hz at baseband. The high freque...