Accurate Doppler Radar-Based Cardiopulmonary Sensing Using Chest-Wall Acceleration

Abstract

This paper presents the theory and experimental results of a new method to significantly increase the detection accuracy of the human heartbeat rate using a continuous-wave Doppler radar. Traditionally, it is assumed that the chest wall displacement signal, which is recorded by the radar and used for heartbeat rate estimation, is a pure sine wave. In this paper, we use the fact that the displacement signal is a complex Gaussian function rather than a pure sine wave. This function shows a declining amplitude versus frequency; therefore, the heartbeat signal will be much weaker than the respiration signal and can be easily buried in the respiration's harmonics. However, by exploiting the chest wall acceleration instead of its displacement, the heartbeat signal is greatly amplified, leading to a significantly higher heartbeat rate detection accuracy. Recorded data from 12 healthy human subjects show an average heartbeat rate detection accuracy of more than 95% when compared with reference electrocardiogram recordings. The proposed technique is robust, simple, and requires minimum calculation resources which are important for online monitoring and power consumption reduction. Measurement results indicate its potential for being used in reliable non-contact heartbeat rate monitoring systems.