LAUSR

This paper proposes a novel real-time human heart rate (HR) estimation method for noncontact vital sign radar detection. The proposed method combines the Hough transform based respiratory harmonics suppression algorithm and linear predictive coding (LPC) based HR estimation algorithm. Since respiration signals can cause serious interference to HR estimation, the Hough transform based respiratory harmonics suppression algorithm is first applied to successively identify and filter the respiration signals and the higher order harmonics by their distributions on the 2-D spectrum. After suppression, HR can be derived in real time based on the linear predictor coefficients of the filtered signals. Compared to the traditional method, the proposed method can estimate HR more precisely by reducing the interference of respiratory harmonics and avoiding the problem of insufficient frequency resolution caused by short-time windows. Experimental results are presented to demonstrate the performance of the proposed method.