Extraction of Cardiopulmonary Rates from 24ghz Doppler Radar Using Time Frequency Analysis

摘要

Vital sign detection using Doppler radar has potential applications in the home and medical healthcare, etc. In this paper, a generalized Warblet transform (GWT) based algorithm is used to reveal the instantaneous frequency (IF) ridge of cardiopulmonary signals from the 24GHz Doppler radar. To improve the accuracy of the estimated IF in GWT, Viterbi algorithm (VA) instead of maximum method is firstly adopted in the iterative process and IF estimation, then, spectrum of IF is calculated to obtain respiration and heartbeat rates. Validation of the proposed algorithm on several simulated signals are conducted. The results show that compared with the original GWT and short time Fourier transform (STFT) combined with maximum method, the proposed algorithm can obtain more accurate IFs especially in a high noise environment.