TY - GEN
T1 - A Map Feature Fusion Based Artifact Removal Method for Non-Contact Vital Sign Detection with a Single FMCW Radar
AU - Qiu, Yuxiang
AU - Yamamoto, Michitaka
AU - Takamatsu, Seiichi
AU - Itoh, Toshihiro
N1 - Publisher Copyright: © 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - The motion artifact is one of the most challenging problems in radar based non-contact vital sign detection, which seriously hampers practical health monitoring for human beings and livestock. In this article, we proposed a map feature fusion method to suppress the artifact conveniently without redesigning the radar system or adding auxiliary devices. The Time-Doppler Map (TDM) and the Time-Range Map (TRM) are initially generated from the radar cube. The phase signal and the motion velocity are then extracted from TRM and TDM, respectively. Subsequently, we utilized the estimated motion velocity to remove the artifact in the extracted phase signal with projection decomposition. Experimental results demonstrated that the artifact was successfully suppressed, enhancing the accuracy of heartrate measurement on moving targets to 97% referring to Electrocardiography (ECG) sensors. Furthermore, the waveform of heartbeat was extracted and the accuracy of intervals between peaks are verified by Bland-Altman test.
AB - The motion artifact is one of the most challenging problems in radar based non-contact vital sign detection, which seriously hampers practical health monitoring for human beings and livestock. In this article, we proposed a map feature fusion method to suppress the artifact conveniently without redesigning the radar system or adding auxiliary devices. The Time-Doppler Map (TDM) and the Time-Range Map (TRM) are initially generated from the radar cube. The phase signal and the motion velocity are then extracted from TRM and TDM, respectively. Subsequently, we utilized the estimated motion velocity to remove the artifact in the extracted phase signal with projection decomposition. Experimental results demonstrated that the artifact was successfully suppressed, enhancing the accuracy of heartrate measurement on moving targets to 97% referring to Electrocardiography (ECG) sensors. Furthermore, the waveform of heartbeat was extracted and the accuracy of intervals between peaks are verified by Bland-Altman test.
KW - TDM
KW - TRM
KW - feature fusion
KW - non-contact vital sign detection
KW - projection decomposition
UR - https://www.scopus.com/pages/publications/85179760614
U2 - 10.1109/SENSORS56945.2023.10324886
DO - 10.1109/SENSORS56945.2023.10324886
M3 - Conference contribution
T3 - Proceedings of IEEE Sensors
BT - 2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE SENSORS, SENSORS 2023
Y2 - 29 October 2023 through 1 November 2023
ER -