Seismocardiographic Signals Detection Based on Circulant Singular Spectrum Analysis Through Millimeter Wave Radar

Haiyang Hu,Yuhao Chen,Shihua Li,Jun Yang,Zhenxing Sun
DOI: https://doi.org/10.1109/icsip52628.2021.9688707
2021-01-01
Abstract:Aiming at the problem that Seismocardiography (SCG) signal detection cannot be promoted due to the bulkiness of the equipment and the SCG containing 7 fiducial points is difficult to extract, this paper proposes a non-contact SCG signal detection algorithm based on the circulant singular spectrum analysis (CiSSA) that runs on a millimeter-wave radar platform. In order to make the detection equipment portable and achieve non-contact detection, this paper uses a millimeter-wave radar working in the 60-64GHz frequency band to collect the signal returned from the surface of the human chest cavity. Then the signal is processed by static clutter elimination and maximum energy distance gate positioning algorithm to improve the signal-to-noise ratio, especially to reduce the interference problem caused by people’s tiny movement. For the key problem that the SCG curve extracted by traditional filtering methods are fuzzy and the fiducial points are difficult to identify, the CiSSA method is used to improve the accuracy and completeness of SCG curve extraction. By CiSSA, the signal is embedded into trajectory matrix, and then decomposed into component matrices containing various frequency characteristics. Next, the component matrices are grouped according to the frequency distribution characteristics of the SCG signal, and finally the result of the grouping is reconstructed into a high-precision SCG signal with clear and distinguishable fiducial points. Experiments show that the proposed SCG detection method can accurately extract the SCG curve, clearly distinguish 7 fiducial points of the SCG, and the success rate of detection can reach 97%. This provides the possibility for the popularization of SCG signal detection.
What problem does this paper attempt to address?