US 12,403,358 B2
Data pre-processing method and exercise vital sign detection radar
Kai-Jen Cheng, New Taipei (TW); and Mn-Yu Chen, New Taipei (TW)
Assigned to WISTRON CORPORATION, New Taipei (TW)
Filed by Wistron Corporation, New Taipei (TW)
Filed on Feb. 23, 2022, as Appl. No. 17/678,492.
Claims priority of application No. 110140489 (TW), filed on Oct. 29, 2021; and application No. 110140492 (TW), filed on Oct. 29, 2021.
Prior Publication US 2023/0136937 A1, May 4, 2023
Int. Cl. A63B 24/00 (2006.01); G01S 7/35 (2006.01); G01S 7/41 (2006.01); G16H 50/20 (2018.01)
CPC A63B 24/0062 (2013.01) [G01S 7/354 (2013.01); G01S 7/356 (2021.05); G01S 7/358 (2021.05); G01S 7/412 (2013.01); G16H 50/20 (2018.01); A63B 2220/89 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A data pre-processing method, performed by a processor in a signal processing apparatus, the method comprising:
obtaining an energy distribution parameter set obtained through beamforming scanning and a digital signal, wherein the digital signal corresponds to a reflected radar signal of an exercise vital sign detection radar;
searching, by using the energy distribution parameter set, for a target in a manner of filtering out background noise;
weighting the digital signal according to the energy distribution parameter set to obtain an optimized signal;
determining a detection range region covering an activity range of the target based on statistical analysis of the energy distribution parameter set within a period;
analyzing the optimized signal only within the detection range region to extract one or more pieces of target phase data corresponding to the target from the optimized signal, wherein the step of analyzing the optimized signal comprises:
obtaining a phase map and a vibration frequency map according to the optimized signal;
selecting at least one candidate position having an energy intensity exceeding an energy threshold from the vibration frequency map;
selecting a target position from the at least one candidate position; and
obtaining the one or more pieces of target phase data in a distance range in the phase map according to the target position; and
inputting the one or more pieces of target phase data into a machine learning model to obtain a vital sign parameter prediction result.