High-Accuracy Heart Rate Estimation By Half/Double BBI Moving Average and Data Recovery Algorithm of 24GHz CW-Doppler Radar
High-Accuracy Heart Rate Estimation By Half/Double BBI Moving Average and Data Recovery Algorithm of 24GHz CW-Doppler Radar
复制标题
24GHz CW多普勒雷达半/双BBI移动平均高精度心率估计及数据恢复算法
DOI：
10.1109/atc55345.2022.9943010
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
K. Ishibashi
中科院分区：
文献类型：
--
作者：
Nguyen Huu Son;H. Yen;G. Sun;K. Ishibashi
由于非接触式生命体征和心肺活动监测的新趋势，多普勒雷达作为一种有前途的技术越来越受到关注。这种医用雷达可以跟踪人体表面的微小运动，以测量心率和呼吸率等几个重要参数。然而，由于雷达信号易受身体运动和呼吸影响，因此使用简单的信号处理方法提供准确的结果可能具有挑战性。在本研究中，我们提出一种结合数位滤波器与过零演算法的时域讯号处理演算法，以准确地撷取出心跳频率。该算法包括两个过程：心率提取的半/双心跳间隔（BBIs）移动平均滤波器和心跳细化IIR滤波器，零交叉检测和数据恢复算法。为了评估算法的准确性，我们测量了10名健康受试者仰卧位的心肺信息，这些受试者还佩戴了心电图（ECG）接触式传感器以提供参考数据。所提出的方法具有相当大的优势，由于紧凑的模型和低的计算成本，同时仍然提供了一个显着的相关性相比，心电图时，系数R = 0.998。
Due to the new trend on non-contact monitoring of vital signs and cardiopulmonary activity, Doppler radar is getting more attention as a promising technology. This medical radar tracks small movements on the human's body-surface to measure several vital parameters such as heart rate and respiration rate. However, because radar signals are susceptible to body movement and breathing effects, using simple signal processing methods can be challenging to provide accurate results. In this study, we propose a time-domain signal processing algorithm that combines Digital filters and a zero-crossing algorithm to extract the heart rate accurately. This algorithm comprises two processes: Heart rate extraction by Half/Double beat-to-beat intervals (BBIs) Moving average filters and Heartbeat refinement by IIR filter, zero-crossing detection and Data recovery algorithm. To evaluate the algorithm's accuracy, we measured the cardiopulmonary information of 10 healthy subjects in the supine position, and these subjects also wore an electrocardiograph (ECG) contact sensor to provide reference data. The proposed method has considerable advantages due to the compact model and low computational cost while still providing a significant correlation compared with ECG when the coefficient R = 0.998.