基于谐波陷波器改进HMLD的呼吸心跳分离算法  

Respiratory and heartbeat separation algorithm based on improved HMLD with harmonic notch filter

作　　者：汪新坤 曹乐 张文艳 WANG Xinkun;CAO Le;ZHANG Wenyan(School of Electrical and Electronic Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620

出　　处：《智能计算机与应用》2023年第1期142-148,共7页Intelligent Computer and Applications

基　　金：国家自然科学基金(61703270)。

摘　　要：针对毫米波雷达生命体征信号检测存在的呼吸谐波及心跳信号提取精度低的问题,本文提出了基于谐波陷波器改进的谐波倍数循环检测(HMLD)的呼吸心跳分离方法。首先,设计了单通道的毫米波雷达信号采集系统,采集人体胸廓雷达回波信号,并进行相位信号提取、相位解缠、滤波等预处理提取胸廓微动信号;其次,通过HMLD方法提取呼吸基波频率、高次谐波频率,并采用陷波器消除呼吸谐波;最后,根据HMLD方法提取心跳基波频率,提取的心跳频率误差率在11.5%以内,并与变分模态分解算法(VMD)提取结果对比,心跳信号信噪比平均提高了2.66 dB。实验结果表明,基于谐波陷波器改进的HMLD方法能够有效的分离呼吸心跳。Aiming at the problem of low extraction accuracy of respiratory harmonic and heartbeat signal in the detection of vital signs signal of millimeter wave radar, a respiratory heartbeat separation method based on improved harmonic multiple cycle detection(HMLD) based on harmonic notch filter is proposed. Firstly, a single channel millimeter wave radar signal acquisition system was designed to collect the human thoracic radar echo signal, and preprocess the phase signal extraction, phase unwrapping and filtering to extract the thoracic micro motion signal;Then the respiratory fundamental frequency and higher harmonic frequency were extracted by HMLD method, and the respiratory harmonic is eliminated by notch filter;Finally, the heartbeat fundamental frequency was extracted according to the HMLD method, and the error rate of the extracted heartbeat frequency was less than 11.5%. Compared with the extraction results of variable empirical mode decomposition(VMD), the heartbeat signal-to-noise ratio is increased by 2.66db on average. The experimental results showed that the improved HMLD method based on a harmonic notch filter can effectively separate respiratory and heartbeat.

关 键 词：毫米波雷达 生命体征信号 谐波陷波器 谐波倍数循环检测 

分 类 号：TN957.51[电子电信—信号与信息处理]