基于EMD的提取心跳呼吸信号的研究  

Research on Extracting Heartbeat and Respiratory Signals Based on Emd

作　　者：李安 LI An(Guilin university of electronic science and technology guilin 541004,guangxi)

机构地区：[1]桂林电子科技大学,广西桂林541004

出　　处：《长江信息通信》2024年第12期33-36,共4页Changjiang Information & Communications

摘　　要：在采集人体生命体征信号的数据时,传统方法监测生命体征时有受测者体感舒适度差、检测结果受环境影响大等缺点。文章提出一种基于毫米波雷达的无接触式检测方式,针对多频率、非平稳雷达回波信号,提出用经验模态分解(EMD)提取人体的心跳呼吸信号。首先对雷达回波信号做快速傅里叶变换(FFT)确定目标距离单元,然后对回波的距离单元使用EMD,观察得到的每个本征模态函数(IMF)频域范围并提取心跳呼吸信号所在的IMF信号,将所提取的IMF信号叠加以完成心跳呼吸信号的重构,最后将重构得到的心跳呼吸信号进行平均恒虚警检测,将结果通过带通滤波器,最后获取心跳呼吸信号。实验结果表明,本文方法能够准确找到目标,提取出心跳呼吸信号,稳定性良好。When collecting the data of human vital signs,the traditional monitoring of vital signs has poor physical comfort of the subjects,and the detection results are greatly affected by the environment.This paper proposes a contact-free detection method based on millimeter-wave radar,for multi-frequency and non-stationary radar to extract the heartbeat respiration signal with empirical mode decomposition(EMD).First of the radar echo signal do fast Fourier transform(FFT)to determine the target distance unit,then use EMD,observe each intrinsic mode function(IMF)frequency domain range and extract the IMF signal,the extracted IMF signal superposition to complete the reconstruction of heartbeat breath signal,finally to the reconstructed heartbeat breath signal average constant virtual alarm detection,the results through the bandpass filter,finally get the heartbeat breath signal.The experimental results show that the method can accurately find the target and extract the heartbeat and respiration signal with good stability.

关 键 词：经验模态分解 生命体征 毫米波雷达 无接触式检测 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]