Research On Millimeter Wave Radar Vital Sign Detection Algorithm

Millimeter wave radar,as an important part of wireless sensing technology,plays an important role in the field of non-contact vital sign detection.It also has a broad application prospect because of its small size,low radiation,high accuracy,and non-intrusion of user privacy,especially in family elderly and child monitoring.Improving the detection capability and accuracy of millimeter wave radar for human vital signs in various scenarios is a key focus of intelligent medical systems.Therefore,this thesis focuses on the research of millimeter wave radar in resting state and motion state for respiratory heartbeat detection methods,and the main research contents are as follows.1.the millimeter wave radar range and velocity measurement principles and system components are introduced,and the principles of respiratory heartbeat waveform extraction are explained in relation to them.The phase deconvolution method is studied for the common phase winding problem in the phase information extraction process,and the method can extract the respiratory heartbeat waveform without distortion.2.For the problem that the echo amplitude is too small when the target is far from the radar,the processing method of coherent accumulation in the resting state is used to enhance the amplitude intensity of the long-range target and expand the effective detection distance of the radar.The advantages and shortcomings of the two breathing heartbeat detection methods are analyzed,and an optimized filter detection method based on the principle of minimum mean square error is proposed according to the relevant problems.The processing results of the measured data show that the method with the trap filter can effectively extract the respiration and heartbeat waveforms,and the average error of heartbeat rate extraction is ±1-4 times per minute.3.The detection scheme of respiratory heartbeat in different postures is designed,and the overall detection performance of the algorithm is verified by the measured data.The scenarios of two people sitting and standing,and interference from moving people are also designed to verify the detection performance and accuracy of the algorithm in different scenarios.4.In the problem of large errors caused by targets spanning multiple distance units that appear in the motion state,this thesis proposes a motion state detection method based on distance compensation after comparing common constant false alarm detection schemes and analyzing their shortcomings.The measured data show that the method can accurately locate the target motion path,and with the optimized filter algorithm can get better results.By comparing the experimental results at different motion speeds,it is shown that the average error of heart rate extraction is ± 3-6 times per minute in the speed range of 0.3-1.2m/s,which meets the expected requirements.