Research On Vital Signs Detection Algorithm Based On FMCW Radar

Respiration and heartbeat are vital signs of human body,and their values can reflect the health status of human body.The existing way of vital signs detection by contact often causes many inconveniences to users,but the FMCW radar enjoys the advantages of small size and high resolution.It can realize non-contact vital signs detection by detecting the micro-motion of the chest caused by breathing and heartbeat,and has a promising prospect for application.However,because the heartbeat signal is relatively weak,the existing vital signs detection algorithm for heart rate estimation is susceptible to interference from respiratory harmonics and noise,which affects the accuracy of heart rate estimation.Therefore,this paper uses the 77 GHz AWR1642FMCW radar to study the vital signs detection algorithm.The details are as follows:In terms of theoretical basis,this paper derives the theoretical formula for the detection of vital signs and targets according to the FMCW radar transmitting signal,receiving signal and IF signal and the feasibility of the vital signs detection based on the basic principle of extracting the phase is demonstrated.In dealing with the problem of respiratory harmonics and noise interference,this paper proposes a Coarse-to-Fine respiration and heartbeat estimation algorithm,and uses the heart rate detection formula proposed by predecessors for peak determination.Coarse estimation of heart rate is performed in the time domain of the heartbeat signal,the first step is to set a for the interval between adjacent waveforms,and waveforms with intervals smaller than the threshold are judged as false waveforms.The coarse estimate of heart rate is obtained by calculating the average of the remaining true wave intervals.The heart rate is finely estimated in the frequency domain of the heartbeat signal,and the spectral peak closest to the coarse estimate in the time domain is defined as the heartbeat frequency of the human body through spectral peak search.For the vital signs detection algorithm,This paper uses Empirical Mode Decomposition(EMD)and Modified Ensemble Empirical Mode Decomposition(MEEMD),two traditional vital signs detection algorithms,to compare with the Coarse-to-Fine algorithm.In order to compare the performance of the three vital signs detection algorithms,we have conducted multiple experiments with different subjects.Before the experiment,an experimental site was built to simulate the possible distance and orientation of the human body for vital signs detection in a real scene.Through experiments,the three algorithms of EMD,MEEMD and Coarse-to-Fine are compared,and the average error rates of heartbeat detection when the human body and the radar are at different distances and angles are 4.18%,3.03% and 2.34%,respectively.The experimental results show that the Coarse-to-Fine estimation algorithm has the lowest heartbeat detection error rate and the shortest average data processing time of 0.71 s,with the best performance.In addition,the effect of distance and angle on the Coarseto-Fine estimation algorithm is also investigated in this paper.From the analysis of the results,it can be concluded that within the range of 1.5m,with the increase of the distance between the human body and the radar,the detection error rate of the Coarseto-Fine estimation algorithm has increased to a certain extent.In the range of 45°,the detection error rate of the algorithm has no obvious trend with the increase of the angle.Finally,the paper designs a host computer interface of vital signs detection system and connects it with an FMCW radar.Then the system was tested under the two conditions of normal heart rate and high heart rate.The test results show that the system can operate stably,and the detection error rate is within 5%.