Design And Implementation Of Multi-person Heart Rate And Respiratory Vital Sign Monitoring System Based On Millimeter-wave Radar

Under the strategy of healthy country,monitoring vital signs is an important research.Vital signs information provides a reliable diagnostic basis and safety guarantee for modern health care and medical applications.According to the monitoring method of vital signs,it can be divided into contact type and non-contact type.The traditional contact measurement method has certain limitations,for example,it will increase patients(burn patients)pain and cause secondary infection;for chronic respiratory diseases,the treatment is time-consuming and laborious.As a simpler and more convenient method of health monitoring,non-contact measurement will play an increasingly important role.At present,most non-contact measurements are based on technologies such as Wi Fi,UWB,Doppler radar and cameras.However,these technologies have low measurement resolution,are susceptible to multipath effects,cannot detect heartbeats,and have privacy issues.Compared with these technologies,millimeter-wave radar technology has the advantages of high-precision resolution,multi-target detection capability,and stronger anti-interference.But the previous vital signs detection based on millimeter-wave radar only supports monitoring of a single-user in a specific postures and due to the influence of harmonics,the measurement accuracy is so low that cannot meet the application demands.In order to solve the limitations of the current non-contact measurement method and improve the performance of the vital signs monitoring system based on millimeter-wave radar,this paper has done the following work.1.This paper proposes a multiple signal classification algorithm based on Gaussian smoothing,which can accurately determine the angle of multi-target vital signs and improves the target detection accuracy and multi-target discrimination ability of the noncontact vital sign detection system.It can accurately distinguish target position,solve the problem of false detection of target overlap,and increase the reliability of multi-target recognition.2.This paper proposes a novel algorithm for extracting vital signals based on principal component analysis.It extracts multi-reflection point signals to perform principal component analysis instead of target peak point signals analysis to find the component that contains the largest displacement of the target chest cavity.This method enhances the signal energy to cope with different human postures and improves the robustness of the system.3.This paper proposes an innovative life signal separation algorithm based on variational modal decomposition.It avoids the influence of harmonics and environmental noise,the respiration and heartbeat signals obtained by this method have more obvious sine property and better signal-to-noise ratio,so that the system can be applied to complex indoor environments.4.This paper builds a real multi-target vital signs monitoring prototype system on the system on chip based on the proposed algorithm to realize real-time monitoring of multitarget vital signs.The system not only monitors the target’s normal vital signs,but also supports the monitoring of abnormal vital signs(such as respiratory arrest)and alarms.Experiments show that the constructed system can complete the real-time monitoring requirements of 4 human targets,and display the real-time breathing rate,heartbeat frequency and location of the tested target.It shows that the median error of the measurement is within the range of 0.25 bpm.