| With the development of the ageing population and diseases,vital signs detection plays an important role in evaluating human health status.Traditional vital signs detection methods usually used contact sensors to detect vital signs such as temperature,heart rate,and respiration.However,these methods not only cause uncomfortable,but are not suitable for long-term monitoring.Recently,millimeter-wave radar is a novelty non-contact vital sign detection system that can detect vital signs without the user interference,for example,heart rate,respiration,and movement.The millimeter-wave radar is widely used in many fields,such as medical diagnosis,health management and smart homes,and so on.However,it has been the remaining challenge for vital signs detection systems based on millimeter-wave radar.Firstly,due to the influence of the environmental interference and noise,the signal quality is affected.Secondly,accurate extraction and separation of vital signs signals need the extraction of relevant information from complex mixed signals.Errors can easily arise when noise frequencies fall within the range of respiratory and heart rates,particularly when weak heart signals are susceptible to respiratory harmonic interference.Based on the above,the design and implementation of signal processing algorithms are subject to higher requirements.Therefore,there are urgent issues that need to be addressed,which include accurate extraction of target information,reduction of noise and harmonic interference,ensuring the safety and reliability of the system,and avoiding potential effects on human health.To solve the above problems,in this paper,we proposed a novelty non-contact vital signs monitoring system based on 77 GHz millimeter-wave radar to achieve accurate extraction of the target breathing and heartbeat signals:(1)A 77 GHz frequency modulated continuous wave(FMCW)millimeter-wave radar-based vital signs monitoring system is developed,which mainly consists of four parts: front-end signal transmission and acquisition,signal simulation and processing,algorithm implementation and display interface for display.The system can continuously monitor the user’s health condition and has the advantages of non-contact,non-intrusive and anti-interference.(2)The Constant False-Alarm Rate(CFAR)algorithm is used to pre-process the echo signal,remove static clutter and extract the target signal.Phase extraction,phase expansion and phase differencing are used to obtain the phase change information containing the vital signs signals.Initial separation of the respiratory heartbeat signal is achieved by the designed wireless impulse response filter IIR.(3)The Variational Mode Decomposition(VMD)algorithm(RESE-VMD)is proposed to optimize the VMD by using relative entropy and sample entropy to search for the ideal parameters of the variational model and continuously update each mode component function and center frequency to effectively avoid mode mixing.At the same time,the appropriate signal components are selected to reconstruct the signal according to the sample entropy values and correlation coefficients.The optimization algorithm RESE-VMD achieves further removal of noise interference from the initially separated respiratory heartbeat signals and improves the efficiency as well as the accuracy of the detection.(4)To verify the feasibility of the built system,the performance of the system is tested and analyzed through simulations and real-life scenario experiments.By comparing the system with conventional signal separation methods such as wavelet transform and complementary ensemble empirical modal decomposition methods,in terms of different Gaussian white noise backgrounds and response times,it is verified that the proposed optimized algorithm,RESE-VMD,achieves better performance in separating respiratory and heartbeat signals,effectively reducing the computational effort while having a higher signal-to-noise ratio.Compared with contact detection device,the experimental results show that the accuracy of heart rate detection using the millimeterwave radar based non-contact detection system can reach about 98%,verifying the accuracy and feasibility of the system. |
