Research On Vital Sign Signal Detection Algorithm Based On FMCW Millimeter Wave Radar

Millimeter wave radar-based non-contact vital sign detection technology is different from traditional optical and infrared detection,and has a very wide application prospect not only in the medical field,but also in the fields of motion monitoring,security prevention and disaster rescue.Frequency-Modulated Continuous Wave(FMCW)radar is widely used in applications that require high accuracy due to its high range accuracy and ability to provide continuous distance measurement.Since the vitals signal is a weak signal and easily disturbed by the environment and system,the study of the vitals signal processing algorithm is the focus of current research.Firstly,in this thesis,the basic principle of vitals detection using FMCW radar in detail are introduced,and the vitals signal and FMCW signal are separately modeled and simulated.On this basis,the theoretical derivation of IF signal expression,FMCW radar ranging principle and distance resolution is done,and the sensitivity of the phase of IF signal to thoracic micromotion is analyzed to determine the use of the phase of IF signal for vital sign detection.The simulated IF signal is put through the process of Range-FFT,target phase extraction and deconvolution to verify the feasibility of the vital sign detection principle.Secondly,the band-pass filter is used to solve the problems that cannot be solved by the band-pass filter,such as the weak random signal of vital signs,which is susceptible to noise interference and the heartbeat signal frequency is easily covered by the respiratory harmonics.The simulation results demonstrate the good noise reduction effect of VMD algorithm.The Least Mean Square(LMS)adaptive filtering algorithm is used to cancel the respiratory harmonics,which can effectively solve the situation that the heartbeat signal is drowned by the respiratory harmonics,and the simulation results show that the LMS algorithm can well suppress the respiratory harmonics and extract the heartbeat signal.Thirdly,the Multiple Signal Classification(MUSIC)algorithm is improved by introducing the idea of simulated annealing to solve the raster search problem of MUSIC algorithm,which improves the accuracy of the algorithm and can estimate the respiration frequency and heartbeat frequency more accurately.The superiority of the proposed algorithm is verified by comparing the root-mean-square error and success probability of the algorithm through simulation,and the effectiveness of the proposed algorithm is verified by using the data collected in real scenes.Finally,the FMCW radar-based vital sign detection system is built,the AWR1642 radar sensor hardware system is introduced in detail,and the upper computer program including serial port configuration and control,data return and parsing and GUI display interface design are written using simulation software,which can display the changes of respiration and heartbeat signals in real time,and the respiration and heartbeat frequencies are estimated using SA-MUSIC The respiration and heart rate are estimated using SA-MUSIC algorithm,and the estimated results are displayed on the GUI interface.