Doppler Radar Sensor And Applications In Non-contact Bio-signal Detection

In the field of wireless Doppler sensing,bio-signals refer to heartbeat,respiratory,walking,gestures,gait and other human-motion-related signals.Recently,with the rapid development of RFIC,miniaturized Doppler radar sensor for bio-signal detection attracts more and more interests.Compared with traditional methods,the approach based on radar sensor has promising application prospect with natural advantage of non-contact detection.Nevertheless,there have been many limitations in previous methods by employing radar sensor before this thesis,resulting in inevitable nonlinearity when recovering Doppler information.These problems severely hinder the practical application of Doppler radar sensor in non-contact bio-signal detection.In order to overcome above problems,this thesis covers a series of researches on the Doppler radar sensor architecture and the linear phase demodulation algorithms,and solves previous difficulties progressively.Moreover,experimental results also show the proper linear reconstruction of high-dynamic-range and weak motions respectively based on the same architecture.The main contributions in this thesis are as follows:Firstly,the nonlinear and noise sources of Doppler transceiver are studied systematically in this thesis.An optimal Doppler sensor architecture based on low-IF receiver is proposed.In the meanwhile,the quadrature imbalance is eliminated in the digital domain.Thus,main problems existing in the previous receivers are theoretically solved.Based on such architecture,24GHz and 5.8 GHz radar sensors are also designed and manufactured.Next,the dynamic center tracking algorithm is proposed for linear demodulation of high-dynamic-range motion.The extended differential and cross-multiply algorithm works with the process mentioned earlier to solve phase wrapping problem.In the experiments part,the same system detects both high-dynamic-range human walking,and micrometer-scale weak motion.Finally,to solve the enormous challenge of acquiring small-scale heartbeat and respiration signals from random body movements,we further take advantages of match filter algorithm in this thesis.The experiments show that vital signs can be accurately recovered by this non-contact detection approach.The above researches in this thesis effectively solve some practical problems of non-contact Doppler radar,showing a wide range of potential applications in the future.