Non-Contact Vital Signs Detection Technology Based On Millimeter Wave Radar Subtitle

With people paying more and more attention to the health of life,the vital signs detection technique is of great importance.Respiration rate and heart rate are the most basic vital signs of the human body.Currently,the conventional detection methods for these two frequencies need to be in direct contact with the monitored object,which will not only cause discomfort to the human body,but also greatly restrict its application occasions.Therefore,the non-contact detection method has broad development prospects and value.Doppler radar has the advantages of all-weather,all-weather,and strong penetration,in recent years,non-contact vital signs detection technology based on millimeter wave radar has gradually become a hot research direction.Although this technology has been initially applied at this stage,there are still some challenges hindering this technology from being widely used,the most typical of which is the unconscious body movements during the detection process.This kind of random body movements is equivalent to or even greater than the displacement of the chest,and can be a very strong noise source,which will greatly reduce the accuracy of vital signs detection.Focusing on the random body movements,this thesis proposes a fast and robust vital signs detection algorithm system and builds a set of non-contact vital signs detection system based on a 77 GHz millimeter wave radar.The main work and innovations of this thesis are as follows:(1)During the data collection,there may exist unconscious random body movements,which reflect as outliers on the vital signs signal.This thesis proposes a multi-channel-weighted Kalman smoother(MCKS)algorithm for the life signal feature enhancement.With the received signals from multiple channels made full use of,this thesis uniformly weights the variance of the observation vector at each time,and reduces the weight corresponding to the observation vector with outliers through smoothing,thereby suppressing the influence of random body movements.In this case,the features of life signal are effectively enhanced.In the environment with random body movements,this method can increase the local signal-to-noise ratio around the respiratory rate and heart rate to more than 7.5d B and 2.5d B,respectively.(2)A regional hidden Markov model(RHMM)is proposed to carry out accurate estimates of respiratory rate and heart rate by exploiting the underlying slowly-varying characteristics of these vital signs.The hidden Markov chain is introduced to model the frequencies of vital signs,based on the optimized signal acquired by MCKS.Unlike the traditional HMM with tiresome computation burden,the proposed RHMM,which takes full advantages of the effective frequency ranges of respiratory rate and heart rate,has the capacity of realizing the fast acquisition of vital signs.Results of multiple objects at five even moments can be updated every 7.5s,and the average error under the environment of random body movements is less than 2bpm.(3)In view of the accuracy,safety,convenience,high efficiency,real-time and scalability that the human vital signs detection system needs to meet,relevant architecture design and specific demonstration system construction are carried out.The whole system is composed of LFMCW signal transceiver module,data storage module on the radar side,with the signal processing module and result displayed module on the PC side.During the data collection,when the transmitted signal realizes multi-target perception,the radar will sample and store the received signal.At the same time,the PC side will start the vital signs detection after receiving the relevant data.