Human Vital Signs Monitoring Based On UWB Radar Sensors

The Ultra Wide Band(UWB)signal has high resolution,strong penetrability and anti-interference.It can penetrate many non-metallic building materials(bricks,wood,drywall,concrete,etc.).UWB signals are favored by a wide range of applications such as military equipment and urban security.With the rapid development of cities,people are increasingly demanding medical health.The existing medical methods can not fully meet all the needs of people in different scenarios.Nowadays,the monitoring methods of vital signs such as respiratory characteristics and heartbeat characteristics are mostly electrode contact monitoring,which makes it impossible to monitor in special scenes.Therefore,based on the significance of UWB radar sensors for non-contact monitoring of human signs,this thesis has carried out a series of research contents.The content is mainly summarized as follows:1.We designed a portable UWB radar sensor human body monitoring system and and select the Gaussian single-cycle pulse UWB radar as the signal transceiving device.We analyze UWB antennas including planar Yagi ultra-wideband antennas and Vivaldi UWB antennas and select the smaller and better directional transmit and receive antenna.In order to verify the reliability of the system,we use the highly accurate medical ECG monitor to extract weak physical data at the same time.The results of the two kind of experiments are compared to obtain a quality evaluation.2.For the non-stationary and nonlinearity of human vital sign signals,the Empirical Mode Decomposition(EMD)algorithm and the Variational Mode Decomposition(VMD)algorithm are applied to such signals.Since the non-contact monitored echo contains more noise and clutter,a series of preprocessing methods including removal of coherent background noise and clutter filtering are adopted.Comparing the raw data with the preprocessed data,the preprocessed signal is more pure.3.A series of experiments were carried out on the extraction of respiratory characteristics and heartbeat characteristics,which were completed by the UWB radar sensor non-contact monitoring system.The experiment included the monitoring of a single UWB radar sensor and multiple UWB radar sensors.The single UWB radar sensor monitoring experiment also includes wall monitoring.The experimental results show that the VMD algorithm can extract and separate the signs to obtain the respiratory characteristics and heartbeat characteristics at the same time,without modal aliasing.4.Using the principle of control variables,for the leakage judgment phenomenon that occurs when monitoring multiple targets in the experiment,changing the distance of the multi-target arm to perform multiple experiments.Experimental conditions to avoid missing judgments are obtained.5.Since the human body target always has different motion states at different times,different time-frequency analysis methods are used to obtain the motion law of the human body changing with time.Time-frequency analysis methods include short-time Fourier transform,combined modal decomposition method and Hilbert transform method,which are EMD-Hilbert Transform Method and VMD-Hilbert Transform Method.A series of experiments are carried out on various situations of the human body’s.The time-frequency algorithms effects under different motion states are analyzed.