Research On Health Monitoring Based On Wireless Communication Signal

With the rapid development of wireless communication technology,device-free intelligent health monitoring technology has gradually become a popular research direction.Compared with traditional perception methods based on sensor devices and radar,wireless communication signal perception technology,represented by Wi-Fi signals,has the advantages of solid portability,high privacy,and the ability to utilize the deployed infrastructure communication facilities fully,and has received widespread attention.Domestic and foreign scholars have researched health monitoring technology based on Wi-Fi signals.While achieving rich results,some problems and challenges still need to be addressed.This article focuses on the issue of improving the accuracy and robustness of respiratory detection in complex scenarios.It conducts in-depth research on Wi-Fi-based respiration detection technology,making full use of signal processing technology and statistical methods,aiming to improve the detection accuracy and robustness of the respiratory detection system.The main research work and achievements can be summarized as follows:(1)By studying the Fresnel zone model and dynamic and static propagation models in wireless signal propagation,a respiratory model of the human body is constructed,and a respiratory detection system is built based on this.Different data collection environments are also built to establish a dataset.(2)To address the difficulty of obtaining subcarriers sensitive to human respiration in complex scenarios,this article proposes a respiratory signal extraction method based on multiple selection mechanisms,designs an antenna link selection method based on dual variance and a subcarrier selection method based on similarity to extract subcarriers most sensitive to respiratory signals,and use them to construct respiratory signal waveforms that are more in line with natural respiration.(3)To address the problem that using a single subcarrier selection method in different scenarios has certain limitations and leads to low detection accuracy,this article adopts three subcarrier selection methods to obtain three different respiratory signal waveforms,and designs a dynamic time warping-based process for selecting the optimal respiratory signal waveform to evaluate the periodicity of different respiratory waveforms in the time domain and select the most periodic respiratory waveform.This thesis conducts many comparative experiments on the established datasets to verify the performance of the proposed methods.The experimental results show that in all complex scenarios,the average detection error of the proposed method is 0.725bpm,which is better than the state-of-the-art single-target respiration detection system,reducing the average detection error by 0.21bpm and improving the accuracy of respiration detection in complex scenarios to a certain extent.