Small-scaled Change Detection And Application Research For Wireless Body Area Network

Nowadays the technology advancement improves people quality of life which makes researchers and scientists to paid more attention to make their lives healthy and long lasting.As a result,a huge investment in research on health monitoring is dedicated.The human-centric wireless network,also called wireless body area network(WBAN)provides a method for health monitoring.With the development of wireless communication technology,non-contact WBAN based on radio frequency(RF)signal has been more widely used in health monitoring due to its non-invasive characteristics in recent years.Due to the wide deployment of Wi-Fi infrastructure and fine-grained channel state information(CSI)on commercial Wi-Fi devices,Wi-Fi has become a promising solution for contactless sensing based on RF signals.In this thesis,we use Wi-Fi CSI based contactless sensing to detect the small-scale change of human breathing.The study and analysis describe Wi-Fi CSI-based non-contact respiratory detection method.The main job is to expound the principle of CSI-based respiration detection,to design breathing information extraction method and to detect various different breathing.The main contribution of this thesis can be summarized as follows:(1)Expounding the principle of CSI-based respiratory detection: By modeling the human respiratory process,the small change of chest and abdomen caused by breathing is analyzed.This small movement has an effect on the propagation path of wireless signal,which is recorded by the received signal and is depicted in the form of CSI.(2)Designing a breathing information extraction method capable of detecting different breathing: Comparing the CSI amplitude information and phase information,the CSI amplitude information is selected as the representation of respiration.And then the CSI data preprocessing methods are executed for denoising.Specifically,we propose a subcarrier selection method based on the scoring mechanism to select the subcarrier sensitivity to the breathing activity,and then we leverage Pauta criterion to remove outliers and the wavelet filtering to denoise.After preprocessing,peak detection algorithm or short time Fourier transform(STFT)technique are utilized for breathing information extraction.(3)A variety of experiments are designed to evaluate the performance of the proposed Wi-Fi CSI based respiratory detection method.The results show that our proposed respiratory detection method can not only detect normal breathing,but also detect abnormal breathing conditions,and also identify sleep apnea syndrome.In summary,the proposed method is a preferred nonwearable alternative to the conventional contact sensing methods,and provides an effective mean for long-term home respiratory monitoring.In addition,the research preliminarily reveals the potential of wireless sensing in the application of small-scale change detection.