Research on regional personnel monitoring technology based on channel state information

Abstract

In view of the shortcomings of traditional camera and sensor type monitoring, such as blind spots, limited recognition distance and sensitive scene limitations, this paper proposes a human activity detection and monitoring method based on channel-state-information (CSI). The CSI information of the WiFi signal in the monitored area. Next, use the Butterworth low-pass filter to detect and remove abnormal data. And then use the principal component analysis (PCA) to extract the features of the human body posture, gait information, and number of people model; Learn to build a number recognition model for CSI data; because everyone is different, gait information can be used as an ID for human identification to identify different identities, and the human gait information based on Dynamic Time Warping (DTW) can be Effective identification, so as to play the effect of regional environmental monitoring. In the experiment, this method can achieve 92% capture performance for human gesture recognition, more than 93% error in indoor area recognition is less than 1, and the correct rate of gait recognition is up to 95.2%.