Research On Behavior Perception Technology Based On WiFi

Perceptual recognition technology is a combination of hardware and data processing technology,so as to perceive and recognize the surrounding environment.With the development of Internet of things technology,perceptual recognition technology is gradually applied to human life,such as radar based population density estimation method,RFID based human activity recognition technology,WiFi based object recognition and positioning,etc.Wireless signal can not only be used for communication,but also for perceptual recognition because of its modulation of surrounding environment information.At present,the perception recognition technology based on WiFi mainly includes: the perception recognition technology based on signal strength RSSI and the perception recognition technology based on channel state information CSI.RSSI is a coarse-grained description of wireless signals,which has poor stability,which makes the accuracy of the perception recognition system based on RSSI not high.Therefore,more and more researchers begin to pay attention to CSI based perceptual recognition.In this thesis,the CSI based perceptual recognition technology is discussed in depth.Firstly,the related research status at home and abroad is summarized,and the classification of wireless signal based perceptual recognition technology is introduced.On this basis,two CSI based perceptual recognition systems are designed and implemented: CSI based outdoor human behavior recognition system and CSI based basic sign language recognition system.In the CSI based outdoor personnel behavior discrimination system,because different activities will lead to different CSI signals,we extract CSI signals under different activities of people outside the door to filter and extract features,so as to realize the activity identification of people outside the door;In the basic sign language recognition system based on CSI,we use the CSI signal which contains the information of frequency domain and time domain as the feature,and design the denoising method to preprocess the original CSI signal.Finally,we use DTW algorithm to recognize the meaning of sign language,so as to realize the auxiliary traffic of language barrier.The effectiveness of the proposed scheme is verified by system experiments.