Pre-Signal Processing of WiFi Sensing for Motion Detection <Abstract> Wi-Fi sensing, which utilizes Channel State Information (CSI) obtained from Wi-Fi signals, to predict human behavior and room states, has gained attention due to its low implementation cost and privacy considerations. This study constructs a system for motion detection using machine learning, utilizing CSI data acquired with commercially available Wi-Fi routers and the Nexmon software. By analyzing the CSI data, signals that may adversely affect motion detection are identified, and methods for their removal are proposed. Experimental results demonstrate the validity of the proposed feature analysis and motion detection system. |