Recently, the Network and Cloud Computing team at Tianjin University has made significant progress in the field of wireless sensing. They turn ordinary WiFi signals into "butlers" in the home environment, analyzing changes in human activities through WiFi signals to determine people's needs and enabling smart home devices to provide corresponding services. The relevant research results were recently published in the academic journal "Interactive, Mobile, Wearable, and Ubiquitous Technologies" of the International Computer Society. At present, most smart home devices rely on user commands or specific sensors to work, making it difficult to continuously and imperceptibly obtain real-time user status and needs. If you want to obtain more user data, you need to use cameras or wearable devices. In response to these issues, the team has developed a new high-precision sensing application: without the need for users to wear any devices, it can accurately determine where the user is, what state they are in, and what they are doing based solely on WiFi signals at home, thereby enabling smart home devices to provide more user-friendly services. The principle is that human movement will disturb WiFi signals, and by analyzing these disturbances through intelligent algorithms, one can perceive the position, posture, and behavior of the human body. However, to make this application "fly into people's homes", two major challenges need to be addressed. Firstly, deployment is difficult. The traditional deployment method requires professionals to hold devices, enter the room to collect WiFi signals, and manually input the location information of devices such as routers and smart speakers. The related operations are complex and time-consuming, making it difficult for ordinary users to complete. Secondly, there is inaccurate perception. WiFi signals will reflect or refract when encountering walls, furniture, etc. In real home environments, signal propagation paths are complex and diverse. However, researchers design applications based on ideal laboratory environments, resulting in significant perception errors in practical applications. In response to the difficulty of deployment, the team has transformed the common household sweeping robot into a "data collector" to specifically collect environmental information at home. During daily cleaning, the robotic vacuum cleaner will draw two maps for the aforementioned system: one is the physical space map of the home, and the other is the WiFi signal map. At the same time, the robotic vacuum cleaner can automatically locate and record the positions of devices such as routers and smart speakers with an accuracy of 10 centimeters. In response to the issue of inaccurate perception, the team has constructed a model for real family environments. This model enables the above applications to "see" the propagation path of WiFi signals in complex environments. Tong Xinyu, a member of the team and associate professor at the School of Computer Science and Technology, Tianjin University, said that in the future, if users have routers equipped with the above applications and let the floor sweeping robot sweep the floor once, they will be able to enjoy attentive service. (New Society)