Non-invasive human thermal adaptive behavior recognition based on privacy-friendly WiFi sensing in buildings: A review

By analyzing thermal adaptive behavior (TAB), we can access the occupant’s thermal comfort in real time and control the heating, ventilation, and air conditioning (HVAC) system accordingly to reduce energy consumption in buildings. Most existing methods are based on wearable devices or cameras to collect occupant behavioral information. Although these methods can effectively identify occupant behavior, they have the problem of violating user privacy. With the development of wireless technologies, human activity recognition using WiFi has the advantages of being non-invasive, privacy-friendly, and light-independent. Therefore, non-invasive TAB recognition based on WiFi technology holds great promise in human thermal comfort. However, existing research on TAB recognition based on WiFi technology lacks comprehensive and consistent conclusions. Thus, in this paper, we have surveyed the literature in recent years to guide in this area. In addition, we present the challenges and future perspectives faced by existing WiFi-based TAB technologies, e.g., developing high-quality WiFi sensing datasets to advance the field of human thermal comfort. We hope this review will guide researchers in recognizing the great promise of WiFi sensing applications for TAB recognition in smart buildings.