Wi-Fi based occupancy detection in a complex indoor space under discontinuous wireless communication: A robust filtering based on event-triggered updating

Demand-driven control of building heating, ventilation and air conditioning (HVAC) and lighting systems warrants an attractive energy saving potential. A reliable and accurate occupancy detection technique should be available for estimating the demand. To this, many occupancy detection methods have been developed either based on physical sensors, cameras or information & communication technology (ICT). Wi-Fi based detection is one of the ICTs that uses Wi-Fi signals (e.g. from smartphones) to represent occupants. Nowadays, new smartphone models all have the battery-saving function that will switch off Wi-Fi communication in the idle mode, causing significant detection uncertainties. This challenge has not been addressed in the current Wi-Fi based occupancy detection methods that still assume or manipulate a continuous Wi-Fi communication. Moreover, most studies are conducted in relatively simple settings with small-scale occupancy that are far from a complex indoor space. Thus, this study investigates the Wi-Fi based occupancy detection in a complex indoor space with large occupancy. An event-triggered updating method is proposed to cater for the discontinuity in Wi-Fi communication of the smartphones. A location filter and a non-human media access control (MAC) address filter are proposed to screen out irrelevant Wi-Fi devices. The proposed event-triggered updating method is able to improve the detection accuracy from 77.3% to 96.8%. The proposed location filter and non-human MAC address filter prove to be effective in removing irrelevant outside and non-human devices. The proposed event-triggered updating method can be applied to existing installations with minimum cost regardless of the Wi-Fi communication continuity.