Localising crowds through Wi-Fi probes

Abstract Most of us carry mobile devices that routinely disseminate radio messages, as is the case with Wi-Fi scanning and Bluetooth beaconing. We investigate whether it is possible to examine these digital crumbs and have them reveal useful insight on the presence of people in indoor locations, as the literature lacks any answers on this topic. Wi-Fi probes are generated sparsely and often anonymised, which hinders the possibility of using them for targeted localisation or tracking. However, by experimenting in three different indoor environments, we demonstrate for the first time that it is possible to extract from them some positioning information. Possible applications include identifying frequented regions where many people are gathered together. In the described experimentation with sniffing devices we adopted fingerprinting interpolation, which requires no survey phase and automatically adapts to changes in the environment. The same process can be carried out using the Wi-Fi access points already installed in the environment, thus allowing for operation free of installation, surveying and maintenance.