Wi-Fi-based Positioning in a Complex Underground Environment

Underground mining tunnels constitute a very particular environment for radio wave propagation – with characteristics of both indoor and outdoor “regular” environments. This paper shows how a general-purpose WiFi-based indoor positioning system, OwlPS (Owl Positioning System) was adapted to work in that particular environment. A series of experiments, conducted in a formerly exploited gold mine at 70 metres below the surface, across about 400 metres of drifts, is then introduced; it primarily aims at determining the positioning accuracy that can be reached in such a context with a Wi-Fi-based positioning system using the signal strength at 2.4 GHz. The results obtained are improved with a filter, and the mean Euclidean distance error is under 10 metres in most cases when the terminal is carried by an operator; this makes OwlPS usable as is for asset management and emergency positioning of workers underground.