WiFi based Indoor Positioning System using Machine Learning and Multi-Node Triangulation Algorithms

The Global Positioning systems are limited by their precision considering the unreliable dependence on the number of satellites received at the instance. Indoor Positioning Systems require much more reliable systems which can be precise up to 2 ft hence ruling out the option of using GPS. The paper proposes a technique to utilize the existing infrastructure of Wi-Fi routers and mobile hotspots in the indoor environment to build a precise localization system. The paper integrates the correlation between RSSI values and transmitter-receiver distance with a model for coordinate system of indoor environment to find most probable coordinates of the receiver when multiple transmitters are present using novel multi-node triangulation algorithms. The IOT-enabled system is affordable, optimized and stand-alone, along with being non-line-of-sight, and offers higher precision than conventional GPS. Applications may be extended to localization system of UAVs (Unmanned Aerial Vehicle) and UGVs (Unmanned Ground Vehicles).