WiFi RSSI and inertial sensor based indoor localisation system: A simplified hybrid approach

Indoor Localisation has been an interesting and significant area of research in the recent times because of its diverse applications. There is always a trade-off between the number of devices used and the accuracy obtained for any indoor localisation system. In our work we make use of WiFi access points, a smart phone with inertial sensors and a server. Since RSSI value is not very much reliable, we go with a hybrid approach where fingerprinted RSSI values of different access points are used to initialise the position using Maximum Likelihood estimate and thereafter the inertial sensors takes over the localisation. After certain distance is travelled, the error accumulated is corrected by access points' RSSI values. The proposed simple hybrid system
¹
yields an accuracy of minimum 2m consistently, with only 3 access points in the indoor space. Hence, the system is overall better than most of the indoor localisation systems present.