RSS and sensor fusion algorithms for indoor location systems on smartphones

Abstract

Location-based applications require knowing the user position constantly in order to find out and provide information about user’s context. They use GPS signals to locate users, but unfortunately GPS location systems do not work in indoor environments. Therefore, there is a need of new methods that calculate the location of users in indoor environments using smartphone sensors. There are studies that propose indoor positioning systems but, as far as we know, they neither run on Android devices, nor can work in real environments. The goal of this chapter is to address that problem by presenting two methods that estimate the user position through a smartphone. The first method is based on euclidean distance and use Received Signal Strength (RSS) from WLAN Acces Points present in buildings. The second method uses sensor fusion to combine raw data of accelerometer and magnetometer inertial sensors. An Android prototype that implements both methods has been created and used to test both methods. The conclusions of the test are that RSS technique works efficiently in smartphones and achieves to estimate the position of users well enough to be used in real applications. On the contrary, the test results show that sensor fusion technique can be discarded due to bias errors and low frequency readings from accelerometers sensor.