A Scheme for Indoor Localization through RF Profiling

We present a certain approach to indoor location tracking based on sensing the strength of received RF signal. Our scheme employs a database of signal strength readings from known locations collected during a prerequisite profiling stage. Subsequently, the problem of estimating the location of a node emitting an RF signal from an unknown place boils down to data mining in the database-to select a best matched set of profile points and then average the coordinates of those points into an approximate location of the tracked sender. We show preliminary experimental results which confirm our hope that this approach will result in a better accuracy than, e.g., triangulation based schemes.