Simulation of a first prototypical 3D solution for Indoor Localization based on Directed and Reflected Signals

Abstract: We introduce a solution for a specific case of Indoor Localization which involves a directed signal, a reflected signal from the wall and the time difference between them. This solution includes robust localization with a given wall, finding the right wall from a group of walls, obtaining the reflecting wall from measurements, using averaging techniques for improving measurements with errors and successfully grouping measurements regarding reflecting walls. It also includes performing self-calibration by computation of wall distance and direction introducing algorithms such as All pairs, Disjoint pairs and Overlapping pairs and clustering walls based on Inversion and Gnomonic Projection. Several of these algorithms are then compared in order to ameliorate the effects of measurement errors.