Modeling of precise indoor positioning system

Abstract

Precision indoor positioning system intended for fire fighters and members of rescue corps is currently developed at the University of Pardubice. Each team member is equipped with a transmitter, generating a unique UWB OFDM signal in periodic pulses. Around the surveyed building several receiver stations are deployed receiving the UWB signals and measuring their time of arrivals. Several TDOA methods are used namely 2D iterative, 3D iterative and the Ralph Bucher method to extract transmitters' positions. Unfortunately each received signal contains not only a direct signal but also its echoes due to reflections at walls. That is why the signals are at first correlated with the known transmitted signal to obtain a rough estimation of the pulses time positions and then a spectral analysis method (SAM) is applied to separate the direct signal from its echoes to get precise time delay of the direct signal (the pseudorange). In this paper a virtual experiment with a model of this system is described and position accuracy is analyzed.