Study Of Ultra-wideband Localization Under NLOS Environments

The Ultra-Wideband (UWB) signal is a leading candidate for the application oflocalization under Non-line of Sight (NLOS) environments because of its high timeresolution and capability to penetrate the wall. The thesis concerns the application oflocating indoor target with through-the-wall measurements, as well as the study ofUWB location technique under NLOS Environments.Due to the NLOS propagation, a direct path (DP) excess delay exits in the time ofarrival (TOA) of the received signals. With the conventional location approaches, thelocation estimate of the target contains a considerable ambiguity caused by the NLOSerror. Considering the NLOS error mitigation for the through-the-wall localization, thethesis analyzes the multipath component (MPC) of the received signals. A model of thewaveform distance that the signal traveled along the through-the-wall propagation pathsis established. Then, a computation of the approximate TOA between the target and thebase station (BS) is presented based on the aforementioned through-the-wallpropagation model. For the application of locating indoor target with NLOSmeasurements, in order to mitigate the NLOS error, a location estimation based onthrough-the-wall propagation model is proposed in the thesis.To a stationary target in the surveilliance area, the localization error is related tothe geometrical arrangement of the base stations. The location precision in theorical alsodepends on the geometrical arrangements of the base stations. Considering the error inthe reference location of the base station, the Geometic Dilution of Precision (GDOP) isformulated based on the through-the-wall propagation model. The GDOP in the thesurveilliance area provides an important reference for the system design. With thedistribution of the GDOP in the the surveilliance area, an optimization of thegeometrical arrangements of the base stations is proposed, which will further improvethe performance of the location systems.Finally, numerical simulations are performed to evaluate the approaches proposedin the thesis. Simulation results verify the validity of improving positioning accuracywith the location estimation based on through-wall propagation model.