UWB Multicell Indoor Localization Experiment System with Adaptive TDOA Combination

This paper proposes an accurate UWB multiple-cell localization system which adopts an adaptive TDOA combination algorithm to reduce the localization uncertainty. The time difference of arrival (TDOA) is adaptively combined in such a way that only makes use of the best directional motion increments to reduce both the random channel noise and the geometric errors. Furthermore, maximum signal to noise ratio (SNR) TDOA combining and tracking error controlling are employed to reduce the localization uncertainty. Unlike the conventional TDOA equation solutions, the algorithm avoids using the nonlinear triangular calculation and the TDOA matrix inversions to improve the performance with minimizing dilution of precision (DOP). This method shows a very good competition compared to the existing methods in tracking/filtering process and even achieves a better Cramer-Rao bound (CRB). The hardware including the moving target (tag) and the receiver system had been designed with a very low cost, and long battery life. Finally, after giving the analysis of DOP, the real experimental system reveals that it outperforms the other methods in term of CRB evaluation and error (bias) mitigation.