Location estimation in NLOS environments using geometric centering optimization methods

Non-Line-of-Sight (NLOS) propagation problem is a main difficulty which severely degrades the accuracy of localization techniques in wireless networks. When very few Line-of-Sight (LOS) data are available, many known NLOS positioning algorithms are of bad performance. This paper proposes several geometric centering optimization methods for mobile station (MS) positioning under severe NLOS environments. These centering methods consider several center points inside a feasible region, which is defined by the Time-of-Arrival (TOA) measurements, as the estimated positions. Chebyshev center, analytic center and improved analytic center are studied and discussed respectively. Simulation results show the improved analytic center performs best comparing with other known algorithms.