Two NLOS error elimination and TOA location algorithms
Duan Kai-yu · Signal Processing · 2008
In cellular network location,the excessive time delay caused by NLOS environment (NLOS error) is the main reason which degrades the location accuracy mainly.This paper separates the combined noise of the NLO$ error and the system measurement er- ror into mean value and stochastic part.The Kalman filtering algorithm has the property that its' output has no relationship with the vari- ance of the noise.According to this property,the stochastic part can be eliminated by the Kalman filtering algorithm needing only the va- fiance of the system measurement rather than the exact variance of the combined noise.Then the location of the mobile station (MS) can be estimated by using a proposed least square (LS) method or optimization algorithm according to the relationship of the mean value and the distance between the MS and the base station(BS).This paper also summarizes the a priori information about filtered distance- error in the light of the results of computer simulations.The second NLOS error elimination algorithm is proposed based on the a priori information.The computer simulations indicate the proposed algorithms can eliminate the NLOS error effectively with higher accuracy and robustness.