Switched model sets-based estimators for mobile localization in rough NLOS conditions

In this paper, we first present a novel switched model sets-based interacting multiple-model (SMS-IMM) algorithm for urban mobile location estimation. Two state-space model sets are considered. The model set 1 only covers the dynamics of a mobile station without taking the non-line-of-sight (NLOS) bias variation into account. The model set 2 consists of the modeling of the MS dynamics and the NLOS bias variation expressed as a random walk process. The IMM using the model set 1 can perform better than the IMM using the model set 2 when the line-of-sight (LOS) condition takes place. This phenomenon can be reversed when the NLOS condition occurs. The proposed SMS-IMM algorithm takes the advantage of the switching between the two model sets so that the sight conditions and the MS locations can be estimated more accurately. Next, we extend the SMS-IMM to a SMS-fuzzy-tuned-IMM for further performance improvement. Simulation results demonstrate the superior performance of the proposed algorithms.