Adaptive AR model based robust mobile location estimation approach in NLOS environment

One of the major problems to accurate mobile location estimation is the non-line-of-sight (NLOS) propagation caused by intermittent blocking of the direct path from transmitter to the receiver, which can be considered as measurement outliers. We first briefly discuss some new NLOS error mitigation algorithms. We then present a robust mobile location estimation algorithm, which uses a robust data cleaner to produce cleaned measurement data. We further use a novel higher order AR model to predict the residuals and achieve more accurate measured range signals with variable order and parameters. The simulation results show that the proposed approach has quite good robustness and efficiency.