Bayesian ranging for radio localization with and without line-of-sight detection

We consider Bayesian ranging methods for localization in wireless communication systems. Based on a channel model and given priors for the range and the line-of-sight (LOS) condition, we propose range estimators with and without LOS detection. Since the pdf of the received frequency-domain signals is unknown, we approximate the maximum-a-posteriori (MAP) and the minimum mean-squared error (MMSE) estimators. The promising ranging accuracy obtained with the proposed estimators is demonstrated by Monte Carlo simulations. We observe that the approximate MMSE estimators outperform the approximate MAP estimators. In addition, we find that including LOS detection in the approximate estimators, while adding a higher computational complexity, has no major impact on the ranging performance.