On Achievable Bound for Non-line-of-sight Geolocation

ABSTRACT Non-line-of-sight (NLOS) geolocation becomes an important issue with the fast development of mobile communications in recent years. Several methods have been proposed to address this problem. But we believe that a more comprehensive study on the best geolocation accuracy that these methods may possibly achieve is imperative. In [1], we developed a unified analysis of the Cramer-Rao Lower Bound (CRLB) applicable to NLOS geolocation. However, our further study reported here shows that the CRLB is not achievable in general cases. In this paper, we first present a new result on achievable bounds for NLOS geolocation with time-of-arrival (TOA), maximum likelihood estimation (MLE) and signal strength (SS) based positioning methods. Their physical interpretation is consistent with that of the CRLB and provides an insightful direction towards construction of NLOS geolocation algorithms. We then evaluate the difference between the achievable bound and the CRLB obtained earlier, and verify quantitatively that the CRLB can only be achieved in some specific cases, e.g., when the signal-to-noise ration (SNR) is same for the signals received at each base station. Some numerical results are reported based on simulation experiments.