Geometric sensitivity of localization using airborne mobile anchors with volume probabilistic multilateration

Locating fixed sensing devices with a mobile anchor is attractive for covering larger deployment areas. Probabilistic localization has been shown to reduce error over deterministic approaches, yet the performance sensitivity to the geometric arrangement of anchor beacon positions remains unexplored. This paper presents a detailed description of a new RSSI-based localization algorithm which uses a volumetric probability distribution function to find the most likely position of a node by information fusion from several mobile beacon radio packets. Results show a significant performance improvement of up to 80% over existing approaches. Secondly, simulations are conducted to determine the effect of the 3-D geometry of localization broadcast packets. Compared to both a symmetric circular arrangement of beacons, and an arrangement of beacons in a line, if beacon positions are allowed to move to a more asymmetric arrangement with a range of angles and heights, the localization error is reduced to approximately 1 metre or by up to 96% compared to a poor geometry. This is despite individual ranging errors of much more than 1 metre.