Performances of an RSSI-based positioning and tracking algorithm

This paper reports the results of a positioning and tracking algorithm for indoor environments based on simulated and pre-computed attenuation map values. The localization is performed through a global optimization that minimizes a cost function computed in the data-space, which is the attenuation reference map relative to the environment under test. The tracking is implemented introducing a correlation between the current position and the previous ones. Two environments of different size, shape and characteristics are chosen for the algorithm validation. In the worst case the localization is performed with a median bias error of 0.71m. The overall median bias error when using tracking features is below 0.75m computed considering 4 distinct trajectories for each environment.