Impact of Chosen Error Criteria in RSS-based Localization: Power vs Distance vs Relative Distance Error Minimization

The paper focus is on the minimum mean square error (MSE) estimation methods in RSS-based localization techniques. We in particular discuss three different error criteria: i) minimizing the mean square distance error (distance-based MSE), ii) minimizing the mean square power error (power-based MSE), iii) minimizing the mean square relative distance error (relative distance-based MSE). The goal is to assess the impact of the chosen error criterion on the estimate accuracy. Simulation results show that i) flip ambiguities in the position estimate may occur when the node to be positioned is very close to the signal source, ii) in presence of flip ambiguities, provided that RSS measurements are combined with information about the signal arrival direction, relative distance-based MSE minimization offers better accuracy in a bi-dimensional localization scenario¹.