In this paper, we give an analysis of the performances of two fingerprinting-based localization schemes using IR-UWB signaling. The first scheme uses channel’s power delay profile as fingerprints and least square error as matching function. The second one use channel 2nd. order statistics as fingerprints and maximum likelihood estimation as matching function. The analysis allows us to model some key parameters as the grid dimensions and the number of samples in building the database, which help in selecting optimal setup of the localization system offline. The performances limitations are given in term of upper bound on the mean square error (MSE) which is more interesting for practical purpose.