This work is a study about radiolocation of mobile terminals in TDMA systems. In agreement with the American legislation and also because of the extreme competitiveness among wireless operators in order to conquer new markets and provision new services, techniques that allow the localization of the mobile terminal have been intensely researched. One of these techniques uses time of arrival (TOA) to estimate the mobile position. Since the original TOA measurement is not accessible, TOA should be estimated from TA (Timing Advance). The value of TA is sent by the BTS to the MS by means of a signaling channel, which is assigned from a set of discrete values that allows one to draw circular contours around each receiving antenna, the location of the mobile is the best intersection of all those estimated contours. The resultant error in estimated TOA might be large by random error and if the wrong TA will be selected. Given that only TA is know, the problem is to find an estimator and its error for the actual TOA. Although literature mentions diverse processes to solve this problem, in order to distinguish between these and the considered here, we will call the process discussed here of Successive Refinement of the Uncertainty.