GPS Ambiguity Resolution On-The-Fly (AROF) procedures process DGPS code and carrier phase measurements to deliver in real-time a very accurate position estimate. They are very attractive to the civil aviation community, but questions still remain about their reliability. The aim of this article is to present a state of the art analysis of the characteristics of the GPS AROF procedures through comparison of several methods. It constitutes a contribution to the evaluation of the applicability of these procedures to precision landings. Using requirements proposed for CAT II/III landings, constraints are extracted on four identified sets of parameters, such as performances, processing modes, means of control and working assumptions. Then, these parameters are determined for four particular procedures, namely the LSAST (Least Squares Ambiguity Search Procedure), MAPAS (Maximum A Posteriori Ambiguity Search), DIAS (Direct Integer Ambiguity Search) and FASF (Fast Ambiguity Search Filter) methods, following principles presented in corresponding publications and after an adaptation of algorithms. This determination is done on a theoretical and practical basis in several configurations. Mathematical developments provide an analysis of performances, and simulations of data for L1 measurements, pseudolites and perturbations such as multipath errors are used to assess the values of the parameters.