The integration of mapping techniques with suitable methods for the characterization and visualization of propagation patterns may enhance the targeting of critical arrhythmic areas, thus optimizing the ablative treatment of atrial arrhythmias. In this study, we tested the feasibility of an innovative approach for the automatic determination of activation and velocity maps from sparse data as provided by electro-anatomic mapping systems. The proposed algorithm reconstructed the activation process by a radial basis function (RBF) interpolation of mapping point latencies. Velocity vectors were analytically determined by differentiation of the interpolation function. The method was tested by a multistate cellular automaton simulation model, implemented on a CARTO-derived atrial endocardial surface, and reconstruction accuracy was evaluated as a function of the number of mapping points. The RBF algorithm accurately reconstructed wave propagation patterns in simulated tissues with homogeneous and heterogeneous conduction properties, consistently with the data access afforded by clinical practice. These preliminary results suggest the possible integration of the method with clinically-used mapping systems to favor the identification of specific propagation patterns and conduction disturbances.