Modulation-style recognition of radar signals is an important aspect in electronic warfare. A hybrid RBF training algorithm combined with artificial immune clustering and evolutionary programming is proposed and applied in modulation-style recognition of radar signals. In the algorithm, artificial immunology for data clustering was used to adaptively determine the amount and the positions of initial RBF centers according to input data set. Then the RBF network was trained with evolutionary programming. Computer simulations demonstrate that the recognition rate of different analog modulation styles designed by the RBF network in this method is high.