An effective approach to behavioral modeling of RF power amplifier based on recursive polynomials

A new approach based on recursive polynomials for behavioral modeling of RF power amplifiers is presented. The orthogonal least squares algorithm is introduced for the identification of the structure of the recursive polynomial. It provides an efficient means for removing the unnecessary terms from the general recursive polynomial and guarantees the model so obtained as simple as possible. This modeling approach does not require the complicated nonlinear parameter estimation procedure. The results in section TV prove that the model can predict the behavior of the amplifier under different signal excitations accurately.