Modeling and identification of continuous-time system for RF amplfiers

In this paper, we present a new identification procedure for radio frequency Power Amplifier (PA) in the presence of nonlinear distortion and memory effects. The proposed procedure uses a continuous-time model where PA dynamics are modeled with a multivariable filter and at general polynomial function. Using the baseband input and output data, model parameters are obtained by an iterative identification algorithm. Finally, the proposed estimation method is tested and validated on experimental data by comparison of the quadrature IQ signals in time domain.