Power Amplifier Behavioral Modeling Strategies Using Neural Network and Memory Polynomial Models

This paper discusses the performance comparison of an artificial neural network (ANN) model and a memory polynomial (MP) model for modeling the dynamic nonlinear input-output characteristics of power amplifier (PA) with memory. The ANN model was based on time delay neural network (TDNN) and the memory polynomial model was developed using analytical polynomial function. Both models were developed to fit the dynamic AM-AM and AM-PM conversions of the PA obtained from QPSK digital modulated signal. Furthermore, the conventional TDNN model topology was extended by introducing an additional input to take into account the frequency tone spacing (for two-tone excitation condition) of the stimulus signal, to incorporate memory-effect behavior. The comparison results show that the two variants of PA models are applicable to model the PA, however, the TDNN model, compared to the memory polynomial model, can give better modeling results.