Split augmented Hammerstein behavioral model with additional distortion path

This paper proposes the split augmented Hammerstein (SAH) model with an additional distortion path. The error signal between the source signal and the output signal of the augmented Hammerstein (AH) model is employed to enhance the accuracy of the AH model. The 13^th memoryless polynomial and a finite impulse response (FIR) filter are used in the memoryless and memory effect subsystems, respectively. The coefficients of FIR filters are obtained by the recursive least square algorithm. To validate the performance of the SAH model, the Doherty power amplifier and a 2-FA WCDMA signal with 10-MHz carrier spacing at 2.14 GHz are employed. The validation results demonstrate that the SAH model can characterize the memory effects as well as the nonlinearity more accurately than the AH model.