A modified bilinear polynomial model for digital predistortion of RF power amplifiers

This paper proposed a novel model to characterize digital predistortion (DPD) of radio-frequency (RF) power amplifiers (PA). The model is derived from bilinear polynomials and has the capability of modeling DPD with high nonlinearity and deep memory effect. Compared with traditional models, e.g., memory polynomial (MP), generalized memory polynomial (GMP), this model shows better performance with fewer coefficients. Recursive least squares (RLS) is selected to estimate coefficients. Simulation results show that the model can effectively linearize PA.