Forward modeling assisted 1-bit data acquisition based model extraction for digital predistortion of RF power amplifiers

Abstract

In this paper, a low-cost data acquisition approach with novel forward modeling is proposed for model extraction of digital predistortion (DPD) of RF power amplifiers (PA). Compared to the existing 1-bit DPD solution, the proposed approach employs a new loss function using only 1-bit resolution error signal to extract the forward model and the coefficients of DPD model can be extracted afterwards based on the estimated PA output. The proposed solution avoids the calculation of step size in the existing approach, and the convergence is guaranteed due to the convexity of the loss function. The experimental results show that the proposed algorithm achieves equivalent linearization performance as that using the conventional method.