A Low-complexity DPD Coefficient Update Method for Varying Transmission Configurations

In this paper, a novel, effective coefficient extraction algorithm is proposed to address the challenges of digital pre-distortion (DPD) for power amplifiers operating under varying transmission conditions. It updates DPD coefficients quickly in response to transmission changes. Singular value decomposition (SVD) method is employed here to extract pretrained feature, thereby accelerating the update of DPD coefficients. To fully capitalize SVD's orthogonalization, a low-complexity parameter extraction method based on limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) is introduced to minimize computational and storage demands. Experimental validations on a Doherty PA indicate that the proposed method achieves an 83% reduction in computational complexity compared to conventional LS method, while still maintaining satisfactory linearization performance across varying power and bandwidth conditions.