Combined Kernel Function and Coefficient Sharing for Digital Predistortion of RF Power Amplifiers With Dynamic Resource Block Allocation
Hang Yin, Anding Zhu
Abstract
Flexible allocation of spectrum resources will introduce transmitted signals with different spectral energy distribution patterns. To make sure that high efficiency and good linearity in radio frequency (RF) systems can be achieved when handling all such signals, the digital predistortion (DPD) technique must be used and is required to generate different behavioral characteristics to compensate the distortions arising from these signals. To reduce the complexity for DPD hardware implementation, this article proposes an approach that combines kernel function and coefficient sharing. Substantial experimental results with 60- and 100-MHz 5G new radio (NR) signals of 63 resource block (RB) allocation patterns on two different types of power amplifiers (PAs) confirm that the proposed method can achieve significant reduction of kernel function diversity and coefficients while maintaining comparable linearization performance.