Behavioral modeling and predistortion of nonlinear power amplifiers based on adaptive filtering techniques

Abstract

In this paper, the use of some of the most popular adaptive filtering algorithms for the purpose of linearizing power amplifiers by the well-known digital predistortion (DPD) technique is investigated. First, an introduction to the problem of power amplifier linearization is given, followed by a discussion of the model used for this purpose. Next, a variety of adaptive algorithms are used to construct the digital predistorter function for a highly nonlinear power amplifier and their performance is comparatively analyzed. Based on the simulations presented in this paper, conclusions regarding the choice of algorithm are derived.IndexTerms-Adaptive filtering, power amplifier nonlinearities, behavioral modeling, nonlinear system identification.

Linearization Approach of Combined Digital Predistortion with Crest Factor Reduction

“…LMS algorithm [12], based on the minimum mean square error rule with steepest descent, is widely used in adaptive algorithm due to its low complexity, fast convergence and small error.The common target function is the mean square error (MSE) criterion, which is expressed as,  ( is function of n. When the state reached steady, the system is convergent. Assumed initial coefficients w (0)…”

Fast Adaptive Digital Pre-Distortion Scheme for Application in Cellular Base Station Power Amplifiers

“…Adaptive DPD schemes can employ adaptive filtering algorithms that estimate the pre-distorters coefficient .Adaptation can also be performed through the use of a look-up table (LUT) [37,38]. For all adaptive algorithms, coefficients of the PD model are updated by comparison between the original excitation signal at the PDs input, and the PA output.…”