Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:D1L-B
Session:
Number:D1L-B1
A Predistorter Based on the Least-Mean Squares Newton Algorithm
Azzedine Zerguine, Oualid Hammi, Abubakr H. Abdelhafiz, Fadhel Ghannouchi,
pp.660-663
Publication Date:2014/9/14
Online ISSN:2188-5079
Summary:
This work presents an efficient digital predistorter based on a modified Least Mean Squares Newton (LMSN) adaptive filtering algorithm. The proposed predistorter is shown to consistently outperform the Recursive Least Squares (RLS) algorithm by 1dB in terms of error performance while requiring on average 10% less computational time. The performance of the proposed predistorter is validated through linearizing a Doherty power amplifier driven by two 20MHz wideband code division multiple access (WCDMA) signals having different carrier configurations. The results presented in this work demonstrate the desirable performance of the algorithm compared to the well-known RLS algorithm.