©2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

This paper compares and discusses four techniques for model order reduction based on compressed sensing (CS), less relevant basis removal (LRBR), principal component analysis (PCA) and partial least squares (PLS). CS and PCA have already been used for reducing the order of power amplifier (PA) behavioral models for digital predistortion (DPD) purposes. While PLS, despite being popular in some signal processing areas, to the best author’s knowledge, still has not been used in the PA linearization field. Finally, the LRBR is an iterative search algorithm proposed by the authors in this paper for the sake of comparison. Experimental results are presented and the advantages and drawbacks of each method discussed.

Peer Reviewed