Memory Feature-Based Sample Selection Strategy for Few-Sample Learning Digital Predistortion

DOI number:10.1109/TMTT.2022.3199482

Affiliation of Author(s):中国科学技术大学

Teaching and Research Group:电子工程与信息科学系

Journal:IEEE Transactions on Microwave Theory and Techniques

Key Words:Behavioral modeling, digital predistortion (DPD), few-sample learning (FSL), power amplifiers (PAs), sample selection method (SSM).

Abstract:This article proposed a novel sample selection strategy for reducing the computational complexity of digital predistortion (DPD). Due to the memory effect of the power amplifier (PA), the PA's output is affected by the memory term. Thus, unlike existing sample selection methods (SSMs) that consider signal amplitude as the only feature, the proposed method regards signal points and their lagged terms (memory terms) as features of each sample point. We also introduce representative subset selection methods to further increase the selected samples' diversity, and these methods are improved to reduce their storage and computational complexity. By expanding the diversity among the selected samples, even a few samples for training can obtain satisfactory performance. In addition, the complexity analysis shows that the proposed method is effective and competitive. Based on the experimental results, the proposed method outperforms the existing techniques in performance, complexity, and stability.

First Author:Guichen Yang (杨贵晨)

Co-author:Wen Qiao,Chengye Jiang,Lei Su,Renlong Han,Jingchao Tan

Indexed by:Journal paper

Correspondence Author:Falin Liu

Document Code:10.1109/TMTT.2022.3199482

Discipline:Engineering

Document Type:J

Volume:71

Issue:2

Page Number:602-612

Translation or Not:no

Date of Publication:2023-02-06

Included Journals:EI、SCI