Vector Decomposition Based Time-Delay Neural Network Behavioral Model for Digital Predistortion of RF Power Amplifiers

DOI number:10.1109/ACCESS.2019.2927875

Journal:IEEE Access

Key Words:Nonlinear RF PA, digital predistortion, artificial neural network, vector decomposition, behavioral modeling.

Abstract:This article presents two novel neural network models for radio frequency (RF) power amplifiers (PAs): vector decomposed time-delay neural network (VDTDNN) model and augmented vector decomposed time-delay neural network (AVDTDNN) model. In contrast to conventional neural network based models, VDTDNN and AVDTDNN comply with the physical characteristics of RF PAs by employing carefully designed network structures. In particular, the nonlinear operations are conducted only on the magnitude of the input signals while the phase information is recovered with the linear weighting. Linear terms with shortcut connection, as well as high order terms, can be used to further boost the modeling performance. The complexity analysis shows that the proposed models have significantly lower complexity than existing neural network models. A wideband GaN RF PA excited by 40MHz and 60MHz OFDM signals was employed to evaluate the performance. Extensive experimental results reveal that the proposed VDTDNN and AVDTDNN models can achieve better linearization performance with lower computational complexity, compared with the existing neural network based models.

First Author:Yikang Zhang (张益康)

Co-author:Yue Li,Anding Zhu

Indexed by:Journal paper

Correspondence Author:Falin Liu

Volume:7

Page Number:91559-91568

Translation or Not:no

Date of Publication:2019-07-12