Residual Structure-Based Multimodel Neural Network With Physically Inspired Core for Digital Predistortion in 6G Intelligent Radio

View at publisher Summarise with VIRAT
Authors: Xin Wei, Xin Liu, Huanhuan Jia, Tong Shen, Ting Feng, Yang Lu, Xiaohua Ma, Wenhua Chen
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Digital predistortion (DPD) is the core technology for mitigating nonlinear distortion of power amplifiers (PAs) in the radio frequency (RF) front-end, especially in 6G integrated sensing and communication systems, where rapidly varying signal parameters demand real-time adaptive correction. To meet this challenge, this article proposes a dynamic DPD neural network (NN) model based on a multimodel residual structure NN (MRS-NN). Inspired by the PA’s bandpass equivalent circuits model, a low-complexity bandpass feedback recurrent NN (BPF-RNN) is developed as the core nonlinear fitting model in MRS-NN. Meanwhile, a multilayer fully connected network is integrated as the feature-extraction module to capture the dynamic characteristics of PA. Experimental results obtained over a range of operating scenarios—covering 64 trained and 16 untrained states—demonstrate that the proposed MRS-NN outperforms existing dynamic DPD models, delivering superior linearization performance while significantly reducing parameter count and computational complexity.

Fields of Study & Subjects

Advanced Power Amplifier Design PAPR reduction in OFDM Neural Networks and Applications

Publication details

Journal / source
IEEE Microwave and Wireless Technology Letters
Volume
36
Issue
6
Pages
898–901
Publication date
30 March 2026
Language
EN