Neural Network based Nonlinear Forward Model Identification for Digital MIMO Arrays Under Load Modulation
Paper in proceeding, 2025
In multi-antenna transmitters, antenna coupling may lead to load modulation of the PAs, making their nonlinear responses beam-dependent and compromising traditional digital predistortion solutions. This paper presents a methodology for nonlinear PA array modeling using time-delay neural networks (TDNNs), with specific focus on digital MIMO systems. RF measurements on an emulated 4×1 array at 2.1 GHz demonstrate the effectiveness of the proposed approach in accurately modeling the PA array under load modulation, while showing clear gains over current polynomial-based and TDNN-based state-of-the-art models.
digital predistortion
MIMO
5G
neural networks
load modulation
power amplifiers
6G
behavioral modeling
antenna crosstalk
active array transmitters
Author
Joel Fernandez
University of Tampere
Lauri Anttila
University of Tampere
Koen Buisman
University of Surrey
Chalmers, Microtechnology and Nanoscience (MC2), Microwave Electronics
Vesa Lampu
University of Tampere
Christian Fager
Chalmers, Microtechnology and Nanoscience (MC2), Microwave Electronics
Thomas Eriksson
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
M. Valkama
University of Tampere
IEEE MTT-S International Microwave Symposium Digest
0149645X (ISSN)
356-3599798331514099 (ISBN)
2025 IEEE/MTT-S International Microwave Symposium, IMS 2025
San Francisco, USA,
San Francisco, USA,
Eureka CELTIC: Energy-Efficient Radio Systems at 100 GHz and beyond: Antennas, Transceivers and Waveforms
VINNOVA (2020-02889), 2021-01-01 -- 2024-02-07.
Subject Categories (SSIF 2025)
Communication Systems
Telecommunications
Signal Processing
DOI
10.1109/IMS40360.2025.11103765