AI-Assisted Deep-Learning-Based Design of High-Efficiency Class F Power Amplifiers
Saved in:
| Title: | AI-Assisted Deep-Learning-Based Design of High-Efficiency Class F Power Amplifiers |
|---|---|
| Authors: | Zhou, Han, 1994, Chang, Haojie, 1994, Widén, David, 2002, Fornstedt, Ludvig, 2000, Melin, Gabriel, 2002, Fager, Christian, 1974 |
| Source: | IEEE Microwave and Wireless Technology Letters. 35(6):690-693 |
| Subject Terms: | gallium nitride (GaN), energy efficiency, waveform engineering, deep learning, Artificial intelligence (AI), Class F, harmonic tuning, machine learning, power amplifier (PA) |
| Description: | This article presents a deep-learning-based approach for designing Class F power amplifiers (PAs). We use convolutional neural networks (CNNs) to predict the scattering parameters of pixelated electromagnetic (EM) layouts. Using a CNN-based surrogate model and an evolutionary algorithm, we synthesize complex Class F output networks. As a proof of concept, we implement a gallium nitride (GaN) HEMT Class F PA, achieving a measured output power of 41.6 dBm and a drain efficiency of 74% at 2.9 GHz. The prototype also linearly reproduces a 20-MHz modulated signal with an 8.5-dB peak-to-average power ratio (PAPR), achieving an adjacent channel leakage ratio (ACLR) of −50.7 dBc with digital predistortion (DPD). To the best of our knowledge, this is the first deep-learning-based Class F PA design using pixelated layout structures. |
| File Description: | electronic |
| Access URL: | https://research.chalmers.se/publication/545797 https://research.chalmers.se/publication/545980 https://research.chalmers.se/publication/545980/file/545980_Fulltext.pdf |
| Database: | SwePub |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | This article presents a deep-learning-based approach for designing Class F power amplifiers (PAs). We use convolutional neural networks (CNNs) to predict the scattering parameters of pixelated electromagnetic (EM) layouts. Using a CNN-based surrogate model and an evolutionary algorithm, we synthesize complex Class F output networks. As a proof of concept, we implement a gallium nitride (GaN) HEMT Class F PA, achieving a measured output power of 41.6 dBm and a drain efficiency of 74% at 2.9 GHz. The prototype also linearly reproduces a 20-MHz modulated signal with an 8.5-dB peak-to-average power ratio (PAPR), achieving an adjacent channel leakage ratio (ACLR) of −50.7 dBc with digital predistortion (DPD). To the best of our knowledge, this is the first deep-learning-based Class F PA design using pixelated layout structures. |
|---|---|
| ISSN: | 27719588 2771957X |
| DOI: | 10.1109/LMWT.2025.3552495 |