EMOptimizer: An AI-Driven Modular Design and Optimization Framework for RFICs
IEEE Transactions on Microwave Theory and Techniques · 2026-06 · 74 (6) · 5016-5028
DOI: 10.1109/tmtt.2026.3672724 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Microwave Theory and Techniques, 2026-06. Verfasst von Gaofeng Wang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
To address the critical challenges of poor design reusability and prolonged design cycle in radio frequency integrated circuit (RFIC) designs, a modular design and optimization framework is herein proposed with inspiration from the digital design methodology. For a given fabrication process node, standardized RF library cells are first constructed by virtue of an artificial intelligence (AI)-based modeling technique. Then, the set of library cells can be reused across multiple RF designs on the same process node. The AI-based library cells are trained on accurate full-wave electromagnetic (EM) data across a broad sweep of design parameters, thereby leading to parametric RF models that are virtually as accurate as the full-wave EM data. The standardized RF library cells are reusable building blocks for the construction of RF circuits, while parametric RF models provide fast and accurate performance predictions, enabling efficient design-space exploration and optimization. The parametric RF models can deliver accurate EM data within seconds, eliminating the need for time-consuming full-wave simulations at each iteration of the design optimization. This revolutionary approach to RFIC designs boosts the design efficiency by hundreds to thousands of times. As case studies, this proposed method is applied to design two integrated passive device (IPD) chips, including a bandpass filter and a balun. It shows that this approach can significantly improve the design efficiency and shorten the design cycle.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Wang, G., Fang, J., Qi, Y., Guan, Z., Wang, X., Zhang, Q., Zhao, P., & Cao, Y. (2026). EMOptimizer: An AI-Driven Modular Design and Optimization Framework for RFICs. IEEE Transactions on Microwave Theory and Techniques, 74(6), 5016-5028. https://doi.org/10.1109/tmtt.2026.3672724
Vancouver
Wang G, Fang J, Qi Y, Guan Z, Wang X, Zhang Q, et al. EMOptimizer: An AI-Driven Modular Design and Optimization Framework for RFICs. IEEE Transactions on Microwave Theory and Techniques. 2026;74(6):5016-5028. doi:10.1109/tmtt.2026.3672724
BibTeX
@article{Wang2026,
author = {Wang, Gaofeng and Fang, Jiayi and Qi, Yanzhu and Guan, Zhimin and Wang, Xianbing and Zhang, Qi and Zhao, Peng and Cao, Yazi},
title = {EMOptimizer: An AI-Driven Modular Design and Optimization Framework for RFICs},
journal = {IEEE Transactions on Microwave Theory and Techniques},
year = {2026},
volume = {74},
number = {6},
pages = {5016--5028},
doi = {10.1109/tmtt.2026.3672724}
}
EUVIMED