AI-Enabled End-to-End Design in RFICs with Controllable Architectural Style from ‘Classical’ to ‘Non-Intuitive’ for mm-Wave/sub-THz LNAs

📋 论文概要

本文提出了一种统一的AI驱动算法设计流程，用于毫米波低噪声放大器（LNA），将规格到布局的拓扑、架构、电路和电磁设计集成在一个框架中。该流程支持从经典到非直观的可控架构风格，增强了可解释性和调试能力，并在24-150GHz频段展示了两个具有先进性能的LNA设计，包括算法发现的新型架构。

💡 主要创新点

• 首次实现拓扑、架构、电路和电磁设计一体化的端到端算法流程，覆盖LNA从规格到布局的全过程。
• 提出可控架构风格设计，允许设计者在'经典'到'非直观'之间灵活选择，提升设计可解释性和易用性。
• 算法自动探索并发现新型LNA架构，在24-150GHz宽频带内实现领先性能。

🏷 关键词

📄 原文摘要

*Equally Credited Authors (ECAs) 1 Abstract This paper introduces a unified algorithmic design flow for low-noise amplifiers (LNAs), spanning specifications to layout and integrating topology, architecture, circuit, and electromagnetic (EM) design in one framework. The algorithm supports designer input from classical to unconventional architectures, enhancing interpretability, debugging, and usability. We demonstrate two LNAs covering 24 to 150GHz with state-of-the-art performance, including novel architectures uncovered by the algorithm. This paper introduces an unified algorithmic design flow for low-noise amplifiers (LNAs) that spans from specifications to layout, integrating topology, architecture, circuit, and electromagnetic (EM) design in an end-to-end framework. The approach incorporates controllable EM interfaces ranging from classical transmission-line shapes to arbitrary pixelated structures, each offering distinct design trade-offs. Unlike prior AI-enabled inverse

👥 作者与机构

Jonathan Zhou*1, Emir Ali Karahan*2, Juho Park1, Sherif Ghozzy1, Kaushik Sengupta1

Princeton University, Princeton, NJ, 2now with Marvell, Irvine, CA