LithoGAN: End-to-End Lithography Modeling with Generative Adversarial Networks
LithoGAN: End-to-End Lithography Modeling with Generative Adversarial Networks
复制标题
DOI：
10.1145/3316781.3317852
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Wei Ye;M. Alawieh;Yibo Lin;David Z. Pan
中科院分区：
文献类型：
--
作者：
Wei Ye;M. Alawieh;Yibo Lin;David Z. Pan
光刻仿真是工艺建模和物理验证中最基本的步骤之一。传统的仿真方法要达到较高的精度，需要耗费大量的计算量。最近，机器学习被引入，通过加速仿真流的抗阻建模阶段来权衡准确性和运行时间。在这项工作中，我们提出了基于生成对抗网络（GAN）的端到端光刻建模框架LithoGAN，将输入掩模模式直接映射到输出抗蚀剂模式。我们的实验结果表明，与传统的光刻模拟和先前基于机器学习的方法相比，LithoGAN可以高精度地预测抗蚀图案，同时实现了数量级的加速。
Lithography simulation is one of the most fundamental steps in process modeling and physical verification. Conventional simulation methods suffer from a tremendous computational cost for achieving high accuracy. Recently, machine learning was introduced to trade off between accuracy and runtime through speeding up the resist modeling stage of the simulation flow. In this work, we propose LithoGAN, an end-to-end lithography modeling framework based on a generative adversarial network (GAN), to map the input mask patterns directly to the output resist patterns. Our experimental results show that LithoGAN can predict resist patterns with high accuracy while achieving orders of magnitude speedup compared to conventional lithography simulation and previous machine learning based approach.