trudyniisiТруды НИИСИSRISA Proceedings2225-73493033-6422НИЦ «КУРЧАТОВСКИЙ ИНСТИТУТ» - НИИСИtrudyniisi-24Research ArticleПРОЕКТИРОВАНИЕ И МОДЕЛИРОВАНИЕ СБИСDESIGN AND MODELING OF VLSIИсследование эффективности применения различных архитектур нейросетей в расчете маски в задаче инверсной фотолитографииApplication of Various Neural Networks Architectures for Mask Calculation in Problem of Inverse PhotolithographyКарандашевЯ. М.KarandashevYa. M.

Москва

karandashev@niisi.ras.ruТепловГ. С.TeplovG. S.

Зеленоград

Долгопрудный

gteplov@niime.ruКереметВ. В.KeremetV. V.

Москва

КармановА. А.KarmanovA. A.

Долгопрудный

КузовковА. В.KuzovkovA. V.

Зеленоград

МальсаговМ. Ю.MalsagovM. Yu.

Москва

ФГУ ФНЦ НИИСИ РАН; Российский университет дружбы народовРоссияАО «НИИМЭ»; Московский физико-технический институтРоссияМосковский государственный университет им. М.В.ЛомоносоваРоссияМосковский физико-технический институтРоссияАО «НИИМЭ»РоссияФГУ ФНЦ НИИСИ РАНРоссия2022151020251247380Copyright © Карандашев Я.М., Теплов Г.С., Керемет В.В., Карманов А.А., Кузовков А.В., Мальсагов М.Ю., 20252025Карандашев Я.М., Теплов Г.С., Керемет В.В., Карманов А.А., Кузовков А.В., Мальсагов М.Ю.Karandashev Y.M., Teplov G.S., Keremet V.V., Karmanov A.A., Kuzovkov A.V., Malsagov M.Y.This work is licensed under a Creative Commons Attribution 4.0 License.https://www.t-niisi.ru/jour/article/view/24

В данной работе нами решается обратная задача вычислительной фотолитографии. Расчет топологии маски производился глубокими нейронными сетями. Исследование было направлено на сравнение эффективности нейросетевых архитектур U-Net, Erf-Net и Deep Lab v3, а также встроенных алгоритмов Calibre Workbench в решении задачи инверсной фотолитографии. Обучение искусственных нейронных сетей было выполнено на специально сгенерированном и размеченном наборе данных. Случайные фигуры генерировались с помощью САПР Calibre Workbench для маски затворов транзисторов технологии 90 нм. Сравнение производилось по параметрам точности и скорости. В качестве метрик в работе были использованы edge placement error (EPE) и intersection over union (IOU). Применение нейронных сетей позволило в 100 раз ускорить расчет маски при сохранении точности в 92% на метрике IOU.

In this work, we solve the inverse problem of computational photolithography. The mask topology was calculated using deep neural networks. The study was aimed at comparing the effectiveness of the neural network architectures U-Net, Erf-Net and Deep Lab v3, as well as the built-in Calibre Workbench algorithms in solving the problem of inverse photolithography. Training of artificial neural networks was performed on a specially generated and labeled data set. Random shapes were generated using the Calibre Workbench CAD mask for the 90nm transistor gate mask. The comparison was made in terms of accuracy and speed. Edge placement error (EPE) and intersection over union (IOU) were used as metrics in the work. The use of neural networks made it possible to speed up the calculation of the mask by 100 times while maintaining an accuracy of 92% on the IOU metric.

вычислительная фотолитографияинверсная литографиясверточные нейронные сетикоррекция оптической близостиU-NetErf-NetDeep Lab v3перевод картинки в картинкуавтоматизация электронного проектированияискусственный интеллектEPEIOUcomputational photolithographyinverse lithography technologyconvolutional neural networksML-OPCU-NetErf-NetDeep Lab v3image-to-imageelectronic design automationartificial intelligenceedge placement errorintersection over unionReferencesChien P., Chen M. Proximity effects in submicron optical lithography //Optical microlithography VI. – International Society for Optics and Photonics, 1987. – Т. 772. – С. 35-41.Chien P., Chen M. Proximity effects in submicron optical lithography //Optical microlithography VI. – International Society for Optics and Photonics, 1987. – Т. 772. – С. 35-41.Balasinski A. et al. A novel approach to simulate the effect of optical proximity on MOSFET parametric yield //International Electron Devices Meeting 1999. Technical Digest (Cat. No. 99CH36318). – IEEE, 1999. – С. 913-916.Balasinski A. et al. A novel approach to simulate the effect of optical proximity on MOSFET parametric yield //International Electron Devices Meeting 1999. Technical Digest (Cat. No. 99CH36318). – IEEE, 1999. – С. 913-916.Wong A. K. K. Resolution enhancement techniques in optical lithography. – SPIE press, 2001. – Т. 47.Wong A. K. K. Resolution enhancement techniques in optical lithography. – SPIE press, 2001. – Т. 47.Балан Н. Н. и др. Основные подходы к моделированию формирования фоторезистивной маски в вычислительной литографии //Известия высших учебных заведений. Материалы электронной техники. – 2020. – Т. 22. – №. 4. – С. 279-289.Балан Н. Н. и др. Основные подходы к моделированию формирования фоторезистивной маски в вычислительной литографии //Известия высших учебных заведений. Материалы электронной техники. – 2020. – Т. 22. – №. 4. – С. 279-289.Otto Oberdan W, Garofalo Joseph G, Low KK et al. Automated optical proximity correction: a rulesbased approach // Optical/Laser Microlithography VII / International Society for Optics and Photonics. –– Vol. 2197. –– 1994. –– P. 278–293.Otto Oberdan W, Garofalo Joseph G, Low KK et al. Automated optical proximity correction: a rulesbased approach // Optical/Laser Microlithography VII / International Society for Optics and Photonics. –– Vol. 2197. –– 1994. –– P. 278–293.Li J. et al. Model-based optical proximity correction including effects of photoresist processes //Optical Microlithography X. – International Society for Optics and Photonics, 1997. – Т. 3051. – С. 643-651.Li J. et al. Model-based optical proximity correction including effects of photoresist processes //Optical Microlithography X. – International Society for Optics and Photonics, 1997. – Т. 3051. – С. 643-651.Hung C. Y. et al. First 65nm tape-out using inverse lithography technology (ILT) //25th Annual BACUS Symposium on Photomask Technology. – International Society for Optics and Photonics, 2005. – Т. 5992. – С. 59921U.Hung C. Y. et al. First 65nm tape-out using inverse lithography technology (ILT) //25th Annual BACUS Symposium on Photomask Technology. – International Society for Optics and Photonics, 2005. – Т. 5992. – С. 59921U.Красников Г. Я., Синюков Д. В. Проблемы и перспективы развития методов коррекции оптической близости для современных уровней технологии //Труды научного совета РАН" Фундаментальные проблемы элементной базы информационно-вычислительных и управляющих систем и материалов для ее создания". – 2019. – Т. 1. – №. 3. – С. 17.Красников Г. Я., Синюков Д. В. Проблемы и перспективы развития методов коррекции оптической близости для современных уровней технологии //Труды научного совета РАН" Фундаментальные проблемы элементной базы информационно-вычислительных и управляющих систем и материалов для ее создания". – 2019. – Т. 1. – №. 3. – С. 17.Spence C., Goad S. Computational requirements for OPC //Design for Manufacturability through Design-Process Integration III. – International Society for Optics and Photonics, 2009. – Т. 7275. – С. 72750U.Spence C., Goad S. Computational requirements for OPC //Design for Manufacturability through Design-Process Integration III. – International Society for Optics and Photonics, 2009. – Т. 7275. – С. 72750U.Choi S., Shim S., Shin Y. Machine learning (ML)-guided OPC using basis functions of polar Fourier transform //Optical Microlithography XXIX. – International Society for Optics and Photonics, 2016. – Т. 9780. – С. 97800H.Choi S., Shim S., Shin Y. Machine learning (ML)-guided OPC using basis functions of polar Fourier transform //Optical Microlithography XXIX. – International Society for Optics and Photonics, 2016. – Т. 9780. – С. 97800H.Choi S., Shim S., Shin Y. Neural network classifier-based OPC with imbalanced training data //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2018. – Т. 38. – №. 5. – С. 938-948.Choi S., Shim S., Shin Y. Neural network classifier-based OPC with imbalanced training data //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2018. – Т. 38. – №. 5. – С. 938-948.Shi B. et al. Fast OPC repair flow based on machine learning //Design-Process-Technology Cooptimization for Manufacturability XIV. – International Society for Optics and Photonics, 2020. – Т. 11328. – С. 113281B.Shi B. et al. Fast OPC repair flow based on machine learning //Design-Process-Technology Cooptimization for Manufacturability XIV. – International Society for Optics and Photonics, 2020. – Т. 11328. – С. 113281B.Shin Y. Computational Lithography Using Machine Learning Models //IPSJ Transactions on System LSI Design Methodology. – 2021. – Т. 14. – С. 2-10.Shin Y. Computational Lithography Using Machine Learning Models //IPSJ Transactions on System LSI Design Methodology. – 2021. – Т. 14. – С. 2-10.Tryasoguzov, P.E., Kuzovkov, A.V., Karandashev, I.M. et al. Using Machine Learning Methods to Predict the Magnitude and the Direction of Mask Fragments Displacement in Optical Proximity Correction (OPC). Opt. Mem. Neural Networks 30, 291–297 (2021). https://doi.org/10.3103/S1060992X21040056Tryasoguzov, P.E., Kuzovkov, A.V., Karandashev, I.M. et al. Using Machine Learning Methods to Predict the Magnitude and the Direction of Mask Fragments Displacement in Optical Proximity Correction (OPC). Opt. Mem. Neural Networks 30, 291–297 (2021). https://doi.org/10.3103/S1060992X21040056Ye W. et al. LithoGAN: End-to-end lithography modeling with generative adversarial networks //2019 56th ACM/IEEE Design Automation Conference (DAC). – IEEE, 2019. – С. 1-6.Ye W. et al. LithoGAN: End-to-end lithography modeling with generative adversarial networks //2019 56th ACM/IEEE Design Automation Conference (DAC). – IEEE, 2019. – С. 1-6.Yang H. et al. GAN-OPC: Mask optimization with lithography-guided generative adversarial nets //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2019. – Т. 39. – №. 10. – С. 2822-2834.Yang H. et al. GAN-OPC: Mask optimization with lithography-guided generative adversarial nets //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2019. – Т. 39. – №. 10. – С. 2822-2834.Sun X. et al. U-Net convolutional neural network-based modification method for precise fabrication of three-dimensional microstructures using laser direct writing lithography //Optics Express. – 2021. – Т. 29. – №. 4. – С. 6236-6247.Sun X. et al. U-Net convolutional neural network-based modification method for precise fabrication of three-dimensional microstructures using laser direct writing lithography //Optics Express. – 2021. – Т. 29. – №. 4. – С. 6236-6247.Shao H. C. et al. From IC Layout to Die Photograph: A CNN-Based Data-Driven Approach //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2020. – Т. 40. – №. 5. – С. 957-970.Shao H. C. et al. From IC Layout to Die Photograph: A CNN-Based Data-Driven Approach //IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. – 2020. – Т. 40. – №. 5. – С. 957-970.Ma X. et al. Model-driven convolution neural network for inverse lithography //Optics express. – 2018. – Т. 26. – №. 25. – С. 32565-32584.Ma X. et al. Model-driven convolution neural network for inverse lithography //Optics express. – 2018. – Т. 26. – №. 25. – С. 32565-32584.Ma X., Zheng X., Arce G. R. Fast inverse lithography based on dual-channel model-driven deep learning //Optics Express. – 2020. – Т. 28. – №. 14. – С. 20404-20421.Ma X., Zheng X., Arce G. R. Fast inverse lithography based on dual-channel model-driven deep learning //Optics Express. – 2020. – Т. 28. – №. 14. – С. 20404-20421.Ronneberger O., Fischer P., Brox T. U-net: Convolutional networks for biomedical image segmentation //International Conference on Medical image computing and computer-assisted intervention. – Springer, Cham, 2015. – С. 234-241.Ronneberger O., Fischer P., Brox T. U-net: Convolutional networks for biomedical image segmentation //International Conference on Medical image computing and computer-assisted intervention. – Springer, Cham, 2015. – С. 234-241.Romera E. et al. Erfnet: Efficient residual factorized convnet for real-time semantic segmentation //IEEE Transactions on Intelligent Transportation Systems. – 2017. – Т. 19. – №. 1. – С. 263-272.Romera E. et al. Erfnet: Efficient residual factorized convnet for real-time semantic segmentation //IEEE Transactions on Intelligent Transportation Systems. – 2017. – Т. 19. – №. 1. – С. 263-272.Chen L. C. et al. Encoder-decoder with atrous separable convolution for semantic image segmenta-Chen L. C. et al. Encoder-decoder with atrous separable convolution for semantic image segmenta-tion //Proceedings of the European conference on computer vision (ECCV). – 2018. – С. 801-818.tion //Proceedings of the European conference on computer vision (ECCV). – 2018. – С. 801-818.Google Colab, https://colab.research.google.comGoogle Colab, https://colab.research.google.comPytorch Documentation, https://pytorch.org/docs/stable/index.htmlPytorch Documentation, https://pytorch.org/docs/stable/index.htmlRonneberger, Olaf, Philipp Fischer, and Thomas Brox. "U-net: Convolutional networks for biomedical image segmentation." In International Conference on Medical image computing and computer-assisted intervention, pp. 234-241. Springer, Cham, 2015.Ronneberger, Olaf, Philipp Fischer, and Thomas Brox. "U-net: Convolutional networks for biomedical image segmentation." In International Conference on Medical image computing and computer-assisted intervention, pp. 234-241. Springer, Cham, 2015.Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A. Efros. "Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks", in IEEE International Conference on Computer Vision (ICCV), 2017Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A. Efros. "Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks", in IEEE International Conference on Computer Vision (ICCV), 2017“Algorithm and methodology for increasing the OPC-recipe efficiency” Medvedev Konstantin A., Kuzovkov Alexey V., Ivanov Vladimir V. 10.22184/NanoRus.2019.12.89.368.372“Algorithm and methodology for increasing the OPC-recipe efficiency” Medvedev Konstantin A., Kuzovkov Alexey V., Ivanov Vladimir V. 10.22184/NanoRus.2019.12.89.368.372

The authors declare that there are no conflicts of interest present.