Infocommunications and Radio Technologies Infocommunications and Radio Technologies ИНФОКОММУНИКАЦИОННЫЕ И РАДИОЭЛЕКТРОННЫЕ ТЕХНОЛОГИИ 2587-9936 53200 10.29039/2587-9936.2022.05.1.03 Электроника, фотоника, приборостроение и связь (2.2) ELECTRONICS, PHOTONICS, INSTRUMENTATION AND COMMUNICATIONS (2.2) Электроника, фотоника, приборостроение и связь (2.2) Application of Neural Networks in the Construction of Nonlinear Models of Field-Effect Transistors Применение нейронных сетей при построении нелинейных моделей полевых транзисторов Богданов Сергей Александрович Bogdanov Sergey Aleksandrovich bogdanov_sa@mail.ru

кандидат технических наук;

candidate of technical sciences;

 АО «НПП “Исток” имени А. И. Шокина» Фрязино Россия Scientific and production enterprise “Istok” n. a. A. I. Shokin Fryazino Russian Federation 25 03 2022 25 03 2022 5 1 45 53 20 03 2022 05 06 2022 https://rusjbpc.ru/en/nauka/article/53200/view

На основе математического аппарата теории искусственных нейронных сетей разработана нелинейная модель полевого транзистора, особенностью которой является возможность обучения нейронных сетей, используемых для аппроксимации вольтамперной характеристики и емкостей затвор-сток и затвор-исток нелинейной модели транзистора встроенными в популярные СВЧ САПР алгоритмами оптимизации. Это позволяет использовать известные преимущества нейронных сетей в задачах аппроксимации функций для повышения достоверности результатов нелинейного моделирования СВЧ-устройств на основе полевых транзисторов.

The nonlinear model of a field-effect transistor based on the mathematical apparatus of the theory of artificial neural networks was developed. The main feature of this model is the possibility of training neural networks used to approximate the current-voltage characteristic and the gate-drain and gate-source capacitances of a nonlinear transistor model by using optimization algorithms built into popular microwave CAD systems. This makes it possible to use the well-known advantages of neural networks in the problems of function approximation to increase the reliability of the results of non-linear modeling of microwave devices based on field-effect transistors.

 нелинейная модель полевого транзистора искусственные нейронные сети радиальная базисная сеть функция активации «SoftPlus» nonlinear model of a field-effect transistor; artificial neural networks; radial basic network; activation function; SoftPlus

Коколов А. А., Шеерман Ф. И., Бабак Л. И. Обзор математических моделей СВЧ поле-вых транзисторов с высокой подвижностью электронов // Доклады ТУСУР. 2010. № 2-1 (22). C. 118-126. A. A. Kokolov, F. I. Sheerman, L. I. Babak, “Review of mathematical models of microwave field-effect transistors with high electron mobility,” Doklady TUSUR, no. 2-1 (22), pp. 118-126, 2010. (In Russ.). Qi-Jun Zhang, Gupta K. C., Devabhaktuni V. K. Artificial neural networks for RF and mi-crowave design - from theory to practice // IEEE Transactions on Microwave Theory and Techniques. 2003. Vol. 51, no. 4. P. 1339-1350. Qi-Jun Zhang, K. C. Gupta, and V. K. Devabhaktuni, “Artificial neural networks for RF and microwave design - from theory to practice,” IEEE Transactions on Microwave Theory and Techniques, vol. 51, no. 4, pp. 1339-1350, 2003, doi: 10.1109/TMTT.2003.809179. Gao J., Zhang L., Xu J., Zhang Q. -J. Nonlinear HEMT Modeling Using Artificial Neural Network Technique. In: IEEE MTT-S International Microwave Symposium Digest. 2005. P. 469-472. J. Gao, L. Zhang, J. Xu, and Q. -J. Zhang, “Nonlinear HEMT Modeling Using Artificial Neu-ral Network Technique,” in IEEE MTT-S International Microwave Symposium Digest, pp. 469-472, 2005, doi: 10.1109/MWSYM.2005.1516631. Alam M. S., Farooq O., Izharuddin, Armstrong G. A.Artificial Neural Network Based Mod-eling of GaAs HBT and Power Amplifier Design for Wireless Communication System. In: 2006 International Conference on Microelectronics. 2006. P. 103-106. M. S. Alam, O. Farooq, Izharuddin, and G. A. Armstrong, “Artificial Neural Network Based Modeling of GaAs HBT and Power Amplifier Design for Wireless Communication System,” in 2006 International Conference on Microelectronics, pp. 103-106, 2006, doi: 10.1109/ICM.2006.373277. Liu W., Na W., Zhu L., Ma J., Zhang, Q. -J. A Wiener-Type Dynamic Neural Network Ap-proach to the Modeling of Nonlinear Microwave Devices // IEEE Transactions on Microwave Theory and Techniques. Vol. 65, no. 6, P. 2043-2062. W. Liu, W. Na, L. Zhu, J. Ma, and Q.-J. Zhang, “A Wiener-Type Dynamic Neural Network Approach to the Modeling of Nonlinear Microwave Devices,” IEEE Transactions on Micro-wave Theory and Techniques, vol. 65, no. 6, pp. 2043-2062, June 2017, doi: 10.1109/TMTT.2017.2657501. Hu W., Luo H., Yan X. Guo Y. -X. An Accurate Neural Network-Based Consistent Gate Charge Model for GaN HEMTs by Refining Intrinsic Capacitances // IEEE Transactions on Microwave Theory and Techniques. 2021. Vol. 69, no. 7. P. 3208-3218. W. Hu, H. Luo, X. Yan and Y.-X. Guo, “An Accurate Neural Network-Based Consistent Gate Charge Model for GaN HEMTs by Refining Intrinsic Capacitances,” IEEE Transactions on Microwave Theory and Techniques, vol. 69, no. 7, pp. 3208-3218, 2021, doi: 10.1109/TMTT.2021.3076064. Angelov I., Zirath H., Rorsman N. A new empirical nonlinear model for HEMT-devices. In: 1992 IEEE MTT-S Microwave Symposium Digest. 1992. Vol. 3. P. 1583-1586. I. Angelov, H. Zirath, and N. Rorsman, “A new empirical nonlinear model for HEMT-devices,” in 1992 IEEE MTT-S Microwave Symposium Digest, vol. 3, pp. 1583-1586, 1992, doi: 10.1109/MWSYM.1992.188320. Angelov I., Andersson K., Schreurs D. [et al]. In: Large-signal modelling and comparison of AlGaN/GaN HEMTs and SiC MESFETs. Asia-Pacific Microw. Conf. 2006. P. 279-282. I. Angelov, K. Andersson, D. Schreurs, D. Xiao, N. Rorsman, V. Desmaris, M. Sudow, and H. Zirath, “Large-signal modelling and comparison of AlGaN/GaN HEMTs and SiC MESFETs,” in Asia-Pacific Microw. Conf., pp. 279-282, 2006, doi: 10.1109/APMC.2006.4429422. Sayed A., Boeck G. An empirical large signal model for silicon carbide MESFETs. In: Galli-um Arsenide and Other Semicond. Appl. Symp. 2005. P. 313-316. A. Sayed and G. Boeck, “An empirical large signal model for silicon carbide MESFETs,” in Gallium Arsenide and Other Semicond. Appl. Symp., pp. 313-316, 2005. Yuk K. S., Branner G. R. An empirical large-Signal model for SiC MESFETs with self-heating thermal model // IEEE Trans. MTT. 2008. Vol. 56, no. 11. P. 2671-2680. K. S. Yuk and G. R. Branner, “An empirical large-Signal model for SiC MESFETs with self-heating thermal model,” IEEE Trans. MTT., vol. 56, no. 11, pp. 2671-2680, 2008, doi: 10.1109/TMTT.2008.2005922. Галушкин А. И. Нейронные сети : основы теории. М. : Горячая линия - Телеком, 2012. 496 с. A. I. Galushkin, Neural networks: fundamentals of theory. Moscow: Goryachaya Liniya -Telekom, 2012. (In Russ.). Медведев В. С., Потемкин В. Г. Нейронные сети. MATLAB 6. М. : ДИАЛОГ-МИФИ, 2001. 630 с. V. S. Medvedev, V. G. Potemkin, Neural networks. MATLAB 6. Moscow: Dialog - MIFI, 2001. (In Russ.). Baggenstoss P. M. A Neural Network Based on First Principles. In: ICASSP 2020-2020. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020. P. 4002-4006. P. M. Baggenstoss, “A Neural Network Based on First Principles,” in ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 4002-4006, 2020, doi: 10.1109/ICASSP40776.2020.9054549. Statz H., Newman P., Smith I. W., Pucel R. A., Haus H. A. GaAs FET device and circuit simula-tion in SPICE // IEEE Transactions on Electron Devices. 1987. Vol. 34, no. 2. P. 160-169. H. Statz, P. Newman, I. W. Smith, R. A. Pucel, and H. A. Haus, “GaAs FET device and cir-cuit simulation in SPICE,” IEEE Transactions on Electron Devices, vol. 34, no. 2, pp. 160-169, 1987, doi: 10.1109/T-ED.1987.22902. Rudolph M., Fager C., Root D. E. Nonlinear transistor model parameter extraction techniques. Cambridge University Press, 2012. 366 p. Matthias Rudolph, Christian Fager, and David E. Root, Nonlinear transistor model parameter extraction techniques, Cambridge University Press, 2012.