https://scholars.lib.ntu.edu.tw/handle/123456789/630384
標題: | Self-Heating of FinFET Circuitry Simulated by Multi-Correlated Recurrent Neural Networks | 作者: | Chung, Chia Che Huang, Bo Wei Lin, Hsin Cheng Chou, Tao Tsen, Chia Jung CHEN-WUING LIU |
關鍵字: | chain circuit | FinFETs | neural network | self-heating | 公開日期: | 1-八月-2022 | 出版社: | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | 卷: | 43 | 期: | 8 | 起(迄)頁: | 1179 | 來源出版物: | IEEE Electron Device Letters | 摘要: | A series of multi-correlated recurrent neural networks is used to predict the relative temperature of inverter chains folded in 3 rows. The circuit hotspot temperature is predicted by a fully connected neural network. The correlated recurrent neural networks trained by the SPICE data within 17 stages can predict T up to 37 stages (2.2× SPICE complexity) with the error as low as 0.9 °C, outperforming the previous fully connected neural network (1.9× SPICE, 3 °C error) and non-correlated recurrent neural network (2.2× SPICE, 3 °C error) by considering the thermal coupling between rows. The precise prediction of temperature profiles and hotspot positions indicate that the thermal physics is learned by correlated recurrent neural networks. Therefore, an 82-stage folded inverter chain can be predicted and optimized confidently by neural networks, while SPICE can only simulate a 37-stage chain due to the high computational cost. A 100-stage chain is also predicted. |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/630384 | ISSN: | 07413106 | DOI: | 10.1109/LED.2022.3183382 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。