Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network
Interconnect parasitic capacitance extraction is crucial in analyzing VLSI circuits’ delay and crosstalk. This paper uses the deep neural network (DNN) to predict the parasitic capacitance matrix of a two-dimensional pattern. To save the DNN training time, the neural network’s output includes only c...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/6/1440 |
| _version_ | 1797612166000410624 |
|---|---|
| author | Yaoyao Ma Xiaoyu Xu Shuai Yan Yaxing Zhou Tianyu Zheng Zhuoxiang Ren Lan Chen |
| author_facet | Yaoyao Ma Xiaoyu Xu Shuai Yan Yaxing Zhou Tianyu Zheng Zhuoxiang Ren Lan Chen |
| author_sort | Yaoyao Ma |
| collection | DOAJ |
| description | Interconnect parasitic capacitance extraction is crucial in analyzing VLSI circuits’ delay and crosstalk. This paper uses the deep neural network (DNN) to predict the parasitic capacitance matrix of a two-dimensional pattern. To save the DNN training time, the neural network’s output includes only coupling capacitances in the matrix, and total capacitances are obtained by summing corresponding predicted coupling capacitances. In this way, we can obtain coupling and total capacitances simultaneously using a single neural network. Moreover, we introduce a mirror flip method to augment the datasets computed by the finite element method (FEM), which doubles the dataset size and reduces data preparation efforts. Then, we compare the prediction accuracy of DNN with another neural network ResNet. The result shows that DNN performs better in this case. Moreover, to verify our method’s efficiency, the total capacitances calculated from the trained DNN are compared with the network (named DNN-2) that takes the total capacitance as an extra output. The results show that the prediction accuracy of the two methods is very close, indicating that our method is reliable and can save the training workload for the total capacitance. Finally, a solving efficiency comparison shows that the average computation time of the trained DNN for one case is not more than 2% of that of FEM. |
| first_indexed | 2024-03-11T06:37:27Z |
| format | Article |
| id | doaj.art-082136efb07b45af99b063a8a9090206 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T06:37:27Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-082136efb07b45af99b063a8a90902062023-11-17T10:45:34ZengMDPI AGElectronics2079-92922023-03-01126144010.3390/electronics12061440Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural NetworkYaoyao Ma0Xiaoyu Xu1Shuai Yan2Yaxing Zhou3Tianyu Zheng4Zhuoxiang Ren5Lan Chen6Institute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaInstitute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, ChinaInstitute of Microelectronics of the Chinese Academy of Sciences, Beijing 100029, ChinaInterconnect parasitic capacitance extraction is crucial in analyzing VLSI circuits’ delay and crosstalk. This paper uses the deep neural network (DNN) to predict the parasitic capacitance matrix of a two-dimensional pattern. To save the DNN training time, the neural network’s output includes only coupling capacitances in the matrix, and total capacitances are obtained by summing corresponding predicted coupling capacitances. In this way, we can obtain coupling and total capacitances simultaneously using a single neural network. Moreover, we introduce a mirror flip method to augment the datasets computed by the finite element method (FEM), which doubles the dataset size and reduces data preparation efforts. Then, we compare the prediction accuracy of DNN with another neural network ResNet. The result shows that DNN performs better in this case. Moreover, to verify our method’s efficiency, the total capacitances calculated from the trained DNN are compared with the network (named DNN-2) that takes the total capacitance as an extra output. The results show that the prediction accuracy of the two methods is very close, indicating that our method is reliable and can save the training workload for the total capacitance. Finally, a solving efficiency comparison shows that the average computation time of the trained DNN for one case is not more than 2% of that of FEM.https://www.mdpi.com/2079-9292/12/6/1440interconnect wireparasitic capacitance matrixdata augmentationDNNResNet |
| spellingShingle | Yaoyao Ma Xiaoyu Xu Shuai Yan Yaxing Zhou Tianyu Zheng Zhuoxiang Ren Lan Chen Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network Electronics interconnect wire parasitic capacitance matrix data augmentation DNN ResNet |
| title | Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network |
| title_full | Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network |
| title_fullStr | Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network |
| title_full_unstemmed | Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network |
| title_short | Extraction of Interconnect Parasitic Capacitance Matrix Based on Deep Neural Network |
| title_sort | extraction of interconnect parasitic capacitance matrix based on deep neural network |
| topic | interconnect wire parasitic capacitance matrix data augmentation DNN ResNet |
| url | https://www.mdpi.com/2079-9292/12/6/1440 |
| work_keys_str_mv | AT yaoyaoma extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT xiaoyuxu extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT shuaiyan extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT yaxingzhou extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT tianyuzheng extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT zhuoxiangren extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork AT lanchen extractionofinterconnectparasiticcapacitancematrixbasedondeepneuralnetwork |