Izvestiya Vysshikh Uchebnykh Zavedenii. Elektronika home

Publication of the journal

The section is currently being updated

Nowadays the IC design process is highly manual and relies on engineer’s field knowledge for reasonable productivity. Technology shrinking has led to higher impact of secondary effects on circuit performance characteristics, which ultimately complicated the design process even further, giving a rise to need for design automation tools. High accuracy simulation based heuristic swarm intelligence methods combined with machine learning and artificial intelligence techniques allow the replacement of costly simulations with performance evaluations via neural networks, the accuracy of which relies on diversity and large quantities of training data that may not be readily available for most design problems. In this work, the IC design automation system is proposed incorporating a generative adversarial network with Wasserstein loss and gradient penalty coupled with Spearman’s rank correlation coefficient matrix for guided generation of predictive network’s training data. It was shown that the proposed system doesn’t require huge amounts of data presence in advance; it uses deep neural network occasionally retrained during the optimization process, for performance evaluation and a genetic algorithm for global exploration of design search space. The designing of high-performance mixed logic line decoder and two-stage operational amplifier has demonstrated the viability of automated IC design system.

Key words: artificial neural network, generative adversarial network, automated circuit optimization, circuit performance modeling
Published in: CIRCUIT ENGINEERING AND DESIGN
Bibliography link: Melikyan V. Sh., Vardumyan A. V., Harutyunyan A. G., Asatryan N. A., Melikyan Sh. V., Karapetyan E. Y. System for automated IC design based on generative adversarial and artificial deep neural networks. Proc. Univ. Electronics, 2023, vol. 28, no. 5, pp. 612–620. https://doi.org/ 10.24151/1561-5405-2023-28-5-612-620. – EDN: SXFJLC.
Financial source: the work has been supported by the Science Committee of the Republic of Armenia (project no. 21T-2B002).

Vazgen Sh. Melikyan
National Polytechnic University of Armenia, Armenia, 0009, Yerevan, Teryan st., 105; “Synopsys Armenia” CJSC, Armenia, 0026, Yerevan, Arshakunyats ave., 41

Arman V. Vardumyan
National Polytechnic University of Armenia, Armenia, 0009, Yerevan, Teryan st., 105; “Synopsys Armenia” CJSC, Armenia, 0026, Yerevan, Arshakunyats ave., 41

Ashot G. Harutyunyan
National Polytechnic University of Armenia, Armenia, 0009, Yerevan, Teryan st., 105

Narek A. Asatryan
National Polytechnic University of Armenia, Armenia, 0009, Yerevan, Teryan st., 105; “Synopsys Armenia” CJSC, Armenia, 0026, Yerevan, Arshakunyats ave., 41

Shavarsh V. Melikyan
University of Bristol, Bristol, UK, BS8 1UB, Bristol, Clifton, Woodland Road, Merchant Venturers Building

Erik Y. Karapetyan
National Polytechnic University of Armenia, Armenia, 0009, Yerevan, Teryan st., 105; “Synopsys Armenia” CJSC, Armenia, 0026, Yerevan, Arshakunyats ave., 41

1. Razavi B. Design of Analog CMOS Integrated Circuits. 2nd ed. Noida, McGraw Hill India, 2017. 782 p.

2. Fayazi M., Colter Z., Afshari E., Dreslinski R. G. Applications of artificial intelligence on the modeling and optimization for analog and mixed-signal circuits: A review. IEEE Trans. Circuits Syst. I: Regul. Pap., 2021, vol. 68, no. 6, pp. 2418–2431. https://doi.org/10.1109/TCSI.2021.3065332

3. Takai N., Fukuda M. Prediction of element values of OP Amp for required specifications utilizing deep learning. 2017 International Symposium on Electronics and Smart Devices (ISESD). Yogyakarta, IEEE, 2017, pp. 300–303. https://doi.org/10.1109/ISESD.2017.8253353

4. Lourenço N., Rosa J., Martins R., Aidos H., Canelas A., Póvoa R., Horta N. On the exploration of promising analog IC designs via artificial neural networks. 2018 15th International Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design (SMACD). Prague, IEEE, 2018, pp. 133–136. https://doi.org/10.1109/SMACD.2018.8434896

5. Haupt R. L. An introduction to genetic algorithms for electromagnetics. IEEE Antennas and Propagation Magazine, 1995, vol. 37, no. 2, pp. 7–15. https://doi.org/10.1109/74.382334

6. Li Ya., Wang Y., Li Yu., Zhou R., Lin Z. An artificial neural network assisted optimization system for analog design space exploration. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, vol. 39, no. 10, pp. 2640–2653. https://doi.org/10.1109/TCAD.2019.2961322

7. Goodfellow I. J., Pouget-Abadie J., Mirza M., Xu B., Warde-Farley D., Ozair S., Courville A., Bengio Y. Generative adversarial networks. arXiv. 2014. Available at: https://arxiv.org/abs/1406.2661 (accessed: 10.07.2023).

8. Gulrajani I., Ahmed F., Arjovsky M., Dumoulin V., Courville A. Improved training of Wasserstein GANs. arXiv. 2017. Available at: https://arxiv.org/abs/1704.00028 (accessed: 10.07.2023).

9. Abadi M., Barham P., Chen J., Chen Z., Davis A., Dean J., Devin M., Ghemawat S. et al. TensorFlow: A system for large-scale machine learning. OSDI’16: Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation. Savannah, GA, USENIX, 2016, pp. 265–283.

10. Savalam C. S., Prasanti K., Haranath A. S. Design and implementation of high performance and low power mixed logic line decoders. International Journal of Innovative Technology and Exploring Engineering (IJITEE), 2019, vol. 8, iss. 6S4, pp. 635–640. https://doi.org/10.35940/ijitee.F1131.0486S419

11. HSPICE Reference Manual. Synopsys Inc., 2018. 968 p.

Information about the publication

Downloads: 0

UDK: 621.382.049.77
Publication type: Scientific article
Source lang: English
DOI: 10.24151/1561-5405-2023-28-5-612-620
RISC id: SXFJLC

Download: File not found (690.04 Kb) JATS XML