MIKRODALGA TASARIMINDA BİLGİ TABANLI YAPAY SİNİR AĞLARINI KULLANARAK BİLGİNİN GÖMÜLDÜĞÜ OTOMATİK MODEL ÜRETİMİ

Yıl 2017, Cilt: 19 Sayı: 57, 742 - 756, 01.09.2017

Murat Sımsek

Öz

Artificial neural networks have emerged as a powerful technique for RF/microwave modeling and design. Artificial neural network parameters as number of neurons, sampling data, which are necessary for training can be utilized through automatic model generation without extra effort of user and can provide an efficient model with desired accuracy. In this work, an efficient modeling strategy combining a prior knowledge with automatic model generation technique is proposed. The aim of this combination is to decrease the need for time consuming fine model response and to increase the performance of automatic model generation algorithm using coarse model during the modeling process. Automatic model generation requires less neuron and training data compared to former methods via prior knowledge input method. Spiral inductor model is considered to demonstrate both the advantages and the validity of this technique in terms of accuracy and time consumption

Anahtar Kelimeler

Computer Aided Design (CAD), Modeling, Optimization, Artificial Neural Networks, Knowledge Based Modeling, Design Automation

Mikrodalga Tasarımında Bilgi Tabanlı Yapay Sinir Ağlarını Kullanarak Bilginin Gömüldüğü Otomatik Model Üretimi

Öz

Yapay sinir ağları RF / mikrodalga modelleme ve tasarımı için güçlü bir teknik olarak ortaya çıkmıştır. Yapay sinir ağlarının nöron sayısı ve örnekleme verileri gibi gerekli eğitim parametrelerini kullanıcının ekstra çabası olmadan kullanabilen otomatik model üretimi, arzu edilen doğrulukta etkili bir model sağlayabilir. Bu çalışmada, ön bilgi ile otomatik model üretim tekniğini birleştiren verimli bir modelleme stratejisi önerilmiştir. Bu kombinasyonun amacı, zaman alan iyi model cevabına olan ihtiyacı azaltmak ve kaba modeli kullanan otomatik model üretim algoritmasının performansını modelleme esnasında arttırmaktır. Otomatik model üretimi, ön bilgi giriş yöntemi sayesinde önceki yöntemlere kıyasla daha az nöron ve eğitim verisi gerektirmektedir. Sarmal endüktans modeli, bu tekniğin doğruluk ve zaman tüketimi açısından hem avantajlarını hem de geçerliliğini kanıtlamak için hesaba katılmıştır

Anahtar Kelimeler

Bilgisayar Destekli Tasarım (CAD), Modelleme, Optimizasyon, Yapay Sinir Ağları, Bilgi Tabanlı Modelleme, Tasarım Otomasyonu

Kaynakça

• [1] P. Burrascano, S. Fiori, and M. Mongiardo, 1999. A review artificial neural networks applications in microwave computer-aided design, Int. J. RF Microwave Computer-Aided Eng., vol. 9, no. 3, pp. 158–174. DOI: 10.1002/(SICI)1099- 047X(199905)9:3<158::AIDMMCE3>3.0.CO;2-V
• [2] V. K. Devabhaktuni, M. C. E. Yagoub, F. Yang, J. Xu, and Q. J. Zhang, 2001. Computer-aided design of RF and microwave circuits and systems, Int.J. RF Microwave CAE, vol. 11, pp. 4–21. DOI: 10.1109/22.989983
• [3] M. B. Steer, J. W. Bandler, and C. M. Snowden, 2002. Computer-aided design of RF and microwave circuits and systems, IEEE Trans. Microwave Theory and Tech., vol. 50, pp. 996–1005. DOI: 10.1109/22.989983
• [4] V. Rizzoli, A. Costanzo, D. Masotti, A. Lipparini, and F. Mastri, 2004. Computer-aided optimization of nonlinear microwave circuits with the aid of electromagnetic simulation, IEEE Trans. Microwave Theory and Tech., vol. 52, pp. 362– 377. DOI: 10.1109/TMTT.2003.820898
• [5] Q. J. Zhang, K. C. Gupta, and V. K. Devabhaktuni, 2003. Artificial neural networks for RF and microwave design from theory to practice, IEEE Trans. Microwave Theory and Tech., vol. 51, no. 4, pp. 1339–1350. DOI: 10.1109/TMTT.2003.809179
• [6] A. H. Zaabab, Q. J. Zhang, and M. S. Nakhla, 1995. Neural network modeling approach to circuit optimization and statistical design, IEEE Trans. Microwave Theory and Tech., vol. 43, pp. 1349–1358. DOI: 10.1109/22.390193
• [7] Q. J. Zhang and K. C. Gupta, 2000. Neural networks for RF and microwave design, Boston, London: Artech House.
• [8] J. E. Rayas-Sanchez, 2004. EM-based optimization of microwave circuits using artificial neural networks: the state-of-the-art, IEEE Trans. Microwave Theory and Tech., vol. 52, no. 1, pp. 420–435. DOI: 10.1109/TMTT.2003.820897
• [9] X. Ding, V. K. Devabhaktuni, B. Chattaraj, M. C. E. Yagoub, M. Deo, J. Xu, and Q. J. Zhang, 2004. Neuralnetwork approaches to electromagnetic based modeling of passive components and their applications to high frequency and high-speed nonlinear circuit optimization, IEEE Trans. Microwave Theory and Tech., vol. 52, no. 1, pp. 436–449. DOI: 10.1109/TMTT.2003.820889
• [10] M. Simsek, 2014. Knowledge based three-step modeling strategy exploiting artificial neural network, in Solving Computationally Expensive Engineering Problems, ser. Springer Proceedings in Mathematics Statistics, S. Koziel, L. Leifsson, and X.-S. Yang, Eds. Springer International Publishing, , vol. 97, pp. 219–239. DOI:10.1007/978-3-319-08985- 0_9
• [11] S. Haykin, 1999. Neural network - a comprehensive foundation, 2nd edition, Prentice Hall Inc., New Jersey, USA.
• [12] J. Bandler, M. A. Ismail, J. E. RayasSanchez, and Q. J. Zhang, 1999. New directions in model development for RF/microwave components utilizing artificial neural networks and space mapping, vol. 4. Antennas and Propagation Society International Symposium, 11-16 July, pp. 2572–2575. DOI: 10.1109/APS.1999.789334
• [13] M. H. Bakr, J. Bandler, M. A. Ismail, J. E. Rayas-Sanchez, and Q. J. Zhang, 2000. Neural space mapping optimization for EM-based design, IEEE Transactions on Microwave Theory and Techniques, vol. 48, no. 12, pp. 2307–2314. DOI: 10.1109/22.898979
• [14] H. Kabir, Y. Wang, M. Yu, and Q. J. Zhang, 2008. Neural network inverse modeling and applications to microwave filter design, IEEE Trans. Microwave Theory and Tech., vol. 56, pp. 867–879. DOI: 10.1109/TMTT.2008.919078
• [15] M. S. A. Aoad and Z. Aydin, 2015. Development of knowledge based response correction for a reconfigurable n-shaped microstrip antenna design, in IEEE MTT-S International Conference on Numerical Electromagnetics and Multiphysics Modeling and Optimization, 11-14 August, Ottawa,Canada. DOI:10.1109/NEMO.2015.7415078
• [16] P. M. Watson, K. C. Gupta, and R. L. Mahajan, 1998. Development of knowledge based artificial neural network models for microwave components, in IEEE Int. Microwave Symp. Digest. pp. 9–12. DOI:10.1109/MWSYM.1998.68931 2
• [17] M. Simsek, Q. J. Zhang, H. Kabir, Y. Cao, and N. S. Sengor, 2011. The recent developments in microwave design, International Journal of Mathematical Modelling and Numerical Optimization, vol. 2, no. 2, pp. 213 – 228. DOI:10.1504/IJMMNO.2011.03942 9
• [18] P. M. Watson and K. C. Gupta, 1996. EM-ANN models for microstrip vias and interconnects in dataset circuits, IEEE Trans. Microwave Theory and Tech., vol. 44, pp. 2495– 2503. DOI: 10.1109/22.554584
• [19] G. G. Towell and J. W. Shavlik, 1994. Knowledge-based artificial neural networks, Artificial Intelligence, vol. 70, pp. 119–165. DOI:10.1016/004-3702(94)90105- 8
• [20] F. Wang and Q. J. Zhang, 1997. Knowledge based neural models for microwave design, IEEE Trans. Microwave Theory and Tech., vol. 45, pp. 2333–2343. DOI: 10.1109/22.643839
• [21] M. Simsek and N. S. Sengor, 2008. A knowledge-based neuromodeling using space mapping technique: compound space mapping-based neuromodeling, Int. J. Numer. Model, vol. 21, no. 1-2, pp. 133–149 DOI: 10.1002/jnm.656
• [22] M. Simsek and N. Tezel, 2012. The reconstruction of shape and impedance exploiting space mapping with inverse difference method, IEEE Transactions on Antennas and Propagation, vol. 60, no. 4, pp. 1868–1877. DOI: 10.1109/TAP.2012.2186225
• [23] J. W. Bandler, M. A. Ismail, J. E. Rayas-Sanchez, and Q. J. Zhang, 1999. Neuromodeling of microwave circuits exploiting space mapping technology, IEEE Trans. Microwave Theory and Tech., vol. 47, no. 12, pp. 2417–2427. DOI: 10.1109/22.808989
• [24] Y. Cao and G. Wang, 2007. A wideband and scalable model of spiral inductors using spacemapping neural network, IEEE Trans. Microwave Theory and Tech., vol. 55, no. 12, pp. 2473– 2480. DOI: 10.1109/TMTT.2007.909602
• [25] S. Koziel and J. W. Bandler, 2007. A space-mapping approach to microwave device modeling exploiting fuzzy systems, IEEE Trans. Microwave Theory and Tech., vol. 55, no. 12, pp. 2539– 2547. DOI: 10.1109/TMTT.2007.909605
• [26] S. Koziel and J. W. Bandler 2007. Microwave device modeling using space-mapping and radial basis functions, in IEEE Int. Microwave Symp. Digest. pp. 799–802, Honolulu, Hawaii. DOI: 10.1109/MWSYM.2007.38007 9 [27] L. Zhang and Q. J. Zhang, 2008. Neuro-space mapping technique for semiconductor device modeling, Optimization and Engineering, vol. 9, no. 4, pp. 393–405. DOI: 10.1007/s11081-007-9024-0
• [28] L. Zhang, Q. J. Zhang, and J. Wood, 2008. Statistical neuro-space mapping technique for large signal modeling of nonlinear devices, IEEE Trans. Microwave Theory and Tech., vol. 56, pp. 2453–2467. DOI:10.1109/TMTT.2008.2004894
• [29] Y. Cao, L. Simonovich, and Q. J. Zhang, 2009. A broadband and parametric model of differential via holes using space-mapping neural network, IEEE Microwave and Wireless Components Lett., vol. 19, pp.533–535. DOI:10.1109/LMWC.2009.202704 8
• [30] J. Rayas-Sanchez and Q. J. Zhang, 2012. On knowledge-based neural networks and neuro-space mapping, in IEEE Microwave Symposium Digest (MTT), pp. 1–3, 17-22 June, Montreal, QC, Canada. DOI: 10.1109/MWSYM.2012.62583 54
• [31] V. K. Devabhaktuni, M. C. E. Yagoub, and Q. J. Zhang, 2001. A robust algorithm for automatic development of neural-network models for microwave applications, IEEE Transactions on Microwave Theory and Techniques, vol. 49, no. 12, pp. 2282–2291. DOI: 10.1109/22.971611
• [32] V. K. Devabhaktuni, B. Chattaraj, M. C. E. Yagoub, and Q. J. Zhang, 2003. Advanced microwave modeling framework exploiting automatic model generation, knowledge neural networks, and space mapping, IEEE Trans. Microwave Theory and Tech., vol. 51, no. 7, pp. 1822–1833. DOI: 10.1109/TMTT.2003.814318
• [33] ADS Momentum, 2014. Agilent Technologies, Palo, Alto, CA.