Training a Neural Network for a New Node Element Design

• Autorzy: Skorpil V. , Precechtel R.

Warianty tytułu

PL

Proces trenowania sieci neuronowej do tworzenia nowego węzła sieci

• Języki publikacji: EN

Abstrakty

EN

This paper describes training a neural network for a new node element design. Neural networks are used to control network nodes instead of the control by sequential circuits. This article primarily discusses the management of routers. Routers are generally able to use all available topologies of a network. Simulation in MATLAB-SIMULINK has tested how the training networks route the given packets. It has been shown that Artificial Neural Networks (ANN) are suitable for the described aim.

PL

W artykule opisano proces trenowania sieci neuronowej, wykorzystanej na potrzeby budowy nowego węzła sieci. Głównym zastosowaniem, omówionym w publikacji jest zarządzanie routerem. Badania symulacyjne przeprowadzone w programie Matlab-Simulink, potwierdziły skuteczność działania sztucznych sieci neuronowych w tej aplikacji.

Słowa kluczowe

EN

artificial neural network; network node; MATLAB Simulink; routing

PL

sztuczne sieci neuronowe; węzły sieci; MATLAB/Simulink; routowanie

• Wydawca: Wydawnictwo SIGMA-NOT
• Czasopismo: Przegląd Elektrotechniczny
• Rocznik: 2013
• Tom: R. 89, nr 2b
• Strony: 187--192
• Opis fizyczny: Bibliogr. 14 poz., rys., tab., wykr.

Twórcy

autor

Skorpil V.

• Brno University of Technology

autor

Precechtel R.

• Brno University of Technology

Bibliografia

• [1] Skorpil,V.,Precechtel,R. Management of Routing Using Artificial Intelligence. In Proceedings of the 11th International Conference Knowledge in Telecommunication Technologies and Optics – KTTO 2011. pp. 29-32, June 22-24, 2011, Szczyrk, Poland, ISBN 978-80-248-2399-7
• [2] Precechtel, R. Optimalization Drive of the Active Network Element. BUT, Brno 2009
• [3] Demuth,H., Beale,M., Neural Network Toolbox for Use with MATLAB. Natick (USA): The MathWorks, Inc., 1994.
• [4] Takizava and a. Fukasawa, Novel neural Network Scheme Composed of Prediction and Studies. Proceedings of the 13 th WSEAS International Conference on systems. Rodos, Greece, Iceland, WSEAS, pp. 611-615, Rodos, 2009, ISBN: 978-960- 474-097-0, ISSN: 1790-2769 |
• [5] Tuckova,J. Introduction to the theory and application of Artificial neural networks. CTU, Prague 2003. 103s. ISBN 80- 01-02800-3.
• [6] L.P.S. Fernandez and A.R. Ruiz and J.de J.M. Juarez, “Urban Noise Permanent Monitoring and Pattern Recognition”. Proceedings of the European Conference of Communications – ECCOM´10. NAUN, Tenerife, Spain, Puerto De La Cruz 2010, pp. 143 - 148 ISBN: 978-960-474-250-9
• [7] Bishop Artificial Intelligence I, II, III, IV. Academia, Prague, 1993, 1997, 2001, 2003.
• [8] Danilo, P. Mandic,P., Recurrent neural networks for prediction, learning algorithms, architectures and stability. John Wiley, Chichester, 2001.
• [9] Norgaard, M., Neural Networks for Modelling and Control of Dynamic Systems. Springer, London 2000.
• [10] Drabek, O., Taufer,I. Seidl, P., Artificial neural networks-- Theory and Application. CHEMagazin 1 (XVI), 2006, p. 12-14. ISSN 1210-7409.
• [11] Drabek,P., Taufer,I., Seidl,P., Artificial neural networks-- Theory and Application. CHEMagazin 5 (XVI), 2006, p. 6-8. ISSN 1210-7409.
• [12] Snorek,M. Jirina,M., Neural Networks and Neural-copmputers. CTU, Prague 1996.124 s. ISBN 80-01--01455-X.
• [13] Susnea and a. Filipescu and v. Minzu and g., Vasil, Virtual Pheromones and Neural Networks Based Wheeled Mobile Robot Control. Proceedings of the 13 th WSEAS International Conference on systems. Rodos, Greece, Iceland, WSEAS, pp. 511-516, Rodos, 2009, ISBN: 978-960-474-097-0, ISSN: 1790- 2769 |
• [14] Bogdamov and r. Mirsu and v. Tiponut, Matlab Model for Spiking Neural Networks, Proceedings of the 13 th WSEAS International Conference on systems. Rodos, Greece, Iceland, WSEAS, pp. 533-537, Rodos, 2009, ISBN: 978-960-474-097-0, ISSN: 1790-2769 |