Buradasınız

Anasayfa » Content

NÖRAL VE BULANIK SİSTEM HÜCRE AKTİVASYON YAKLAŞIMLARI VE FPGA’DA DONANIMSAL GERÇEKLENMESİ

NEURAL AND FUZZY SYSTEM CELL ACTIVATION APPROXIMATIONS AND HARDWARE IMPLEMENTATIONS ON FPGA

Journal Name:

  • Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi

Publication Year:

  • 2011

Key Words:

Keywords (Original Language):

Author NameUniversity of AuthorFaculty of Author
Abstract (2. Language): 
Currently, neural and fuzzy systems are methods which have found wide application area. Implementation of these systems on a hardware platform providing their own features is important. With parallel data streaming and processing features, FPGAs have become a preferable hardware platform for implementing of neural and fuzzy systems. In hardware implementation of these systems, the cell activation function, which is the most important unit, is of key importance. In this study, implementation mathematical approximations of commonly used logarithmic sigmoid, hyperbolic tangent sigmoid and Gaussian activation functions on FPGA using single-precision floating-point number format is investigated. For the each function, suitable approach to implement on FPGAs is comparatively given and obtained the actual synthesis results from the Xilinx Virtex 5 xc5vlx110- 3ff1153 FPGA are presented. Obtained experimental results show that proposed implementation approaches have consumed very little hardware resources. Using these proposed approaches, neural and fuzzy systems in various structures can be implemented.
Bookmark/Search this post with
Abstract (Original Language): 
Günümüzde nöral ve bulanık sistemler, çok geniş bir alanda kullanılan yöntemlerdir. Bu sistemlerin kendi özelliklerini sağlayan bir donanım ortamında gerçeklenmesi önemlidir. FPGA’lar paralel veri akışı ve paralel işlem yapma özellikleri ile nöral ve bulanık sistemlerin gerçeklenmesinde tercih edilen donanım olmaya başlamıştır. Bu sistemlerde kilit role sahip olan hücre aktivasyon fonksiyonunun, donanım üzerinde gerçeklenmesi önemlidir. Bu çalışmada, sıklıkla kullanılan logaritmik sigmoidal, hiperbolik tanjant sigmoidal ve Gauss tipi aktivasyon fonksiyonlarının matematiksel yaklaşımlarının tek duyarlıklı kayan noktalı sayı formatıyla FPGA’de gerçeklenmesi irdelenmiştir. Her bir fonksiyon için FPGA’de gerçeklemeye uygun yaklaşımı karşılaştırmalı olarak verilmiş, Xilinx firmasına ait Virtex 5 xc5vlx110-3ff1153 FPGA’sında gerçeklenerek elde edilen sentez sonuçları sunulmuştur. Elde edilen deneysel sonuçlar, önerilen gerçekleme yaklaşımlarının çok az donanımsal kaynak tükettiğini göstermiştir. Önerilen bu yaklaşımlar kullanılarak çeşitli yapılarda nöral ve bulanık sistemler gerçeklenebilir.
FULL TEXT (PDF): 
PDF icon arastirmax-noral-bulanik-sistem-hucre-aktivasyon-yaklasimlari-fpgada-donanimsal-gerceklenmesi.pdf
  • 1
8 - 16
  • Turkish

REFERENCES

References: 

[1]. Martinez, J., Toledo, F.J., Fernandez, E. ve Ferrandez, J.M., “A retinomorphic architecture based on discrete-time cellular neural networks using reconfigurable
computing”, Neurocomputing, Cilt 71, No 4-6, 766-775, 2008.
[2]. Nedjah, N, Silva, R.M.D., Mourelle, L.M.M, ve Silva, M.V.C.D., “Dynamic MAC-based architecture of artificial neural networks suitable for hardware implementation on FPGAs”, Neurocomputing, Cilt 72, No 10-12, 2171-2179, 2009.
[3]. Ferreira, P., Ribeiro, P., Antunes, A. , ve Dias, F.M., “A high bit resolution FPGA implementation of a FNN with a new algorithm for the activation function”, Neurocomputing, Cilt 71, No 1-3, 71-77, 2006.
[4]. Won E., “A hardware implementation of articial neural nertworks using field programmable gate arrays”, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Cilt 581, No 3, 816-820, 2007.
[5]. Ferrer, D., Gonzalez, R., Fleitas, R., Acle, J.P., ve Canetti, R., “NeuroFPGA - Implementing Artificial Neural Networks on Programmable Logic Devices”, Proceedings of the Design, Automation and Test in Europe Conference and Exhibition Designers’ Forum (DATE’04), Cilt 3, 218-223, 2004.
[6]. Çavuşlu, M.A., Karakuzu, C., Şahin, S., “Neural Networkwork Hardware Implementation Using FPGA”, ISEECE 2006 3rd International Symposium on Electrical, Electronic and Computer Engineering Symposium Proceedings, Nicosia, TRNC, 287-290, 2006.
[7]. Savich, A.W., Moussa, M., ve Areibi, S., “The Impact of Arithmetic Representation on Implementing MLP-BP on FPGAs: A Study”, IEEE Transactions on Neural Networks, Cilt 18, No 1, 240 – 252, 2007.
[8]. Farmahini-Farahani, A., Fakhraie, S.M., ve Safari, S., “Scalable Architecture for on-Chip Neural Network Training using Swarm Intelligence”, Proc. of the Design, Automation and Test in Europe Conf. (DATE’08), Munich, Germany, 1340-1345, 2008.
[9]. Lin , C.J. ve Lee, C.Y. “FPGA Implementation of a Recurrent Neural Fuzzy Network with On-Chip Learning for Prediction and Identification Applications” , Journal of
Information Science and Engineering, Cilt 25, No. 2, 575-589, 2009.
[10]. Blake, J. J. ve Maguire, L. P., “The implementation of fuzzy systems, neural Networks and fuzzy neural networks using FPGAs”, Information Sciences, Cilt 112, No 1-4, 151-168, 1998.
[11]. Lin, C.-J. ve Tsai, H.-T. “ FPGA implemengtation of a wavelet neural network with particle swarm optimization learning”, Mathematical and Computer Modeling, Cilt 47, 982–996, 2008.
[12]. Juang, C.-F. ve Hsu, C.-H., “Temperature control by chip-implemented adaptive recurrent fuzzy controller designed by evolutionary algorithm,” IEEE Transactions on Circuits and Systems I: Regular Papers, Cilt. 52, No. 11, 2376–2384, 2005.
[13]. Glackin, B., Maguire, L. P., ve McGinnity, T. M., “Intrinsic and extrinsic implementation of a bio-inspired hardware system”, Information Sciences, Cilt 161, No 1-2, 1-19, 2004.
[14]. Campo, I., Echanobe, J., Bosque, G., ve Tarela, J. M., “Efficient hardware/software implementation of an adaptive neuro-fuzzy system,” IEEE Transactions on Fuzzy Systems, Cilt 16, No.3, 761-778, 2008.
[15]. Mahyuddin, M.N., Wei, C.Z. ve Arshad, M.R., “FPGA as an Embedded System of a Mobile Robot with incorporated Neuro-Fuzzy Algorithm for Obstacle Avoidance Mission”, MASAUM Journal of Basic and Applied Sciences (MJBAS), Cilt 1, No 3, 361-367, 2009.
[16]. Echanobe, J., del Campo, I., ve Bosque, G., “An adaptive neuro-fuzzy system for efficient implementations”, Information Sciences, Cilt 178, No 9, 2150-2162, 2008.
[17]. Jang, J.-S. R., “ANFIS: adaptive-network-based fuzzy inference system”, IEEE Trans. on Systems, Man, and Cybernetics, Cilt 23, No 3, 665-684, 1993.
[18]. Çavuşlu, M. A. , Karakuzu, C., Şahin, S., Yakut, M., “Neural Network Training Based on FPGA with Floating Point Number Format and It‟s Performance”, Neural Computing & Applications, Cilt 20, No 2, 195–202, 2011.

Thank you for copying data from http://www.arastirmax.com