Thesis supervisor: István Oniga
Location of studies (in Hungarian): Debreceni Egyetem Informatikai Kar Abbreviation of location of studies: DE IK
Description of the research topic:
Syllabus The aim of the research is to implement artificial neural networks using reconfigurable devices (FPGA). In first phase of the research the implementation is done using high level programming language (Matlab, C, C++). In next phase the PhD. student can chose between possible hardware implementation techniques: using hardware description language (HDL), implementation using System Generator or with high level synthesis (HLS) tools. Case studies: application possibilities of hardware implemented neural networks in pattern recognition (activity recognition).
Bibliography
1. Dennis Silage, Trends in Embedded Design Using Programmable Gate Arrays, Bookstand Publishing 2013, 320 oldal, ISBN 978-1-61863-541-9
2. Editors: Omondi, Amos R., Rajapakse, Jagath C. (Eds.), FPGA Implementations of Neural Networks, Publisher Springer (2006)
3. Jin, Zhanpeng, Autonomously Reconfigurable Artificial Neural Network on a Chip. Doctoral Dissertation, University of Pittsburgh, 2010.
4. S. Oniga, et al., Optimizing FPGA Implementation of Feed-Forward Neural Networks, Proceedings of the 11th International Conference on Optimization of Electrical and Electronic Equipment, OPTIM 2008, Brasov, Romania, pp.31-36, http://dx.doi.org/10.1109/OPTIM.2008.4602494.
5. Avvaru Srinivasulu, FPGA Implementation of Hopfield Neural Network, LAP Lambert Academic Publishing, 2012, ISBN 3848435454, 9783848435456.
6. Kiran Kintali and Yongfeng Gu, Model-Based Design with Simulink, HDL Coder, and Xilinx System Generator for DSP, MathWorks, White paper, 2012.
Recommended language skills (in Hungarian): angol Number of students who can be accepted: 1