Title :
Configurable multi-layer CNN-UM emulator on FPGA
Author :
Nagy, Zoltan ; Szolgay, Peter
Author_Institution :
Dept. of Image Process. & Neurocomputing, Univ. of Veszprem, Hungary
Abstract :
A new emulated digital multi-layer CNN-UM chip architecture called Falcon has been developed. In this paper the main steps of the FPGA implementation are introduced. Main results are as follows: CNN-UM architecture emulated on Xilinx Virtex series FPGA, 3D non-linear spatio-temporal dynamics can be implemented on this architecture. The critical parameters of the implementation in single layer configuration are 55 million cell update/second/processor core or equivalently 1 GOPS computing performance. In face of the high performance the power requirements of the architecture are relatively low only ∼3 W per processor core. Using re-configurable devices to implement emulated digital architectures provides more flexibility compared to the custom VLSI designs because different Falcon architectures can be used on the same FPGA device.
Keywords :
cellular neural nets; field programmable gate arrays; multilayer perceptrons; neural chips; neural net architecture; reconfigurable architectures; 3D nonlinear spatio-temporal dynamics; CNN Universal Machine; CNN-UM chip architecture; FPGA; Falcon; Xilinx Virtex series FPGA; cellular neural network; computing performance; configurable multilayer CNN-UM emulator; emulated digital architectures; power requirements; reconfigurable devices; Analog computers; Cellular neural networks; Computer architecture; Equations; Field programmable gate arrays; Laboratories; Signal processing; State feedback; Turing machines; Very large scale integration;
Conference_Titel :
Cellular Neural Networks and Their Applications, 2002. (CNNA 2002). Proceedings of the 2002 7th IEEE International Workshop on
Print_ISBN :
981-238-121-X
DOI :
10.1109/CNNA.2002.1035049