Title :
FPGA implementation of pulse coupled oscillator
Author :
Maeda, Yutaka ; Nakatsuka, Makito
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Kansai Univ., Suita, Japan
fDate :
31 July-4 Aug. 2005
Abstract :
This paper proposes a learning scheme for pulse coupled oscillators using the simultaneous perturbation optimization method. Moreover, hardware implementation of the pulse coupled oscillator with learning capability using field programmable gate array is described. It is complicated for ordinary method to find proper parameter values in the pulse coupled oscillator for a proper period, since the pulse coupled oscillator is a kind of recurrent network. The simultaneous perturbation optimization method can give a simple solution to this problem. In addition, this scheme is suitable for hardware realization. Some simulation results of the learning method for pulse coupled oscillators are shown. At the same time, we fabricated FPGA pulse coupled oscillator system with learning mechanism. Results by the hardware system are also shown.
Keywords :
field programmable gate arrays; optimisation; oscillators; perturbation techniques; recurrent neural nets; FPGA implementation; FPGA pulse coupled oscillator system; field programmable gate array; learning scheme; perturbation optimization; recurrent network; Coupling circuits; Electronic circuits; Electronics industry; Field programmable gate arrays; Hardware; Learning systems; Neurons; Optimization methods; Oscillators; Time to market;
Conference_Titel :
Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on
Print_ISBN :
0-7803-9048-2
DOI :
10.1109/IJCNN.2005.1555905