Title :
Novel impulse neural circuits for pulsed neural network
Author_Institution :
Dept. of Res., Electron. & Telecommun. Res. Inst., Daejeon, South Korea
Abstract :
Two novel neural circuits based on stochastic pulse coding (SPC) and noise feedback pulse coding (NFPC) are presented. The artificial axon hillock circuits to realize these codings are compared with respect to the coding efficiency. The axon hillock circuit using NFPC needs much less number of pulses than that using SPC to convert analog neuron body voltage into a pulse sequence under the same signal to noise ratio. This result suggests that the NFPC neural circuit can improve the speed of computation and communication, compared with the SPC neural circuit. Simulation and experimental results are included.
Keywords :
MOS integrated circuits; circuit feedback; encoding; mixed analogue-digital integrated circuits; neural nets; probabilistic logic; pulse circuits; analog neuron voltage; axon hillock circuits; impulse neural circuits; noise feedback pulse coding; pulse sequence; pulsed neural network; stochastic pulse coding; Artificial neural networks; Circuit noise; Feedback circuits; Nerve fibers; Neural networks; Neurofeedback; Neurons; Pulse circuits; Stochastic resonance; Voltage;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714356
