A Vlsi Neural Network Processor Based on Hippocampal Model

Authors:
Richard H. Tsai;Bing J. Sheu;Theodore W. Berger
Affiliations:
Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089;Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089;Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90089
Venue:
Analog Integrated Circuits and Signal Processing
Year:
1998

Citing 7
Cited 0

Analog VLSI and neural systems

Analog VLSI and neural systems
1994 Special Issue: A biologically based model of functional properties of the hippocampus

Neural Networks - Special issue: models of neurodynamics and behavior
The computational eye

IEEE Spectrum
Neuromorphic vision chips

IEEE Spectrum
Ocular implants for the blind

IEEE Spectrum
Neural Information Processing and VLSI

Neural Information Processing and VLSI
An Introduction to Neural and Electronic Networks

An Introduction to Neural and Electronic Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

The hippocampal region of the brain system can be analyzedwith the nonlinear system modeling approach. The input-outputrelationship of the neural units is best represented by the kernelfunctions of different complexities. The modeling expressionof the first and second order kernels are computed in analogcurrent-mode instead of digital data processing in order to fullyexplore massively parallel processing capability of the neuralnetworks. Two distinct methods are utilized: the table-look-upapproach and the model-based approach. The former can achievehigh accuracy but consumes large silicon area while the lattersaves silicon area and maintains moderately high accuracy. Circuit-levelsimulation results and experimental data from two test structuresare presented.