Brainway computer implementation availability using reconfiguration

This paper describes a study of the brainway computer implementation availability. As an example of this brainway reconfigurable computer, it is proposed a digital hardware implementation of an artificial neural network, neocognitron, which represents the human neural system. This model may represen...

Full description

Saved in:
Bibliographic Details
Main Authors: Ide, A.N., Saito, J.H., Abib, S.
Format: Conference Proceeding
Language:English
Subjects:
Arithmetic
Artificial neural networks
Brain modeling
Circuits
Computer networks
Concurrent computing
Field programmable gate arrays
Humans
Neural network hardware
Pattern analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper describes a study of the brainway computer implementation availability. As an example of this brainway reconfigurable computer, it is proposed a digital hardware implementation of an artificial neural network, neocognitron, which represents the human neural system. This model may represent all the human brain system as the sensory information inputs, their association and actions. It is an FPGA based reconfigurable cluster-processing platform, where the whole artificial neural network is processed. It is described the architecture of each arithmetic circuit, which composes the neocognitron structure, and how they are processed. Finally, an analysis of the space occupied for each circuit is done and the processing time is measured, considering the neocognitron model applied to visual pattern recognition. In conclusion, the results showed the availability of the brainway computer: fast design hardware implementation; pipeline and parallel computations; and reconfiguration.
DOI:10.1109/ICONIP.2002.1198206