A CONTENT-ADDRESSABLE POLYCHROMATIC NEURAL-NET USING A (CEFE)-DOPED LINBO3 PHOTOREFRACTIVE CRYSTAL

Authors
Citation
Fts. Yu et al., A CONTENT-ADDRESSABLE POLYCHROMATIC NEURAL-NET USING A (CEFE)-DOPED LINBO3 PHOTOREFRACTIVE CRYSTAL, Optics communications, 107(3-4), 1994, pp. 300-308
Citations number
8
Categorie Soggetti
Optics
Journal title
ISSN journal
00304018
Volume
107
Issue
3-4
Year of publication
1994
Pages
300 - 308
Database
ISI
SICI code
0030-4018(1994)107:3-4<300:ACPNUA>2.0.ZU;2-U
Abstract
A polychromatic neural net using a (Ce:Fe)-doped LiNbO3 photorefractiv e crystal is presented. This neural net is a two-level high-content ad dressable memory. The first level is a polychromatic Hamming net for c olor image classification, and the second level is a mapping net, base d on a photorefractive crystal associative memory. The major advantage of this neural net is the large storage capacity and it requires a fe wer interconnection links. Experimental demonstrations are provided, i n which is shown that the proposed neural net is consistent with the t heoretical model.