Skip to Main content Skip to Navigation
Theses

Contribution à l'étude de l'impact des nanotechnologies sur les Architectures : Apprentissage d'inspiration neuronale de fonctions logiques pour circuits programmables

Abstract : My thesis work firstly deals with issues coming from the semiconductor evolution. Then I have done some investigations in the properties of novel nanodevices. They are all different from CMOS according to several criteria. Secondly, having knowledge overall architectures, I chose to develop neural networks for emulating logic circuit. Indeed, the logic gates can be efficiently emulated by neural networks made from carbon nanotubes and multilevel memories. At last, based on neural redundancy, the robustness of neural gate architecture has been assessed by simulation, showing that kind of circuit can be reconfigurable through on chip learning process.
Document type :
Theses
Complete list of metadatas

Cited literature [53 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00422144
Contributor : Michel He <>
Submitted on : Tuesday, October 6, 2009 - 8:45:22 AM
Last modification on : Friday, October 23, 2020 - 4:51:27 PM
Long-term archiving on: : Tuesday, June 15, 2010 - 9:11:41 PM

Identifiers

  • HAL Id : tel-00422144, version 1

Collections

Citation

Michel He. Contribution à l'étude de l'impact des nanotechnologies sur les Architectures : Apprentissage d'inspiration neuronale de fonctions logiques pour circuits programmables. Autre. Université Paris Sud - Paris XI, 2008. Français. ⟨tel-00422144⟩

Share

Metrics

Record views

380

Files downloads

740