A Spiking Neural Network implemented with Single-Electron Transistors and NoCs

Pês, Beatriz dos Santos; Guimarães, Janaína Gonçalves; Oroski, Elder; Bonfim, Marlio

doi:10.1016/j.nancom.2018.06.001

Cited by 4 publications

(2 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The best advantage of artificial neural networks (ANNs) lies in their automatic learning capability, which allows solving problems without requiring writing complex rules. The results in this paper, in excellent agreement with precedent results Simulated with different methods (technique) applied to various Single Electron Transistors (SETs) models [16,17,18,19,20,21,22]. This last observation shows the applicability of the artificial neural network to the study of nanoscale CMOS circuits.…”

Section: B the Neural Network Transistor Modelsupporting

confidence: 85%

Modeling and Simulation of Single-Electron Transistor (SET) with Aluminum Island Using Neural Network

Abdelkrim¹

2019

Carpathian Journal of Electronic and Computer Engineering

View full text Add to dashboard Cite

SET is important in the field of nanoelectronics since a decade. This paper presents electrical characteristic of Single-Electron Transistor (SET) with Aluminum Island using Neural Network. The I-V characteristic of the Single-Electron Transistor (SET) is predicted according to different parameters (VG, T, VD, C, and R). The simulation process is based on analytical transistor model and neural network transistor model. Single Electron Transistor (SET) is the simplest device in which the effect of Coulomb blockade can be observed.

show abstract

Section: B the Neural Network Transistor Modelsupporting

confidence: 85%

Modeling and Simulation of Single-Electron Transistor (SET) with Aluminum Island Using Neural Network

Abdelkrim¹

2019

Carpathian Journal of Electronic and Computer Engineering

View full text Add to dashboard Cite

show abstract

“…The intrinsically non-linear response, mesoscopic dimension, low energy demand, and fast operation make SET devices appealing for neuromorphic computing. − In contrast to the traditional von Neuman architecture, it employs bioinspired networks of non-linear devices and weighting elements for data processing and storage, allowing computation orders of faster magnitude, as demonstrated in several remarkable systems. − This opens the possibility of overcoming the present limitations faced by Moore’s law and Dennard’s scaling . Of great importance for neuromorphic computing is the plasticity, related to memory effects in both short- and long-term regimes, for which many resort to memristive devices .…”

Section: Introductionmentioning

confidence: 99%

Single-Electron Tunneling PbS/InP Heterostructure Nanoplatelets for Synaptic Operations

Jarschel

Kim

Biadala

et al. 2021

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Power consumption, thermal management, and wiring challenge of the binary serial architecture drive the search for alternative paradigms to computing. Of special interest is neuromorphic computing, in which materials and device structures are designed to mimic neuronal functionalities with energy-efficient non-linear responses and both short-and long-term plasticities. In this work, we explore and report on the enabling potential of single-electron tunneling (SET) in PbS nanoplatelets epitaxially grown in the liquid phase on InP, which present these key features. By extrapolating the experimental data in the SET regime, we predict and model synaptic operations. The low-energy (

show abstract