A Solid-State Neuron for Spiking Neural Network Implementation

Yajie Yajie Chen, Steve Hall, Liam McDaid, Octavian Buiu, Peter Kelly

Research output: Contribution to journalArticle

Abstract

This paper presents a compact analog neuron cell incorporating an array of charge-coupledsynapses connected via a common output terminal. The novel silicon synapse is based on a two stage charge-coupled device where the weighting functionality can be integrated into the first stage. A pre-synaptic spike to the second gate allows the charge under the first gate to drift onto the floating diffusion output stage to produce a current, or voltage spike. Parallel defined synapses are each assigned to the left hand side of a current mirror gate where the right hand side feeds into a thresholding inverter. The decay of the membrane potential is mimicked by the charge leakage through a reverse-biased diode, whose model is verified by comparing the simulations and measured data. Spice simulation results show that the proposed neuron cell is capable of capturing the summing and thresholding dynamics of biological neurons.
LanguageEnglish
Pages83-89
Number of pages7
JournalEngineering Letters
Volume16
Issue number1
Publication statusPublished - 19 Feb 2008

Fingerprint

Neurons
Neural networks
Charge coupled devices
Diodes
Mirrors
Membranes
Silicon
Electric potential

Cite this

Yajie Chen, Y., Hall, S., McDaid, L., Buiu, O., & Kelly, P. (2008). A Solid-State Neuron for Spiking Neural Network Implementation. Engineering Letters, 16(1), 83-89.
Yajie Chen, Yajie ; Hall, Steve ; McDaid, Liam ; Buiu, Octavian ; Kelly, Peter. / A Solid-State Neuron for Spiking Neural Network Implementation. In: Engineering Letters. 2008 ; Vol. 16, No. 1. pp. 83-89.
@article{8281e170840b4eb0aecb566e2a22fab1,
title = "A Solid-State Neuron for Spiking Neural Network Implementation",
abstract = "This paper presents a compact analog neuron cell incorporating an array of charge-coupledsynapses connected via a common output terminal. The novel silicon synapse is based on a two stage charge-coupled device where the weighting functionality can be integrated into the first stage. A pre-synaptic spike to the second gate allows the charge under the first gate to drift onto the floating diffusion output stage to produce a current, or voltage spike. Parallel defined synapses are each assigned to the left hand side of a current mirror gate where the right hand side feeds into a thresholding inverter. The decay of the membrane potential is mimicked by the charge leakage through a reverse-biased diode, whose model is verified by comparing the simulations and measured data. Spice simulation results show that the proposed neuron cell is capable of capturing the summing and thresholding dynamics of biological neurons.",
author = "{Yajie Chen}, Yajie and Steve Hall and Liam McDaid and Octavian Buiu and Peter Kelly",
year = "2008",
month = "2",
day = "19",
language = "English",
volume = "16",
pages = "83--89",
journal = "Engineering Letters",
issn = "1816-093X",
number = "1",

}

Yajie Chen, Y, Hall, S, McDaid, L, Buiu, O & Kelly, P 2008, 'A Solid-State Neuron for Spiking Neural Network Implementation', Engineering Letters, vol. 16, no. 1, pp. 83-89.

A Solid-State Neuron for Spiking Neural Network Implementation. / Yajie Chen, Yajie; Hall, Steve; McDaid, Liam; Buiu, Octavian; Kelly, Peter.

In: Engineering Letters, Vol. 16, No. 1, 19.02.2008, p. 83-89.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Solid-State Neuron for Spiking Neural Network Implementation

AU - Yajie Chen, Yajie

AU - Hall, Steve

AU - McDaid, Liam

AU - Buiu, Octavian

AU - Kelly, Peter

PY - 2008/2/19

Y1 - 2008/2/19

N2 - This paper presents a compact analog neuron cell incorporating an array of charge-coupledsynapses connected via a common output terminal. The novel silicon synapse is based on a two stage charge-coupled device where the weighting functionality can be integrated into the first stage. A pre-synaptic spike to the second gate allows the charge under the first gate to drift onto the floating diffusion output stage to produce a current, or voltage spike. Parallel defined synapses are each assigned to the left hand side of a current mirror gate where the right hand side feeds into a thresholding inverter. The decay of the membrane potential is mimicked by the charge leakage through a reverse-biased diode, whose model is verified by comparing the simulations and measured data. Spice simulation results show that the proposed neuron cell is capable of capturing the summing and thresholding dynamics of biological neurons.

AB - This paper presents a compact analog neuron cell incorporating an array of charge-coupledsynapses connected via a common output terminal. The novel silicon synapse is based on a two stage charge-coupled device where the weighting functionality can be integrated into the first stage. A pre-synaptic spike to the second gate allows the charge under the first gate to drift onto the floating diffusion output stage to produce a current, or voltage spike. Parallel defined synapses are each assigned to the left hand side of a current mirror gate where the right hand side feeds into a thresholding inverter. The decay of the membrane potential is mimicked by the charge leakage through a reverse-biased diode, whose model is verified by comparing the simulations and measured data. Spice simulation results show that the proposed neuron cell is capable of capturing the summing and thresholding dynamics of biological neurons.

M3 - Article

VL - 16

SP - 83

EP - 89

JO - Engineering Letters

T2 - Engineering Letters

JF - Engineering Letters

SN - 1816-093X

IS - 1

ER -

Yajie Chen Y, Hall S, McDaid L, Buiu O, Kelly P. A Solid-State Neuron for Spiking Neural Network Implementation. Engineering Letters. 2008 Feb 19;16(1):83-89.