FPGA based silicon spiking neural array

Andrew Cassidy, Susan Denham, Patrick Kanold, Andreas Andreou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Rapid design time, low cost, flexibility, digital precision, and stability are characteristics that favor FPGAs as a promising alternative to analog VLSI based approaches for designing neuromorphic systems. High computational power as well as low size, weight, and power (SWAP) are advantages that FPGAs demonstrate over software based neuromorphic systems. We present an FPGA based array of Leaky-Integrate and Fire (LIF) artificial neurons. Using this array, we demonstrate three neural computational experiments: auditory Spatio-Temporal Receptive Fields (STRFs), a neural parameter optimizing algorithm, and an implementation of the Spike Time Dependant Plasticity (STDP) learning rule.

Original languageEnglish (US)
Title of host publicationConference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007
Pages75-78
Number of pages4
DOIs
StatePublished - Dec 1 2007
Externally publishedYes
EventIEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007 - Montreal, QC, Canada
Duration: Nov 27 2007Nov 30 2007

Publication series

NameConference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007

Conference

ConferenceIEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007
CountryCanada
CityMontreal, QC
Period11/27/0711/30/07

Fingerprint

Field programmable gate arrays (FPGA)
Silicon
Neurons
Plasticity
Fires
Costs
Experiments

ASJC Scopus subject areas

  • Hardware and Architecture
  • Biomedical Engineering

Cite this

Cassidy, A., Denham, S., Kanold, P., & Andreou, A. (2007). FPGA based silicon spiking neural array. In Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007 (pp. 75-78). [4463312] (Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007). https://doi.org/10.1109/BIOCAS.2007.4463312

FPGA based silicon spiking neural array. / Cassidy, Andrew; Denham, Susan; Kanold, Patrick; Andreou, Andreas.

Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007. 2007. p. 75-78 4463312 (Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cassidy, A, Denham, S, Kanold, P & Andreou, A 2007, FPGA based silicon spiking neural array. in Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007., 4463312, Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007, pp. 75-78, IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007, Montreal, QC, Canada, 11/27/07. https://doi.org/10.1109/BIOCAS.2007.4463312
Cassidy A, Denham S, Kanold P, Andreou A. FPGA based silicon spiking neural array. In Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007. 2007. p. 75-78. 4463312. (Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007). https://doi.org/10.1109/BIOCAS.2007.4463312
Cassidy, Andrew ; Denham, Susan ; Kanold, Patrick ; Andreou, Andreas. / FPGA based silicon spiking neural array. Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007. 2007. pp. 75-78 (Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007).
@inproceedings{a66994c69ff0465f971ef8aec806b8ac,
title = "FPGA based silicon spiking neural array",
abstract = "Rapid design time, low cost, flexibility, digital precision, and stability are characteristics that favor FPGAs as a promising alternative to analog VLSI based approaches for designing neuromorphic systems. High computational power as well as low size, weight, and power (SWAP) are advantages that FPGAs demonstrate over software based neuromorphic systems. We present an FPGA based array of Leaky-Integrate and Fire (LIF) artificial neurons. Using this array, we demonstrate three neural computational experiments: auditory Spatio-Temporal Receptive Fields (STRFs), a neural parameter optimizing algorithm, and an implementation of the Spike Time Dependant Plasticity (STDP) learning rule.",
author = "Andrew Cassidy and Susan Denham and Patrick Kanold and Andreas Andreou",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/BIOCAS.2007.4463312",
language = "English (US)",
isbn = "142441525X",
series = "Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007",
pages = "75--78",
booktitle = "Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007",

}

TY - GEN

T1 - FPGA based silicon spiking neural array

AU - Cassidy, Andrew

AU - Denham, Susan

AU - Kanold, Patrick

AU - Andreou, Andreas

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Rapid design time, low cost, flexibility, digital precision, and stability are characteristics that favor FPGAs as a promising alternative to analog VLSI based approaches for designing neuromorphic systems. High computational power as well as low size, weight, and power (SWAP) are advantages that FPGAs demonstrate over software based neuromorphic systems. We present an FPGA based array of Leaky-Integrate and Fire (LIF) artificial neurons. Using this array, we demonstrate three neural computational experiments: auditory Spatio-Temporal Receptive Fields (STRFs), a neural parameter optimizing algorithm, and an implementation of the Spike Time Dependant Plasticity (STDP) learning rule.

AB - Rapid design time, low cost, flexibility, digital precision, and stability are characteristics that favor FPGAs as a promising alternative to analog VLSI based approaches for designing neuromorphic systems. High computational power as well as low size, weight, and power (SWAP) are advantages that FPGAs demonstrate over software based neuromorphic systems. We present an FPGA based array of Leaky-Integrate and Fire (LIF) artificial neurons. Using this array, we demonstrate three neural computational experiments: auditory Spatio-Temporal Receptive Fields (STRFs), a neural parameter optimizing algorithm, and an implementation of the Spike Time Dependant Plasticity (STDP) learning rule.

UR - http://www.scopus.com/inward/record.url?scp=77956356646&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956356646&partnerID=8YFLogxK

U2 - 10.1109/BIOCAS.2007.4463312

DO - 10.1109/BIOCAS.2007.4463312

M3 - Conference contribution

AN - SCOPUS:77956356646

SN - 142441525X

SN - 9781424415250

T3 - Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007

SP - 75

EP - 78

BT - Conference Proceedings - IEEE Biomedical Circuits and Systems Conference Healthcare Technology, BiOCAS2007

ER -