FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware

Shvan Karim, Jim Harkin, Liam McDaid, Bryan Gardiner, Junxiu Liu, David Halliday, Andy Tyrrell, Jon Timmis, Alan Millard, Anju Johnson

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

Abstract

Spiking Astrocyte-neuron Networks (SANNs) model the adaptive/repair feature of the human brain. They integrate astrocyte cells with spiking neurons to facilitate a distributed and fine-grained self-repair capability at the synapse level. SANNs are more complex with the addition of astrocyte cells and require longer simulation times, as they are dynamic over much longer time-scales than traditional neural networks. Therefore, dedicated FPGA accelerators offer reductions in simulation times. To support the acceleration of SANNs, the capability of fault injection to synapses and monitoring significant levels of neuron and astrocyte data for off-chip transmission to PC-based analysis, are required. This paper presents an FPGA-based monitoring platform (FMP) for injecting faults and capturing and analyzing data acquired from the SANN FPGA accelerator, Astrobyte. The FMP uses custom logic and a NIOS II based system to control fault injection and data monitoring on the FPGA. Results show accurate accelerated simulations of fault injection scenarios using FMP with speedups up to 65 times greater compared with equivalent Matlab implementations.

LanguageEnglish
Title of host publication2018 IEEE International Symposium on Circuits and Systems (ISCAS)
PublisherIEEE
Number of pages5
ISBN (Electronic)9781538648810
ISBN (Print)9781538648827
DOIs
Publication statusPublished - 4 May 2018
EventIEEE International Symposium on Circuits and Systems
- Florence, Italy
Duration: 27 May 201830 May 2018
http://www.iscas2018.org/ (Link to Conference Website)

Conference

ConferenceIEEE International Symposium on Circuits and Systems
Abbreviated titleISCAS
CountryItaly
CityFlorence
Period27/05/1830/05/18
Internet address

Fingerprint

Field programmable gate arrays (FPGA)
Data acquisition
Neurons
Neural networks
Hardware
Monitoring
Particle accelerators
Repair
Astrocytes
Brain

Cite this

Karim, S., Harkin, J., McDaid, L., Gardiner, B., Liu, J., Halliday, D., ... Johnson, A. (2018). FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware. In 2018 IEEE International Symposium on Circuits and Systems (ISCAS) [8351512] IEEE. https://doi.org/10.1109/ISCAS.2018.8351512
Karim, Shvan ; Harkin, Jim ; McDaid, Liam ; Gardiner, Bryan ; Liu, Junxiu ; Halliday, David ; Tyrrell, Andy ; Timmis, Jon ; Millard, Alan ; Johnson, Anju. / FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware. 2018 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2018.
@inproceedings{f7e8ad90eb914b07aa04d7db158281d7,
title = "FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware",
abstract = "Spiking Astrocyte-neuron Networks (SANNs) model the adaptive/repair feature of the human brain. They integrate astrocyte cells with spiking neurons to facilitate a distributed and fine-grained self-repair capability at the synapse level. SANNs are more complex with the addition of astrocyte cells and require longer simulation times, as they are dynamic over much longer time-scales than traditional neural networks. Therefore, dedicated FPGA accelerators offer reductions in simulation times. To support the acceleration of SANNs, the capability of fault injection to synapses and monitoring significant levels of neuron and astrocyte data for off-chip transmission to PC-based analysis, are required. This paper presents an FPGA-based monitoring platform (FMP) for injecting faults and capturing and analyzing data acquired from the SANN FPGA accelerator, Astrobyte. The FMP uses custom logic and a NIOS II based system to control fault injection and data monitoring on the FPGA. Results show accurate accelerated simulations of fault injection scenarios using FMP with speedups up to 65 times greater compared with equivalent Matlab implementations.",
keywords = "Astrocytes, Data Acquisition, FPGA acceleration, Fault injection, Self repair, Spiking neural network",
author = "Shvan Karim and Jim Harkin and Liam McDaid and Bryan Gardiner and Junxiu Liu and David Halliday and Andy Tyrrell and Jon Timmis and Alan Millard and Anju Johnson",
year = "2018",
month = "5",
day = "4",
doi = "10.1109/ISCAS.2018.8351512",
language = "English",
isbn = "9781538648827",
booktitle = "2018 IEEE International Symposium on Circuits and Systems (ISCAS)",
publisher = "IEEE",

}

Karim, S, Harkin, J, McDaid, L, Gardiner, B, Liu, J, Halliday, D, Tyrrell, A, Timmis, J, Millard, A & Johnson, A 2018, FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware. in 2018 IEEE International Symposium on Circuits and Systems (ISCAS)., 8351512, IEEE, IEEE International Symposium on Circuits and Systems
, Florence, Italy, 27/05/18. https://doi.org/10.1109/ISCAS.2018.8351512

FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware. / Karim, Shvan; Harkin, Jim; McDaid, Liam; Gardiner, Bryan; Liu, Junxiu; Halliday, David; Tyrrell, Andy ; Timmis, Jon; Millard, Alan; Johnson, Anju.

2018 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2018. 8351512.

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

TY - GEN

T1 - FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware

AU - Karim, Shvan

AU - Harkin, Jim

AU - McDaid, Liam

AU - Gardiner, Bryan

AU - Liu, Junxiu

AU - Halliday, David

AU - Tyrrell, Andy

AU - Timmis, Jon

AU - Millard, Alan

AU - Johnson, Anju

PY - 2018/5/4

Y1 - 2018/5/4

N2 - Spiking Astrocyte-neuron Networks (SANNs) model the adaptive/repair feature of the human brain. They integrate astrocyte cells with spiking neurons to facilitate a distributed and fine-grained self-repair capability at the synapse level. SANNs are more complex with the addition of astrocyte cells and require longer simulation times, as they are dynamic over much longer time-scales than traditional neural networks. Therefore, dedicated FPGA accelerators offer reductions in simulation times. To support the acceleration of SANNs, the capability of fault injection to synapses and monitoring significant levels of neuron and astrocyte data for off-chip transmission to PC-based analysis, are required. This paper presents an FPGA-based monitoring platform (FMP) for injecting faults and capturing and analyzing data acquired from the SANN FPGA accelerator, Astrobyte. The FMP uses custom logic and a NIOS II based system to control fault injection and data monitoring on the FPGA. Results show accurate accelerated simulations of fault injection scenarios using FMP with speedups up to 65 times greater compared with equivalent Matlab implementations.

AB - Spiking Astrocyte-neuron Networks (SANNs) model the adaptive/repair feature of the human brain. They integrate astrocyte cells with spiking neurons to facilitate a distributed and fine-grained self-repair capability at the synapse level. SANNs are more complex with the addition of astrocyte cells and require longer simulation times, as they are dynamic over much longer time-scales than traditional neural networks. Therefore, dedicated FPGA accelerators offer reductions in simulation times. To support the acceleration of SANNs, the capability of fault injection to synapses and monitoring significant levels of neuron and astrocyte data for off-chip transmission to PC-based analysis, are required. This paper presents an FPGA-based monitoring platform (FMP) for injecting faults and capturing and analyzing data acquired from the SANN FPGA accelerator, Astrobyte. The FMP uses custom logic and a NIOS II based system to control fault injection and data monitoring on the FPGA. Results show accurate accelerated simulations of fault injection scenarios using FMP with speedups up to 65 times greater compared with equivalent Matlab implementations.

KW - Astrocytes

KW - Data Acquisition

KW - FPGA acceleration

KW - Fault injection

KW - Self repair

KW - Spiking neural network

UR - http://www.scopus.com/inward/record.url?scp=85057103074&partnerID=8YFLogxK

U2 - 10.1109/ISCAS.2018.8351512

DO - 10.1109/ISCAS.2018.8351512

M3 - Conference contribution

SN - 9781538648827

BT - 2018 IEEE International Symposium on Circuits and Systems (ISCAS)

PB - IEEE

ER -

Karim S, Harkin J, McDaid L, Gardiner B, Liu J, Halliday D et al. FPGA-based Fault-injection and Data Acquisition of Self-repairing Spiking Neural Network Hardware. In 2018 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE. 2018. 8351512 https://doi.org/10.1109/ISCAS.2018.8351512