Energy-efficient soft error-tolerant digital signal processing

Byonghyo Shim, Naresh R. Shanbhag

Research output: Contribution to journalArticle

73 Citations (Scopus)

Abstract

In this paper, we present energy-efficient soft error-tolerant techniques for digital signal processing (DSP) systems. The proposed technique, referred to as algorithmic soft error-tolerance (ASET), employs low-complexity estimators of a main DSP block to achieve reliable operation in the presence of soft errors. Three distinct ASET techniques-spatial, temporal and spatio-temporal-are presented. For frequency selective finite-impulse response (FIR) filtering, it is shown that the proposed techniques provide robustness in the presence of soft error rates of up to P er = 10 -2 and P er = 10 -3 in a single-event upset scenario. The power dissipation of the proposed techniques ranges from 1.1 X to 1.7 X (spatial ASET) and 1.05 X to 1.17 X (spatio-temporal and temporal ASET) when the desired signal-to-noise ratio SNR des = 25 dB. In comparison, the power dissipation of the commonly employed triple modular redundancy technique is 2.9 X.

Original languageEnglish
Article number1637464
Pages (from-to)336-348
Number of pages13
JournalIEEE Transactions on Very Large Scale Integration (VLSI) Systems
Volume14
Issue number4
DOIs
Publication statusPublished - 2006 Apr 1
Externally publishedYes

Fingerprint

Digital signal processing
Energy dissipation
Impulse response
Redundancy
Signal to noise ratio

Keywords

  • Digital signal processing (DSP)
  • Low-power
  • Reduced precision redundancy (RPR)
  • Reliability
  • Soft error tolerance
  • Triple modular redundancy (TMR)

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Cite this

Energy-efficient soft error-tolerant digital signal processing. / Shim, Byonghyo; Shanbhag, Naresh R.

In: IEEE Transactions on Very Large Scale Integration (VLSI) Systems, Vol. 14, No. 4, 1637464, 01.04.2006, p. 336-348.

Research output: Contribution to journalArticle

@article{babbfc8532b34be39136f903ffcf4dff,
title = "Energy-efficient soft error-tolerant digital signal processing",
abstract = "In this paper, we present energy-efficient soft error-tolerant techniques for digital signal processing (DSP) systems. The proposed technique, referred to as algorithmic soft error-tolerance (ASET), employs low-complexity estimators of a main DSP block to achieve reliable operation in the presence of soft errors. Three distinct ASET techniques-spatial, temporal and spatio-temporal-are presented. For frequency selective finite-impulse response (FIR) filtering, it is shown that the proposed techniques provide robustness in the presence of soft error rates of up to P er = 10 -2 and P er = 10 -3 in a single-event upset scenario. The power dissipation of the proposed techniques ranges from 1.1 X to 1.7 X (spatial ASET) and 1.05 X to 1.17 X (spatio-temporal and temporal ASET) when the desired signal-to-noise ratio SNR des = 25 dB. In comparison, the power dissipation of the commonly employed triple modular redundancy technique is 2.9 X.",
keywords = "Digital signal processing (DSP), Low-power, Reduced precision redundancy (RPR), Reliability, Soft error tolerance, Triple modular redundancy (TMR)",
author = "Byonghyo Shim and Shanbhag, {Naresh R.}",
year = "2006",
month = "4",
day = "1",
doi = "10.1109/TVLSI.2006.874359",
language = "English",
volume = "14",
pages = "336--348",
journal = "IEEE Transactions on Very Large Scale Integration (VLSI) Systems",
issn = "1063-8210",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Energy-efficient soft error-tolerant digital signal processing

AU - Shim, Byonghyo

AU - Shanbhag, Naresh R.

PY - 2006/4/1

Y1 - 2006/4/1

N2 - In this paper, we present energy-efficient soft error-tolerant techniques for digital signal processing (DSP) systems. The proposed technique, referred to as algorithmic soft error-tolerance (ASET), employs low-complexity estimators of a main DSP block to achieve reliable operation in the presence of soft errors. Three distinct ASET techniques-spatial, temporal and spatio-temporal-are presented. For frequency selective finite-impulse response (FIR) filtering, it is shown that the proposed techniques provide robustness in the presence of soft error rates of up to P er = 10 -2 and P er = 10 -3 in a single-event upset scenario. The power dissipation of the proposed techniques ranges from 1.1 X to 1.7 X (spatial ASET) and 1.05 X to 1.17 X (spatio-temporal and temporal ASET) when the desired signal-to-noise ratio SNR des = 25 dB. In comparison, the power dissipation of the commonly employed triple modular redundancy technique is 2.9 X.

AB - In this paper, we present energy-efficient soft error-tolerant techniques for digital signal processing (DSP) systems. The proposed technique, referred to as algorithmic soft error-tolerance (ASET), employs low-complexity estimators of a main DSP block to achieve reliable operation in the presence of soft errors. Three distinct ASET techniques-spatial, temporal and spatio-temporal-are presented. For frequency selective finite-impulse response (FIR) filtering, it is shown that the proposed techniques provide robustness in the presence of soft error rates of up to P er = 10 -2 and P er = 10 -3 in a single-event upset scenario. The power dissipation of the proposed techniques ranges from 1.1 X to 1.7 X (spatial ASET) and 1.05 X to 1.17 X (spatio-temporal and temporal ASET) when the desired signal-to-noise ratio SNR des = 25 dB. In comparison, the power dissipation of the commonly employed triple modular redundancy technique is 2.9 X.

KW - Digital signal processing (DSP)

KW - Low-power

KW - Reduced precision redundancy (RPR)

KW - Reliability

KW - Soft error tolerance

KW - Triple modular redundancy (TMR)

UR - http://www.scopus.com/inward/record.url?scp=33746638860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746638860&partnerID=8YFLogxK

U2 - 10.1109/TVLSI.2006.874359

DO - 10.1109/TVLSI.2006.874359

M3 - Article

AN - SCOPUS:33746638860

VL - 14

SP - 336

EP - 348

JO - IEEE Transactions on Very Large Scale Integration (VLSI) Systems

JF - IEEE Transactions on Very Large Scale Integration (VLSI) Systems

SN - 1063-8210

IS - 4

M1 - 1637464

ER -