Adaptive Approximate Computing in Arithmetic Datapaths

Sana Mazahir; Osman Hasan; Muhammad Shafique

doi:10.1109/MDAT.2017.2772874

Adaptive Approximate Computing in Arithmetic Datapaths

Sana Mazahir, Osman Hasan, Muhammad Shafique

Electrical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Editor's note: Approximate arithmetic units are used in error tolerant applications, like image processing, to minimize energy, and resource usage in general, by keeping accuracy to the minimum level required. However, in a datapath errors may accumulate over successive approximate operations. This article proposes a scheme to bound the size of the errors with adaptive error detection and correction as an integral part of the datapath. - Axel Jantsch, TU Wien.

Original language	English (US)
Pages (from-to)	65-74
Number of pages	10
Journal	IEEE Design and Test
Volume	35
Issue number	4
DOIs	https://doi.org/10.1109/MDAT.2017.2772874
State	Published - Aug 2018

Keywords

accuracy configurability
adaptive computing
approximate adders
Approximate computing
approximate multipliers
error detection and correction (EDC)
self-aware architectures

ASJC Scopus subject areas

Software
Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/MDAT.2017.2772874

Cite this

@article{834e3e57fce8481797e98dab46c80b2c,

title = "Adaptive Approximate Computing in Arithmetic Datapaths",

abstract = "Editor's note: Approximate arithmetic units are used in error tolerant applications, like image processing, to minimize energy, and resource usage in general, by keeping accuracy to the minimum level required. However, in a datapath errors may accumulate over successive approximate operations. This article proposes a scheme to bound the size of the errors with adaptive error detection and correction as an integral part of the datapath. - Axel Jantsch, TU Wien.",

keywords = "accuracy configurability, adaptive computing, approximate adders, Approximate computing, approximate multipliers, error detection and correction (EDC), self-aware architectures",

author = "Sana Mazahir and Osman Hasan and Muhammad Shafique",

year = "2018",

month = aug,

doi = "10.1109/MDAT.2017.2772874",

language = "English (US)",

volume = "35",

pages = "65--74",

journal = "IEEE Design and Test",

issn = "2168-2356",

publisher = "IEEE Computer Society",

number = "4",

}

TY - JOUR

T1 - Adaptive Approximate Computing in Arithmetic Datapaths

AU - Mazahir, Sana

AU - Hasan, Osman

AU - Shafique, Muhammad

PY - 2018/8

Y1 - 2018/8

N2 - Editor's note: Approximate arithmetic units are used in error tolerant applications, like image processing, to minimize energy, and resource usage in general, by keeping accuracy to the minimum level required. However, in a datapath errors may accumulate over successive approximate operations. This article proposes a scheme to bound the size of the errors with adaptive error detection and correction as an integral part of the datapath. - Axel Jantsch, TU Wien.

AB - Editor's note: Approximate arithmetic units are used in error tolerant applications, like image processing, to minimize energy, and resource usage in general, by keeping accuracy to the minimum level required. However, in a datapath errors may accumulate over successive approximate operations. This article proposes a scheme to bound the size of the errors with adaptive error detection and correction as an integral part of the datapath. - Axel Jantsch, TU Wien.

KW - accuracy configurability

KW - adaptive computing

KW - approximate adders

KW - Approximate computing

KW - approximate multipliers

KW - error detection and correction (EDC)

KW - self-aware architectures

UR - http://www.scopus.com/inward/record.url?scp=85034272206&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034272206&partnerID=8YFLogxK

U2 - 10.1109/MDAT.2017.2772874

DO - 10.1109/MDAT.2017.2772874

M3 - Article

AN - SCOPUS:85034272206

SN - 2168-2356

VL - 35

SP - 65

EP - 74

JO - IEEE Design and Test

JF - IEEE Design and Test

IS - 4

ER -

Adaptive Approximate Computing in Arithmetic Datapaths

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this