Repository logo
  • English
  • العربية
  • বাংলা
  • Català
  • Čeština
  • Deutsch
  • Ελληνικά
  • Español
  • Suomi
  • Français
  • Gàidhlig
  • हिंदी
  • Magyar
  • Italiano
  • Қазақ
  • Latviešu
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Srpski (lat)
  • Српски
  • Svenska
  • Türkçe
  • Yкраї́нська
  • Tiếng Việt
Log In
New user? Click here to register.Have you forgotten your password?
  1. Home
  2. Scholalry Output
  3. Publications
  4. Single Exact Single Approximate Adders and Single Exact Dual Approximate Adders
 
  • Details

Single Exact Single Approximate Adders and Single Exact Dual Approximate Adders

Source
IEEE Transactions on Very Large Scale Integration VLSI Systems
ISSN
10638210
Date Issued
2023-07-01
Author(s)
Jha, Chandan Kumar
Nandi, Ankita
Mekie, Joycee  
DOI
10.1109/TVLSI.2023.3268275
Volume
31
Issue
7
Abstract
In this article, we present the design of approximate adders which provide dynamic runtime configurability between exact and approximate modes at the circuit level. We propose the single exact single approximate (SESA) adders that allow for fine grain configurability between exact and approximate modes. We also propose the single exact dual approximate (SEDA) adder that allows for coarse grain configurability between exact and approximate modes. Unlike SESA adders, the SEDA adder allows for two approximate computations at a time. Both the SESA and SEDA adders have a maximum bounded error. We implemented SESA and SEDA adders using UMC 28-nm technology node and evaluated them using the Cadence virtuoso tool. On average, SESA and SEDA adders consume 40% and 51% lesser energy when compared with the exact mirror adder when used in approximate mode. We have evaluated our result on image addition and image enhancement using 16-bit SESA and SEDA adders. We also evaluated 32-bit SESA and SEDA adders on the Moby benchmarks to highlight their use in approximate processors.
Unpaywall
URI
https://d8.irins.org/handle/IITG2025/26749
Subjects
Adders | approximate adders | configurable | image processing | \textit{gem5}
IITGN Knowledge Repository Developed and Managed by Library

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science

  • Privacy policy
  • End User Agreement
  • Send Feedback
Repository logo COAR Notify