Skip to Main content Skip to Navigation
Conference papers

Testing Approximate Digital Circuits: Challenges and Opportunities

Marcello Traiola 1 Arnaud Virazel 1 Patrick Girard 1 Mario Barbareschi 2 Alberto Bosio 1
1 TEST - TEST
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Résumé : Approximate Computing (AxC) is based on the observation that a significant class of applications can inherently tolerate a certain amount of errors (i.e., the output quality is still acceptable to the user). AxC exploits this characteristic in order to apply selective approximations or occasional relaxations of the specifications. The benefit is a significant gain in energy efficiency and area reduction for Integrated Circuits (ICs). During the mission-mode, the IC can be affected by faults caused by environmental perturbations (e.g., radiations, electromagnetic interference), or aging-related phenomena. These faults may be propagated through the IC structure to the outputs and thus lead to observable errors. These errors (due to faults) may worsen the accuracy reduction - already introduced by the AxC - and possibly lead it to become unacceptable. This paper aims at investigating the challenges and the opportunities related to the test of AxC ICs.
Complete list of metadata

https://hal-lirmm.ccsd.cnrs.fr/lirmm-03033024
Contributor : Isabelle Gouat <>
Submitted on : Tuesday, December 1, 2020 - 10:43:15 AM
Last modification on : Monday, December 7, 2020 - 7:28:05 PM
Long-term archiving on: : Tuesday, March 2, 2021 - 6:41:41 PM

File

testing-approximate-digital.pd...
Files produced by the author(s)

Identifiers

Collections

Citation

Marcello Traiola, Arnaud Virazel, Patrick Girard, Mario Barbareschi, Alberto Bosio. Testing Approximate Digital Circuits: Challenges and Opportunities. IEEE 19th Latin-American Test Symposium (LATS), Mar 2018, Sao Paulo, Brazil. pp.1-6, ⟨10.1109/LATW.2018.8349681⟩. ⟨lirmm-03033024⟩

Share

Metrics

Record views

34

Files downloads

33