Soft-Error Analysis of RRAM 1T1R Compute-In-Memory Core for Artificial Neural Networks

Ruolan Jia, Stefan Pechmann, Fritscher Markus, Christian Wenger, Lei Zhang, Amelie Hagelauer

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This work analyses SEU-induced soft-errors in analog compute-in-memory cores using resistive random-Access memory (RRAM) for artificial neural networks, where their bitcells utilize one-Transistor-one-RRAM (1T1R) structure. This is modeled by combining the Stanford-PKU RRAM Model and the model of the radiation-induced photocurrent in access transistors. As results, this work derives the maximal RRAM crossbar size without occurring any logic flip and indicates the requirements for RRAM technology to achieve a SEU-resilient 1T1R compute-in memory cores.

Original languageEnglish
Title of host publication2024 39th Conference on Design of Circuits and Integrated Systems, DCIS 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350364392
DOIs
StatePublished - 2024
Event39th Conference on Design of Circuits and Integrated Systems, DCIS 2024 - Catania, Italy
Duration: 13 Nov 202415 Nov 2024

Publication series

Name2024 39th Conference on Design of Circuits and Integrated Systems, DCIS 2024

Conference

Conference39th Conference on Design of Circuits and Integrated Systems, DCIS 2024
Country/TerritoryItaly
CityCatania
Period13/11/2415/11/24

Keywords

  • Compute-In-Memory
  • IT1R
  • MAC
  • Modeling
  • Radiation Hardening
  • RHBD
  • RRAM
  • Single Effect Upsets

Fingerprint

Dive into the research topics of 'Soft-Error Analysis of RRAM 1T1R Compute-In-Memory Core for Artificial Neural Networks'. Together they form a unique fingerprint.

Cite this