A Low-Power RRAM Memory Block for Embedded, Multi-Level Weight and Bias Storage in Artificial Neural Networks

Please always quote using this URN: urn:nbn:de:bvb:29-opus4-174878
  • Pattern recognition as a computing task is very well suited for machine learning algorithms utilizing artificial neural networks (ANNs). Computing systems using ANNs usually require some sort of data storage to store the weights and bias values for the processing elements of the individual neurons. This paper introduces a memory block using resistive memory cells (RRAM) to realize this weight and bias storage in an embedded and distributed way while also offering programming and multi-level ability. By implementing power gating, overall power consumption is decreased significantly without data loss by taking advantage of the non-volatility of the RRAM technology. Due to the versatility ofPattern recognition as a computing task is very well suited for machine learning algorithms utilizing artificial neural networks (ANNs). Computing systems using ANNs usually require some sort of data storage to store the weights and bias values for the processing elements of the individual neurons. This paper introduces a memory block using resistive memory cells (RRAM) to realize this weight and bias storage in an embedded and distributed way while also offering programming and multi-level ability. By implementing power gating, overall power consumption is decreased significantly without data loss by taking advantage of the non-volatility of the RRAM technology. Due to the versatility of the peripheral circuitry, the presented memory concept can be adapted to different applications and RRAM technologies.show moreshow less

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Metadaten
Author:Stefan Pechmann, Timo Mai, Julian Potschka, Daniel Reiser, Peter Reichel, Marco Breiling, Marc Reichenbach, Amelie Hagelauer
Persistent identifiers - URN:urn:nbn:de:bvb:29-opus4-174878
Persistent identifiers - DOI:https://doi.org/10.3390/mi12111277
ISSN:2072-666X
Title of the journal / compilation (English):Micromachines
Publisher:MDPI
Document Type:Article
Language:English
Year of publication:2021
Date of first Publication:2021/10/20
Publishing Institution:Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)
Release Date:2021/11/09
Tag:ANN; RRAM; embedded memory; low-power; memory block; multi-level
Volume/year:12
Issue:11
Article Number:1277
Original publication:Micromachines 12.11 (2021): 1277. <https://www.mdpi.com/2072-666X/12/11/1277>
Institutes:Technische Fakultät
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 000 Informatik, Informationswissenschaft, allgemeine Werke
open_access (DINI-Set):open_access
Collections:Universität Erlangen-Nürnberg / Eingespielte Open Access Artikel / Eingespielte Open Access Artikel 2021
Licence (German):Creative Commons - CC BY - Namensnennung 4.0 International
Einverstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.