The search result changed since you submitted your search request. Documents might be displayed in a different sort order.
  • search hit 9 of 10
Back to Result List

Low Conductance State Drift Characterization and Mitigation in Resistive Switching Memories (RRAM) for Artificial Neural Networks

  • The crossbar structure of Resistive-switching random access memory (RRAM) arrays enabled the In-Memory Computing circuits paradigm, since they imply the native acceleration of a crucial operations in this scenario, namely the Matrix-Vector-Multiplication (MVM). However, RRAM arrays are affected by several issues materializing in conductance variations that might cause severe performance degradation. A critical one is related to the drift of the low conductance states appearing immediately at the end of program and verify algorithms that are mandatory for an accurate multi-level conductance operation. In this work, we analyze the benefits of a new programming algorithm that embodies Set and Reset switching operations to achieve better conductance control and lower variability. Data retention analysis performed with different temperatures for 168 hours evidence its superior performance with respect to standard programming approach. Finally, we explored the benefits of using our methodology at a higher abstraction level, throughThe crossbar structure of Resistive-switching random access memory (RRAM) arrays enabled the In-Memory Computing circuits paradigm, since they imply the native acceleration of a crucial operations in this scenario, namely the Matrix-Vector-Multiplication (MVM). However, RRAM arrays are affected by several issues materializing in conductance variations that might cause severe performance degradation. A critical one is related to the drift of the low conductance states appearing immediately at the end of program and verify algorithms that are mandatory for an accurate multi-level conductance operation. In this work, we analyze the benefits of a new programming algorithm that embodies Set and Reset switching operations to achieve better conductance control and lower variability. Data retention analysis performed with different temperatures for 168 hours evidence its superior performance with respect to standard programming approach. Finally, we explored the benefits of using our methodology at a higher abstraction level, through the simulation of an Artificial Neural Network for image recognition task (MNIST dataset). The accuracy achieved shows higher performance stability over temperature and time.show moreshow less

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author: Andrea Baroni, Artem Glukhov, Eduardo Perez, Christian WengerORCiD, Daniele Ielmini, Piero Olivo, Cristian Zambelli
DOI:https://doi.org/10.1109/TDMR.2022.3182133
ISSN:1530-4388
Title of the source (English):IEEE Transactions on Device and Materials Reliability
Document Type:Scientific journal article peer-reviewed
Language:English
Year of publication:2022
Tag:Multilevel switching; RRAM; neural network
Volume/Year:22
Issue number:3
First Page:340
Last Page:347
Faculty/Chair:Fakultät 1 MINT - Mathematik, Informatik, Physik, Elektro- und Informationstechnik / FG Halbleitermaterialien
Einverstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.