Detection of Read-Disturb Effects in RRAM-Based Computation-in-Memory Architectures for Neural Networks
IEEE Transactions on Circuits and Systems I: Regular Papers · 2026-08 · 73 (8) · 5648-5661
DOI: 10.1109/tcsi.2026.3675336 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems I: Regular Papers, 2026-08. Verfasst von Mohammad Amin Yaldagard und 5 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Gefördert von: TU Delft AI Labs & Talent Programme, European Union (EU) for the projects NEUROKIT2E and CONVOLVE, Dutch Research Council (NWO) for the project Self-Healing Neuromorphic Systems, Federal Ministry of Education and Research (BMBF), Germany, through the NEUROTEC II.
Abstract (Originalsprache)
Resistive random-access memory (RRAM)-based computation-in-memory (CIM) architectures offer a promising solution to meet the stringent energy efficiency demands of executing artificial intelligence (AI) algorithms directly on edge devices. However, these architectures suffer from the read-disturb problem, which can lead to accumulated computational errors over time. To maintain the required level of computational accuracy, conventional approaches rely on a static reprogramming process after a predefined number of read cycles, necessitating large counters and resulting in inefficiencies. This paper presents experimental results using real RRAM devices to analyze the read-disturb effect and builds on these insights to propose a circuit-level detection methodology for real-time monitoring of conductance drifts. The proposed method initiates reprogramming only when the device drift exceeds a defined threshold and reprogramming is actually needed. Additionally, an analytical method is developed to determine the minimum conductance state ratio needed to meet reliable detection criteria. Based on this foundation, the proposed detection technique is further optimized for dynamic identification of read-disturb effects. Experiment-augmented SPICE simulation results, using a calibrated model implemented in TSMC 40 nm CMOS technology, validate the functionality and effectiveness of the proposed detection approach. These results demonstrate its potential to improve both the reliability and efficiency of RRAM-based CIM architectures that provide up to a 4x improvement in energy-efficiency compared to traditional periodic reprogramming methods.
Förderung
- TU Delft AI Labs & Talent Programme, European Union (EU) for the projects NEUROKIT2E and CONVOLVE
- Dutch Research Council (NWO) for the project Self-Healing Neuromorphic Systems
- Federal Ministry of Education and Research (BMBF), Germany, through the NEUROTEC II
Zitieren
APA
Yaldagard, M. A., Bende, A., Diware, S., Rana, V., Hamdioui, S., & Bishnoi, R. (2026). Detection of Read-Disturb Effects in RRAM-Based Computation-in-Memory Architectures for Neural Networks. IEEE Transactions on Circuits and Systems I: Regular Papers, 73(8), 5648-5661. https://doi.org/10.1109/tcsi.2026.3675336
Vancouver
Yaldagard MA, Bende A, Diware S, Rana V, Hamdioui S, Bishnoi R. Detection of Read-Disturb Effects in RRAM-Based Computation-in-Memory Architectures for Neural Networks. IEEE Transactions on Circuits and Systems I: Regular Papers. 2026;73(8):5648-5661. doi:10.1109/tcsi.2026.3675336
BibTeX
@article{Yaldagard2026,
author = {Yaldagard, Mohammad Amin and Bende, Ankit and Diware, Sumit and Rana, Vikas and Hamdioui, Said and Bishnoi, Rajendra},
title = {Detection of Read-Disturb Effects in RRAM-Based Computation-in-Memory Architectures for Neural Networks},
journal = {IEEE Transactions on Circuits and Systems I: Regular Papers},
year = {2026},
volume = {73},
number = {8},
pages = {5648--5661},
doi = {10.1109/tcsi.2026.3675336}
}
EUVIMED