EUVIMEDEUVIMED entdecken
Crossref

Accelerated Modeling of Parasitic Effect and Electromigration-Aware Reliability Evaluation for Interconnects in RRAM-Based Crossbar Array

Wenbo Wang, Siwei Tan, Yizhang Liu, Hao Xie, Lunyao Wang, Yinshui Xia, Wenchao Chen

IEEE Transactions on Electron Devices · 2026-08 · 73 (8) · 5092-5098

DOI: 10.1109/ted.2026.3703390 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Electron Devices, 2026-08. Verfasst von Wenbo Wang und 6 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Zhejiang Province, State Key Laboratory of Radio Frequency Heterogeneous Integration.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The parasitic effects and reliability challenges in resistive random access memory (RRAM) crossbar arrays are critical factors constraining the computational accuracy of in-memory computing (IMC) systems. In this work, the characteristics of electromigration-induced void formation are investigated through multiphysics simulations that couple electrical, thermal, and stress effects, and the reliability is assessed by tracking the temporal evolution of interconnect resistance. To enhance the efficiency of parasitic parameter extraction, a ResNet-based neural network is developed for predicting RLC equivalent circuit parameters. In addition, the Partial Element Equivalent Circuit (PEEC) method is utilized instead of full-wave simulations, significantly accelerating the generation of datasets Finally, the reliability evolution characteristics of nanoscale interconnects are incorporated into a neural computing circuit composed of an RRAM crossbar array. Both the circuit’s output behavior for single-image inference and the temporal evolution of recognition accuracy on the MNIST test dataset are analyzed. This study provides an effective solution for parasitic parameter extraction and reliability evaluation in high-precision IMC systems.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Zhejiang Province
  • State Key Laboratory of Radio Frequency Heterogeneous Integration

Zitieren

APA

Wang, W., Tan, S., Liu, Y., Xie, H., Wang, L., Xia, Y., & Chen, W. (2026). Accelerated Modeling of Parasitic Effect and Electromigration-Aware Reliability Evaluation for Interconnects in RRAM-Based Crossbar Array. IEEE Transactions on Electron Devices, 73(8), 5092-5098. https://doi.org/10.1109/ted.2026.3703390

Vancouver

Wang W, Tan S, Liu Y, Xie H, Wang L, Xia Y, et al. Accelerated Modeling of Parasitic Effect and Electromigration-Aware Reliability Evaluation for Interconnects in RRAM-Based Crossbar Array. IEEE Transactions on Electron Devices. 2026;73(8):5092-5098. doi:10.1109/ted.2026.3703390

BibTeX
@article{Wang2026,
  author = {Wang, Wenbo and Tan, Siwei and Liu, Yizhang and Xie, Hao and Wang, Lunyao and Xia, Yinshui and Chen, Wenchao},
  title = {Accelerated Modeling of Parasitic Effect and Electromigration-Aware Reliability Evaluation for Interconnects in RRAM-Based Crossbar Array},
  journal = {IEEE Transactions on Electron Devices},
  year = {2026},
  volume = {73},
  number = {8},
  pages = {5092--5098},
  doi = {10.1109/ted.2026.3703390}
}
Kostenlos starten →