EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Configurable Kernel Map Implementation in Memristor Crossbar for Convolution Neural Network

Gyeonghae Kim, Sangwook Youn, Jinwoo Park, Hyungjin Kim

Advanced Intelligent Systems · 2025-11 · 7 (11)

DOI: 10.1002/aisy.202500032 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Intelligent Systems, online erschienen am 15.05.2025. Verfasst von Gyeonghae Kim und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Research Foundation of Korea.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study experimentally verifies a configurable convolutional neural network implementation in a 48 × 48 memristor crossbar array. It allows the number of convolution reading cycles to be adjusted based on design demands, such as area or time constraints. This adaptability can be achieved by modifying the area of kernel weights on the array, enabling a balance between time and area efficiency. A detailed algorithm for the configurable convolution scheme is presented, taking into account factors such as array size, feature map size, and cycle limits. Also, the experimental validation using a binary 12 × 12 MNIST dataset demonstrates the practical applicability of our scheme. By varying the number of cycles per inference ( N cycle ), this study analyzes how different array sizes can affect performance, confirming the relationship between the total number of cells ( N cell ) and cycles. Larger N cycle values can result in the increased usage of faulty devices, leading to more distorted feature map pixels. Additionally, it analyzes the robustness of the scheme to weight errors, finding that longer cycles are less robust due to more frequent reads on critical fault devices within the crossbar, causing accuracy degradation.

Förderung

  • National Research Foundation of Korea

Zitieren

APA

Kim, G., Youn, S., Park, J., & Kim, H. (2025). Configurable Kernel Map Implementation in Memristor Crossbar for Convolution Neural Network. Advanced Intelligent Systems, 7(11). https://doi.org/10.1002/aisy.202500032

Vancouver

Kim G, Youn S, Park J, Kim H. Configurable Kernel Map Implementation in Memristor Crossbar for Convolution Neural Network. Advanced Intelligent Systems. 2025;7(11). doi:10.1002/aisy.202500032

BibTeX
@article{Kim2025,
  author = {Kim, Gyeonghae and Youn, Sangwook and Park, Jinwoo and Kim, Hyungjin},
  title = {Configurable Kernel Map Implementation in Memristor Crossbar for Convolution Neural Network},
  journal = {Advanced Intelligent Systems},
  year = {2025},
  volume = {7},
  number = {11},
  doi = {10.1002/aisy.202500032}
}
Kostenlos starten →