EUVIMEDEUVIMED entdecken
Crossref

An SOT-MRAM-Based Δ-Bias-Σ Computing Paradigm for High Energy Efficiency Computer Vision Applications

Shuyu Wang, Zhenghan Fang, Haoran Du, Ruiqi Zhang, Hongjin Zhu, Bo Liu, Hao Cai

IEEE Transactions on Circuits and Systems I: Regular Papers · 2026-07 · 73 (7) · 4748-4761

DOI: 10.1109/tcsi.2026.3674531 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems I: Regular Papers, 2026-07. Verfasst von Shuyu Wang und 6 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Spin–orbit torque (SOT)-MRAM offers a promising solution for computer vision (CV) applications through its superior throughput and bit-cell energy efficiency. Meanwhile, redundant features lead to dominant energy consumption in memory-access peripherals and analog-to-digital converters (ADCs). Existing delta-sigma in-memory computing ( $\Delta \Sigma $ IMC) only reduces logic switching power, which has limited energy benefit for non-volatile memory (NVM)-based computing cores. This paper proposes an SOT-MRAM-based $\Delta $ -bias- $\Sigma $ computing macro to address the critical energy bottlenecks in CV applications. Voltage-delta-sensitive multiplier (VDM) realizes in-SOT-MRAM multiplication of incremental values, while eliminating redundant memory-access power caused by repetitive features. Delta-sensitive analog biasing (DS-biasing) scheme reduces the required ADC dynamic range, while delta-input-adaptive (DA) variable resolution ADC eliminates unnecessary quantization energy from repetitive inputs. Simulations based on 32kb SOT-MRAM computing macro demonstrate that the VDM achieves $6.45\times $ energy reduction over conventional MRAM compute unit, while the DS-biasing combined with DA-ADC achieves $1.8\times $ A/D conversion energy reduction compared with $\Delta \Sigma $ IMC. The CV computing system achieves 546TOPS/W/b on MNIST image classification task with 97.9% accuracy, while improving 34% and 5% energy efficiency in video edge detection task and MobileViT inference task, respectively.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Jiangsu Province

Zitieren

APA

Wang, S., Fang, Z., Du, H., Zhang, R., Zhu, H., Liu, B., & Cai, H. (2026). An SOT-MRAM-Based Δ-Bias-Σ Computing Paradigm for High Energy Efficiency Computer Vision Applications. IEEE Transactions on Circuits and Systems I: Regular Papers, 73(7), 4748-4761. https://doi.org/10.1109/tcsi.2026.3674531

Vancouver

Wang S, Fang Z, Du H, Zhang R, Zhu H, Liu B, et al. An SOT-MRAM-Based Δ-Bias-Σ Computing Paradigm for High Energy Efficiency Computer Vision Applications. IEEE Transactions on Circuits and Systems I: Regular Papers. 2026;73(7):4748-4761. doi:10.1109/tcsi.2026.3674531

BibTeX
@article{Wang2026,
  author = {Wang, Shuyu and Fang, Zhenghan and Du, Haoran and Zhang, Ruiqi and Zhu, Hongjin and Liu, Bo and Cai, Hao},
  title = {An SOT-MRAM-Based Δ-Bias-Σ Computing Paradigm for High Energy Efficiency Computer Vision Applications},
  journal = {IEEE Transactions on Circuits and Systems I: Regular Papers},
  year = {2026},
  volume = {73},
  number = {7},
  pages = {4748--4761},
  doi = {10.1109/tcsi.2026.3674531}
}
Kostenlos starten →