Robust power-law computing for edge intelligence using spin-orbit torque magnetic tunnel junctions
Applied Physics Reviews · 01.12.2026 · 13 (4)
DOI: 10.1063/5.0352180 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Physics Reviews, online erschienen am 02.11.2026. Verfasst von Shilong Xiong und 11 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Chinese Academy of Sciences President's International Fellowship Initiative.
Abstract (Originalsprache)
Power-law transformations are widely used in edge perception, where compact implementations must balance numerical precision, circuit resources, and device variability. Here, we demonstrate a stochastic power-law architecture based on spin–orbit-torque magnetic tunnel junctions (SOT-MTJs). An event-driven feedback loop adjusts the write voltage to regulate the probability of an all-1 event in an n-bit block. Under the fixed-point sampling model, this establishes a switching probability p = b1/n, from which m-bit event statistics yield the power-law target bm/n. Online probability locking removes the need for offline per-device probability-voltage characterization or stored calibration tables. Measurements on W-based SOT-MTJs, combined with digital analysis of the acquired bitstreams, demonstrate representative power-law operations with absolute deviations of 0.003 or less at the reported precision, as well as parallel decoding of multiple exponents from one source. A 4-bit gamma-correction example preserves dark-region image detail and reaches approximately 35 dB peak signal-to-noise ratio relative to the theoretical gamma-domain reference near 1000 raw samples. A 5 Oe device-level field perturbation and a joint ±90% simulation centered on the measured parameter means of 25 devices examine adaptation and the safe voltage operating boundary under the tested conditions. These results establish a nonlinear computing primitive with accuracy controlled by sampling depth and an operating point maintained through feedback.
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Chinese Academy of Sciences President's International Fellowship Initiative
- Russian Science Foundation
Zitieren
APA
Xiong, S., Wan, C., Li, X., Xu, Y., Kong, D., Xia, S., Zhang, T., Davydenko, A. V., Du, G., Samardak, A. S., Yu, G., & Han, X. (2026). Robust power-law computing for edge intelligence using spin-orbit torque magnetic tunnel junctions. Applied Physics Reviews, 13(4). https://doi.org/10.1063/5.0352180
Vancouver
Xiong S, Wan C, Li X, Xu Y, Kong D, Xia S, et al. Robust power-law computing for edge intelligence using spin-orbit torque magnetic tunnel junctions. Applied Physics Reviews. 2026;13(4). doi:10.1063/5.0352180
BibTeX
@article{Xiong2026,
author = {Xiong, Shilong and Wan, Caihua and Li, Xiaohan and Xu, Yingqian and Kong, Dehao and Xia, Shengkai and Zhang, Tianyi and Davydenko, Aleksandr V. and Du, Guanxiang and Samardak, Alexander S. and Yu, Guoqiang and Han, Xiufeng},
title = {Robust power-law computing for edge intelligence using spin-orbit torque magnetic tunnel junctions},
journal = {Applied Physics Reviews},
year = {2026},
volume = {13},
number = {4},
doi = {10.1063/5.0352180}
}
EUVIMED