EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Coupled Ferroelectric‐Photonic Memory in a Retinomorphic Hardware for In‐Sensor Computing

Ngoc Thanh Duong, Yufei Shi, Sifan Li, Yu-Chieh Chien, Heng Xiang, Haofei Zheng, Peiyang Li, Lingqi Li, Yangwu Wu, Kah-Wee Ang iD

Advanced Science · 2024-03 · 11 (12)

DOI: 10.1002/advs.202303447 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Science, online erschienen am 17.01.2024. Verfasst von Ngoc Thanh Duong und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, Singapore, 117583, Singapore. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education Tier-1 and Tier-2, National Research Foundation, Singapore, A*STAR Science and Engineering Research Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The development of all‐in‐one devices for artificial visual systems offers an attractive solution in terms of energy efficiency and real‐time processing speed. In recent years, the proliferation of smart sensors in the growth of Internet‐of‐Things (IoT) has led to the increasing importance of in‐sensor computing technology, which places computational power at the edge of the data‐flow architecture. In this study, a prototype visual sensor inspired by the human retina is proposed, which integrates ferroelectricity and photosensitivity in two‐dimensional (2D) α‐In 2 Se 3 material. This device mimics the functions of photoreceptors and amacrine cells in the retina, performing optical reception and memory computation functions through the use of electrical switching polarization in the channel. The gate‐tunable linearity of excitatory and inhibitory functions in photon‐induced short‐term plasticity enables to encode and classify 12 000 images in the Mixed National Institute of Standards and Technology (MNIST) dataset with remarkable accuracy, achieving ≈94%. Additionally, in‐sensor convolution image processing through a network of phototransistors, with five convolutional kernels electrically pre‐programmed into the transistors is demonstrated. The convoluted photocurrent matrices undergo straightforward arithmetic calculations to produce edge and feature‐enhanced scenarios. The findings demonstrate the potential of ferroelectric α‐In 2 Se 3 for highly compact and efficient retinomorphic hardware implementation, regardless of ambipolar transport in the channel.

Schlagwörter

Stichwörter der Autor:innen

convolution image processingferroelectric semiconductorsin‐sensor computingoptoelectronic memoryretinomorphic sensors

Autor:innen und Institutionen

  • Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, Singapore, 117583, Singapore.

ORCID-Profile: Kah-Wee Ang

Förderung

  • Ministry of Education Tier-1 and Tier-2
  • National Research Foundation, Singapore
  • A*STAR Science and Engineering Research Council
  • National Research Foundation Singapore
  • Science and Engineering Research Council

Zitieren

APA

Duong, N. T., Shi, Y., Li, S., Chien, Y. C., Xiang, H., Zheng, H., Li, P., Li, L., Wu, Y., & Ang, K. W. (2024). Coupled Ferroelectric‐Photonic Memory in a Retinomorphic Hardware for In‐Sensor Computing. Advanced Science, 11(12). https://doi.org/10.1002/advs.202303447

Vancouver

Duong NT, Shi Y, Li S, Chien YC, Xiang H, Zheng H, et al. Coupled Ferroelectric‐Photonic Memory in a Retinomorphic Hardware for In‐Sensor Computing. Advanced Science. 2024;11(12). doi:10.1002/advs.202303447

BibTeX
@article{Duong2024,
  author = {Duong, Ngoc Thanh and Shi, Yufei and Li, Sifan and Chien, Yu-Chieh and Xiang, Heng and Zheng, Haofei and Li, Peiyang and Li, Lingqi and Wu, Yangwu and Ang, Kah-Wee},
  title = {Coupled Ferroelectric‐Photonic Memory in a Retinomorphic Hardware for In‐Sensor Computing},
  journal = {Advanced Science},
  year = {2024},
  volume = {11},
  number = {12},
  doi = {10.1002/advs.202303447}
}
Kostenlos starten →