Ultralow‐power in‐memory computing based on ferroelectric memcapacitor network
Exploration · 2023-06 · 3 (3)
DOI: 10.1002/exp.20220126 ↗At a glance
Journal article in Exploration, published online on 2023-05-11. Written by Bobo Tian and 23 co-authors. Institution: Key Laboratory of Polar Materials and Devices, Ministry of Education, Shanghai Center of Brain-inspired Intelligent Materials and Devices, Department…. Cited 55 times so far (Crossref); 33 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Key Research and Development Program of China, National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities.
Abstract (original language)
Analog storage through synaptic weights using conductance in resistive neuromorphic systems and devices inevitably generates harmful heat dissipation. This thermal issue not only limits the energy efficiency but also hampers the very‐large‐scale and highly complicated hardware integration as in the human brain. Here we demonstrate that the synaptic weights can be simulated by reconfigurable non‐volatile capacitances of a ferroelectric‐based memcapacitor with ultralow‐power consumption. The as‐designed metal/ferroelectric/metal/insulator/semiconductor memcapacitor shows distinct 3‐bit capacitance states controlled by the ferroelectric domain dynamics. These robust memcapacitive states exhibit uniform maintenance of more than 10 4 s and well endurance of 10 9 cycles. In a wired memcapacitor crossbar network hardware, analog vector‐matrix multiplication is successfully implemented to classify 9‐pixel images by collecting the sum of displacement currents ( I = C × d V /d t ) in each column, which intrinsically consumes zero energy in memcapacitors themselves. Our work sheds light on an ultralow‐power neural hardware based on ferroelectric memcapacitors.
Subjects
Author keywords
Authors and institutions
- Key Laboratory of Polar Materials and Devices, Ministry of Education, Shanghai Center of Brain-inspired Intelligent Materials and Devices, Department of Electronics East China Normal University Shanghai China.
- Zhejiang Lab Hangzhou China.
- School of Materials Science and Engineering Shanghai University of Engineering Science Shanghai China.
- CentraleSupélec, CNRS-UMR8580, Laboratoire SPMS Université Paris-Saclay Gif-sur-Yvette France.
- State Key Laboratory of Infrared Physics, Chinese Academy of Sciences Shanghai Institute of Technical Physics Shanghai China.
- Frontier Institute of Chip and System Fudan University Shanghai China.
Funding
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Fundamental Research Funds for the Central Universities
Cite
APA
Tian, B., Xie, Z., Chen, L., Hao, S., Liu, Y., Feng, G., Liu, X., Liu, H., Yang, J., Zhang, Y., Bai, W., Lin, T., Shen, H., Meng, X., Zhong, N., Peng, H., Yue, F., Tang, X., Wang, J., …, & Duan, C. (2023). Ultralow‐power in‐memory computing based on ferroelectric memcapacitor network. Exploration, 3(3). https://doi.org/10.1002/exp.20220126
Vancouver
Tian B, Xie Z, Chen L, Hao S, Liu Y, Feng G, et al. Ultralow‐power in‐memory computing based on ferroelectric memcapacitor network. Exploration. 2023;3(3). doi:10.1002/exp.20220126
BibTeX
@article{Tian2023,
author = {Tian, Bobo and Xie, Zhuozhuang and Chen, Luqiu and Hao, Shenglan and Liu, Yifei and Feng, Guangdi and Liu, Xuefeng and Liu, Hongbo and Yang, Jing and Zhang, Yuanyuan and Bai, Wei and Lin, Tie and Shen, Hong and Meng, Xiangjian and Zhong, Ni and Peng, Hui and Yue, Fangyu and Tang, Xiaodong and Wang, Jianlu and Zhu, Qiuxiang and Ivry, Yachin and Dkhil, Brahim and Chu, Junhao and Duan, Chungang},
title = {Ultralow‐power in‐memory computing based on ferroelectric memcapacitor network},
journal = {Exploration},
year = {2023},
volume = {3},
number = {3},
doi = {10.1002/exp.20220126}
}
EUVIMED