Browsing by Author "Hong, Ming-Chun"

Now showing 1 - 2 of 2

A Fluorographene-Based Synaptic Transistor
(2019-08-13)
Liu, Bo
;
Hong, Ming-Chun
;
Sahoo, Mamina
;
Ong, Bin Leong
;
Tok, Eng Soon
Exploring brain-inspired synaptic devices has recently become a new focus of research in nanoelectronic communities. In this emerging field, incorporating 2D materials into three-terminal synaptic transistors has brought various advantages. However, achieving a stable and long-term weight-modulation in these synaptic transistors, which are typically based on interface charge storage, is still a challenge due to the nature of their spontaneous relaxation. The application of an atomically thin fluorographene layer into the synaptic junction region suppresses this issue and improves the efficiency, tunability, and symmetry of the synaptic plasticity as well as establishing a stable weight-regulation paradigm. These unique properties can be attributed to the dipolar rotation of C-F in fluorographene. To obtain a better physical understanding, a vacancy-dependent C-F dipolar rotation model is proposed and supported by hysteresis analysis and density functional theory calculations. As proposed and demonstrated, the unique fluorographene-based synaptic transistor may be a promising building block for constructing efficient neuromorphic computing hardware.
Extremely Compact Integrate-and-Fire STT-MRAM Neuron: A Pathway toward All-Spin Artificial Deep Neural Network
(2019-01-01)
Wu, Ming-Hung
;
Hong, Ming-Chun
;
Chang, Chih-Cheng
;
Sahu, Paritosh
;
Wei, Jeng-Hua
This work reports the complete framework from device to architecture for deep learning acceleration in an all-spin artificial neural network (ANN) built by highly manufacturable STT-MRAM technology. The most compact analog integrate-and-fire neuron reported to date is developed based on the back-hopping oscillation in magnetic tunnel junctions. This novel device is unique because it performs numerous essential neural functions simultaneously, including current integration, voltage spike generation, state reset, and 4-bit precision. The device itself is also a stochastic binary synapse, and thus eases the implementation of the compact all-spin ANN with high accuracy for online training.