Enabling Artificial Spiking Sensory Neurons with a Single Flexible VO <sub>2</sub> Mott Memristor for Neuromorphic Sensing
In this work, we present flexible VO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> Mott memristors with excellent electrical properties, i.e. low operating current (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$<100 \mu ~\mathrm{A}$</tex>), good endurance (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$>10^{9}$</tex>) and good flexibility (bending radius <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\sim 5 ~\text{mm}$</tex>) for constructing artificial spiking sensory neurons. As a result, an artificial spiking thermoreceptor (AST) and an artificial spiking photoreceptor (ASP) are both realized by using a single flexible VO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> Mott memristor. With the stimulus temperature elevated from <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$20^{\circ} \mathrm{C}$</tex> to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$40^{\circ} \mathrm{C}$</tex>, the spiking frequency of AST first increases, then falls off rather quickly to zero at the higher temperature of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$45{}^{\circ} \mathrm{C}$</tex>, resembling the response characteristics of biological thermoreceptor. And the ASP has high spike-encoded photosensitivity and ultra-wide photosensing range (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$405 \sim 808 ~\text{nm}$</tex>). Finally, the spike-based electronic retina is proposed by connecting an array of artificial spiking photoreceptors with multiple-input floating gate MOS transistors acting as optical nerves. Consequently, the color image is successfully segmented and the edges of objects in the image are clearly extracted through the electronic retina by simulation, thus providing a solid foundation for neuromorphic sensing system.
Authors
- Weihua Liu (ORCID: https://orcid.org/0000-0001-5748-3021)
- Chuanyu Han (ORCID: https://orcid.org/0000-0002-4732-0429)
- Sheng Li Fang
- Xin Li (ORCID: https://orcid.org/0000-0002-3811-2062)
- Geng Li (ORCID: https://orcid.org/0000-0002-7239-9351)
- Shujing Zhao (ORCID: https://orcid.org/0000-0003-1087-1662)
- Shi Quan Fan
Institutions
- Xi'an Jiaotong University (CN)
Publication Details
- Published
- 2025-03-09
- DOI
- https://doi.org/10.1109/edtm61175.2025.11041398
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China