High-Linearity Ta2O5 Memristor and Its Application in Gaussian Convolution Image Denoising DOI
Yucheng Wang,

Hexin Wang,

Dingyun Guo

и другие.

ACS Applied Materials & Interfaces, Год журнала: 2024, Номер 16(36), С. 47879 - 47888

Опубликована: Авг. 27, 2024

In the image Gaussian filtering process, convolving with a matrix is essential due to numerous arithmetic computations involved, predominantly multiplications and additions. This can heavily tax system's memory, particularly frequent use. To address this issue, W/Ta2O5/Ag memristor was employed substantially mitigate computational overhead associated convolution operations. Additionally, an interlayer of ZnO subsequently introduced into memristor. The resulting Ta2O5/ZnO heterostructure layer exhibited improved linearity in pulse response, which enhanced facilitates easy adjustment conductance magnitude through linear mapping number pulses conductance. Subsequently, W/Ta2O5/ZnO/Ag bilayer as weights for kernel noise removal processing achieved by assembling 5 × array kernel. When denoising performed using arrays, compared convolution, average loss less than 5% observed. provided memristors demonstrate significant potential convolutional computations, subsequent applications neural networks (CNNs).

Язык: Английский

Mechanism and Application of Optical Modulation of Terahertz Waves in a Perovskite Heterojunction DOI
Wen Li,

Hengzhen Cheng,

Caiyu Feng

и другие.

ACS Applied Electronic Materials, Год журнала: 2024, Номер 6(11), С. 8351 - 8359

Опубликована: Ноя. 13, 2024

We investigated the mechanism and application of modulation terahertz waves using perovskite heterostructures. At interface between PEDOT:PSS layers under excitation by an external light source, photogenerated carriers transferred from layer to layer, which was accompanied charge accumulation. A large number scattered waves, thereby modulating signal in sample. The process recovery a material is analogous principle biological synapses, are involved memory learning animal brains. This enabled us develop perovskite-based synaptic devices. properties heterojunction materials open up applications technology.

Язык: Английский

Процитировано

2

Research Progress of Photoelectric Resistive Switching Mechanism of Halide Perovskite DOI Open Access

Huajun Guo,

Shuailing AN,

Jie Meng

и другие.

Journal of Inorganic Materials, Год журнала: 2023, Номер 38(9), С. 1005 - 1005

Опубликована: Янв. 1, 2023

Язык: Английский

Процитировано

4

Piezotronics and its applications in artificial intelligent sensing DOI Open Access
Qijun Sun,

Yanqiang Lei,

Zhong Lin Wang

и другие.

Scientia Sinica Technologica, Год журнала: 2023, Номер 53(6), С. 860 - 880

Опубликована: Май 9, 2023

The coupling between piezoelectric polarization and semiconductor properties (for example, electronic transport) in low-dimensional nanomaterials, such as ZnO GaN, gives rise to unprecedented device characteristics. This has increased research interest the emerging fields of piezotronics, which offers new means manipulating charge-carrier transport, generation, recombination or separation controlled operation flexible devices through application external mechanical stimuli. We review recent progress understanding fundamental theory, investigation piezotronics different materials systems, development transistors, generalized provide an in-depth discussion future directions.

Язык: Английский

Процитировано

1

Uncovering the structural and photoelectric properties of perovskite crystals by joint multifaceted characterization and theoretical analysis DOI Creative Commons
Aochen Du, Xingke Zheng, Genzhuang Li

и другие.

Journal of Materials Research and Technology, Год журнала: 2024, Номер 28, С. 3988 - 4000

Опубликована: Янв. 1, 2024

The potential of organic-inorganic hybrid perovskites, owing to their distinct photoelectric attributes and boosted photoluminescence stability, has garnered substantial attention. Despite comprehensive examinations the synthesis optical characteristics FAPbBr3 crystals, understanding microstructure's implication on applications, electronic configuration, electron-phonon interplay remains vague. In this study, we detailed phonon dispersion, vibratory properties using density functional theory. We further calculated formation enthalpy stability diagram. Characterization calculation crystalline structure were accomplished via Rietveld refinement. Phonon dispersion curves, as well total partial densities states, also calculated. theoretical results align with experimental Fourier infrared Raman spectra. Beyond elementary characterization, explored qualities thereby reaffirming outstanding extensive application potential. A non-contact fixed-point lighting device was designed based principle electromagnetic induction coils, which proved uniformity crystal luminescence. According first principle, possesses a Poisson's ratio 0.32, indicative high flexibility, emphasizing superb vast scope.

Язык: Английский

Процитировано

0

High-Linearity Ta2O5 Memristor and Its Application in Gaussian Convolution Image Denoising DOI
Yucheng Wang,

Hexin Wang,

Dingyun Guo

и другие.

ACS Applied Materials & Interfaces, Год журнала: 2024, Номер 16(36), С. 47879 - 47888

Опубликована: Авг. 27, 2024

In the image Gaussian filtering process, convolving with a matrix is essential due to numerous arithmetic computations involved, predominantly multiplications and additions. This can heavily tax system's memory, particularly frequent use. To address this issue, W/Ta2O5/Ag memristor was employed substantially mitigate computational overhead associated convolution operations. Additionally, an interlayer of ZnO subsequently introduced into memristor. The resulting Ta2O5/ZnO heterostructure layer exhibited improved linearity in pulse response, which enhanced facilitates easy adjustment conductance magnitude through linear mapping number pulses conductance. Subsequently, W/Ta2O5/ZnO/Ag bilayer as weights for kernel noise removal processing achieved by assembling 5 × array kernel. When denoising performed using arrays, compared convolution, average loss less than 5% observed. provided memristors demonstrate significant potential convolutional computations, subsequent applications neural networks (CNNs).

Язык: Английский

Процитировано

0