A5100415069- Optical Network Technologies
- Photonic and Optical Devices
- Neural Networks and Reservoir Computing
- Advanced Measurement and Detection Methods
- Semiconductor Quantum Structures and Devices
- Electron and X-Ray Spectroscopy Techniques
- GaN-based semiconductor devices and materials
- Photonic Crystal and Fiber Optics
- Quantum and electron transport phenomena
- Spectroscopy and Laser Applications
- Optical Systems and Laser Technology
- Infrared Target Detection Methodologies
- Advanced Algorithms and Applications
- CCD and CMOS Imaging Sensors
- Quantum Information and Cryptography
- Particle Detector Development and Performance
- Color Science and Applications
- Quantum Computing Algorithms and Architecture
- Advanced Fiber Optic Sensors
- Advanced Fiber Laser Technologies
- Topological Materials and Phenomena
- Currency Recognition and Detection
- Radio Frequency Integrated Circuit Design
- Statistical Distribution Estimation and Applications
- Organic Electronics and Photovoltaics
Heilongjiang University
2025
Tianjin University
2023-2024
Chongqing University
2024
Dalian University of Technology
2024
Anhui University
2023
Donghua University
2023
Shandong University of Science and Technology
2022
Tiangong University
2018-2020
Xiangtan University
2018
Shenyang Jianzhu University
2014
Temporal knowledge graph completion is a technique that uses existing to predict or infer the missing information in temporal graph. It combines technical features of and time series analysis deal with entities relationships change over time. The technology fails make effective use special relationship between relations information, it difficult fully represent complex In order solve above problems, model based on probability box embedding (TPBoxE) was proposed. Firstly, are represented...
A simple photonic crystal fiber (PCF) design with a rectangular array of four airholes in the core region and traditional circular-airhole cladding is proposed this work. The modal birefringence induced by asymmetry distribution and/or elliptical shape holes. symmetric structure results good confinement loss performance limiting light region. Therefore, enables simultaneous realization high low loss. Simulations based on full-vector finite element method (FEM) anisotropic (PML) show that...
An array of passive silicon-on-insulator optical devices is laid out in repeating patterns on four foundry-fabricated wafers. The physical and characterization these microrings, racetrack resonators, directional couplers are found to exhibit significant variation response. A device-heating experiment carried a number different demonstrates that thermal effects independent the device's location wafer. analysis responses room-temperature used determine process variation. We find if we form...
Optical packet header recognition is an important signal processing task of optical communication networks. In this work, we propose all-optical reservoir, consisting integrated double-ring resonators (DRRs) as nodes, for fast and accurate recognition. As the delay-bandwidth product (DBP) node a key figure-of-merit in adopt deep reinforcement learning algorithm to maximize DBPs various types DRRs, which has advantage full parameter space optimization convergence speed. Intriguingly,...
High-efficient GaN-based laser diodes (LDs) with tunnel junction are designed by replacing conventional p-type AlGaN cladding layers and GaN contact lower-resistant n-type contact. In addition, the characteristics of LDs numerically investigated using commercial software lastip. It is found that performance these greatly improved. As a comparison, absorption loss non-radiative recombination reduced. The threshold current series resistance decreased 12% 59%, respectively, slope efficiency...
Self-organized InGaN quantum dots (QDs) with emission wavelength from green to red range have been grown on GaN templated c-plane sapphire substrates by metal-organic chemical vapor deposition (MOCVD). The effects of matrix layer composition the structural and optical properties QDs investigated. A continued growth is observed during In0.1Ga0.9N layer, which results in an increase QDs' size. By using instead one, annealing induced blue-shift energy can be suppressed. After top cap a larger...
We propose a reinforcement-learning-optimized nonlinear physical diffractive neural network, which can simultaneously perform OAM-mode and LP-mode conversion with Gaussian noise removal. The PSNR SSIM of the converted modes reach 27.94 dB 0.838, respectively.
Abstract In daily life, the color of matter will change in some cases. This paper mainly studies factors that affect powder. The effects grinding particle size, transparency material and solvent medium on colour powder were investigated. innovation this lies quantitative measurement analysis by method measuring wavelength light intensity with spectrometer.
An improved technique of continuous shaping current-injected waveforms based on the single-mode rate equations is proposed to suppress relaxation oscillations (ROs) from direct modulation distributed feedback laser (DFB). The signal expression current deduced theoretically dependence DFB desired output in detail, and specific parameters derivation different polynomial degree also discussed necessarily. Furthermore, a p-function with inverse operation adopted construct Fourier series...
We report a novel design of photonic crystal fiber (PCF) with rectangular array four closely-spaced, highly elliptical air holes in the core region and circular-air-hole cladding. The proposed PCF is able to support ultra-wideband singlepolarization single-mode (SPSM) transmission from visible band near infrared band. With aid inner cladding formed by central holes, one polarization fundamental mode can be cut off at very short wavelengths SPSM propagation achieved. also suppresses higher...
Abstract The methods of multi-DOF actuator orientation detection based on machine vision and magnetic field sensor neural network multisensor fusion are presented in this paper. Special grid pattern is printed the surface rotor according to pseudo-random encoder for camera capturing image rotor. sensors adopted properly placed circling shaft information. By analyzing rotor, coordinates characteristic points coordinate system fixed derived. Combined with scheme reduce measurement error...
Superconducting flux qubits with three Josephson junctions are promising candidates for the building blocks of a quantum computer. We have applied imaginary time evolution method to study model this qubit accurately by calculating its wave functions and eigenenergies. Because such manipulated magnetic microwave pulses they might be irradiated into non-computational states which is called leakage effect. Through density matrix under either hard-shaped pulse or Gaussian-shaped carry out NOT...
In order to reduce dead zone of Closed Shell-Electrode Silicon Detector (CSESD), this paper introduces an optimal scheme 3D shell electrode type detector– a novel Honeycomb Shell Electrode Detector– HSED (Chinese patent #zl 201710 676008.3). paper, we first study the new structure and find non-elongated cell is smaller than that CSESD. An array arrangement method CSESD also proposed, which makes virtually no in comprised by elongated detector cells. electrical simulations with SILVACO TCAD,...
We propose an approach to probe coherent spin-state dynamics of molecules using circularly polarized hard x-ray pulses. For the dynamically aligned nitric oxide in a superposition spin-orbit coupled electronic state that can be prepared through stimulated Raman scattering, we demonstrate capability ultrafast diffraction not only reveal quantum beating wave packet, but also image spatial spin density molecule. With signal, show matrix retrieved. The spatio-temporal resolving power paves way...
For the hierarchical gamma and model, we calculate Bayes estimator of rate parameter distribution under Stein's loss function which penalizes gross overestimation underestimation equally corresponding Posterior Expected Loss (PESL). We also obtain squared error PESL. Moreover, empirical estimators with a conjugate prior by moment MLE methods. In numerical simulations, have illustrated five aspects: two inequalities PESLs for oracle method; Maximum Likelihood Estimators (MLEs) are consistent...
The delay-bandwidth product in double-ring resonators (DRRs) is optimized using reinforcement learning. Then, the DRRs are used to build an all-optical reservoir for optical packet header recognition, enabling a word-error rate as low 9×10 -4 .
Over the past few decades, optical manipulation has emerged as a highly successful tool in various fields such biology, micro/nanorobotics, and physics. Among different techniques, transverse slot waveguide shown remarkable potential enhancing field significantly improving trapping capabilities. Additionally, microring resonators have demonstrated ability to enhance at specific resonance wavelengths, enabling capture of particles. In this study, we investigated impact structure on...
The neural network, as a non-linear system with large-scale parallel distributed processing, has been widely used in the field of prediction. This paper proposes stock forecasting method based on combination Stacked Denoising AutoEncoders (SDAE) and Back Propagation Neural Network (BPNN). Using more than 4100 historical data Sinopec Group from 2001/8/8 to 2018/8/16 for Seventeen years, then process select six features them training experiment first uses SDAE pre-training stage obtain...
An optimal effective current operator for flux qubit has been investigated by taking account of the inductive effects circuit loop. The whole system is treated as two interacting subsystems: one inductance-free consisting three Josephson junctions and other a high frequency $LC$ oscillator. As composite hardly affords excessively high-energy photon, an theory providing its physical variable operators achieved, which can take but does not include additional degree freedom Considering tradeoff...