A5100737174- Terahertz technology and applications
- Microwave Engineering and Waveguides
- Microwave and Dielectric Measurement Techniques
- Stock Market Forecasting Methods
- Stochastic Gradient Optimization Techniques
- Transition Metal Oxide Nanomaterials
- Time Series Analysis and Forecasting
- Sparse and Compressive Sensing Techniques
- Complex Systems and Time Series Analysis
- Ultrasonics and Acoustic Wave Propagation
- Geophysical Methods and Applications
- Ferroelectric and Piezoelectric Materials
- Magnetic and transport properties of perovskites and related materials
- Phase-change materials and chalcogenides
- Inertial Sensor and Navigation
- Image and Signal Denoising Methods
- Privacy-Preserving Technologies in Data
- Polysaccharides and Plant Cell Walls
- Advanced materials and composites
- Diamond and Carbon-based Materials Research
- Gyrotron and Vacuum Electronics Research
- Liquid Crystal Research Advancements
- Maritime Navigation and Safety
- Glass properties and applications
- Advanced Cellulose Research Studies
Jilin University
1995-2022
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry
2022
Xi'an Shiyou University
2021
Nagoya University
2012
The dielectric properties of Shanxi anthracite and Shandong bituminous coals in China are investigated the low-terahertz (THz), W-band frequency from 75 GHz to 110 for first time. In this range, complex constant coal samples is obtained using free space method. It found that both real parts decrease considerably with increasing GHz. exhibit higher imaginary part values than coals. exhibits a more significantly decreasing trend range 90 compared frequencies below all strongly dependent on...
Stochastic gradient descent (SGD) holds as a classical method to build large scale machine learning models over big data. A stochastic is typically calculated from limited number of samples (known mini-batch), which potentially incurs high variance and causes the estimated parameters bounce around optimal solution. To improve stability gradient, recent years have witnessed proposal several semi-stochastic algorithms, distinguish themselves standard SGD by incorporating global information...
Noise suppression is a crucial step before seismic data analysis. The noise in desert areas has the characteristics of nonstationary, non-Gaussian, and low-frequency, which makes some traditional methods cannot suppress well. weighted nuclear norm minimization (WNNM), one most effective to all low-rank matrix approximation methods, assigns different weights singular values. good denoising performance WNNM based on accuracy block matching, theoretical basis matching nonlocal self-similarity...
Millimeter wave (mmW) propagation parameters including transmission, attenuation, and reflection coefficients of coal were obtained experimentally in the frequency region W-band (75-110 GHz). In addition, a raster scanning mmW imaging system transmission modalities was developed, which forms images according to difference between dielectric cracks those bulk samples, using spectrum module vector network analyzer radiate signal object obtain from S-parameters. The electromagnetic simulation...
Depolarizer is an important optical element for high-precision fiber optic gyroscopes in space and long-haul high-bit-rate transmission communication. Carboxylated cellulose nanocrystals prepared by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-mediated oxidation prefer to form free-standing films with nematic-like mesostructures whose potential application remains underexploited. Herein, we show that carboxylated hydrochloride acid hydrolysis TEMPO-mediated self-assemble into mesophase...
A low-loss terahertz (THz) hollow-core pipe waveguide constructed of plastic material is demonstrated. The structure designed to be especially effective in transmitting THz waves 110 GHz, which has important applications communications, imaging, and sensing. Guiding the electromagnetic wave based on principle antiresonant reflection. Through careful theoretical analyses systematic modeling simulation, followed by a thorough experimental investigation, we show that proposed can successfully...
Stochastic gradient descent (SGD) holds as a classical method to build large scale machine learning models over big data. A stochastic is typically calculated from limited number of samples (known mini-batch), so it potentially incurs high variance and causes the estimated parameters bounce around optimal solution. To improve stability gradient, recent years have witnessed proposal several semi-stochastic algorithms, which distinguish themselves standard SGD by incorporating global...
In this paper, given a news article referring to one company, we decide whether it is piece of good that followed by moving-up trend in the company's stock market or bad reversely. Additionally, predict how will fluctuation price be influenced article. The existing research work did not support flexible identification trends series, take account case temporal consecutive articles may influence sensitively. our proposed methods, realize more and accurate investigation correlation between...
Abstract Kalman filter processes the input and observation signals with noise on basis of linear state space representation to obtain system or real signal. In one-dimensional model, due lack multi-dimensional description target, prior estimate is often measured value previous moment. When target changes, divergence phenomenon will occur. Aiming at problem that traditional lacks dimension, which leads imprecision filter, this paper focuses improving estimation method state, proposes a...