A5101937658- Advanced Wireless Communication Techniques
- Anaerobic Digestion and Biogas Production
- Error Correcting Code Techniques
- Wastewater Treatment and Nitrogen Removal
- Wireless Signal Modulation Classification
- Advanced Neural Network Applications
- Metallurgical Processes and Thermodynamics
- Machine Fault Diagnosis Techniques
- Advanced MIMO Systems Optimization
- Gear and Bearing Dynamics Analysis
- Direction-of-Arrival Estimation Techniques
- Advanced Wireless Communication Technologies
- Wireless Communication Networks Research
- Nanocluster Synthesis and Applications
- Advanced Nanomaterials in Catalysis
- Water Quality Monitoring and Analysis
- Biofuel production and bioconversion
- GABA and Rice Research
- Geochemistry and Geochronology of Asian Mineral Deposits
- Vehicle emissions and performance
- Physical Unclonable Functions (PUFs) and Hardware Security
- Fault Detection and Control Systems
- Statistical Methods and Bayesian Inference
- Industrial Technology and Control Systems
- Molten salt chemistry and electrochemical processes
Xinjiang Production and Construction Corps
2025
Shihezi University
2025
Southeast University
2013-2024
Xi'an University of Technology
2024
National University of Defense Technology
2022
Southwest Jiaotong University
2018-2022
Qilu University of Technology
2022
Fudan University
2016-2020
Massachusetts Institute of Technology
2019
Shenyang Aluminum & Magnesium Engineering & Research Institute (China)
2012-2018
Cancer treatment has a far greater chance of success if the neoplasm is diagnosed before onset metastasis to vital organs. Hence, cancer early diagnosis extremely important and remains major challenge in modern therapeutics. In this contribution, facile new method for rapid multimodal tumor bioimaging reported by using biosynthesized iron complexes gold nanoclusters via simple introduction AuCl4- Fe2+ ions. The observations demonstrate that Au may act as fluorescent computed tomography...
Recent research has shown that deep learning models are vulnerable to adversarial attacks, including gradient which can lead incorrect outputs. The existing attack methods typically rely on repetitive multistep strategies improve their success rates, resulting in longer training times and severe overfitting. To address these issues, we propose an adaptive perturbation-based method with dual-loss optimization (APDL). This adaptively adjusts the single-step perturbation magnitude based...
A vine copula model is a flexible high-dimensional dependence which uses only bivariate building blocks. However, the number of possible configurations grows exponentially as variables increases, making selection major challenge in development. In this work, we formulate structure learning problem with both vector and reinforcement representation. We use neural network to find embeddings for best generate structure. Throughout experiments on synthetic real-world datasets, show that our...
In semantic segmentation, researchers face the shortage of pixel-level annotated data. And it is particularly severe in medical images. On other hand, unlabeled data are abundantly produced diagnosis routine. paper, we introduced Parasitic GAN for brain tumor segmentation to exploit more efficiently. composed three parts: segmentor S, generator G, and discriminator V. With label maps by supplementary synthesized generator, could learn a precise boundary ground truth. Thus, benefits from...
A novel intercarrier interference (ICI)-aware orthogonal frequency division multiplexing (OFDM) channel estimation network ICINet is presented for rapidly time-varying channels. consists of two components: a preprocessing deep neural subnetwork (PreDNN) and cascaded residual learning-based (CasResNet). By fully taking into account the impact ICI, proposed PreDNN first refines initial estimates in subcarrier-wise fashion. In addition, CasResNet designed to further enhance accuracy. The...
In this paper, we investigate the design of statistically robust detectors for multi-input multi-output (MIMO) systems subject to imperfect channel state information (CSI). A maximum likelihood (ML) detection problem is formulated by taking into consideration CSI uncertainties caused both estimation error and variation. To address challenging discrete optimization problem, propose an efficient alternating direction method multipliers (ADMM)-based algorithm, which only requires calculating...
As a kind of high efficient equipment waster treatment, Up-flow Anaerobic sludge Blanket(UASB)has many advantages, such as effluent quality, steady operation, broad application on treatment concentration organic wasters. Due to the COD consistency wastewater from starch production, expense are always very high. The paper started with advantages Sludge Bed(UASB), feasible generating marsh gas UASB was analyzed and utilization condition use were discussed. At last, it pointed that could be...
As the machine tool becomes more and complex, shallow model represented by learning SVM is difficult to characterize complex mapping relationship between measured signal health status of equipment, it faced with problem dimensionality disaster. In view feature extraction process uncertainty traditional intelligent recognition, a method fault recognition based on deep residual neural network proposed in this paper. This uses original time domain train complete classification type without...
A new Gibbs sampling DOA estimator based on Bayesian method (GSDB) is proposed to estimate the directions of multiple sources. The combines sampler and high-resolution method. formulation derived. not only possesses performance direction finding in original but also provides reduced computational complexity one from O(L/sup K/) O(K/spl times/J/spl times/N/sub s/). Comparison with MUSIC shows that has higher resolution better low SNR.
In order to disscuss the ability of H2-production and wastewater treatment, a up-flow anaerobic sludge bed (UASB) using synthesize substrate with brown sugar was conducted investigate hydrogen yield, producing rate, fermentation type biohydrogen production, chemical oxygen demand (COD) removal respectively. this paper, UASB reactor seeded from Harbin Wenchang Sewage treatment plant dewatered sludge. Successful start-up achieved within 40 days at 35±1°C.The concentration in influent is...
This paper proposes a method to evaluate the degradation stages of stability screw with data fusion technology and deep residual neural network. Firstly, provided by multi-sensors are fused, then time domain images signals input into network for training testing. The effectiveness proposed is verified using sets collected from test bench ball screw. contain massive samples involving 3 under 9 working conditions obtained accelerometers. From results comparison, it can be found that...
The receiver design for multi-input multi-output (MIMO) ultra-reliable and low-latency communication (URLLC) systems can be a tough task due to the use of short channel codes few pilot symbols. Consequently, error propagation occur in traditional turbo receivers, leading performance degradation. Moreover, processing delay induced by information exchange between different modules may also undesirable URLLC. To address issues, we advocate perform joint estimation, detection, decoding (JCDD)...
This paper proposes 3D-MedGAN, MLU-Net and Info-Max-Net models for overcoming the lack of labeled data extracting multi-level feature images in medical image segmentation. 3D-MedGAN is aimed at dealing with images. It uses a generative adversarial network to simulate then draws newly generated samples from distribution learned by model. Training segmentation model mixing real can effectively improve effect multiple layers different levels convolutional angles extract information By adopting...
The emergence of ultra-reliable and low-latency communication (URLLC) poses challenges on the receiver design. Classical turbo receivers, which exhibit excellent performance under long data packet transmissions, can suffer from non-negligible error propagation in context URLLC due to use short correction codes. To address this issue, we advocate a novel joint detection decoding (JDD) for multi-input multi-output (MIMO) systems with low-density parity-check (LDPC) Specifically, first...
A novel Bayesian high-resolution direction-of-arrival (DOA) estimator is proposed based on the maximum a posteriori principle. The statistical performance of DOA also investigated. Comparison with MUSIC and likelihood (MLE) shows that has highest resolution more accurate estimation for either incoherent or coherent sources. It robust in case low SNR.