A5103076352- Hydraulic flow and structures
- Hydrology and Sediment Transport Processes
- Advanced MIMO Systems Optimization
- Anomaly Detection Techniques and Applications
- Antenna Design and Optimization
- Speech and Audio Processing
- Millimeter-Wave Propagation and Modeling
- Advanced Vision and Imaging
- COVID-19 epidemiological studies
- Wireless Communication Security Techniques
- Wireless Signal Modulation Classification
- COVID-19 diagnosis using AI
- Advanced Adaptive Filtering Techniques
- Acoustic Wave Phenomena Research
- Matrix Theory and Algorithms
- Visual Attention and Saliency Detection
- Technology and Data Analysis
- Maritime Navigation and Safety
- Ship Hydrodynamics and Maneuverability
- Scientific Computing and Data Management
- Full-Duplex Wireless Communications
- Virtual Reality Applications and Impacts
- Landslides and related hazards
- Simulation and Modeling Applications
- Robotics and Sensor-Based Localization
University of Electronic Science and Technology of China
2022-2025
State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering
2023-2024
Nanjing Hydraulic Research Institute
2023-2024
University College London
2024
Communications Technology Laboratory
2023
San Jose State University
2021
Chongqing University of Posts and Telecommunications
2020
Purdue University West Lafayette
2020
Beihang University
2018-2019
University of Jinan
2019
360° omnidirectional images are very helpful in creating immersive multimedia contents, which enables a huge demand their efficient generation and effective assessment. In this paper, we leverage an attentive idea to meet by addressing two concerns: how generate good image fast robust way what is for human. To end, propose deep stitching approach facilitate the of images, composed modules. The low-resolution deformation module aims learn rules from dual-fisheye with joint implicit explicit...
Semantic communications can significantly improve transmission efficiency, especially in the low signal-to-noise (SNR) regime. However, two-way semantic still remain an unexplored topic, which is a critical issue many machine communication scenarios. Simply extending existing systems to situation requires bidirectional information feedback during training process, resulting significant overhead. To fill this gap, we investigate (TW-SC) system, where be omitted by exploiting weight...
Despite many exciting innovations in computer vision, recent studies reveal a number of risks existing vision systems, suggesting results such systems may be unfair and untrustworthy. Many these can partly attributed to the use training image dataset that exhibits sampling biases thus does not accurately reflect real visual world. Being able detect potential prior model development is essential for mitigating fairness trustworthy concerns vision. In this paper, we propose three-step...
In this paper, we present a novel method to stitch two fisheye images into panoramic image by using color calibration box. The box is new tool with five planes for double fisheyes camera calibrate. With the box, obtain parameters used in our map back same logical sphere and them according imaging model of cameras aligned back-to-back. Experimental results show that can not only produce 360-degree scenes but also eliminate brightness difference from cameras. Moreover, approach achieves...
Millimeter-wave massive MIMO can effectively improve the signal-to-noise ratio, but high-dimensional channel matrix significantly increases complexity of classic estimation algorithm. On other hand, millimeter-wave has low rank and sparsity properties in angle domain. Combining these two accuracy. This article proposes a novel sparse method based on joint nuclear norm ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1-2</sub> -regularization....
Intelligent wireless networks have long been expected to self-configuration and self-optimization capabilities adapt various environments demands. In this paper, we develop a novel distributed hierarchical deep reinforcement learning (DHDRL) framework with two-tier control in different times cales optimize the long-term spectrum efficiency (SE) of downlink cell-free multiple-input single-output (MISO) network, consisting multiple access points (AP) user terminals (UT). To realize proposed...
In the era of Internet, massive data are coming. The big mining and analysis is vital to guide people in their work. this paper, we introduce properties library borrowing distribution books. By data, bring forward some opinions three aspects. Firstly, according number books borrowed from different stacks, come up with ideas optimize floor set make reasonable use resources. Secondly, Based on time, suggestions management Finally, types most borrowed, put optimized collection library.
COVID-19 has resulted in a worldwide pandemic, leading to "lockdown" policies and social distancing. The pandemic profoundly changed the world. Traditional methods for observing these historical events are difficult because sending reporters areas with many infected people can put reporters' lives danger. New technologies needed safely responses policies. This paper reports using thousands of network cameras deployed purpose witnessing activities response continuously provide real-time...
The enormous energy carried by discharged water poses a serious threat to the Piano Key Weir (PKW) and its downstream hydraulic structures. However, previous research on dissipation in PKWs has mainly focused effects, methods have been largely limited physical model experiments. To deeply investigate discharge capacity characteristics of PKW, this study established PKW with universally applicable geometric parameters. By combining experiments numerical simulations, flow pattern at overflow...
Sea-crossing bridges face critical challenges due to scour, which can destabilize foundations. This study investigates the scour characteristics of large-diameter main pier Haiwen Bridge, with a pile diameter 4.3 meters. Seabed changes were monitored over 630 days using field tests and numerical simulations. The analyzed relationship between flow velocity depth, revealing that maximum recorded depth was 3.65 meters at upstream side. A linear regression model, developed from 75 sets data,...
This paper presents an algorithm for nonnegative matrix factorization 2D (NMF-2D) with the flexible β-Divergence. The β-Divergence is a group of cost functions parametrized by single parameter β. Least Squares divergence, Kullback-Leibler divergence and Itakura-Saito are special cases (β=2,1,0). more complete which uses range β, instead be limited to just cases. We describe maximization-minimization (MM) lead multiplicative updates. proposed decomposes information-bearing into...
This paper presents an algorithm for nonnegative matrix factorization 2D (NMF-2D) with the flexible s-Divergence. The s-Divergence is a group of cost functions parameterized by single parameter s. Least Squares divergence, Kullback-Leibler divergence and Itakura-Saito are special cases (s=2,1,0).This more complete holistic which uses range s, instead being limited to cases. We describe maximization minimization (MM) lead multiplicative updates. proposed decomposes information-bearing into...
Intelligent wireless networks have long been expected to self-configuration and self-optimization capabilities adapt various environments demands. In this paper, we develop a novel distributed hierarchical deep reinforcement learning (DHDRL) framework with two-tier control in different timescales optimize the long-term spectrum efficiency (SE) of downlink cell-free multiple-input single-output (MISO) network, consisting multiple access points (AP) user terminals (UT). To realize proposed...
As a competitive technology for 6G, semantic communications can significantly improve transmission efficiency. However, many existing communication systems require information feedback during the training coding process, resulting in significant overhead. In this article, we consider two-way (TW-SC) system, where be omitted by exploiting weight reciprocity transceiver. Particularly, channel simulator and transceiver are implemented on both TW-SC nodes distribution is modeled conditional...
Surface vortices is a common hydraulic phenomenon at water intakes, which has adverse effects on the safe operation of engineering. The complexity vortex motion, diversity influencing factors and limitation research methods make formation development mechanism not completely clear. This paper reports results an experiment morphologies dynamic characteristics intermittent air-entraining different moments during entire evolution processes. It found that processes involve inception stage, rapid...