A5009136223- Advanced Adaptive Filtering Techniques
- Target Tracking and Data Fusion in Sensor Networks
- Blind Source Separation Techniques
- Distributed Sensor Networks and Detection Algorithms
- Advanced Computational Techniques and Applications
- Robotics and Sensor-Based Localization
- Sparse and Compressive Sensing Techniques
- Robotic Path Planning Algorithms
- Indoor and Outdoor Localization Technologies
- Machine Fault Diagnosis Techniques
- Computational Drug Discovery Methods
- Network Security and Intrusion Detection
- Image and Signal Denoising Methods
- Innovative Microfluidic and Catalytic Techniques Innovation
- Software Engineering and Design Patterns
- Guidance and Control Systems
- Distributed Control Multi-Agent Systems
- Educational Technology and Assessment
- Peer-to-Peer Network Technologies
- Security in Wireless Sensor Networks
- Electrical and Bioimpedance Tomography
- Control and Dynamics of Mobile Robots
- Underwater Acoustics Research
- Ultrasonics and Acoustic Wave Propagation
- Advanced Sensor and Control Systems
Northwestern Polytechnical University
2021-2024
China Pharmaceutical University
2020-2021
Southwest University
2018-2020
Shanghai Center for Brain Science and Brain-Inspired Technology
2019
Wuhan University
2014
PLA Information Engineering University
2011
Rongsheng Petrochemical (China)
2003
Well-known compressed sensing (CS) is widely used in image acquisition and reconstruction. However, accurately reconstructing images from measurements at low sampling rates remains a considerable challenge. In this paper, we propose novel Transformer-based hybrid architecture (dubbed TransCS) to achieve high-quality CS. the module, TransCS adopts trainable matrix strategy that gains better reconstruction by learning structural information training images. inspired powerful long-distance...
The convolutional neural network (CNN)-based reconstruction methods have dominated the compressive sensing (CS) in recent years. However, existing CNN-based approaches show potential restrictions capturing non-local similarity of images, because intrinsic characteristic layers, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathit{i.e.}$</tex-math></inline-formula> , locality and weight sharing. In...
Under false data-injection (FDI) attacks, the data of some agents are tampered with by FDI attackers, which causes that distributed algorithm cannot estimate ideal unknown parameter. Due to concealment malicious many detection algorithms against attacks often have poor results or low efficiencies. To solve these problems, a conveniently diffusion least-mean-square (DLMS) cross-verification (CV) is proposed attacks. The DLMS CV (DLMS-CV) comprised two subsystems: one subsystem provides test...
In the electronic information era, wireless sensor network (WSN) has always been an essential foundation for collection, processing and communication. WSN with multi-task estimation, distributed cooperation estimation cluster learning attractive topic. When unknown parameters become complex, some algorithms may not work, their performance could degrade. addition, problems of time delay, caused by synchronous data fusion, different sampling rates between types sensors are usually neglected in...
In recent years distributed estimation has attracted much attention. traditional algorithms, each node performs data fusion over synchronous data, which causes lots of time consumptions in the actual situations and performance degradation. To deal with this problem, we propose a new one-step asynchronous strategy algorithms. Moreover, proposed algorithms or without measurement sharing are studied to provide different cooperation strategies. particular, convergence behavior is analyzed, why...
With the continuous development of big data processing technology, selection algorithms gradually attract attention researchers. In this paper, a distributed data-selection diffusion least mean square (DLMS) algorithm, which can improve estimation accuracy traditional and also censor packets that do not bring enough innovation in wireless sensor networks, is proposed to valid iterative updates. And adaption-then-combination strategy algorithm obtained. Meanwhile, system, channel attacks are...
Semi-supervised encrypted traffic anomaly detection models in zero-positive scenarios are susceptible to human labeling errors or poisoning attacks, thereby compromising the stability and reliability of model. However, existing methods insufficient address challenge reduced inter-class distance caused by attacks inability reconstruction error serve as a reliable criterion. To alleviate these challenges, framework called Poison-Resistant Anomaly Detection (PRAD) is proposed mitigate enhance...
In this paper, we propose an adaptive malicious punishment DLMS algorithm to mitigate the adversarial attack. Two types of attacks are proposed in WSNs. The threshold is designed detect attacks. Then corresponding weight nodes can be reduced, due punishing factor. simulations demonstrate that robust
Nonlinear approximation is widely used in signal processing. Real-life signals can be modeled as functions of bounded variation. Thus the variable knot approximating function could self- adaptively chosen by balancing total variation target function. In this paper, we adopt continuous piecewise linear instead existing constants approximation. The results experiments show that new method superior to old one.
Compressed sensing is a revolutionary signal processing technique, which allows the signals of interest to be acquired at sub-Nyquist rate, meanwhile still permitting from highly incomplete measurements reconstructed perfectly. As well known, construction matrix one key technologies promote compressed theory application. Because Toeplitz can support fast algorithm and corresponds discrete convolution operation, it has essential research significance. However, conventional random matrix, due...
Drug-induced cardiotoxicity has become one of the major reasons leading to drug withdrawal in past decades, which is closely related blockade human Ether-a-go-go-related gene (hERG) potassium channel. Developing reliable hERG predicting model and optimizing can greatly reduce risk faced discovery. In this study, we constructed eight classification models, best shows desirable generalization ability on low-similarity clinical compounds, as well advantages perceiving activity gap caused by...
Considering the problem of noise in damage monitoring signal aviation connection structure, a new denoising algorithm based on combination EEMD and wavelet adaptive threshold function is proposed, which can effectively solve shortcomings single de-noising algorithm. In this algorithm, designed, poor continuity constant deviation traditional function. At same time, variable adjustment factor further adopted to realize different decomposition scales. The simulation compared soft hard...
With the development of computer technology, importance attached to security application systems is increasing day by day. Access control, as an important has been applied in operating system, database management and network system etc. This paper adopt ideology high cohesion low coupling software engineering, achieve design login component improving RBAC model under .NET platform, increases mechanism direct authorize users shield part users’ authority based on main role, enriches old mode...
This paper pointed out information resource construction of cartography teaching is the guarantee improving quality, important component course construction; according to characteristics classify resource, laying foundations for integration; Besides, with years experience in system, progress fundamental principle, method and achievements finally it discussed system.