A5104131986- Network Security and Intrusion Detection
- Anomaly Detection Techniques and Applications
- Artificial Immune Systems Applications
- EEG and Brain-Computer Interfaces
- Advanced Malware Detection Techniques
- Software System Performance and Reliability
- Smart Grid Security and Resilience
- Blind Source Separation Techniques
- Internet Traffic Analysis and Secure E-voting
- Advanced Memory and Neural Computing
- CCD and CMOS Imaging Sensors
- Reinforcement Learning in Robotics
- Opportunistic and Delay-Tolerant Networks
- Neural dynamics and brain function
- Advanced Control Systems Optimization
- Advanced Algorithms and Applications
- Software-Defined Networks and 5G
- Surface Modification and Superhydrophobicity
- Service-Oriented Architecture and Web Services
- Computational Geometry and Mesh Generation
- Data Management and Algorithms
- AI in Service Interactions
- Mechanical Circulatory Support Devices
- Industrial Vision Systems and Defect Detection
- Elevator Systems and Control
East China University of Science and Technology
2003-2025
Zhengzhou University of Light Industry
2024
Institute of Information Engineering
2023-2024
Beijing University of Posts and Telecommunications
2022-2024
Switch
2024
University of Chinese Academy of Sciences
2024
South China University of Technology
2024
Northwestern Polytechnical University
2023
Tsinghua University
2022
For least squares support vector machines (LS-SVM) based intrusion detection method, there is a big obstacle that the amount of audit data for modelling very large even small network scale, so it's impractical to directly train LS-SVM using original training datasets. Furthermore, adopts equality constraints and squared functions, which leads solve an easy-to-compute linear system, however followed lack sparseness, all will become LS-SVM, cause low speed. This paper proposed novel model...
In this work, polypropylene (PP) and PP/poly(ethylene-
Event-related potentials (ERPs) reflect electropotential changes within specific cortical regions in response to events or stimuli during cognitive processes. The P300 speller is an important application of ERP-based brain-computer interfaces (BCIs), offering potential assistance individuals with severe motor disabilities by decoding their electroencephalography (EEG) communicate. This study introduced a novel paradigm using dynamically growing bubble (GB) visualization as the stimulus,...
In view of the problem that computer generated Forces tools VR - has a stronger ability dynamic deployment but lacks communication network simulation, while simulation and tool VRNET's is strong lacked. With development interface, functions VRNET are integrated with matching model terrain two pivotal technologies as research point, by building an equipment mapping library to achieve through boundary, scale transformation coordinate system terrain. After functional integration, can provide...
To Solve the problem of low accuracy and high false alarm, a construction method Bagging ensemble based on random subspace PCA (Principle Component Analysis) was proposed. create training data for base classifier, feature set is randomly split into several subsets applied to each subset. all principal components are retained keep variety information in data; increase diversity classifiers ensemble, sampling with replacement used choose non-empty sample subset class; avoid performance...
Solving multiobjective optimization problems in many real-world domains is complicated because of insufficient data, multi-objection and multi-measurement, etc. In this paper, a grey model suggested to describe the situation new method, based on relational clustering analysis genetic algorithms, proposed deal with model. The simulation results demonstrate that suitable method effective.
This paper provides two concepts, i.e. generalized minimum dominating set and dynamic optimum rate, defines fitness function solved by genetic algorithm as well analyses the complexity of k-order adjacent matrix algorithm. Simple produced good results in simulation.
This paper starts from the concept of multi-Agent pinning control algorithm applied to UAV flocking control. Layered is proposed optimize tracking results for problem target loss that may occur when clusters are large and there external interference. According idea breadth-first search, speed position feedback its parent node added input ordinary agents, matching term improved by proposing a correction factor, which determines degree being affected neighborhood judging state each other, so...
Fixed broadband internet service can provide a stable network of up to 100 megabits or even gigabits and users at home use fixed for all kinds surfing, including website application access, watching videos, playing games, etc. Traditional maintenance networks primarily uses human manual meth-ods, supplemented by some low-level semi-automation operations. Since the long processes with numerous elements in network, it is difficult traditional operation support effectively high quality. When...