A5100422993- Interconnection Networks and Systems
- Mobile Ad Hoc Networks
- Advanced Graph Theory Research
- Cooperative Communication and Network Coding
- Network Security and Intrusion Detection
- Energy Efficient Wireless Sensor Networks
- Anomaly Detection Techniques and Applications
- Caching and Content Delivery
- Optimization and Search Problems
- Traffic Prediction and Management Techniques
- Complexity and Algorithms in Graphs
- Software Testing and Debugging Techniques
- Internet Traffic Analysis and Secure E-voting
- Advanced Malware Detection Techniques
- Opportunistic and Delay-Tolerant Networks
- Advanced Data Storage Technologies
- Software Reliability and Analysis Research
- Energy Load and Power Forecasting
- Adversarial Robustness in Machine Learning
- Graph Labeling and Dimension Problems
- Context-Aware Activity Recognition Systems
- Transportation Planning and Optimization
- Peer-to-Peer Network Technologies
- Privacy-Preserving Technologies in Data
- Cloud Computing and Resource Management
China Institute of Water Resources and Hydropower Research
2023-2025
Beihang University
2024
Ministry of Transport
2024
Yanshan University
2023-2024
Hainan Agricultural School
2024
Zhejiang Normal University
2015-2024
Beijing Institute of Technology
2024
Beijing Institute of Petrochemical Technology
2024
Dalian Maritime University
2024
Beijing University of Posts and Telecommunications
2014-2023
We all depend on mobility, and vehicular transportation affects the daily lives of most us. Thus, ability to forecast state traffic in a road network is an important functionality challenging task. Traffic data often obtained from sensors deployed network. Recent proposals spatial-temporal graph neural networks have achieved great progress at modeling complex correlations data, by as diffusion process. However, intuitively, encompasses two different kinds hidden time series signals, namely...
Traditional human activity recognition (HAR) based on a motion sensor adopts sliding window labeling and prediction. This method faces the multi-class problem, which mistakenly labels different classes of sampling points within as class. In this paper, we propose novel HAR U-Net to overcome performing prediction each point. The data collected from wearable sensors are mapped into an image with single-pixel column multi-channel, then, it is input network complete pixel-level function. We...
Spinel-type MnCo2O4 nanofibers (MCFs) were successfully synthesized by electrospinning and sequential calcination. The crystal structure, composition morphology of the MCFs characterized X-ray diffraction, Fourier transform infrared spectroscopy, energy-dispersive photoelectron scanning electron microscopy transmission microscopy. Due to outstanding electron-transfer ability spinel large surface area nanofibers, employed as electrocatalysts for oxidation glucose. Cyclic voltammetry...
With the rapid development of Internet Things technology, a large number devices are connected to Things, and at same time, network attacks security threats introduced. Intrusion detection system (IDS) is one effective methods for protecting network. rise artificial intelligence intrusion based on ML/DL widely applied. However, neural vulnerable adversarial perturbation. Most existing cannot guarantee basic function traffic data. In this paper, we propose an improved attack model Generated...
In this paper, we study the problem of computing quality fault-tolerant virtual backbone in homogeneous wireless network, which is defined as <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex Notation="TeX">$k$</tex></formula> -connected Notation="TeX">$m$</tex></formula> -dominating set a unit disk graph. This NP-hard, and thus many efforts have been made to find constant factor approximation algorithm for it, but...
In this paper, we study the problem of constructing quality fault-tolerant Connected Dominating Sets (CDSs)in homogeneous wireless networks, which can be defined as minimum k-Connected m-Dominating Set ((k,m)-CDS) in Unit Disk Graphs (UDGs). We found that every existing approximation algorithm for is incomplete k ¿3 a sense it does not generate feasible solution some UDGs. Based on these observations, propose new polynomial time computing (3,m)-CDSs. also show our correct and its ratio constant.
Abstract With the technical support of 5G, drone network plays a critical role in autonomous and digital era. However, due to wireless autonomy characteristics, is prone diverse malicious attacks, so it's vital deploy intrusion detection system detect attacks. For real open environment, unknown attacks will occur constantly, but existing methods are usually designed for static closed‐set scenario fail recognize correctly, threatening security network. Therefore, we design an Based on...
The size of datasets is growing exponentially as information technology advances, and it becoming more crucial to provide efficient learning algorithms for neural networks handle massive amounts data. Due their potential handling huge datasets, feed-forward with random weights (FNNRWs) have drawn a lot attention. In this paper, we introduced an network scheme (FNNS) processing weights. FNNS divides large-scale data into subsets the same size, each subset derives corresponding submodel....
Wireless Sensor Networks (WSNs), the major technique in sensing layer of Internet Things. Because WSN is extensively regarded as a data-centric network, more and researchers start to exploit information-centric networking (ICN), branch future network architecture, design WSNs. In this paper, we propose collaborative caching strategy for wireless sensor (ICN-WSN). The proposed consists three parts: node betweenness based cache size adjustment, data replacement frequency decision, content...
This paper evaluates the potential gains a workflow-aware storage system can bring. Two observations make us believe such is crucial to efficiently support workflow-based applications: First, workflows generate irregular and application-dependent data access patterns. These patterns render existing systems unable harness all optimization opportunities as this often requires conflicting options or even design decision at level of system. Second, when scheduling, workflow runtime engines...