- Complex Network Analysis Techniques
- Speech and Audio Processing
- Network Traffic and Congestion Control
- Software-Defined Networks and 5G
- Interconnection Networks and Systems
- Cooperative Communication and Network Coding
- Advanced Adaptive Filtering Techniques
- Graph theory and applications
- Blind Source Separation Techniques
- Speech Recognition and Synthesis
- Network Security and Intrusion Detection
- Mobile Ad Hoc Networks
- Caching and Content Delivery
- IPv6, Mobility, Handover, Networks, Security
- Cloud Computing and Resource Management
- Opportunistic and Delay-Tolerant Networks
- Mobile Agent-Based Network Management
- Multimedia Communication and Technology
- Cryptography and Data Security
- Advanced Data Compression Techniques
- Peer-to-Peer Network Technologies
- Anomaly Detection Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Hearing Loss and Rehabilitation
- Energy Efficient Wireless Sensor Networks
Beijing Jiaotong University
2015-2024
Applied Radar (United States)
2021
Signal Processing (United States)
2021
Cloud Computing Center
2021
Beijing Information Science & Technology University
2013-2017
Microsoft Research Asia (China)
2008
Transmission efficiency and robustness are two important properties of various networks a number optimization strategies have been proposed recently. We propose scheme to enhance the network performance by adding small fraction links (or edges) currently existing topology, we present four edge addition for edges efficiently. aim at minimizing maximum node betweenness any in improve its transmission efficiency, experiments on both Barabási–Albert (BA) Erdös–Rényi (ER) confirmed effectiveness...
Since the betweenness of nodes in complex networks can theoretically represent traffic load under currently used routing strategy, we propose an improved efficient (IE) strategy to enhance network capacity based on centrality. Any node with highest is susceptible congestion. An way improve redistribute heavy from these central non-central nodes, so this paper, firstly give a path cost function by considering sum tunable parameter β along actual path. Then, minimizing cost, our IE achieved...
Global static routing is one kind of important algorithms for complex networks, especially in large communication networks. In this paper, we propose a heuristic global algorithm to mitigate traffic congestion on two-layer The proposed extends the relevant weighted literature [Y. Zhou, Y. F. Peng, X. L. Yang and K. P. Long, Phys. Sci. 84, 055802 (2011)]. Our path constructed from proper assignment edge weights by considering information both layers an adjustable parameter α. When adopted...
The study of traffic dynamics on multilayered networks is a hot issue, where the network topology composed two layers subnetworks, such as wired–wireless and peer-to-peer networks. Virtual links logical layer can be changed or constructed easily, therefore topological structure upper efficiently by link removal strategy. In this paper, edges linking to nodes with large betweenness will removed. upper-layer optimized freely based our method. Simulation results show that for both shortest path...
In this letter, we propose a new secure and efficient certificateless aggregate signature scheme which has the advantages of both public key cryptosystem signature. Based on computational Diffie-Hellman problem, our can be proven existentially unforgeable against adaptive chosen-message attacks. Most importantly, requires short group elements for constant pairing computations verification, leads to high efficiency due no relations with number signers.
Considering the heterogeneous structure of scale-free networks causing low traffic capacity network, we propose to improve network transport efficiency by rewiring a fraction edges for network. In this paper, six edge strategies are discussed and extensive simulations on Barabási–Albert (BA) confirm effectiveness these strategies. From another perspective, directly reuse removed under some edge-removal [Z. Liu, M. B. Hu, R. Jiang, W. X. Wang Q. S. Wu, Phys. Rev. E76 (2007) 037101; G. Zhang,...
As two-layer or multi-layer network model can more accurately reveal many real structures of complex systems such as peer-to-peer (P2P) networks on IP networks, to better understand the traffic dynamics and improve capacity, we propose efficiently construct structure upper logical layer which be possibly implemented. From beginning, assume that has same lower physical network, then use link-removal strategy in a fraction links with maximal product (k i * k j ) are removed from layer, where...
In real communication systems, each node has a finite queue length to store packets due physical constraints. this paper, we propose resource allocation strategy for traffic dynamics in scale-free networks. With of queue, the on i is based B , where generalized betweenness centrality i. The overall capacity network system can be evaluated by critical packet generating rate (R c ). Through use proposed scheme shortest path protocol and efficient routing protocol, our performs better than...
Abstract Today’s datacenter networks (DCNs) scale is rapidly increasing because of the wide deployment cloud services and rapid rise edge computing. The bandwidth consumption cost a DCN are growing sharply with extensions network size. Thus, how to keep traffic balanced key challenging issue. However, traditional load balancing algorithms such as Equal-Cost Multi-Path routing (ECMP) not suitable for high dynamic in DCNs. In this paper, we propose port-based forwarding scheduling (PFLBS)...
Traffic capacity is critical for various networks and strongly depends on the distribution of link's bandwidth resources. In this paper, we propose a betweenness-based allocation strategy in which each link l ij allocated proportionally to product (1 + B i ) α j , where tunable parameter, are betweenness node j, respectively. The optimal value achieved by extensive simulations slightly increases with network size. Our new achieves highest traffic when compared average previously proposed...
ABSTRACT Intrusion detection has become an indispensable technique to ensure the security and reliability of information systems. Support vector machine (SVM) its many improved algorithms have been successfully applied intrusion systems in recent years. However, training process SVM ignores important prior knowledge, within‐class structure set. In this paper, we propose classification algorithm, which combines minimum scatter Fisher discriminant analysis with traditional SVM. The central...
This paper proposes a novel construction of efficient certificateless aggregate signature (CLAS) scheme.On basis the computational Diffie-Hellman (CDH) assumption, proposed scheme can be proven existentially unforgeable against adaptive chosen-message attacks.The new also requires small constant pairing computations for verification, which is independent number signers.Most importantly, certain synchronization aggregating randomness avoided by scheme.All signers don't need to share same...
Outlier detection, as a type of one-class classification problem, is one important research topics in data mining and machine learning. Its task to identify sample points markedly deviating from the normal data. A reliable outlier detector needs build model which encloses tightly. In this paper, an improved SVM (OC-SVM) classifier proposed for detection problems. We name method OC-SVM with minimum within-class scatter (OC-WCSSVM), exploits inner-class structure training set via minimizing...
We propose network coding schemes for a broadcast system where the source packets exhibit hierarchical structure, such that successful information retrieval of packet at certain layer relies on correct receptions in previous layers. Assuming entire transmission consists two stages, stage and retransmission stage, progressive decoding, we novel combining based coding. The objective is to reduce number retransmissions decoding fraction packets. proposed can be applied scalable video systems.
Considering the link congestion based traffic model, which can more accurately model diffusing process of many real complex systems such as Internet, we propose an efficient weighted routing strategy in each link's weight is assigned with edge betweenness original un-weighted network a tunable parameter α. As links highest are susceptible to congestion, our efficiently redistribute heavy load from central noncentral links. The capacity under this new achieved when compared shortest path and...
Packets transmitting in real communication networks such as the Internet can be classified time-sensitive or time-insensitive. To better support real-time and time-insensitive applications, we propose a two-level flow traffic model which packets are labeled level-1 level-2, those with have higher priority to transmitted. In order enhance capacity of model, expand global dynamic routing strategy new source supports no routing-flaps, high capacity, diverse flows. As shown this paper, proposed...
This paper proposes a new approach for detecting tone errors in continuous Mandarin speech. In the training phase, variations are modeled with context-depended MSD-HMM which considers six contextual factors instead of two traditional triphone HMM. evaluation goodness pronunciation is measured by Kullback-Leibler divergence (KLD) between expected model and most representative model. When KLD models larger than threshold, detected as error. ROC curve, we get equal error rate at 2.6%.
ABSTRACT This paper presents a secure and efficient hierarchical identity‐based access authentication protocol for mobile network. First, new scheme is constructed to improve the registration performance by port‐based vector switching; especially when there are multihops bandwidth between user home network, delay greatly reduced. Second, an proposed combining designed proposal with signature. It has characteristics of identity based, local verification, concurrency hierarchy. The whole...
The forwarding address plays an important role in constructing a communication network. In this paper, new suitable for next generation networks named the vector (VA) is proposed which different from coding methods of current networks. characteristics VA are analyzed. Complex network theory and theoretical analysis method introduced to study average length when used construct global Simulation experiments practical topology model carried out validate results. results show that not only can...
Datacenter networks (DCNs) play a key role in providing cloud services. The energy consumption and cost of DCN are growing sharply with the extensions network bandwidth size. consumption, complexity depend on some design factors such as topology structure, addressing scheme routing mechanism. A novel architecture for cloud-service DCNs regular topologies is proposed this paper. First all, we propose port-based source-routing (PSRA) scheme, which makes table-lookup operation unnecessary...
With the increasing application of object detection algorithms in practice, single-stage detector YOLOv4 excellent both precision and inference speed has been widely used various scenarios. In order to further improve small detectability YOLOv4, this paper proposes combine 8-times downsampling feature map with that is outputted by second residual block CSPDarknet-53 processed via Hybrid Dilated Convolution (HDC), obtain new features. addition, comes up an idea using Focal Loss negative...