检索结果-内蒙古大学图书馆

QoS-aware Traffic Classification Architecture Using Machine Learning and deep Packet Inspection in SdNs

Procedia Computer Science 2018年 131卷 1209-1216页

作者： Changhe Yu Julong Lan JiChao Xie Yuxiang Hu National Digital Switching System Engineering & Technological R&D Center Zhengzhou 450002 China

The QoS-aware traffic classification techniques of SdN networks is the basis for network to provide fine-grained QoS traffic engineering. In this paper, we propose an architecture which combines deep packet detection and semi-supervised machine learning of multi-classifier in SdN. This architecture can classify flows into different QoS categories. Based on this, network can achieve fine-grained adaptive QoS traffic engineering. Moreover, through deep packet detection techniques, network can maintain a dynamic flow database. Classifier can adapt to the rapid emergence of network application and fickle traffic characteristics of current network by periodically re-training with the dynamic flow database. Experiments verify that our classification framework can achieve good classification accuracy.

关键词： traffic classification QoS machine learning SdN

来源：评论

学校读者我要写书评

暂无评论

Survey on quantitative evaluations of moving target defense

引用

网络与信息安全学报 2018年第9期4卷 66-76页

作者： Huanruo LI Yunfei GUO Shumin HUO Guozhen CHENG Wenyan LIU National Digital Switching System Engineering & Technological R&D Center Zhengzhou 450002 China

Quantitative evaluations are of great importance in network security *** recent years,moving target defense(MTd)has appeared to be a promising defense approach that blocks asymmetrical advantage of attackers and favors the defender-notwithstanding,it has a limited deployment due to its uncertain efficiency and effectiveness in *** that case,quantitative metrics and evaluations of MTd are essential to prove its capability and impulse its further *** article presents a comprehensive survey on state-of-the-art quantitative ***,taxonomy of MTd techniques is stated according to the software stack ***,a concrete review and comparison on existing quantitative evaluations of MTd is ***,notice-worthy open issues regarding this topic are proposed along with the conclusions of previous studies.

关键词：计算机网络网络管理应用程序信息安全

来源：评论

学校读者我要写书评

暂无评论

A dynamic Shortest Path deployment of Virtual Network Function

A Dynamic Shortest Path Deployment of Virtual Network Functi...

引用

2nd International Conference on Mechanical, Electronic, Control and Automation engineering (MECAE 2018)

作者： Xiaolei Wang Shiqing Sun Hongbo Tang National Digital Switching System Engineering & Technological R &D Center

In the 5 G mobile communication network virtualization scenario, how to deploy service function chaining of the core network efficiently is the key problem to realize the efficient deployment of virtual Evolved Packet Core network services. In order to solve the problem that the existing deployment methods are difficult to meet the requirement of the mobile communication with low latency, this paper proposed a method for service function chaining deployment based on Q-learning. This method solved the problem by applying establish a Markov decision process model to the latency optimization in the context of VNF deployment, and then design a Q-learning algorithm to found the deployment solutions with minimum delay cost of network services. Simulation results show that the proposed method achieves better performances in terms of average processing time, request acceptance rate, gain and execution time.

关键词： 5G network function virtualization v EPC service function chaining Q-learning

来源：评论

学校读者我要写书评

暂无评论

Artificial Noise Aided Polar Codes for Physical LayerSecurity

引用

China Communications 2017年第12期14卷 15-24页

作者： Huiqing Bai Liang Jin Ming Yi China National Digital Switching System Engineering and Technological R&D Center Zhengzhou 450002 China

The secrecy rates of the existing practical secrecy coding methods are relative low to satisfy the security requirement of 5 G *** propose an artificial noise(AN) aided polar coding algorithm to improve the secrecy ***,a secrecy coding model based on AN is presented,where the confidential bits of last transmission code block are adopted as AN to inject into the current *** this way,the AN can only be eliminated from the jammed codeword by the legitimate *** the AN is shorter than the codeword,we then develop a suboptimal jamming positions selecting algorithm with the goal of maximizing the bit error rate of the *** and simulation results demonstrate that the proposed algorithm outperforms the random selection method and the method without AN.

关键词： physical layer security polarcodes artificial noise jamming positions se-lection

来源：评论

学校读者我要写书评

暂无评论

Person re-Identification based on data Prior distribution

Person Re-Identification based on Data Prior Distribution

引用

2018 International Conference on Advanced Control, Automation and Artificial Intelligence (ACAAI2018)

作者： Yancheng Wu Hongchang Chen Shaomei Li Chao Gao Hongxin Zhi Yuchao Jiang Yanchuan Wang National Digital Switching System Engineering & Technological R&D Center

In order to solve the problem of insufficient accuracy of the existing person re-identification *** propose a neural network model for identifying pedestrian properties and pedestrian Id. Compared with the existing methods, the model mainly has the following three advantages. First, our network adds extra full connection layer, ensure model migration ability. Second,based on the number of samples in each attribute, the loss function of each attribute has been normalized, avoid number unbalanced among the attributes to effect the identification accuracy. Third, we use the distribution of the attribute data in the prior knowledge, through the number to adjust the weight of each attribute in the loss layer, avoid the number of data sets for each attribute of positive and negative samples uneven impact on recognition. Experimental results show that the algorithm proposed in this paper has high recognition rate, and the rank-1 accuracy rate on dukeMTMC dataset is 72.83%,especially on Market1501 dataset. The rank-1 accuracy rate is up to 86.90%.

关键词： person re-identification data prior distribution weight adjustment deep learning neural network

来源：评论

学校读者我要写书评

暂无评论

Context-based Synthetic data for Logo recognition

Context-based Synthetic Data for Logo Recognition

引用

2018 3rd International Conference on Computer Science and Information engineering(ICCSIE 2018)

作者： Yuchao Jiang Lixin Ji Chao Gao Yancheng Wu National Digital Switching System Engineering & Technological R&D Center

In order to solve the problem of sparse training samples in logo recognition task,a multi-type context-based logo data synthesis algorithm is *** algorithm comprehensively utilizes the local and full context of the logo object and the scene image to guide the synthesis of the logo *** experimental results on the FlickrLogos-32 show that the proposed algorithm can greatly improve the performance of the logo recognition algorithm without relying on additional manual annotation,verify the validity of the synthesis algorithm,and further prove that multi-type context can improve the performance of the object recognition algorithm.

关键词： logo recognition context data synthesis deep learning

来源：评论

学校读者我要写书评

暂无评论

An Adaptation Method in Noise Mismatch Conditions for dNN-Based Speech Enhancement

An Adaptation Method in Noise Mismatch Conditions for DNN-Ba...

引用

2018 International Conference on Network, Communication, Computer engineering (NCCE 2018)

作者： Siying Xu dan Qu Xingyan Long China National Digital Switching System Engineering & Technological R&D Center

ISBN: (纸本)9781510871076

The deep learning-based speech enhancement has shown considerable ***,it still suffers performance degradation under mismatch *** this paper,an adaptation method is proposed to improve the performance under noise mismatch ***,we advise a noise aware training by supplying identity vectors(ivectors) as parallel input features to adapt dNN acoustic models with the target ***,given a small amount adaptation data,the noise-dependent dNN is obtained by using Euclidean distance regularization from a noiseindependent dNN,and forcing the estimated masks to be close to the unadapted ***,experiments were carried out on different noise and SNr conditions,and the proposed method has achieved significantly 29% benefits of STOI at most and provided consistent improvement in PESQ and seg SNr against the baseline systems.

关键词： Enhancement adaptation method dNN PESQ segSNr STOI

来源：评论

学校读者我要写书评

暂无评论

Active Perceptive dynamic Scheduling Mechanism Based on Negative Feedback

引用

Procedia Computer Science 2018年 131卷 520-524页

作者： Yingying Lv Yunfei Guo Qiang Chen Guozhen Cheng Yang Chen National Digital Switching System Engineering & Technological R&D Center Zhengzhou 450002 Henan China

With the popularization of the Internet, the production and living of people and even the security development of the country have been more and more dependent on cyberspace, the importance of cyberspace has become increasingly prominent. Network security should be paid more and more attention to. Mimicry defense is one of active defense technologies. We focused on the dHr architecture and give a research on mimicry scheduling mechanism based on negative feedback and we have analyzed the feasibility of this method in theory.

关键词： mimicry defense active defense negative feedback

来源：评论

学校读者我要写书评

暂无评论

A Correlation-aware diverse Variant Placement to Increase Network resilience

A Correlation-aware Diverse Variant Placement to Increase Ne...

引用

International Conference on Computer and Communications (ICCC)

作者： Jianjian Ai Hongchang Chen Guozhen Cheng National Digital Switching System Engineering and Technological R&D Center Zhengzhou China

ISBN: (纸本)9781538683408

due to the high homogeneity of current routing infrastructure, the resilience of the network is facing a serious threat when a defective software upgrade or a denial-of-service attack occurs. Many existing works adopt heterogeneity philosophy to improve the resilience of the network. For example, diverse variants are placed to nodes in the network. However, the existing works assume that diverse variants do not have common vulnerabilities, which is an invalid assumption in some real networks. Therefore, the existing diverse variant placement algorithms could not achieve optimal performance. In this paper, we consider the situation that some variants have common vulnerabilities. We model the correlation-aware diverse variant placement problem as an integer-programming optimization problem. Since the problem is NP-hard, we design a Simulated Annealing-based algorithm to efficiently solve the problem. The simulation results show that compared with baseline algorithms, the proposed algorithms can effectively improve network resilience about 15%.

关键词： resilience Network topology routing Topology Mathematical model Simulated annealing Measurement

来源：评论

学校读者我要写书评

暂无评论

N-Fold Superposition: Improving Neural Networks by reducing the Noise in Feature Maps

N-Fold Superposition: Improving Neural Networks by Reducing ...

引用

International Conference on Audio, Language and Image Processing, ICALIP

作者： Yang Liu Qiang Qu Chao Gao National Digital Switching System Engineering & Technological R&D Center Zhengzhou China

Considering the use of Fully Connected (FC) layer limits the performance of Convolutional Neural Networks (CNNs), this paper develops a method to improve the coupling between the convolution layer and the FC layer by reducing the noise in Feature Maps (FMs). Our approach is divided into three steps. Firstly, we separate all the FMs into n blocks equally. Then, the weighted summation of FMs at the same position in all blocks constitutes a new block of FMs. Finally, we replicate this new block into n copies and concatenate them as the input to the FC layer. This sharing of FMs could reduce the noise in them apparently and avert the impact by a particular FM on the specific part weight of hidden layers, hence preventing the network from overfitting to some extent. Using the Fermat Lemma, we prove that this method could make the global minima value range of the loss function wider, by which makes it easier for neural networks to converge and accelerates the convergence process. This method does not significantly increase the amounts of network parameters (only a few more coefficients added), and the experiments demonstrate that this method could increase the convergence speed and improve the classification performance of neural networks.

关键词： Frequency modulation Convolution Kernel Couplings Convergence Biological neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：