检索结果-南通市图书馆

Rethinking multi-spatial information for transferable adversarial attacks on speaker recognition systems

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第3期9卷 620-631页

作者： Junjian Zhang Hao Tan Le Wang yaguan qian Zhaoquan Gu Guangzhou University GuangzhouChina Department of New Networks Peng Cheng LaboratoryShenzhenChina Harbin Institute of Technology(Shenzhen) ShenzhenChina Zhejiang University of Science and Technology HangzhouChina

Adversarial attacks have been posing significant security concerns to intelligent systems,such as speaker recognition systems(SRSs).Most attacks assume the neural networks in the systems are known beforehand,while black-box attacks are proposed without such information to meet practical *** black-box attacks improve trans-ferability by integrating multiple models or training on multiple datasets,but these methods are *** by the optimisation strategy with spatial information on the perturbed paths and samples,we propose a Dual Spatial Momentum Iterative Fast Gradient Sign Method(DS-MI-FGSM)to improve the transferability of black-box at-tacks against ***,DS-MI-FGSM only needs a single data and one model as the input;by extending to the data and model neighbouring spaces,it generates adver-sarial examples against the integrating *** reduce the risk of overfitting,DS-MI-FGSM also introduces gradient masking to improve *** authors conduct extensive experiments regarding the speaker recognition task,and the results demonstrate the effectiveness of their method,which can achieve up to 92%attack success rate on the victim model in black-box scenarios with only one known model.

关键词： speaker recognition spoofing attacks

Dynamic load distribution with hop-by-hop forwarding based on max-min one-way delay

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2014年第6期57卷 113-124页

作者： CHEN Fei WU ChunMing WANG Bin qian yaguan WU XiaoChun College of Computer Science Zhejiang University College of Science Zhejiang University of Science and Technology College of Information and Electronic Engineering Zhejiang Gongshang University

Load balancing can effectively improve network performance and scalability,but it may cause packet disorder,so worsening the ***,without MPLS to establish the desired end-to-end paths,hop-by-hop routing load balancing is more difficult to achieve than the source routing;however,it can significantly improve network *** this paper,we propose a load balancing scheme with hop-by-hop routing,by using the burstiness features of flows to make sure that the packets of the same flow arrive at the receiving end in *** results show that our algorithm can adapt to the dynamic changes of the end-to-end delay and the routing vector,and also can achieve fine-gained load balancing.

关键词： load balancing adaptive routing multipath routing traffic engineering packet reordering

Network View and Cognitive Mechanism for Virtual Network Resource Management Based Intelligent

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2014年第3期23卷 574-578页

作者： WU Xiaochun WU Chunming WANG Bin qian yaguan LAN Julong Computer Science College Zhejiang University Institute of Network and Communication Engineering Zhejiang Gongshang University National Digital Switching System Engineering & Technology R&D Center

Network virtualization has been a particularly important driving force behind the development of NGN. Finding effective ways to control self-adapting virtual networks has become a research hot spot. This paper tries to find an efficient way in combining dynamic resource sensing with traffic character and requirements in virtual network, based on that an intelligent cognition based virtual network resources management model has been put forward. We design a network topology view by coloring network topology matrix with current resources model. We present an intelligent virtual resource management mechanism based on the proposed network topology view. Experiment results illustrate that the proposed mechanism can improve VN capacity by automatic resource adjustment, which can obtain high utilization of VN resources without increasing computational expenses.

关键词： Intelligent cognition Self-adapting Virtual network

Modeling and Generating Realistic Background Traffic by Hybrid Approach

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2015年第10期12卷 147-157页

作者： qian yaguan GUAN Xiaohui JIANG Ming CEN Gang School of Science Zhejiang University of Science and Technology Zhejiang University of Water Resources and Electric Power College of Computer Science Hang Zhou Dianzi University School of Information and Electronic Engineering Zhejiang University of Science and Technology

One of the key challenges in largescale network simulation is the huge computation demand in fine-grained traffic *** from using high-performance computing facilities and parallelism techniques,an alternative is to replace the background traffic by simplified abstract models such as fluid *** paper suggests a hybrid modeling approach for background traffic,which combines ON/OFF model with TCP *** ON/OFF model is to characterize the application activities,and the ordinary differential equations(ODEs) based on fluid flows is to describe the TCP congestion avoidance *** apparent merits of this approach are(1) to accurately capture the traffic self-similarity at source level,(2) properly reflect the network dynamics,and(3) efficiently decrease the computational *** experimental results show that the approach perfectly makes a proper trade-off between accuracy and complexity in background traffic simulation.

关键词： network simulation background traffic ON/OFF models fluid flows self-similarity

在线全文

学校读者我要写书评

暂无评论

数字视频加密方法研究进展

电信科学 2008年第12期24卷 67-69页

作者：钱亚冠浙江科技学院理学院杭州310023

随着视频数据在网络通信流量中逐渐占据主导地位,其安全性日益引起人们的重视。视频数据与一般的文本数据相比存在诸多不同,同时视频通信对带宽和实时性要求很高,传统的数据加密方法无论在计算复杂度,还是与视频编码标准的兼容性方面,... 详细信息

随着视频数据在网络通信流量中逐渐占据主导地位,其安全性日益引起人们的重视。视频数据与一般的文本数据相比存在诸多不同,同时视频通信对带宽和实时性要求很高,传统的数据加密方法无论在计算复杂度,还是与视频编码标准的兼容性方面,都不能适应现状。本文对目前国内外学者在该领域的研究进展做了较详细的描述,在分析和比较各种加密方法的优点、缺点的基础上提出了自己的观点。

关键词：数字视频加密压缩算法

在线全文

学校读者我要写书评

暂无评论

一种基于不均衡数据的网络入侵流量分类方法

电信科学 2015年第6期31卷 89-95页

作者：关晓惠钱亚冠浙江水利水电学院杭州310018 浙江科技学院理学院杭州310023

在网络入侵流量检测中,普遍存在不同攻击类型的流量分布不均现象,导致少数攻击流量类识别率较低。为解决此类问题,基于不同特征空间的分类器流水线组合方法将多分类问题转化为不同特征空间上的两分类问题,有效地实现少数类重抽样和特征... 详细信息

在网络入侵流量检测中,普遍存在不同攻击类型的流量分布不均现象,导致少数攻击流量类识别率较低。为解决此类问题,基于不同特征空间的分类器流水线组合方法将多分类问题转化为不同特征空间上的两分类问题,有效地实现少数类重抽样和特征空间的优化,避免了少数类受多数类特征的干扰。实验表明,此方法可以有效地提高攻击流量中少数类的分类精度和召回率。

关键词：攻击流量类不均衡分类器流水线组合

在线全文

学校读者我要写书评

暂无评论

网络入侵检测系统中的漂移检测

电信科学 2015年第3期31卷 67-73页

作者：钱亚冠关晓惠浙江科技学院杭州310023 浙江水利水电学院杭州310018

目前基于机器学习的入侵检测系统大都建立在入侵数据始终保持统计平稳的假设之上,无法应对攻击者有意改变数据特性或新型攻击方式的出现,而导致的检测率下降的状况。对于上述问题,即攻击漂移,提出了加权Rényi距离的检测方法。在KDD Cu... 详细信息

目前基于机器学习的入侵检测系统大都建立在入侵数据始终保持统计平稳的假设之上,无法应对攻击者有意改变数据特性或新型攻击方式的出现,而导致的检测率下降的状况。对于上述问题,即攻击漂移,提出了加权Rényi距离的检测方法。在KDD Cup99数据集上的实验证明,Rényi距离可以有效地增强检测效果;在检测到漂移后,通过重新训练模型可以使得对攻击的识别率显著提高。

关键词：入侵检测攻击流量攻击漂移加权R6nyi距离

在线全文

学校读者我要写书评

暂无评论

基于过抽样技术的P2P流量识别方法

电信科学 2014年第4期30卷 109-113页

作者：钱亚冠张旻浙江科技学院理学院杭州310023 杭州电子科技大学计算机学院杭州310018

针对P2P类不平衡问题提出将复杂的多类不平衡问题转化为简单的两类不平衡问题,再通过迭代SMOTE过抽样技术丰富P2P的概念表达,从而提高P2P流量的识别率。实验结果表明,该方法可以显著提高诸如Na觙ve Bayes这样的简单模型在P2P上的识别率... 详细信息

针对P2P类不平衡问题提出将复杂的多类不平衡问题转化为简单的两类不平衡问题,再通过迭代SMOTE过抽样技术丰富P2P的概念表达,从而提高P2P流量的识别率。实验结果表明,该方法可以显著提高诸如Na觙ve Bayes这样的简单模型在P2P上的识别率,由此可证明该方法对于改善P2P类不平衡问题的有效性。

关键词：类不平衡过抽样 P2P流量识别

在线全文

学校读者我要写书评

暂无评论

大规模网络模拟的背景流量建模

电信科学 2011年第9期27卷 66-72页

作者：钱亚冠王滨关晓惠浙江科技学院理学院杭州310023 浙江大学计算机学院杭州310027 浙江水利水电专科学校计算机系杭州310018

大规模网络模拟的一个巨大挑战就是对计算资源的巨大需求,因此建立适合大规模网络模拟的背景流量模型具有十分重要的意义。本文提出的一种背景流量模型可实现3个目标:真实性、高效性和对网络动态的反馈性。该模型利用ON/OFF模型表达两... 详细信息

大规模网络模拟的一个巨大挑战就是对计算资源的巨大需求,因此建立适合大规模网络模拟的背景流量模型具有十分重要的意义。本文提出的一种背景流量模型可实现3个目标:真实性、高效性和对网络动态的反馈性。该模型利用ON/OFF模型表达两种典型的应用(Web和P2P)的用户行为模型;在TCP层上采用fluid的抽象流量表示,以提高模拟效率;用微分方程表达拥塞避免行为,使得流量源具有对网络的响应能力。最后通过DaSSFNet建立了模型原型,通过实验模拟,该模型框架可以很好地表达自相似的流量特征,同时在模拟效率上也远远高于包级流量模型。因此,该模型适合作为大规模的网络模拟环境的背景流量发生器。

关键词：大规模网络模拟背景流量 ON/OFF模型流体流