网络层析成像若干关键技术研究
|
摘要
随着互联网的飞速发展，网络的服务和网络的架构也在发生深刻变化，要成功设计、控制和管理网络，就需要了解和掌握网络的内部特性(包括各种网络特征参数和拓扑结构)。如何在日益复杂的互联网网络环境中实时、精确测量网络性能已成为国内外科学界和工业界共同关注的前沿科学问题之一，该问题是网络行为学的基本研究内容，也是网络优化设计、异常检测与分析、网络监测与评估等工作的基础。传统的网络测量方法需要通过网络节点间的协作来求解，但是随着网络朝着分布化、非协作、异构化管理和基于边缘控制的方向演变，我们发现传统网络测量方法不能完全满足各种应用对网络测量的要求，针对某些特定情况，我们还需要研究能够不依赖网络中间节点协作的网络测量方法。
网络层析成像将医学和地震学等领域成功应用的层析成像理论和方法应用于通信网络的测量问题，能够在没有中间节点协作的条件下，将网络测量问题转化为反问题的求解，估计得到大尺度网络内部的各种性能统计信息，是目前备受国际学术界关注的网络测量新技术之一。本文系统地回顾了网络层析成像的起源与各个阶段所取得的重要成果，系统地对网络层析方法进行了分类，指出了在当前网络环境下进行网络层析成像所面临的一些主要问题；从网络层析成像解的稳定性、唯一性、求解实时性和非线性网络层析成像问题出发，针对网络层析成像中所面临的几个实际问题进行了深入研究，在以下几个方面取得了一些有价值的研究成果：
1．网络链路时延估计的层析成像方法研究
提出了一种基于递归多感知器的网络链路延迟追踪算法，该算法能够在没有任何先验信息条件下追踪非平稳网络链路平均延迟的时变行为，估计链路延迟的概率密度分布，克服了以往方法对网络链路先验模型的依赖，很大程度提高了解的精度和求解的稳定性、实时性。
2．网络链路丢包率估计的层析成像方法研究
提出了基于包群的单播链路丢包率估计层析成像方法，在根节点一次性向所有叶节点发送探测包群，包群中相邻节点或相近节点可以自由组合成包对、三包组、四包组等，获得路径的丢包率和条件丢包率，构成超定方程组；为了精度和计算复杂度考虑，利用基于奇异值分解的广义逆方法求解超定系统方程得到最后链路的丢包率。该方法与现有的三包组方法相比，可以在探测包数量相同的情况下，通过自由组合，构成包含更多方程的超定方程组，提高了解的精度和稳定性。
3．大尺度网络流量矩阵估计方法研究
提出了一种基于递归多感知器大尺度网络流量矩阵的追踪算法，首先引入了黑盒子模型对流量矩阵估计问题进行无先验建模；其次，利用递归神经网络对模型进行求解；最后得到整个网络流量矩阵的追踪算法。该方法克服了对先验模型的依赖，提高了估计精度、稳定性和实时性，还便于采用并行算法或网络分布式算法进一步提高求解的实时性。
4．网络拓扑结构估计方法研究
提出了一种快速的网络拓扑结构的分层聚类估计算法，通过引入了基于高斯混合模型的递归算法，改进了网络拓扑结构估计中相似性量度聚类过程，可以在不降低估计精度的前提下，提高了网络拓扑结构估计的实时性。
5．网络流量分类方法研究
提出了一种基于混合高斯的网络流量半监督分类方法。与有监督分类方法相比，半监督方法只需要利用少部分的标识数据(大约10％)进行分类，由于标识的数据获得经常是困难的、昂贵的或者花费大量的时间，半监督方法非常适合具体的应用，具有更好的实时性；与无监督分类方法相比，由于使用标识数据进行聚类团的类别确定，具有更高的分类精度。同时，分析了影响半监督分类方法性能的几大因素，优化了半监督分类方法的参数配置，进一步提高了半监督分类方法的分类性能。
With the rapid development of the internet, network services and architecture are also undergoing profound changes. To design, control and manage the network successfully, we have to understand and grasp its internal characteristics (including all kinds of network parameters and topology). How to accurate and timely estimate network performance characteristics in more and more complex internet environment have become one of the forefront scientific problems which are focused by the academic community and industry community all over the world; Meantime, this problem is not only the basic research content of network behavior, but also the foundation of network optimize design, anomaly detection and analysis, network monitoring and evaluation and so on. In past, the traditional internet measure method usually resolves such a problem through the cooperation of node, however, as internet toward the evolution of distributive, no corporative, heterogeneous management and edge control, the traditional internet measure method do not meet fully the requirements of various applications on network measurement, for certain specific circumstances, we still need to study the network measurement method without the cooperation of individual servers and routers.
By applying tomography theory which is successfully applied in fields such as medicine and seismology to the problems of network measurement in communication network, network tomography can infer the internal performance without the cooperation of internal node and transform network measurement problem into getting the solution of inversion problem in large-scale networks, which has become one of the focused new technologies. In the thesis, we review systematically the past progresses and important achievements at different phases in network tomography research area. The classification of network tomography methodologies are summarized systematically. We also point out some key issues for large scale network tomography. This thesis study profoundly some practical issues of network tomography from nonlinear, stable, unique and real-time solution point of view and results in following innovative achievements:
1. Research on network tomography method for link delay estimation
This thesis propose a novel link delay tracking algorithm based on the recurrent multilayer perceptron (RMLP) network, this algorithm is capable of tracking time-varying average delay behavior in non-stationary network and estimating the probability density distribution of internal delay characteristics without any prior information. Compare with existing delay tomography, RMLP method don't depend on the queue delay prior model and improve the stability, real-time of solution, obviously.
2. Research on network tomography method for link loss estimation
This thesis propose a method of estimating the packet loss rate on unicast network which is based on multi-packet stripe probe. The multi-packet stripe probe is firstly sent to all receivers to cover all paths in the same time. Then, multi-packet stripe probe to the adjacent nodes or similar nodes are free to form two-packet stripe, three-packet stripe or four-packet stripe and we calculate absolute packet loss rate and conditional packet loss rate of the path to construct an over-determined equations. In order to having better accuracy and reducing computational complexity, we choose singular value decomposition (SVD) to calculate the Moore-Penrose generalized inverse matrix of the large-scale over-determined system equations, and finally get the link loss rate estimation. Compare with existing three-packet stripe method, our method can construct over-determined equations with more equations under the same number of probe packet circumstances, and that obtain more stability, flexibility and accurate estimation.
3. Research on large scale traffic matrix estimation method
This thesis propose a novel method of large scale traffic matrix estimation based on recurrent multilayer perceptron (RMLP) network. First of all we introduce the black-box models to model tracking traffic matrix without any prior information, then we use RMLP to resolve the solution of black-box models, finally, we get the entire based-RMLP traffic matrix tracking algorithm. Compare with existing delay tomography, RMLP method don't depend on the traffic prior model and improve the accuracy of estimates, stability and real-time demand obviously. Meanwhile, by using parallel algorithms and network distributed algorithms, we will further improve the real-time of solution.
4. Research on network topology discovery
By introducing recursive unsupervised learning of Gaussian Mixture Models (GMM) to topology discovery approach, this thesis proposed a Fast Hierarchical Topology Estimation (FHTE) approach. By improving similarity metrics clustering process in network topology estimation, our proposed method evidently reducing the computation time under the same estimation accuracy circumstances compare to the Hierarchical Topology Estimation (HTE) approach.
5. Research on network traffic classification
This thesis propose a GMM-based semi-supervised classification method. Compare with supervised classification method, our approach takes advantage of only a few labeled data (about 10%) to identify different internet applications. Since obtaining the labeled instances is often difficult, expensive, or time consuming, the approach is particularly well suite for practical application and is more real-time. Semi-supervised classification method obtain more accurate classification compare to unsupervised classification method since it use a small amount of labeled instances to help us identify which Gaussian belongs to some class. Meanwhile, by analysis of semi-supervised classification method factors and their impact on classification performance, an optimum configuration is achieved for the GMM-based semi-supervised classification system, which further improve the classification performance of our approach.
网络层析成像将医学和地震学等领域成功应用的层析成像理论和方法应用于通信网络的测量问题，能够在没有中间节点协作的条件下，将网络测量问题转化为反问题的求解，估计得到大尺度网络内部的各种性能统计信息，是目前备受国际学术界关注的网络测量新技术之一。本文系统地回顾了网络层析成像的起源与各个阶段所取得的重要成果，系统地对网络层析方法进行了分类，指出了在当前网络环境下进行网络层析成像所面临的一些主要问题；从网络层析成像解的稳定性、唯一性、求解实时性和非线性网络层析成像问题出发，针对网络层析成像中所面临的几个实际问题进行了深入研究，在以下几个方面取得了一些有价值的研究成果：
1．网络链路时延估计的层析成像方法研究
提出了一种基于递归多感知器的网络链路延迟追踪算法，该算法能够在没有任何先验信息条件下追踪非平稳网络链路平均延迟的时变行为，估计链路延迟的概率密度分布，克服了以往方法对网络链路先验模型的依赖，很大程度提高了解的精度和求解的稳定性、实时性。
2．网络链路丢包率估计的层析成像方法研究
提出了基于包群的单播链路丢包率估计层析成像方法，在根节点一次性向所有叶节点发送探测包群，包群中相邻节点或相近节点可以自由组合成包对、三包组、四包组等，获得路径的丢包率和条件丢包率，构成超定方程组；为了精度和计算复杂度考虑，利用基于奇异值分解的广义逆方法求解超定系统方程得到最后链路的丢包率。该方法与现有的三包组方法相比，可以在探测包数量相同的情况下，通过自由组合，构成包含更多方程的超定方程组，提高了解的精度和稳定性。
3．大尺度网络流量矩阵估计方法研究
提出了一种基于递归多感知器大尺度网络流量矩阵的追踪算法，首先引入了黑盒子模型对流量矩阵估计问题进行无先验建模；其次，利用递归神经网络对模型进行求解；最后得到整个网络流量矩阵的追踪算法。该方法克服了对先验模型的依赖，提高了估计精度、稳定性和实时性，还便于采用并行算法或网络分布式算法进一步提高求解的实时性。
4．网络拓扑结构估计方法研究
提出了一种快速的网络拓扑结构的分层聚类估计算法，通过引入了基于高斯混合模型的递归算法，改进了网络拓扑结构估计中相似性量度聚类过程，可以在不降低估计精度的前提下，提高了网络拓扑结构估计的实时性。
5．网络流量分类方法研究
提出了一种基于混合高斯的网络流量半监督分类方法。与有监督分类方法相比，半监督方法只需要利用少部分的标识数据(大约10％)进行分类，由于标识的数据获得经常是困难的、昂贵的或者花费大量的时间，半监督方法非常适合具体的应用，具有更好的实时性；与无监督分类方法相比，由于使用标识数据进行聚类团的类别确定，具有更高的分类精度。同时，分析了影响半监督分类方法性能的几大因素，优化了半监督分类方法的参数配置，进一步提高了半监督分类方法的分类性能。
With the rapid development of the internet, network services and architecture are also undergoing profound changes. To design, control and manage the network successfully, we have to understand and grasp its internal characteristics (including all kinds of network parameters and topology). How to accurate and timely estimate network performance characteristics in more and more complex internet environment have become one of the forefront scientific problems which are focused by the academic community and industry community all over the world; Meantime, this problem is not only the basic research content of network behavior, but also the foundation of network optimize design, anomaly detection and analysis, network monitoring and evaluation and so on. In past, the traditional internet measure method usually resolves such a problem through the cooperation of node, however, as internet toward the evolution of distributive, no corporative, heterogeneous management and edge control, the traditional internet measure method do not meet fully the requirements of various applications on network measurement, for certain specific circumstances, we still need to study the network measurement method without the cooperation of individual servers and routers.
By applying tomography theory which is successfully applied in fields such as medicine and seismology to the problems of network measurement in communication network, network tomography can infer the internal performance without the cooperation of internal node and transform network measurement problem into getting the solution of inversion problem in large-scale networks, which has become one of the focused new technologies. In the thesis, we review systematically the past progresses and important achievements at different phases in network tomography research area. The classification of network tomography methodologies are summarized systematically. We also point out some key issues for large scale network tomography. This thesis study profoundly some practical issues of network tomography from nonlinear, stable, unique and real-time solution point of view and results in following innovative achievements:
1. Research on network tomography method for link delay estimation
This thesis propose a novel link delay tracking algorithm based on the recurrent multilayer perceptron (RMLP) network, this algorithm is capable of tracking time-varying average delay behavior in non-stationary network and estimating the probability density distribution of internal delay characteristics without any prior information. Compare with existing delay tomography, RMLP method don't depend on the queue delay prior model and improve the stability, real-time of solution, obviously.
2. Research on network tomography method for link loss estimation
This thesis propose a method of estimating the packet loss rate on unicast network which is based on multi-packet stripe probe. The multi-packet stripe probe is firstly sent to all receivers to cover all paths in the same time. Then, multi-packet stripe probe to the adjacent nodes or similar nodes are free to form two-packet stripe, three-packet stripe or four-packet stripe and we calculate absolute packet loss rate and conditional packet loss rate of the path to construct an over-determined equations. In order to having better accuracy and reducing computational complexity, we choose singular value decomposition (SVD) to calculate the Moore-Penrose generalized inverse matrix of the large-scale over-determined system equations, and finally get the link loss rate estimation. Compare with existing three-packet stripe method, our method can construct over-determined equations with more equations under the same number of probe packet circumstances, and that obtain more stability, flexibility and accurate estimation.
3. Research on large scale traffic matrix estimation method
This thesis propose a novel method of large scale traffic matrix estimation based on recurrent multilayer perceptron (RMLP) network. First of all we introduce the black-box models to model tracking traffic matrix without any prior information, then we use RMLP to resolve the solution of black-box models, finally, we get the entire based-RMLP traffic matrix tracking algorithm. Compare with existing delay tomography, RMLP method don't depend on the traffic prior model and improve the accuracy of estimates, stability and real-time demand obviously. Meanwhile, by using parallel algorithms and network distributed algorithms, we will further improve the real-time of solution.
4. Research on network topology discovery
By introducing recursive unsupervised learning of Gaussian Mixture Models (GMM) to topology discovery approach, this thesis proposed a Fast Hierarchical Topology Estimation (FHTE) approach. By improving similarity metrics clustering process in network topology estimation, our proposed method evidently reducing the computation time under the same estimation accuracy circumstances compare to the Hierarchical Topology Estimation (HTE) approach.
5. Research on network traffic classification
This thesis propose a GMM-based semi-supervised classification method. Compare with supervised classification method, our approach takes advantage of only a few labeled data (about 10%) to identify different internet applications. Since obtaining the labeled instances is often difficult, expensive, or time consuming, the approach is particularly well suite for practical application and is more real-time. Semi-supervised classification method obtain more accurate classification compare to unsupervised classification method since it use a small amount of labeled instances to help us identify which Gaussian belongs to some class. Meanwhile, by analysis of semi-supervised classification method factors and their impact on classification performance, an optimum configuration is achieved for the GMM-based semi-supervised classification system, which further improve the classification performance of our approach.
引文
[1]张宏莉，方滨兴，胡铭曾，等．Internet测量与分析综述．软件学报，2003，14(1)：110-116
[2]黄志尧，金宁德，李海青．层析成像技术在多相流检测中的应用．化学反应工程与工艺，1996，12(4)：395-405
[3]胥颐，刘福田，刘建华，等．中国大陆西北造山带及其毗邻盆地的地震层析成像．中国科学D辑，2000，30(2)：113-122
[4]许有生，蒋伟荣，赵富金，等．外伤性脾破裂的CT诊断(附38例分析)．临床放射学杂志，1998，17(3)：153-155
[5]M.Coates A.Hero,R.Nowak,and B.Yu.Internet tomography.IEEE Signal Process.Mag,2002,19(3):47-65
[6]R.Castro,M.J.Coates,G.Liang,R.Nowak and B.Yu.Network tomography:recent developments.Statistical Science,2004,19(3):499-517
[7]Y.Vardi.Network tomography:Estimating source-destination traffic intensities from link data.J.Amer.Stat.Assoc,1996,91(433):365-377
[8]M.Coates and R.Nowak.Network tomography for internal delay estimation.in Proc.IEEE Int.Conf.Acoust,Speech,and Signal Proc,2002:3409-3412
[9]Mark J.Coates,Robert D.Nowak.Sequential Monte Carlo Inference of Internal Delays in Nonstationary Data Networks.IEEE Transactions On Signal Processing,2002,50(2):366-376
[10]M.F.Shih and A.O.Hero.Unicast-based inference of network link delay distributions using mixed finite mixture models.IEEE Transactions On Signal Processing,Special Issue on Signal Processing in Networking,2003,51 (9):2219-2228
[11]Y.Tsang,M.J.Coates and R.Nowak.Network delay tomography.IEEE Transactions On Signal Processing,special issue on Signal Processing in Networking,2003,51(8):2125-2136
[12]Xia,Y.Tse,D.Inference of Link Delay in Communication Networks.IEEE Journal on Selected Areas in Communications,2006,24(12):2235-2248
[13]E.Lawrence,G.Michailidis,and V.Nair.Network delay tomography using flexicast experiments.Journal of the Royal Statistical Society:Series B (Statistical Methodology),2006, 68(29):785-813
[14]V.Arya,N.G.Duffield,D.Veitch.Temporal Delay Tomography.IEEE Infocom 2008,Phoenix,AZ,USA,June 12-16,2008.accepted for publication Nov.2007
[15]F.Lo Presti,N.G.Duffield,J.Horowitz,and D.Towsley.Multicast-based inference of network internal delay distributions.IEEE/ACM Transaction on Networking,2002,10(6):761-775.
[16]G.Liang and B.Yu.Maximum pseudo likelihood estimation in network tomography.IEEE Transactions On Signal Processing.2003,51 (8):2043-2053
[17]M.Coates and R.Nowak.Networks for networks:Internet analysis using Bayesian graphical models.in Proc.IEEE Neural Network for Signal Processing Workshop,vol.2.Sydney,Australia,Dec.2000,755-764
[18]N.Duffield and F.Lo Presti.Multicast inference of packet delay variance at interior network links.in Proc.IEEE INFOCOM 2000,3:1351-1360
[19]K.Lai and M.Baker.Measuring link bandwidths using a deterministic model of packet delay.in Proc.ACM SIGCOMM,Stockholm,Sweden,Aug.2000,283-294
[20]S.Ratnasamy and S.McCanne.Inference of multicast routing trees and bottleneck bandwidths using end-to-end measurements.in Proc.IEEE INFOCOM,New York,Mar.1999,1:353-360
[21]V.Paxson.End-to-end Internet packet dynamics.IEEE/ACM Transactions On Networking,1999,7:277-292
[22]M.J.Coates and R.Nowak.Network loss inference using unicast end-to-end measurement.in Proc.of ITC Conference on IP Traffic Modelling and Management,Monterey,CA,Sept.2000,28:1-9
[23]Nick Duffield,Francesco Lo Presti,Vern Paxson,and Don Towsley.Network Loss Tomography Using Striped Unicast.Probes.IEEE/ACM Transactions On Networking,2006,14(4):697-710
[24]N.G.Duffield,F.Lo Presti,V.Paxson,D.Towsley.Inferring link loss using striped unicast probes.in Proc.IEEE Infocom 2001,Anchorage,Alaska,2001,22-26
[25]T.Bu,N.Duffield,F.Lo Presti,D.Towsley.Network tomography on general topologies.Proceedings ACM Sigmetrics 2002,Marina Del Rey,CA,June 15-19,2002,21-30
[26]Xi,B,Michailidis,G.Andnair.Least squares estimates of network link loss probabilities using end-to-end multicast measurements.In Proceedings of the 37th Conference on Information Sciences and Systems.2003,27-31
[27]R.C(?)ceres,N.G.Duffield,J.Horowitz,and D.Towsley.Multicast-based inference of network internal loss characteristics.IEEE Transactions On Inf.Theory,1999,45(7):2462-2480
[28]N.G.Duffield,J.Horowitz,D.Towsley,W.Wei,and T.Friedman.Multicast-based loss inference with missing data.IEEE J.Sel.Areas Commun,vol.2002,20(4):700-713
[29]Y.Tsang,M.Coates,and R.Nowak.Passive network tomography using EM algorithms.in Proc.2001 IEEE Int.Conf.Acoust.,Speech,and Signal Processing,vol3:1469-1472
[30]A.B.Downey.Using pathchar to estimate Intemet link characteristics.in Proc.ACM SIGCOMM'99,1999,241-250
[31]V.Padmanabhan,L.Qiu,and H.Wang.Server-based inference of Internet link lossiness.in IEEE INFOCOM,2003,1:145-155
[32]V.N.Padmanabhan,L.Qiu,and H.J.Wang.Passive network tomography using bayesian inference.ACM SIGCOMM Internet Measurement Workshop.Nov.2002,93-94
[33]Dong Guo Xiaodong Wang Bayesian inference of network loss and delay characteristics with applications to TCP performance prediction.IEEE Transactions On Signal Processing,2003,51 (8):2205-2218
[34]Medina A,Fraleigh C,Taft N,Bhattacharyya S,Diot C.A taxonomy of IP traffic matrices.In:Proc.Of the SPIE Scalability and Traffic Control in IP Networks Ⅱ.SPIE,Boston:SPIE Publishers,2002,4868:200213
[35]Vardi Y.Network Tomography:.Estimating source-destination traffic intensities from link data.Journal of the American Statistical Association,1996,91 (433):365-377
[36]Kowalski J,Warfield B.Modeling traffic demand between nodes in a telecommunications network.In:Proc.of the Australian Telecommunications and Networks Conf.(ATNAC'95).Sydney:IEEE Press,1995,705-710
[37]Zhang Y,Roughan M,Duffield N.Greenberg.Fast accurate computation of large-scale IP traffic matrices from link loads.ACM Sigmetrics Performance Evaluation Review,2003,31(1):206-217
[38]Medina A,Taft N,Salamatian K,Bhattacharyya S,Diot C.Traffic matrix estimation:existing techniques and new directions.In:Proc.of the ACM SIGCOMM 2002 on Applications,Technologies,Architectures,and Protocols for Computer Communications.Pittsburgh:ACM Press,2002,161-174
[39]Medina A,Salamatian K,Taft N,Matta Ⅰ,Diot C.A two-step statistical approach for inferring network traffic demands.Technical Report,BUCS-2004-011,Boston:Computer Science,Boston University,2004.
[40]Gunnar A,Johaasson M,Telkarnp T.Traffic matrix estimation on a large IP backbone:A comparison on real data.In Proc.of the ACM SIGCOMM Internet Measurement Conf.(IMC).Sicily:ACM Press,2004,149-160
[41]Zhang Y,Roughan M,Lurid C,Donoho D.An information-theoretic approach to traffic matrix estimation.In Proc.of the ACM SIGCOMM 2003,on Applications,Technologies,Architectures,and Protocols for Computer Communications.Karlsruhe:ACM Press,2003,301-312
[42]Tebaldi C,West M.Bayesian inference on network traffic using link count data.Journal of the American Statistical Association,1998,93(442):557-576
[43]Vaton S,Gravey A.Network Tomography:An iterative Bayesian analysis.In:Charzinski J,Lehnert R,TranGia P,eds.Proc.Of the 18th Int'l Teletraffic Congress (ITC).Berlin,2003,261-270
[44]Cao J,Davis D,Wiel SV,Yu B.Time-Varying network Tomography:Router link data.Journal of the American Statistical Association,2000,95(452):1063-1075
[45]Cao J,Vaader Wiel S,Yu B,Zhu Z.A scalable method for estimating network traffic matrices from link counts.Mobile Computing and Networking,2000,200-209
[46]Liang G,Yu B.Maximum pseudo likelihood estimation in network Tomography.IEEE Transactions On Signal Processing,2003,51(8):2043-2053
[47]Juva I,Vaton S,Virtamo J.Quick traffic matrix estimation based on link count covariances.In:Proc.of the IEEE Int'l Conf.on Communications (ICC).Istanbul:IEEE Communications Society,2006,603-608
[48]Eum S,Murphy J,Harris RJ.A fast accurate LP approach for traffic matrix estimation.In:Proc.of the 19th Int'l Teletraffic Congress (ITC19).Beijing:Beijing University of Posts and Telecommunications Press,2005,243-252
[49]Soule A,Nucci A,Cruz R,Leonardi E,Taft N.How to identify and estimate the largest traffic matrix elements in a dynamic environment.ACM Sigmetrics Performance Evaluation Review,2004,32(1):73-84
[50]Soule A,Lakhina A,Taft N,Papagiannaki K,Salamatian K,Nucci A,Crovella M,Diot C.Traffic matrices:Balancing measurements,inference and modeling.ACM Sigrnetrics Performance Evaluation Review,2005,33(1):362-373
[51]Erramill V,Crovella M,Taft N.An independent-connection model for traffic matrices.In:Almeida JM,Almeida VAF,Barford P,eds.Proc.of the ACM SIGCOMM Internet Measurement Conf.(IMC).Rio de Janeriro:ACM Press,2006,251-256
[52]Edoardo Airoldi and Christos Faloutsos.Recovering Latent Time-Series from their Observed Sums:Network Tomography with Particle Filters.ACM SIGKDD Conference,2004,30-39
[53]Vaton S,Bedo JS.Network traffic matrix:How can one learn the prior distributions from the link counts only?.In:Gelfand S,ed.Proc.of the IEEE Int'l Conf.on Communications (ICC).Parris:IEEE Communications Society,2004,2138-2142
[54]Zhang Y,Roughan M,Lund C,Donoho DL.Estimating point-to-point and point-to-multipoint traffic matrices:An information-theoretic approach.Networking.IEEE/ACM Transactions On Networking,2005,13(5):947-960
[55]Dempster AP,Laird NM,Rubin DB.Maximum likelihood from incomplete data via the EM algorithm (with discussion).Journal of the Royal Statistical Society (Series B),1977,39(1):1-38
[56]Soule A,Nucci A,Cruz R,Leopardi E,Taft N.Estimating dynamic traffic matrices by using viable routing changes.IEEE/ACM Transactions On Networking,2007,15(3):485-498
[57]Papagiannaki K,Taft N,Lakhina A.A distributed approach to measure IP traffic mafices.In:Lombardo A,Kurose JF,eds.Proc.of the ACM SIGCOMM Internet Measurement Conf (IMC).Taormina:ACM Press,2004,161-174
[58]Lakhina A,Papagiannaki K,Crovella M,Diot C,Kolaczyk ED,Taft N.Structural analysis of network traffic flows.In:Coffman E,ed.Proc.of the ACM Sigmetrics/Performance.New York:ACM Press,2004,61-72
[59]Soule A,Salamatian K,Taft N.Traffic matrix tracking using Kalman filters.ACM Sigmetrics Performance Evaluation Review,2005,33(3):24-31
[60]Rahman MM,Saha S,Chengan U,Alfa AS.IP traffic matrix estimation methods:Comparisons and improvements.In:Proc.of the IEEE Int'l Conf on Communications (ICC).Istanbul:IEEE Communications Society,2006,90-96
[61]Shioda S,Ohtani K.Proposals on the source-destination traffic matrix estimation for IP-based VPNs.Proc.of IEEE GLOBECOM,2005,1 (28):348-353
[62]Liang G,Taft N,Yu B.A fast lightweight approach to origin-destination IP traffic estimation using partial measurements.IEEE/ACM Transactions On Networking (TON),2006,14(S1):2634-2648
[63]Zhao GF,Wang LJ,Tang H,Cheng DJ.Dynamic traffic matrix measurement model of network based IP/MPLS.Journal of China Institute of Communications,2003,24(10):145-152
[64]赵国锋，王灵矫，唐红，程代杰．基于IP／MPLS网络的动态业务流量矩阵测量模型．通信学报，2003，24(10)：145-152
[65]周静静，杨家海，杨扬，张辉．流量矩阵估算的研究．软件学报，2007，18(11)：2669-2682
[66]N.G.Duffield,J.Horowitz,F.Lo Presti,D.Towsley.Multicast topology inference from end-to-end measurements.ITC Seminar on IP traffic,Measurement and Modelling,Monterey,CA,Sep.2000,1-10
[67]N.G.Duffield,J.Horowitz,F.Lo Presti,D.Towsley.Multicast topology inference from measured end-to-end loss.IEEE Transactions On Info.Theory,2002,48(1):26-45
[68]N.G.Duffield,J.Horowitz,F.Lo Presti.Adaptive multicast topology inference.IEEE Infocom 2001,Anchorage,Alaska,2001,3:1636-1645
[69]R.Castro,M.Coates,R.Nowak.Likelihood based hierarchical clustering.IEEE Transactions On Signal Processing,2004,52(8):2308-2321
[70]M.Shih,A.Hero.Network topology discovery using finite mixture models.IEEE Int'l Conf.on Acoustics,Speech,and Signal Processing (ICASSP) 2003,Montreal,Canada,vol 3:433-436
[71]A.Bestavros,J.Byers,K.Harfoush.Inference and labeling of metric-induced network topologies.IEEE Infocom 2002,New York,NY.vol.2:628-637
[72]M.Coates,R.Castro,R.Nowak.Maximum likelihood network topology identification from edge-based unicast measurements.ACM Sigmetric,Marina Del Rey,CA.2002,11-20
[73]M.Rabbat,R.Nowak,M.Coates.Network Tomography and the Identification of Shared Infrastructure.Asilomar Conference on Signals,Systems,and Computers,Pacific Grove,CA.2002,1:34-38
[74]M.Coates,M.Rabbat,R.Nowak.Merging logical topologies using end-to-end measurements.ACM Internet Measurement Conference,Miami,FL.,2003,192-203
[75]M.Rabbat,R.Nowak,M.Coates.Multiple Source,Multiple Destination Network Tomography.IEEE Infocom,Hong Kong,2004,3:628-1639
[76]M.Shih and A.Hero.Topology Discovery on Unicast Networks:A Hierarchical Approach Based on End-to-End Measurements.CSPL Technical Report TR-357,Department of EECS,University of Michigan March 2005
[77]Duffield N G,Francesco Lo Presti.Network Tomography from Measured End-to-End Delay Covariance.IEEE/ACM Transactions On Networking,2004,12(6):978-992
[78]Shih M F,Hero A O.Unicast inference of network link delay distributions from edge measurements.In Proc.IEEE Int.Conf.Acoust.,Speech,and Signal,2002,2:1305-1308
[79]G.V.Puskorius and L.A.Feldkamp.Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks.IEEE Transactions On Neural Networks,1994,5(2):279-297
[80]L.A.Feldkamp,G.V.Puskorius.Training Controller for robustness:multi-stream DEKF.IEEE International Conference on Neural Network.June 1994,4:2377-2382
[81]P.J.Werbos.Backpropagation through time:What it does and how to do it.Proc.IEEE,1990,78:1550-1560
[82]L.A.Feldkamp and G.V.Puskorius.A signal processing framework based on dynamic neural networks with application to problems in adaptation filtering and classification.Proceedings of the IEEE,1998,86(11):2259-2277
[83]G.V.Puskorius and L.A.Feldkamp.De coupled extended Kalman filter training of feedforward layered networks.in Proc.Int.Joint Conf.Neural Networks,Seattle,WA,1991,vol 1:771-777
[84]Li,S.Wind power prediction using recurrent multilayer perceptron neural networks.Power Engineering Society General Meeting,2003,IEEE vol 4:13-17
[85]Danil V.Prokhorow,L.A.Feldkamp and Ivan Yu.Tyukin.Adaptive behavior with fixed weights in RNN:An overview.IEEE International Joint Conference on Neural Network.Honolulu,2002,3:2018-2022
[86]P.Harihara.Real-time implementations of a neural networks-based motor speed filter using a digital signal processor.M.S.thesis,TX:Texas A&M Univ,College Station,2002,81-87
[87]Figueiredo M.,Jain,A.K.Unsupervised learning of finite mixture models.IEEE Transactions On Pattern Analysis and Machine Intelligence,2002,24(3):381-396
[88]Z.Zivkovic and F.van der Heijden.Recursive Unsupervised Learning of Finite Mixture Models.IEEE Transactions On PAMI,2004,26(5):651-656
[89]Wei Wei,Sharad Jaiswal,Jim Kurose,and Don Towsley.Identifying 802.11 traffic from passive measurements using iterative bayesian inference.IEEE INFOCOM,2006,1-12
[90]Wei Wei,Bing Wang,Chun Zhang,James F.Kurose,Donald F.Towsley.Classification of access network types:Ethernet wireless LAN,ADSL,cable modem or dialup? IEEE Infocom 2005,1060-1071
[91]Saddi W,Azzouna NB,Guillemin F.IP traffic classification via blind source separation based on Jacobi algorithm.In:Freire MM,Chemouil P,Lorenz P,eds.Proc.of the ECUMN 2004.LNCS Heidelberg:Springer-Verlag,2004,3262:287-296
[92]A.Soule,K.Salamatian,N.Taft,R.Emilion,and K.Papagiannaki.Flow Classification by Histograms or How to Go on Safari in the Internet.In Proceedings of ACM Sigmetrics,New York,NY,June 2004,32(1):49-60
[93]Andrew W.Moore,Denis Zuev.Internet traffic classification using Bayesian analysis techniques.ACM Sigmetrics Performance Evaluation Review archive 2005,33(1):50-60
[94]A.W.Moore and D.Papagiannaki.Toward the Accurate Identification of Network Applications.In Proceedings of the Sixth Passive and Active Measurement Workshop (PAM 2005),Springer-Verlag LNCS,March 2005,3431:41-54
[95]A.McGregor,M.Hall,P.Lorier,and J.Brunskill.Flow Clustering Using Machine Learning Techniques.In Proceedings of the Fifth Passive and Active Measurement Workshop (PAM 2004),Springer-Verlag LNCS,April 2004,3015:205-214
[96]Sebastian Zander,Thuy Nguyen,and Grenville Armitage.Automated Traffic Classification and Application Identification using Machine Learning.In 30th Annual IEEE Conference on Local Computer Networks,Sydney,Australia,November 2005,220-227
[97]F.Hern'andez-Carnpos,A.B.Nobel,F.D.Smith,and K.Jeffay.Statistical clustering of internet communication patterns.In Proceedings of the 35th Symposium on the Interface of Computing Science and Statistics,Computing Science and Statistics,2003,1-16
[98]Jeffrey Erman,Martin Arlitt,Anirban Mahanti.Traffic Classification Using Clustering Algorithms.SIGCOMM Workshops September,Pisa,Italy,2006,281-286
[99]M.Roughan,S.Sen,O.Spatscheck,and N.Duffield.Class-of-Service Mapping for QoS:A statistical signature-based approach to IP traffic classification.In ACM SIGCOMM Internct Measurement Conference,Taormina,Sicily,Italy,2004,135-148
[100]Tom Auld,Andrew W.Moore.Bayesian Neural Networks for Internet Traffic Classification,IEEE Transactions On Neural Network,2007,18(1):223-239
[101]Zhu X.Semi-supervised learning literature survey.Technical report TR-1530,Madison:Department of Computer Science,University of Wisconsin-Madison,2005
[102]Dy JG,Brodley CE.Feature selection for unsupervised learning.J Mach Learning Res 2004;5:845-89
[103]Reynolds DA,Durr RB,Quatieri TF.Speaker verification using adapted gaussian mixture models.Digital Signal Process 2000,10:19-41
[104]Yonghong H,Englehart KB,Hudgins B,Chan ADC.A gaussian mixture model based classification scheme for myoelectric control of powered upper limb prostheses.IEEE Trans Biomed Eng 2005,52:1801-1811
[105]Aiyou Chen,Jin Cao,Tian Bu.Network Tomography:Identifiability and Fourier Domain Estimation.IEEE Infocom 2007,1875-1883
[106]Coleman,T.F.and Y.Li.A Reflective Newton Method for Minimizing a Quadratic Function Subject to Bounds on some of the Variables.SIAM Journal on Optimization,1996,6(4):1040-1058
[107]Rivals and L.Personnaz.Black-Box Modeling with State-Space Neural Networks.in Neural Adaptive Control Technology,Edited by R.Zbikowski and K.J.Hunt.World Scientific,1996,87-99
[108]Rabbat,M.G.Treichler,J.R.Wood,S.L.Larimore,M.G.Understanding the topology of a telephone network via internally-sensed network tomography.IEEE International Conference on Acoustics,Speech,and Signal Processing,2005,vol3:977-980
[109]Aouad,H.Tohme,S.Using network tomography for dynamic path adaptation.IEEE Wireless Communications and Networking Conference,2005,vol 3:1743-1748
[110]Yongyi Mao,Kschischang F.R.,Baochun Li,Pasupathy S.A factor graph approach to link loss monitoring in wireless sensor networks.IEEE Journal on Selected Areas in.Communications,2005,23(4):820-829
[111]F.Ricciato,F.Vacirca,W.Fleischer,J.Motz,M.Rupp.Passive tomography ofa 3G network:challenges and opportunities.IEEE Infocom 2006,vol 1:116-122
[112]N Spring,D Wetherall,T Anderson.Reverse-engineering the Internet.ACM SIGCOMM Computer Communication Review,2004,34(1):3-8
[113]N.G.Duffield.Simple network performance tomography.in Proc.ACM SIGCOMM Internet Measurement Conf.,Miami Beach,F1,Oct.2003,210-215
[2]黄志尧，金宁德，李海青．层析成像技术在多相流检测中的应用．化学反应工程与工艺，1996，12(4)：395-405
[3]胥颐，刘福田，刘建华，等．中国大陆西北造山带及其毗邻盆地的地震层析成像．中国科学D辑，2000，30(2)：113-122
[4]许有生，蒋伟荣，赵富金，等．外伤性脾破裂的CT诊断(附38例分析)．临床放射学杂志，1998，17(3)：153-155
[5]M.Coates A.Hero,R.Nowak,and B.Yu.Internet tomography.IEEE Signal Process.Mag,2002,19(3):47-65
[6]R.Castro,M.J.Coates,G.Liang,R.Nowak and B.Yu.Network tomography:recent developments.Statistical Science,2004,19(3):499-517
[7]Y.Vardi.Network tomography:Estimating source-destination traffic intensities from link data.J.Amer.Stat.Assoc,1996,91(433):365-377
[8]M.Coates and R.Nowak.Network tomography for internal delay estimation.in Proc.IEEE Int.Conf.Acoust,Speech,and Signal Proc,2002:3409-3412
[9]Mark J.Coates,Robert D.Nowak.Sequential Monte Carlo Inference of Internal Delays in Nonstationary Data Networks.IEEE Transactions On Signal Processing,2002,50(2):366-376
[10]M.F.Shih and A.O.Hero.Unicast-based inference of network link delay distributions using mixed finite mixture models.IEEE Transactions On Signal Processing,Special Issue on Signal Processing in Networking,2003,51 (9):2219-2228
[11]Y.Tsang,M.J.Coates and R.Nowak.Network delay tomography.IEEE Transactions On Signal Processing,special issue on Signal Processing in Networking,2003,51(8):2125-2136
[12]Xia,Y.Tse,D.Inference of Link Delay in Communication Networks.IEEE Journal on Selected Areas in Communications,2006,24(12):2235-2248
[13]E.Lawrence,G.Michailidis,and V.Nair.Network delay tomography using flexicast experiments.Journal of the Royal Statistical Society:Series B (Statistical Methodology),2006, 68(29):785-813
[14]V.Arya,N.G.Duffield,D.Veitch.Temporal Delay Tomography.IEEE Infocom 2008,Phoenix,AZ,USA,June 12-16,2008.accepted for publication Nov.2007
[15]F.Lo Presti,N.G.Duffield,J.Horowitz,and D.Towsley.Multicast-based inference of network internal delay distributions.IEEE/ACM Transaction on Networking,2002,10(6):761-775.
[16]G.Liang and B.Yu.Maximum pseudo likelihood estimation in network tomography.IEEE Transactions On Signal Processing.2003,51 (8):2043-2053
[17]M.Coates and R.Nowak.Networks for networks:Internet analysis using Bayesian graphical models.in Proc.IEEE Neural Network for Signal Processing Workshop,vol.2.Sydney,Australia,Dec.2000,755-764
[18]N.Duffield and F.Lo Presti.Multicast inference of packet delay variance at interior network links.in Proc.IEEE INFOCOM 2000,3:1351-1360
[19]K.Lai and M.Baker.Measuring link bandwidths using a deterministic model of packet delay.in Proc.ACM SIGCOMM,Stockholm,Sweden,Aug.2000,283-294
[20]S.Ratnasamy and S.McCanne.Inference of multicast routing trees and bottleneck bandwidths using end-to-end measurements.in Proc.IEEE INFOCOM,New York,Mar.1999,1:353-360
[21]V.Paxson.End-to-end Internet packet dynamics.IEEE/ACM Transactions On Networking,1999,7:277-292
[22]M.J.Coates and R.Nowak.Network loss inference using unicast end-to-end measurement.in Proc.of ITC Conference on IP Traffic Modelling and Management,Monterey,CA,Sept.2000,28:1-9
[23]Nick Duffield,Francesco Lo Presti,Vern Paxson,and Don Towsley.Network Loss Tomography Using Striped Unicast.Probes.IEEE/ACM Transactions On Networking,2006,14(4):697-710
[24]N.G.Duffield,F.Lo Presti,V.Paxson,D.Towsley.Inferring link loss using striped unicast probes.in Proc.IEEE Infocom 2001,Anchorage,Alaska,2001,22-26
[25]T.Bu,N.Duffield,F.Lo Presti,D.Towsley.Network tomography on general topologies.Proceedings ACM Sigmetrics 2002,Marina Del Rey,CA,June 15-19,2002,21-30
[26]Xi,B,Michailidis,G.Andnair.Least squares estimates of network link loss probabilities using end-to-end multicast measurements.In Proceedings of the 37th Conference on Information Sciences and Systems.2003,27-31
[27]R.C(?)ceres,N.G.Duffield,J.Horowitz,and D.Towsley.Multicast-based inference of network internal loss characteristics.IEEE Transactions On Inf.Theory,1999,45(7):2462-2480
[28]N.G.Duffield,J.Horowitz,D.Towsley,W.Wei,and T.Friedman.Multicast-based loss inference with missing data.IEEE J.Sel.Areas Commun,vol.2002,20(4):700-713
[29]Y.Tsang,M.Coates,and R.Nowak.Passive network tomography using EM algorithms.in Proc.2001 IEEE Int.Conf.Acoust.,Speech,and Signal Processing,vol3:1469-1472
[30]A.B.Downey.Using pathchar to estimate Intemet link characteristics.in Proc.ACM SIGCOMM'99,1999,241-250
[31]V.Padmanabhan,L.Qiu,and H.Wang.Server-based inference of Internet link lossiness.in IEEE INFOCOM,2003,1:145-155
[32]V.N.Padmanabhan,L.Qiu,and H.J.Wang.Passive network tomography using bayesian inference.ACM SIGCOMM Internet Measurement Workshop.Nov.2002,93-94
[33]Dong Guo Xiaodong Wang Bayesian inference of network loss and delay characteristics with applications to TCP performance prediction.IEEE Transactions On Signal Processing,2003,51 (8):2205-2218
[34]Medina A,Fraleigh C,Taft N,Bhattacharyya S,Diot C.A taxonomy of IP traffic matrices.In:Proc.Of the SPIE Scalability and Traffic Control in IP Networks Ⅱ.SPIE,Boston:SPIE Publishers,2002,4868:200213
[35]Vardi Y.Network Tomography:.Estimating source-destination traffic intensities from link data.Journal of the American Statistical Association,1996,91 (433):365-377
[36]Kowalski J,Warfield B.Modeling traffic demand between nodes in a telecommunications network.In:Proc.of the Australian Telecommunications and Networks Conf.(ATNAC'95).Sydney:IEEE Press,1995,705-710
[37]Zhang Y,Roughan M,Duffield N.Greenberg.Fast accurate computation of large-scale IP traffic matrices from link loads.ACM Sigmetrics Performance Evaluation Review,2003,31(1):206-217
[38]Medina A,Taft N,Salamatian K,Bhattacharyya S,Diot C.Traffic matrix estimation:existing techniques and new directions.In:Proc.of the ACM SIGCOMM 2002 on Applications,Technologies,Architectures,and Protocols for Computer Communications.Pittsburgh:ACM Press,2002,161-174
[39]Medina A,Salamatian K,Taft N,Matta Ⅰ,Diot C.A two-step statistical approach for inferring network traffic demands.Technical Report,BUCS-2004-011,Boston:Computer Science,Boston University,2004.
[40]Gunnar A,Johaasson M,Telkarnp T.Traffic matrix estimation on a large IP backbone:A comparison on real data.In Proc.of the ACM SIGCOMM Internet Measurement Conf.(IMC).Sicily:ACM Press,2004,149-160
[41]Zhang Y,Roughan M,Lurid C,Donoho D.An information-theoretic approach to traffic matrix estimation.In Proc.of the ACM SIGCOMM 2003,on Applications,Technologies,Architectures,and Protocols for Computer Communications.Karlsruhe:ACM Press,2003,301-312
[42]Tebaldi C,West M.Bayesian inference on network traffic using link count data.Journal of the American Statistical Association,1998,93(442):557-576
[43]Vaton S,Gravey A.Network Tomography:An iterative Bayesian analysis.In:Charzinski J,Lehnert R,TranGia P,eds.Proc.Of the 18th Int'l Teletraffic Congress (ITC).Berlin,2003,261-270
[44]Cao J,Davis D,Wiel SV,Yu B.Time-Varying network Tomography:Router link data.Journal of the American Statistical Association,2000,95(452):1063-1075
[45]Cao J,Vaader Wiel S,Yu B,Zhu Z.A scalable method for estimating network traffic matrices from link counts.Mobile Computing and Networking,2000,200-209
[46]Liang G,Yu B.Maximum pseudo likelihood estimation in network Tomography.IEEE Transactions On Signal Processing,2003,51(8):2043-2053
[47]Juva I,Vaton S,Virtamo J.Quick traffic matrix estimation based on link count covariances.In:Proc.of the IEEE Int'l Conf.on Communications (ICC).Istanbul:IEEE Communications Society,2006,603-608
[48]Eum S,Murphy J,Harris RJ.A fast accurate LP approach for traffic matrix estimation.In:Proc.of the 19th Int'l Teletraffic Congress (ITC19).Beijing:Beijing University of Posts and Telecommunications Press,2005,243-252
[49]Soule A,Nucci A,Cruz R,Leonardi E,Taft N.How to identify and estimate the largest traffic matrix elements in a dynamic environment.ACM Sigmetrics Performance Evaluation Review,2004,32(1):73-84
[50]Soule A,Lakhina A,Taft N,Papagiannaki K,Salamatian K,Nucci A,Crovella M,Diot C.Traffic matrices:Balancing measurements,inference and modeling.ACM Sigrnetrics Performance Evaluation Review,2005,33(1):362-373
[51]Erramill V,Crovella M,Taft N.An independent-connection model for traffic matrices.In:Almeida JM,Almeida VAF,Barford P,eds.Proc.of the ACM SIGCOMM Internet Measurement Conf.(IMC).Rio de Janeriro:ACM Press,2006,251-256
[52]Edoardo Airoldi and Christos Faloutsos.Recovering Latent Time-Series from their Observed Sums:Network Tomography with Particle Filters.ACM SIGKDD Conference,2004,30-39
[53]Vaton S,Bedo JS.Network traffic matrix:How can one learn the prior distributions from the link counts only?.In:Gelfand S,ed.Proc.of the IEEE Int'l Conf.on Communications (ICC).Parris:IEEE Communications Society,2004,2138-2142
[54]Zhang Y,Roughan M,Lund C,Donoho DL.Estimating point-to-point and point-to-multipoint traffic matrices:An information-theoretic approach.Networking.IEEE/ACM Transactions On Networking,2005,13(5):947-960
[55]Dempster AP,Laird NM,Rubin DB.Maximum likelihood from incomplete data via the EM algorithm (with discussion).Journal of the Royal Statistical Society (Series B),1977,39(1):1-38
[56]Soule A,Nucci A,Cruz R,Leopardi E,Taft N.Estimating dynamic traffic matrices by using viable routing changes.IEEE/ACM Transactions On Networking,2007,15(3):485-498
[57]Papagiannaki K,Taft N,Lakhina A.A distributed approach to measure IP traffic mafices.In:Lombardo A,Kurose JF,eds.Proc.of the ACM SIGCOMM Internet Measurement Conf (IMC).Taormina:ACM Press,2004,161-174
[58]Lakhina A,Papagiannaki K,Crovella M,Diot C,Kolaczyk ED,Taft N.Structural analysis of network traffic flows.In:Coffman E,ed.Proc.of the ACM Sigmetrics/Performance.New York:ACM Press,2004,61-72
[59]Soule A,Salamatian K,Taft N.Traffic matrix tracking using Kalman filters.ACM Sigmetrics Performance Evaluation Review,2005,33(3):24-31
[60]Rahman MM,Saha S,Chengan U,Alfa AS.IP traffic matrix estimation methods:Comparisons and improvements.In:Proc.of the IEEE Int'l Conf on Communications (ICC).Istanbul:IEEE Communications Society,2006,90-96
[61]Shioda S,Ohtani K.Proposals on the source-destination traffic matrix estimation for IP-based VPNs.Proc.of IEEE GLOBECOM,2005,1 (28):348-353
[62]Liang G,Taft N,Yu B.A fast lightweight approach to origin-destination IP traffic estimation using partial measurements.IEEE/ACM Transactions On Networking (TON),2006,14(S1):2634-2648
[63]Zhao GF,Wang LJ,Tang H,Cheng DJ.Dynamic traffic matrix measurement model of network based IP/MPLS.Journal of China Institute of Communications,2003,24(10):145-152
[64]赵国锋，王灵矫，唐红，程代杰．基于IP／MPLS网络的动态业务流量矩阵测量模型．通信学报，2003，24(10)：145-152
[65]周静静，杨家海，杨扬，张辉．流量矩阵估算的研究．软件学报，2007，18(11)：2669-2682
[66]N.G.Duffield,J.Horowitz,F.Lo Presti,D.Towsley.Multicast topology inference from end-to-end measurements.ITC Seminar on IP traffic,Measurement and Modelling,Monterey,CA,Sep.2000,1-10
[67]N.G.Duffield,J.Horowitz,F.Lo Presti,D.Towsley.Multicast topology inference from measured end-to-end loss.IEEE Transactions On Info.Theory,2002,48(1):26-45
[68]N.G.Duffield,J.Horowitz,F.Lo Presti.Adaptive multicast topology inference.IEEE Infocom 2001,Anchorage,Alaska,2001,3:1636-1645
[69]R.Castro,M.Coates,R.Nowak.Likelihood based hierarchical clustering.IEEE Transactions On Signal Processing,2004,52(8):2308-2321
[70]M.Shih,A.Hero.Network topology discovery using finite mixture models.IEEE Int'l Conf.on Acoustics,Speech,and Signal Processing (ICASSP) 2003,Montreal,Canada,vol 3:433-436
[71]A.Bestavros,J.Byers,K.Harfoush.Inference and labeling of metric-induced network topologies.IEEE Infocom 2002,New York,NY.vol.2:628-637
[72]M.Coates,R.Castro,R.Nowak.Maximum likelihood network topology identification from edge-based unicast measurements.ACM Sigmetric,Marina Del Rey,CA.2002,11-20
[73]M.Rabbat,R.Nowak,M.Coates.Network Tomography and the Identification of Shared Infrastructure.Asilomar Conference on Signals,Systems,and Computers,Pacific Grove,CA.2002,1:34-38
[74]M.Coates,M.Rabbat,R.Nowak.Merging logical topologies using end-to-end measurements.ACM Internet Measurement Conference,Miami,FL.,2003,192-203
[75]M.Rabbat,R.Nowak,M.Coates.Multiple Source,Multiple Destination Network Tomography.IEEE Infocom,Hong Kong,2004,3:628-1639
[76]M.Shih and A.Hero.Topology Discovery on Unicast Networks:A Hierarchical Approach Based on End-to-End Measurements.CSPL Technical Report TR-357,Department of EECS,University of Michigan March 2005
[77]Duffield N G,Francesco Lo Presti.Network Tomography from Measured End-to-End Delay Covariance.IEEE/ACM Transactions On Networking,2004,12(6):978-992
[78]Shih M F,Hero A O.Unicast inference of network link delay distributions from edge measurements.In Proc.IEEE Int.Conf.Acoust.,Speech,and Signal,2002,2:1305-1308
[79]G.V.Puskorius and L.A.Feldkamp.Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks.IEEE Transactions On Neural Networks,1994,5(2):279-297
[80]L.A.Feldkamp,G.V.Puskorius.Training Controller for robustness:multi-stream DEKF.IEEE International Conference on Neural Network.June 1994,4:2377-2382
[81]P.J.Werbos.Backpropagation through time:What it does and how to do it.Proc.IEEE,1990,78:1550-1560
[82]L.A.Feldkamp and G.V.Puskorius.A signal processing framework based on dynamic neural networks with application to problems in adaptation filtering and classification.Proceedings of the IEEE,1998,86(11):2259-2277
[83]G.V.Puskorius and L.A.Feldkamp.De coupled extended Kalman filter training of feedforward layered networks.in Proc.Int.Joint Conf.Neural Networks,Seattle,WA,1991,vol 1:771-777
[84]Li,S.Wind power prediction using recurrent multilayer perceptron neural networks.Power Engineering Society General Meeting,2003,IEEE vol 4:13-17
[85]Danil V.Prokhorow,L.A.Feldkamp and Ivan Yu.Tyukin.Adaptive behavior with fixed weights in RNN:An overview.IEEE International Joint Conference on Neural Network.Honolulu,2002,3:2018-2022
[86]P.Harihara.Real-time implementations of a neural networks-based motor speed filter using a digital signal processor.M.S.thesis,TX:Texas A&M Univ,College Station,2002,81-87
[87]Figueiredo M.,Jain,A.K.Unsupervised learning of finite mixture models.IEEE Transactions On Pattern Analysis and Machine Intelligence,2002,24(3):381-396
[88]Z.Zivkovic and F.van der Heijden.Recursive Unsupervised Learning of Finite Mixture Models.IEEE Transactions On PAMI,2004,26(5):651-656
[89]Wei Wei,Sharad Jaiswal,Jim Kurose,and Don Towsley.Identifying 802.11 traffic from passive measurements using iterative bayesian inference.IEEE INFOCOM,2006,1-12
[90]Wei Wei,Bing Wang,Chun Zhang,James F.Kurose,Donald F.Towsley.Classification of access network types:Ethernet wireless LAN,ADSL,cable modem or dialup? IEEE Infocom 2005,1060-1071
[91]Saddi W,Azzouna NB,Guillemin F.IP traffic classification via blind source separation based on Jacobi algorithm.In:Freire MM,Chemouil P,Lorenz P,eds.Proc.of the ECUMN 2004.LNCS Heidelberg:Springer-Verlag,2004,3262:287-296
[92]A.Soule,K.Salamatian,N.Taft,R.Emilion,and K.Papagiannaki.Flow Classification by Histograms or How to Go on Safari in the Internet.In Proceedings of ACM Sigmetrics,New York,NY,June 2004,32(1):49-60
[93]Andrew W.Moore,Denis Zuev.Internet traffic classification using Bayesian analysis techniques.ACM Sigmetrics Performance Evaluation Review archive 2005,33(1):50-60
[94]A.W.Moore and D.Papagiannaki.Toward the Accurate Identification of Network Applications.In Proceedings of the Sixth Passive and Active Measurement Workshop (PAM 2005),Springer-Verlag LNCS,March 2005,3431:41-54
[95]A.McGregor,M.Hall,P.Lorier,and J.Brunskill.Flow Clustering Using Machine Learning Techniques.In Proceedings of the Fifth Passive and Active Measurement Workshop (PAM 2004),Springer-Verlag LNCS,April 2004,3015:205-214
[96]Sebastian Zander,Thuy Nguyen,and Grenville Armitage.Automated Traffic Classification and Application Identification using Machine Learning.In 30th Annual IEEE Conference on Local Computer Networks,Sydney,Australia,November 2005,220-227
[97]F.Hern'andez-Carnpos,A.B.Nobel,F.D.Smith,and K.Jeffay.Statistical clustering of internet communication patterns.In Proceedings of the 35th Symposium on the Interface of Computing Science and Statistics,Computing Science and Statistics,2003,1-16
[98]Jeffrey Erman,Martin Arlitt,Anirban Mahanti.Traffic Classification Using Clustering Algorithms.SIGCOMM Workshops September,Pisa,Italy,2006,281-286
[99]M.Roughan,S.Sen,O.Spatscheck,and N.Duffield.Class-of-Service Mapping for QoS:A statistical signature-based approach to IP traffic classification.In ACM SIGCOMM Internct Measurement Conference,Taormina,Sicily,Italy,2004,135-148
[100]Tom Auld,Andrew W.Moore.Bayesian Neural Networks for Internet Traffic Classification,IEEE Transactions On Neural Network,2007,18(1):223-239
[101]Zhu X.Semi-supervised learning literature survey.Technical report TR-1530,Madison:Department of Computer Science,University of Wisconsin-Madison,2005
[102]Dy JG,Brodley CE.Feature selection for unsupervised learning.J Mach Learning Res 2004;5:845-89
[103]Reynolds DA,Durr RB,Quatieri TF.Speaker verification using adapted gaussian mixture models.Digital Signal Process 2000,10:19-41
[104]Yonghong H,Englehart KB,Hudgins B,Chan ADC.A gaussian mixture model based classification scheme for myoelectric control of powered upper limb prostheses.IEEE Trans Biomed Eng 2005,52:1801-1811
[105]Aiyou Chen,Jin Cao,Tian Bu.Network Tomography:Identifiability and Fourier Domain Estimation.IEEE Infocom 2007,1875-1883
[106]Coleman,T.F.and Y.Li.A Reflective Newton Method for Minimizing a Quadratic Function Subject to Bounds on some of the Variables.SIAM Journal on Optimization,1996,6(4):1040-1058
[107]Rivals and L.Personnaz.Black-Box Modeling with State-Space Neural Networks.in Neural Adaptive Control Technology,Edited by R.Zbikowski and K.J.Hunt.World Scientific,1996,87-99
[108]Rabbat,M.G.Treichler,J.R.Wood,S.L.Larimore,M.G.Understanding the topology of a telephone network via internally-sensed network tomography.IEEE International Conference on Acoustics,Speech,and Signal Processing,2005,vol3:977-980
[109]Aouad,H.Tohme,S.Using network tomography for dynamic path adaptation.IEEE Wireless Communications and Networking Conference,2005,vol 3:1743-1748
[110]Yongyi Mao,Kschischang F.R.,Baochun Li,Pasupathy S.A factor graph approach to link loss monitoring in wireless sensor networks.IEEE Journal on Selected Areas in.Communications,2005,23(4):820-829
[111]F.Ricciato,F.Vacirca,W.Fleischer,J.Motz,M.Rupp.Passive tomography ofa 3G network:challenges and opportunities.IEEE Infocom 2006,vol 1:116-122
[112]N Spring,D Wetherall,T Anderson.Reverse-engineering the Internet.ACM SIGCOMM Computer Communication Review,2004,34(1):3-8
[113]N.G.Duffield.Simple network performance tomography.in Proc.ACM SIGCOMM Internet Measurement Conf.,Miami Beach,F1,Oct.2003,210-215