Path selection with preemption and re-routing control for multi-protocol label switching networks

Multi Protocol Label Switching (MPLS) networks enhance the services of conventional best-effort IP networks by providing end-to-end Quality of Service (QoS) guaranteed Label Switched Paths (LSP) between customer sites. The LSP has to be set up in advance before carrying the traffic. Contention for network resources may happen if many LSPs try to use a common network link with limited bandwidth. In this paper, we investigate the problem of providing services to high priority LSPs whereby existing LSPs with lower priority may be preempted. The consequent interruption of the services of preempted LSPs would detrimentally affect users’ perception on the QoS provided. Therefore, the preemption strategies may incorporate additional re-routing mechanisms to provide alternative paths for the LSPs which are to-be-preempted so that their services remain unaffected. A newly arrived high priority LSP in an MPLS network may find M possible paths between its source and destination. It may select the shortest path which may trigger preemption or choose a longer path which however utilizes more resources. We begin by formulating preemption strategies with global re-routing. Our investigations include the effects of routing of high priority LSPs on the shortest path and its alternative paths. We show that by persistently routing the high priority LSP on the shortest path, more preempted LSPs can be re-routed which would reduce the negative effects of preemption. However, as excessive re-routing may degrade the network performance as well, a re-routing control strategy is proposed to constrain the length of these re-routed paths. Finally, a decentralized preemption strategy with local re-routing is also presented to approximate the performance of the proposed strategy with significantly lower control overheads. Simulations show that with this approach, high priority LSPs can gain better access to network resources while simultaneously ensuring that, as compared to the existing preemption strategies, the network throughput and the ongoing connection services are not adversely affected.     

多协议标签交换网络中最小干扰路由算法的研究

在最小干扰路由(MIR)算法基础上,提出一种改进的MIR算法。该算法结合了基于多商品流理论的流量描述路由（PBR）算法将节点对带宽预测作为路由信息的优点,以及MIR算法的核心思想,给出了节点对的权值计算公式,并说明了如何在现有MIR算法中利用节点对的权值。改进后的算法能更好地适应实际网络路由需要,提高网络资源利用率和吞吐量。     

基于多协议层联合优化的移动性管理技术

针对下一代网络的通用移动性,从多协议层联合优化的角度提出一种可以有效融合多业务的通用移动性管理方案。该方案以网络层移动性支持协议（Mobile IP）及其微移动管理协议为基础,并在此基础上融合会话初始协议（SIP）,将网络层移动性管理与应用层移动性管理进行联合优化设计,同时考虑链路层的切换,减少功能性和信令数据信息的重复。     

Mobility label based network: Hierarchical mobility management and packet forwarding architecture

Scalability of the network layer mobility management solution is one of the most important requirements for the mobility control plane. Mobility Label Based Network (MLBN) is a new approach to the network layer mobility management problem that relies solely on MPLS to provide both macro- and micro-mobility for IPv4 and IPv6 mobile hosts and routers. This new approach does not rely on the existing IP mobility management protocols such as Mobile IP and is based on the combination of Multi-Protocol BGP (MP-BGP) and MPLS. In the context of the MLBN the scalable control plane should be capable of efficient Mobility Label distribution while allowing the MPLS-based forwarding plane to deliver mobile traffic in an optimal manner. This paper presents a hierarchical mobility management system capable of both macro- and micro-mobility support without the use of Mobile IP and its derivatives and allows scalable Mobility Label distribution and MPLS label stack based packet forwarding in support of optimal traffic delivery between the communicating mobile users.     

多协议标签交换流量工程隧道实时监测方法——MTRM

针对现有工作无法实时动态监测多协议标签交换(MPLS)流量工程(TE)隧道状态变化的问题,提出了一种MPLS TE隧道实时监测方法--MTRM。在网络中植入被动采集探针以采集OSPF-TE信令,以此为基础构建网络模型,使用隧道路径实时监测算法进行实时的隧道路径计算,最终实现动态监测。仿真实验在15个节点的MPLS网络中进行。结果表明,MTRM能够在5s之内监测到隧道变化,准确率超过90%。这种MPLS TE隧道实时监测方法,大大降低了MPLS网络管理和流量工程实施的难度,具有广阔的应用前景。     

IM-ELPR: Influence maximization in social networks using label propagation based community structure

Applied Intelligence - The popularity of social networks has grown manifolds in recent years because of various activities like fast propagation of ideas, publicity, and news. Influence...     

基于聚类标签均值的半监督支持向量机

针对标签均值半监督支持向量机在图像分类中随机选取无标记样本会导致分类正确率不高,以及算法的稳定性较低的问题,提出了基于聚类标签均值的半监督支持向量机算法。该算法修改了原算法对于无标记样本的惩罚项,对选取的无标记样本聚类,使用聚类标签均值替换标签均值。实验结果表明,使用聚类标签均值训练的分类器大大减少了背景与目标的错分情况,提高了分类的正确率以及算法的稳定性,适合用于图像分类。     

基于正交调制的IP-over-WDM光标记交换网络

详细阐述了基于正交调制格式的IP-over-WDM光标记交换网络的网络体系结构,给出了边缘路由器的结构图与核心路由器的实现方法,最后给出了基于正交调制格式的IP-over-WDM光标记传输实验,实验中的标记信号速率为312 Mb/s,净荷速率10Gb/s,在标准单模光纤(SMF)上实现了88km传输,实验结果表明了所述方案的可行性.     

基于OSPF和BGP的网络拓扑实时发现系统的设计

针对当前包括了多个自治系统（AS）的大型网络拓扑结构实时绘制遇到的困难,提出了一个基于OSPF协议和BGP协议的快速拓扑发现系统的设计方法。该系统对AS之间发送的BGP报文和AS内部的OSPF报文进行采集,对采集的报文进行分析,能够迅速、高效、准确地反映出全网的拓扑结构及其实时变化。对系统报文采集,报文分析,图形保存等关键技术做了讨论。     

Rimgnet: A packet switched local network with decentralized control

Ringnet is a commercially available 8 MHz packet switched communication system designed to connect hosts in the same building, usually less than several hundred feet apart, although it can also be used for loosely coupled multiprocessor systems. Control of the network is completely decentralized and depends only on a circulating token for arbitration of the communication medium. This paper describes the hardware and software architecture of the network and discusses some of the design principles used in its implementation.     

Stability of switched neural networks with time-varying delays

This paper investigates the globally exponential stability of switched neural networks with time-varying delays. By virtue of mode-dependent average dwell time, some significant criteria of exponential stability are obtained for delayed switched neural networks with only stable subsystems or with both stable subsystems and unstable subsystems. The proposed theoretical results could be utilized not only to verify the globally exponential stability of switched neural networks, but also to design appropriate switching signal to guarantee the globally exponential stabilization. They can explicitly reflect the effect of mode-dependent average dwell time on the stability of switched neural networks. In contrast to the previous results, the proposed criteria are more straightforward and effective in the real-world application. Three numerical examples are introduced to illustrate the effectiveness of the proposed results.

     

Consensus of switched multi-agent systems with random networks

This paper studies the consensus problem of the switched multi-agent system composed of continuous-time and discrete-time subsystems. Communication among agents is modelled as a random network where the existence of any information channel is probabilistic and independent of other channels. Then, some necessary and sufficient conditions are presented for solving average consensus of the switched multi-agent system under arbitrary switching. Furthermore, we show that the average consensus in different sense (mean square, almost surely and in probability, respectively) are equivalent. Finally, simulations are provided to illustrate the effectiveness of our theoretical results.     

基于层级图标签表示网络的多标签文本分类

多标签文本分类是一项基础而实用的任务,其目的是为文本分配多个可能的标签。近年来,人们提出了许多基于深度学习的标签关联模型,以结合标签的信息来学习文本的语义表示,取得了良好的分类性能。通过改进标签关联的建模和文本语义表示来推进这一研究方向。一方面,构建的层级图标签表示,除了学习每个标签的局部语义外,还进一步研究多个标签共享的全局语义。另一方面,为了捕捉标签和文本内容间的联系并加以利用,使用标签文本注意机制来引导文本特征的学习过程。在三个多标签基准数据集上的实验表明,该模型与其他方法相比具有更好的分类性能。     

Semi-supervised support vector regression based on self-training with label uncertainty: An application to virtual metrology in semiconductor manufacturing

Dataset size continues to increase and data are being collected from numerous applications. Because collecting labeled data is expensive and time consuming, the amount of unlabeled data is increasing. Semi-supervised learning (SSL) has been proposed to improve conventional supervised learning methods by training from both unlabeled and labeled data. In contrast to classification problems, the estimation of labels for unlabeled data presents added uncertainty for regression problems. In this paper, a semi-supervised support vector regression (SS-SVR) method based on self-training is proposed. The proposed method addresses the uncertainty of the estimated labels for unlabeled data. To measure labeling uncertainty, the label distribution of the unlabeled data is estimated with two probabilistic local reconstruction (PLR) models. Then, the training data are generated by oversampling from the unlabeled data and their estimated label distribution. The sampling rate is different based on uncertainty. Finally, expected margin-based pattern selection (EMPS) is employed to reduce training complexity. We verify the proposed method with 30 regression datasets and a real-world problem: virtual metrology (VM) in semiconductor manufacturing. The experiment results show that the proposed method improves the accuracy by 8% compared with conventional supervised SVR, and the training time for the proposed method is 20% shorter than that of the benchmark methods.     

基于类标记扩展的半监督网络流量特征选择算法

针对网络流量特征选择过程中存在的样本标记瓶颈问题,以及现有半监督方法无法选择强相关的特征的不足,提出一种基于类标记扩展的多类半监督特征选择(SFSEL)算法。该算法首先从少量的标记样本出发,通过K-means算法对未标记样本进行类标记扩展;然后结合基于双重正则的支持向量机(MDrSVM)算法实现多类数据的特征选择。与半监督特征选择算法Spectral、PCFRSC和SEFR在Moore数据集进行了对比实验,SFSEL得到的分类准确率和召回率明显都要高于其他算法,而且SFSEL算法选择的特征个数明显少于其他算法。实验结果表明: SFSEL算法能够有效地提高所选特征的相关性,获取更好的网络流量分类性能。     

基于隐藏标签节点挖掘的跨网络用户身份识别

随着各种社交网络不断涌现,以及针对社交网络的安全和商业应用的不断普及,跨网络用户身份识别成为当前的研究热点。针对现有的基于自中心网络环境算法(Ego-UI)对标签节点利用率不高的缺点,该文提出一种基于隐藏标签节点挖掘的跨网络用户身份识别算法(HLNM-UI)。该算法通过给待匹配节点添加社团聚类信息,将挖掘出的隐藏标签节点加入到自中心网络里,通过对潜在的关系信息加以利用,提高待匹配节点的辨识度,然后利用标签节点找寻最佳匹配,最后通过迭代运算实现全网络所有节点的身份识别。在多个人工随机网络和真实社交网络实验结果表明,该文提出的算法相比现有的基于自中心网络算法具有更高的召回率和F-1值。     

ANEMA: Autonomic network management architecture to support self-configuration and self-optimization in IP networks

The increasing scale, technology advances and services of modern networks have dramatically complicated their management such that in the near future it will be almost impossible for human administrators to monitor them. To control this complexity, IBM has introduced a promising approach aiming to create self-managed systems. This approach, called Autonomic Computing, aims to design computing equipment able to self-adapt its configuration and to self-optimize its performance depending on its situation in order to fulfill high-level objectives defined by the human operator.In this paper, we present our autonomic network management architecture (ANEMA) that implements several policy forms to achieve autonomic behaviors in the network equipments. In ANEMA, the high-level objectives of the human administrators and the users are captured and expressed in terms of ‘Utility Function’ policies. The ‘Goal’ policies describe the high-level management directives needed to guide the network to achieve the previous utility functions. Finally, the ‘behavioral’ policies describe the behaviors that should be followed by network equipments to react to changes in their context and to achieve the given ‘Goal’ policies.In order to highlight the benefits of ANEMA architecture and the continuum of policies to introduce autonomic management in a multiservice IP network, a testbed has been implemented and several scenarios have been executed.     

基于网络演算计算交换式工业以太网中的最大时延

为保证周期性实时数据在交换式工业以太网中的服务质量,在交换机中引入非抢占式的优先级调度方法,并在源节点对进入网络的非周期数据流进行漏桶管制;基于网络演算理论推导了交换机对不同类型数据的服务曲线;针对一个树型拓扑的交换式工业以太网模型,利用所推导出的结论计算了实时数据的最大网络时延.     

带有时变时滞的切换基因调控网络的稳定性

研究了带有时变时滞的切换基因调控网络的稳定性问题。和现有关于切换基因调控网络结果不同的是,切换基因调控网络包括稳定子系统和不稳定子系统。利用平均驻时方法和线性矩阵不等式技术,得到了带有时变时滞的切换基因调控网络指数稳定的判据。最后的仿真实例验证了结果的有效性。