基于FARIMA模型的Internet网络业务预报

最近的网络研究发现Internet网络业务同时呈现长相关和短相关特性，因此建立可以同时描述，预报长相关和短相关特性的网络业务模型很有必要。文中给出了利用FARIMA模型进行建模和预报的方法，实验表明这种方法用于实际Internet网络trace是非常有效的，另外提供了简化FARIMA模型拟合的方法和具体步骤，这样大大缩短了模型辨识的时间，对于实际网络预报有很好的实用性。  

基于Elman神经网络的网络流量建模及预测

根据实际网络中测量得到的网络流量数据，建立一个基于Elman神经网络的流量模型，介绍Elman神经网络的架构设计，并提出一种基于正交最小二乘的学习算法，在此基础上对网络流量进行预测。仿真实验结果表明，该模型具有良好的预测效果，相对于传统线性模型及BP神经网络模型具有更高的预测精度和更好的自适应性。  

基于RBF神经网络的网络流量建模及预测

随着计算机网络的迅速发展,目前的网络规模极为庞大和复杂,网络流量预测对于网络管理具有至关重要的意义。根据实际网络中测量的大量网络流量数据,建立了一个基于RBF神经网络的流量模型,给出了RBF神经网络的结构设计及基于正交最小二乘的学习算法,并基于该流量模型对网络流量进行预测。仿真结果表明,该模型具有较高的预测效果,相对于传统线性模型及BP神经网络模型具有更高的预测精度和良好的自适应性。  

Internet 流量模型分析与评述

Internet流量模型对网络性能管理、QoS、准入控制等都有很重要的意义和作用.首先总结了现阶段已发现的主要网络流量的特性及相关度量参数,概要地介绍网络流量建模的意义和分类,然后按照"传统-自相似-流量建模的新发展"这3个阶段阐述网络流量建模的发展历程与最新的研究成果,最后针对目前网络流量建模中存在的难点问题,展望了该领域未来的研究发展方向.  

Compiling network traffic into rules using soft computing methods for the detection of flooding attacks

The ability to dynamically collect and analyze network traffic and to accurately report the current network status is critical in the face of large-scale intrusions, and enables networks to continually function despite of traffic fluctuations. The paper presents a network traffic model that represents a specific network pattern and a methodology that compiles the network traffic into a set of rules using soft computing methods. This methodology based upon the network traffic model can be used to detect large-scale flooding attacks, for example, a distributed denial-of-service (DDoS) attack. We report experimental results that demonstrate the distinctive and predictive patterns of flooding attacks in simulated network settings, and show the potential of soft computing methods for the successful detection of large-scale flooding attacks.  

基于遗传程序设计的IP业务流量长期预测

针对流量行为分析对自动化、智能化和预测精度的进一步需求和存在的不足，提出了基于遗传程序设计(GP)的IP业务流量长期预测算法．论文首先给出了GP的定义和GP个体的二叉树表示，然后重点论述了GP的演化算法设计流程、数据样本获取、GP的演化建模和实验．实验结果表明，通过对Web业务流量进行GP建模实验，搜索到了两个较好的业务流量预测模型，模型所对应的预测曲线与原始数据曲线拟合得较好，说明采用GP建模是有效的，能反映业务流量的变化规律．  

Simulating Network Traffic with a DWT-based Model

Traffic models play a significant role in the analysis and characterization of network traffic and network performance. Thorough research and accurate modeling on network traffic become an efficient way to explore network internal mechanism, control network flux and optimize network performance. In this paper, we put forward a Haar DWT-basod （Discrete Wavelet Transform） traffic model. The scaling analysis on two simulated traces shows that our model can generate multi-fractal traffic data, which is in a close fit to the real trace statistics.  

Stochastic User Equilibrium Traffic Assignment with Turn-delays in Intersections

Turn-delays in intersections contribute significantly to travel times and thus route choices in urban networks. However, turns are difficult to handle in traffic assignment models due to the asymmetric Jacobian in the cost functions. The paper describes a model where turn delays have been included in the solution algorithm of Stochastic User Equilibrium (SUE) traffic assignment. When the Jacobian is symmetric, SUE minimises the road users' 'perceived travel resistances'. This is a probit-model where the links cost-functions of the links are traffic dependent. Hereby, overlapping routes are handled in a consistent way. However, no theoretical proof of convergence has been given if the Jacobian is asymmetric, although convergence can be shown probable for model data representing realistic road-networks. However, according to the authors knowledge SUE with intersection delays have not been tested earlier on a full-scale network. Therefore, an essential part of the paper presents practical tests of convergence. Both geometric delays and delays caused by other turns are considered for each turn. Signalised and non-signalised intersections are handled in different ways, as are roundabouts. In signalised intersections a separate model handles queues longer than one green-period. Green-waves can also be taken into consideration. The model has been tested on a large-scale network for Copenhagen with good results. To make it possible to establish the comprehensive data, a GIS-based 'expert system' was implemented (see Nielsen, O.A., Frederiksen, R. D. and Simonsen, N. (1997). Using expert system rules to establish data on intersections and turns in road networks. International Transactions in Operational Research , 5 , 513–529.  

Correlational and distributional effects in network traffic models

Simulation studies are used to evaluate the impact of the distributional and correlational characteristics of traffic arrival processes on the performance of network routing elements. It is shown that synthetic traffic models that capture only the distributional or the correlational characteristics of real workloads can yield substantially optimistic predictions of queue lengths and drop rate.

A new technique for generating synthetic arrival streams is proposed and evaluated. Arrival streams are generated by the widely used method of sampling from a target distribution. However, the uniform stream used in the sampling is itself derived from fractional Gaussian noise. The resulting synthetic streams are shown to have sample autocorrelation functions that are consistent with long-range dependence and to provide measurably better performance estimates than standard distribution-based and FGN-based techniques.  

基于分段IFS分形重建的流量模型

网络流量模型是网络仿真和网络行为研究中的一个基本问题。在分段IFS分形编码重建的基础上提出了一个新的网络流量模型，它基于实际流量的分形重建，所得重建流量变化趋势接近初始流量。在调节因子θ的作用下，重建流量的Hurst参数可以在一定范围内变化，同时重建流量自相关函数曲线与原始流量类似。因此，为网络流量仿真建模研究提供了一种新的选择。