输入排队Crossbar架构下的矩阵模型 及MM-LQF调度策略

输入排队Crossbar交换是高性能交换设备最为常用而关键的技术之一.本文建立了IQ-Crossbar架构下的矩阵模型,给出了IQ-Crossbar的状态矩阵、队长矩阵、到达矩阵和匹配矩阵的数学定义,并通过分析IQ-Crossbar的信元排队机理,提出和证明了队长矩阵迭代定理和状态矩阵迭代定理.该矩阵模型为分析IQ-Crossbar架构下的调度算法提供了理论依据.基于所建立的矩阵模型,在分析现有LQF调度算法优缺点的基础上,本文提出了一种新的调度策略MM-LQF,该策略的运算效率是LQF的3.72倍,支持的端口门限速率是LQF的2.35倍,在贝努利均匀流量重载条件下平均时延是LQF的1/2;在贝努利Diagonal流量条件下吞吐率为100%.     

可编程路由器中基于缓冲队列长度阈值的处理器调度

可编程路由器除了转发IP分组之外,还需要执行计算任务.如何调度可编程路由器中CPU的处理能力是一个需要解决的重要问题.本文首先建立了一种通用的可编程路由器软件体系结构,在此基础上,提出了一种基于缓冲队列长度阈值的CPU调度算法,采用随机Petri网对算法进行了模型分析和计算.结果表明,该调度算法可以同时保证可编程路由器中的尽力发送流和QoS流的计算需求.     

一种基于节点实时流量的异常检测算法

针对无线传感器节点缺乏移动性和可预知的流量模式等特征,从而难以在传感器网络中进行异常检测的问题,该文提出了一种基于节点的实时异常检测算法.根据节点流量的到达过程,提出了一种新的节点流量到达模型,根据多层次的滑动窗口事件存储原理,将动态统计值进行短时间的保持,对不同时间段的到达过程指标进行比较,包括节点的可计算资源,低复杂度,融合特性等,以此来判别流量到达过程是否发生异常,从而对传感器网络进行有效的异常检测.     

LTE网络测试仪IP流量生成技术的研究与实现

针对LTE网络测试仪的测试需求,本文对IP流量生成技术进行了研究,提出了一种双模式IP流量发生器的设计方案,介绍了总体平台的设计架构及其工作原理,性能模式下流量控制的算法,仿真模式下利用读写索引进行数据边发送、边更新的软件控制方法。测试应用表明,该IP流量发生器能定量、定速发送指定IP数据分组,且能模拟高速网络流量,具有良好的性能,能较好地满足IP数据采集与仿真测试的需求。     

基于BGP增强的流量调度技术

介绍了运营商流量调度领域的发展趋势,概括了运营商IP网络的主要流量调度需求,分析了现网基于BGP的流量调度技术存在的问题,提出了一种基于SDN的集中式流量调度方案.该方案对现有BGPFlowspec协议进行了增强,避免了逐个设备的路由策略配置,并可兼容传统BGP路由协议,为传统运营商提供了一种更为灵活的流量调度控制方法.     

基于流量分布的高速路由表查找算法

IP路由表查找是实现高性能路由器的主要瓶颈。根据IP业务流量分布特性，在现有的路由表查找技术的基础上，提出了基于流量分布的高速路由表查找算法。     

流量工程对IP路由器的要求

流量工程（TE）是目前IETF和大型的Internet服务商正在深入讨论的热门话题,本文介绍了在流量工程环境网络中的IP路由器人有的路由和转发功能,以及目前关于流量工程问题讨论中的主要解决方案。     

多径信号到达角和时延高分辨联合估计算法

本文提出了一种新的高分辨高精度的多径信号到达角和时延联合估计算法.该方法通过对多径信号与参考信号进行相关处理,再根据PRO-ESPRIT算法的思想,实现了DOA(Direction-Of-Arrival)估计;并对相关延迟向量进行变换,从互功率谱相位中提取时延信息,完成了高分辨的TOA(Time-Of-Arrival)估计.最后,本文推导了频域模型下到达角和时延联合估计的Cramer-Rao下界.并将通过仿真对本算法的性能进行了评估.     

带VOQ的输入队列交换网络中的分组调度算法研究

交换技术已经成为高速路由器的核心技术。本文基于目前高速交换技术所采用的主要体系结构,带有虚拟输出缓冲队列(VOQ)的输入队列交换结构,分析已经存在的各种调度算法的性能,并设计基于遗传算法的调度策略,提供IP数据网络的QoS对吞吐量和抖动的保障。     

IPTV网络测试仪中流量生成器的设计与实现

根据IPTV网络性能测试的需求,提出了基于FPGA流量生成解决方案,给出了流量设计规则和详细的流标签设计格式,结合IP报文和自定义流的格式要求,描述了自定义IP包的生成过程。为了模拟网络环境中数据流的多样性和突发性,讨论了用于流调度的GRR算法,重点分析了均匀突发流的流模式和流表设计,最后通过FPGA对系统设计进行仿真验证,结果表明设计方案稳定,在IPTV测试仪的应用中取得良好效果。     

光突发交换中的突发业务流模型及其应用

提出了一种光突发交换中的突发业务流模型,采用该模型对光突发交换中的LAUC-VF输出调度算法在不同的突发业务强度和突发长度下的性能进行了模拟仿真,分析比较了该算法在此突发业务流和普通业务流模型下的性能,仿真结果表明,该突发业务流模型具有一定的合理性。     

基于突发分离的自相似网络流量预测

网络流量预测在网络拥塞控制及资源分配中起着至关重要的作用。对于具有自相似性的网络业务流量,由于其存在较强突发,传统预测方法的预测精度普遍较低。本文针对存在高突发的网络流量数据,提出了一种基于数据分离的流量预测方法。在预测步骤前,本方法首先通过控制图将网络流量中难以预测的突发流量进行有效的分离,从而得到突发流量和非突发流量两部分数据。之后分别采用人工神经网络和自适应模板匹配方法实现对非突发流量和突发流量的预测。最后通过对两部分预测结果的合并得到最终的预测结果。基于实际流量数据的实验结果表明:相较于传统流量预测方法,本文所提出的方法具有更高的流量预测精度。      

自相似网络通信量模型研究综述

越来越多的研究表明网络通信量不是Markov过程,而是在任意时间尺度上都具有突发特性,即自相似特性。描述网络通信量的数学模型主要有自相似和长相关结构。网络的某些参数服从重尾分布,从而导致网络通信量时间尺度上的突发特性。该文分析了传统网络通信量模型和性能分析的弊端,描述了新型网络通信量模型应该具有的基本特征。本文重点研究了网络自相似通信量相关的ON/OFF模型、用户访问概率模型和网络流量闭环模型,讨论了相关的研究方向,并总结了在研究网络通信量模型的过程中应该注意的原则和问题。     

DYNAMIC BANDWIDTH ALLOCATION IN AN ATM MULTIPLEXER WITH MULTICLASS TRAFFIC AND ITS PERFORMANCE ANALYSIS

This paper introduces two source models: MAP(Markovian arrival process) model for the traffic with correlation and burst, e.g., voice, video, etc. and PAP(Poisson arrival process) model for the traffic with non-correlation, such as data, etc. Then a movable boundary bandwidth access policy is chosen.Basing on above model, the performance measures, e.g., mean waiting time and loss probability,especially the queue length time distribution are obtained. Finally, a number of numerical results are provided and shown through simulation.     

Fast blocking probability evaluation of end-to-end optical burst switching networks with non-uniform ON–OFF input traffic model

A novel, accurate, numerically stable and fast mathematical method for network-wide blocking probability evaluation of end-to-end optical burst switching (OBS) networks with heterogeneous link capacity under non-uniform traffic is proposed. Unlike most of previous works, which have used the Poisson model for the burst traffic, this paper considers a non-uniform ON–OFF model being a more realistic choice for OBS networks. Compared with simulation, the method was proved to be very accurate and much faster (up to four orders of magnitude faster than simulation in the studied cases). These features make the proposed method very useful for network analysis, especially for large size networks where simulation time can be prohibitively high.     

业务自相似性对光突发交换调度算法影响分析

使用泊松业务流模型对光突发交换网络进行性能分析不能准确地反映网络状态。该文从理论上对突发包长度进行了推导,对基于时间门限汇聚机制下突发包数据流自相似程度进行了计算机仿真测量,并利用自相似业务流模型对光突发交换网络中比较常用的LAUC和LAUC-VF调度算法进行了性能仿真。仿真结果表明:基于时间门限的汇聚机制能够有效地降低数据流自相似程度,数据流的自相似特性对LAUC算法的影响并不非常明显,但对LAUC-VF算法的性能则产生了比较严重的影响,其突发包丢失率较泊松流平均增加了近3个百分点。     

How to reconstruct end-to-end traffic based on time-frequency analysis and artificial neural network

End-to-end traffic, which describes the inherent characteristics and end-to-end behaviors of communication networks, is the crucial input parameter of network management and network traffic engineering. This paper proposes a new reconstruction algorithm to develop the research on reconstruction of end-to-end traffic in large-scale communication networks. We firstly conduct the time-frequency analysis on end-to-end traffic, and then localize its features to gain its time-frequency properties before decomposing it into the low-frequency and high-frequency components. We find that if decomposing appropriately, the low-frequency component of end-to-end traffic can accurately reflect its change trend, while its high-frequency component can well show the burst and fluctuation nature. This motivates us to find a reasonable time-frequency decomposition strategy to extract the low-frequency and high-frequency components of end-to-end traffic. Moreover, this further inspires us to use the regressive model to model the low-frequency part, exploit artificial neural network to characterize the high-frequency component, and then combine these two parts according to the regressive model and artificial neural network to precisely reconstruct end-to-end traffic. Simulation results show that in contrast to previous methods our algorithm is much more effective and promising.     

A novel transfer mode to reduce burst loss rate for optical burst switching networks

A crucial issue in optical burst switching (OBS) networks is burst loss caused by resource contention. As a result, many methods are currently being proposed to reduce burst loss rate. These methods can be summed up into two categories: burst scheduling algorithms and contention resolutions. Both categories of methods can reduce burst loss rate to a certain degree. However, to make OBS to become a viable solution, the burst loss rate needs to be further reduced. Furthermore, almost all methods ignore the fact that an unfortunately scheduled, locally generated single-hop burst could block a number of future incoming transit bursts, though the burst just travels to its next downstream node. This phenomenon becomes more evident when links are heavily loaded in mesh OBS networks. To eliminate contention caused by single-hop traffic completely, this paper proposes a novel transfer mode called packet calking by differentiating between single-hop traffic from multihop traffic for OBS networks. An analysis model is developed to evaluate the performance of packet calking. Theoretical results are validated through extensive simulations in both ring and mesh networks. These results show that packet calking outperforms the transfer mode without packet calking in terms of burst loss rate and link utilization.     

TELNET的客户端流量模型

本文针对英特网上的一个常用协议－远程登录（ＴＥＬＮＥＴ）协议分析了它的客户端数据包产生的时间间隔分布,发现泊松模型的指数分布不能很好地刻画它的特性。本文依据Ｔｃｐｌｉｂ提供的经验分布曲线,认为Ｐａｒｅｔｏ分布在很大时间范围内能较好地描述ＴＥＬＮＥＴ的客户端流量特性,并对其突发序列的特性进行了分析。本文最后指出在描述网络流量时自相似模型要比泊松模型有效和精确得多,并给出了Ｐａｒｅｔｏ分布在建立自相似     

Characteristics of WAP Traffic

This paper considers the characteristics of Wireless Application Protocol (WAP) traffic. We start by constructing a WAP traffic model by analysing the behaviour of users accessing public WAP sites via a monitoring system. A wide range of different traffic scenarios were considered, but most of these scenarios resolve to one of two basic types. The paper then uses this traffic model to consider the effects of large quantities of WAP traffic on the core network. One traffic characteristic which is of particular interest in network dimensioning is the degree of self-similarity, so the paper looks at the characteristics of aggregated traffic with WAP, Web and packet speech components to estimate its self-similarity. The results indicate that, while WAP traffic alone does not exhibit a significant degree of self-similarity, a combined load from various traffic sources retains almost the same degree of self-similarity as the most self-similar individual source.