TCP-Friendly拥塞控制机制及协议

当前实际网络中传输实时多媒体流使用用户数据报协议(UDP),而UDP不具备拥塞退避机制。在拥塞的网络环境中,UDP流将大量抢占TCP流的网络带宽,同时自身的丢包也迅速增加,并可能带来系统拥塞崩溃的潜在危险,因此提出了TCP友好的概念。文章介绍了TCP-Friendly的提出背景以及实现策略,并对各策略做出评价。     

对TCP/IP计算机网络拥塞控制的研究

为提升计算机的网络性能,更好地避免拥塞现象的发生,需要对其进行必要的技术控制。鉴于此,对基于TCP/IP协议的网络拥塞控制方法进行分析。在TCP拥塞控制中主要采用TCP Tahoe,TCP Reno,TCP New Reno以及TCP Sack四种方法,其中TCP New Reno对快速恢复算法进行了改进,通过对TCP协议中的Reno进行可视化处理,实行对网络拥塞的有效管理。而IP拥塞控制方法则分为FIFO,FQ和WFQ,RED以及ECN四种类型,通过队列调度管理方式实现了对网络拥塞的有效管理。     

随机早期检测算法的参数研究

RED算法较好解决了TCP拥塞控制中延迟和长期拥塞问题,但RED算法对参数的设置很敏感,不同的参数设置对RED的性能影响很大。文章就参数的设置提高了RED算法的性能,克服它的一些缺点进行了探讨和研究。     

通信网络中的拥塞控制

介绍了网络中发生拥塞带来的问题，TCP／IP的拥塞控制和常用的减少拥塞的几种措施；讨论了拥塞管理的算法并介绍了2种拥塞避免的算法，即随机早期检测(RED)算法和加权随机早期检测(WRED)算法；说明了参数选择的依据，并简单描述了在大型TCP／IP网络中运用WRED算法的意义。     

FCA-ItswTCM:区分服务中一种公平的拥塞自适应标记算法

 本文提出了一种新的滑动窗口标记算法——公平的拥塞自适应标记算法(FCA-ItswTCM).算法近似识别TCP流和UDP流,适度区分标记,规避拥塞控制机制对公平性的影响;细粒度描述拥塞,预测拥塞,以此自适应调节各流注入黄包比例,兼顾网络拥塞状态对公平性的影响.仿真实验表明,与其他几种滑动窗口标记算法相比,FCA-ItswTCM对确保TCP流和UDP流带宽共享的公平性、提高资源利用率及系统稳定性有较好的效果.     

利用可变队列长度的AQM改进与研究

AQM（主动队列管理）通过在拥塞的早期阶段丢包来告知信源。AQM改进了响应流（如TCP）和非响应流（如UDP）之间的公平性,它可以带来较高的吞吐量及有效的链路。提出了可变队列长度适应RED算法,它基于每小时的可变队列长度,并且可以更有效地应对突发流。如果将当前可变队列长度与RED及ARED相比,这种方法可以提供更为平缓的平均度列长度。试验结果表明,可变队列长度适应RED比弃尾法提高了公平性,与ARED比较性能也大为改观。这样即使有突发流存在,也可以使数据包稳定地传输。     

基于RTT的TCP流带宽公平性保障机制

TCP端到端的拥塞控制机制使得TCP连接获得的瓶颈带宽反比于RTT(数据包往返时间)。为了缓解TCP对于RTT较小流的偏向,区分服务的流量调节机制在RTT较小的流取得目标速率且获得多余资源的情况下可以确保RTT较大流不至于饥饿。现有的基于RTT的流量调节机制在网络拥塞程度较轻时非常有效,但是当网络拥塞程度较重时,由于对RTT较大流的过分保护而导致RTT较小流饥饿。因此,通过引进自适应的思想提出了改进方法,其主要思想就是根据网络的拥塞程度自适应地调整对RTT较大流的保护程度。大量的仿真试验表明所提的机制能有效保障TCP流的带宽公平性并且比现有的方法具有更好的强壮性。     

TCP网络的自适应非奇异终端滑模控制

针对TCP网络的拥塞控制问题,采用非奇异终端滑模控制理论提出了一种新的主动队列管理算法。采用非奇异终端滑模面以克服传统终端滑模控制的奇异问题,同时确保系统能在有限时间内收敛至平衡点。考虑到UDP流干扰的情况,用Lyapunov稳定性方法给出了一个自适应律来消除UDP流干扰对系统的影响。仿真结果表明,该算法可以使队列长度快速收敛到设定值,同时维持较小的队列振荡,优于传统的滑模控制。     

TCP友好的速率控制

TCP友好的速率控制(TFRC)主要适用于实时数据传输的一种拥塞控制机制,具有突出的TCP友好性即在相同的环回时间(RTT)下可以和TCP流享有近乎相同的带宽,从而避免了由于UDP等传输层协议缺乏拥塞控制而带来的网络拥塞甚至崩溃.本文简要介绍了它的协议机制并通过一些仿真和试验的结果初步讨论了其性能.     

前向主动网络拥塞控制算法及其性能分析

本文提出了一种基于主动式网络(Active Networks)技术的拥塞控制算法FACC(Forward Active Networks Congestion Control).与传统的TCP(Transport Control Protocol)相比,FACC算法通过在网络结点直接提供拥塞检测和拥塞控制机制,大大缩短源端点的拥塞反应时间,从本质上提高了网络拥塞检测和控制的性能,从而提高了终端用户的平均吞吐量.文中还利用计算机仿真研究了FACC算法在各种网络条件下的性能,并与传统的Tahoe,Reno,NewReno及SACK TCP协议做了对比.结果表明无论网络中存不存在非受控数据流时,FACC控制算法均能明显地提高用户终端的平均吞吐量,并且由于采用FACC控制算法而增加的网络结点运算迟延也很小.     

基于不同业务模型的TCP性能分析

TCP稳定性对网络应用性能有直接的影响。本文根据TCP所承载的不同业务类型，即分别加载FTP业务和具有自相似特性的业务，分别从拥塞窗口的变化特征、TCP连接的同步现象，以及RED参数调节对TCP性能的影响等三个方面分析了TCP性能。实验数据表明，带宽的增加只是影响TCP工作稳定性的一个原因，TCP所承载的业务类型特征的差异对TCP的稳定性有直接的影响。而采用RED和TD这两种不同的缓存管理机制对TCP的工作稳定性的影响基本没有差异。实验还进一步研究了业务特征与TCP同步现象的联系。结果表明，同步现象除了跟TCP连接数与链路的带宽之比有关外，与TCP承载的业务类型也有关系。通过调整RED的两个主要参数和，实验结果表明，它们对TCP性能的影响不显著。     

On-demand transmission of 3D models over lossy networks

Three-dimensional (3D) meshes are used intensively in distributed graphics applications where model data are transmitted on demand to users’ terminals and rendered for interactive manipulation. For real-time rendering and high-resolution visualization, the transmission system should adapt to both data properties and transport link characteristics while providing scalability to accommodate terminals with disparate rendering capabilities. This paper presents a transmission system using hybrid unequal-error-protection and selective-retransmission for 3D meshes which are encoded with multi-resolutions. Based on the distortion-rate performance of the 3D data, the end-to-end channel statistics and the network parameters, transmission policies that maximize the service quality for a client-specific constraint is determined with linear computation complexity. A TCP-friendly protocol is utilized to further provide performance stability over time as well as bandwidth fairness for parallel flows in the network. Simulation results show the efficacy of the proposed transmission system in reducing transmission latency and providing smooth performance for interactive applications. For example, for a fixed rendering quality, the proposed system achieves 20–30% reduction in transmission latency compared to the system based on 3TP, which is a recently presented 3D application protocol using hybrid TCP and UDP.     

A Robust Analytical Model of Mobile IP Handoff with Multiple Traffic Profiles

Mobile IP has been developed to provide the continuous information network access to mobile users. The performance of Mobile IP mobility management scheme is dependent on traffic characteristics and user mobility. Consequently, it is important to assess this performance in-depth through these factors. This paper introduces a novel analytical model of handoff management in mobile IP networks. The proposed model focuses on the effect the traffic types and their frame error rates on the handoff latency. It is derived based on general distribution of both successful transmission attempts and the residence time to be applicable in all cases of traffic characteristics and user mobility. The impact of the behavior of wireless connection, cell residence time, probability distribution of transmission time and the handoff time is investigated. Numerical results are obtained and presented for both TCP and UDP traffics. As expected, the reliability of TCP leads to higher handoff latency than UDP traffic. It is shown that, higher values of FER increase the probability of erroneous packet transfer across the link layer. A short retransmission time leads to end the connection most likely in the existing FA; however a long retransmission time leads to a large delivery time. The proposed model is robust in the sense that it covers the impact of all the effective parameters and can be easily extended to any distribution.     

Nonlinear instabilities in TCP-RED

This work develops a discrete-time dynamical feedback system model for a simplified TCP network with RED control and provides a nonlinear analysis that can help in understanding observed parametric sensitivities. The model describes network dynamics over large parameter variations. The dynamical model is used to analyze the TCP-RED operating point and its stability with respect to various RED controller and system parameters. Bifurcations are shown to occur as system parameters are varied. These bifurcations, which involve the emergence of oscillatory and/or chaotic behavior, shed light on the parametric sensitivity observed in practice. The bifurcations arise due to the presence of a nonlinearity in the TCP throughput characteristic as a function of drop probability at the gateway. Among the bifurcations observed in the system are period doubling and border collision bifurcations. The bifurcations are studied analytically, numerically, and experimentally.     

基于H∞控制理论的网络拥塞控制器的设计

针对具有多维性、非线性、动态性、不确定性等复杂特征的计算机网络系统,文章从控制理论出发,提出了一种基于H∞控制理论的主动队列管理（AQM）算法,文章着重分析RED算法。首先,建立了AQM／TCP拥塞控制的动态模型;其次,系统的渐进稳定由线性矩阵不等式（LMI）描述,并应用MATLAB的LMI工具箱求解H∞控制的一般线性时滞系统;然后将求解结果应用到RED中,从而得出了控制器。最后在NS2平台上进行了仿真实验,试验结果表明,该算法在解决系统不确定性引起的鲁棒性问题上是有效的。     

基于多协议的人工影响天气中转服务器

为了满足人工影响天气不同作业决策时段需求,设计了一种基于混合TCP - UDP协议的人工影响天气中转服务器.连接双方(指挥端和客户端)的指令通讯协议采用混合TCPUDP算法的传输协议进行数据连接.该机制使用UDP作为短对话时的传输层协议,而对于大量数据需要传输时则使用TCP作为传输层的协议.这样,对于短对话可以避免TCP的额外开销,而对于长对话又可以得到由TCP提供的可靠传输和拥塞控制.     

改进TCP拥塞控制机制实现网络资源的加权分配

在现有的TCP/IP协议体系下，所有TCP业务完全公平地竞争网络资源，使之难于满足不同用户、不同应用的通信需求。本文提出了一种改进的TCP拥塞控制机制，理论分析和仿真验证表明，此机制实现简单，系统额外开销小，易进行网络资源的加权分配，能有效地实现服务质量的相对保证，改善网络通信性能。     

Multimedia transmission with adaptive QoS based on real‐time protocols

In this paper, we describe a mechanism for adaptive transmission of multimedia data, which is based on real‐time protocols. The proposed mechanism can be used for unicast or multicast transmission of multimedia data over heterogeneous networks, like the Internet, and has the capability to adapt the transmission of the multimedia data to network changes. In addition, the implemented mechanism uses an inter‐receiver fairness function in order to treat the group of clients with fairness during the multicast transmission in a heterogeneous environment. The proposed mechanism uses a ‘friendly’ to the network users congestion control policy to control the transmission of the multimedia data. We implement a prototype application based on the proposed mechanism and we evaluate the proposed mechanism both in unicast and multicast transmission through a number of experiment and a number of simulations in order to examine its fairness to a group of clients and its behaviour against transport protocols (TCP) and UDP data streams. Copyright © 2003 John Wiley & Sons, Ltd.     

Robust optimization of Random Early Detection

Random Early Detection (RED) is the most widely used Adaptive Queue Management (AQM) mechanism in the internet. Although RED shows better performance than its predecessor, DropTail, its performance is highly sensitive to parameter settings. Under non-optimum parameter settings, the performance degrades and quickly approaches that of DropTail gateways. As the network conditions change dynamically and since the optimum parameter settings depend on these, the RED parameters also need to be optimized and updated dynamically. Since the interaction between RED and TCP is not well understood as analytical solutions cannot be obtained, stochastic approximation based parameter optimization is proposed as an alternative. However, simulation based approaches may yield a sub-optimal solution since for these to work, the network needs to be accurately simulated which is, however, infeasible with today’s internet. In this paper, we present an optimization technique for optimizing RED parameters that makes use of direct measurements in the network. We develop a robust two-timescale simultaneous perturbation stochastic approximation algorithm with deterministic perturbation sequences for optimization of RED parameters. A proof of convergence of this algorithm is provided. Network simulations, using direct implementation of the algorithm over RED routers, are carried out to validate the proposed approach. The algorithm presented here is found to show better performance as compared to a recently proposed algorithm that adaptively tunes a RED parameter.     

Design of Active Queue Management System for Scalable TCP in High Speed Networks

The Scalable Transmission Control Protocol (TCP), based on a multiplicative increase, multiplicative decrease congestion avoidance algorithm, has been proposed recently to overcome the inability of Standard TCP to utilize the full bandwidth in high speed networks. This paper employs a novel approach to derive a transfer-function model of Scalable TCP that is then employed in a control-theoretic design of random early detection (RED)-based active queue management (AQM) for such a network. Robust stability of the proposed scheme is established under prescribed conditions, and the design is validated by discrete-event simulations using the ns2 tool. This work was supported by a Gledden Visiting Senior Fellowship for the first author.