动态权重RED网关

随着网络流量的增加,拥塞控制的重要性日益突出,尤其是链路拥塞控制。IETF推荐RED(Random Early Detection)作为下一代Internet路由器的拥塞控制算法。该文在分析了RED的优点和缺点后,在现在RED算法的基础上通过改进RED算法的平均队列长度计算,使用平均队列长度的均值,判断拥塞的发生和结束,动态调整权重,提高RED对拥塞发生和拥塞结束的反应速度,缩短拥塞时间,维持较低的实际队列长度。最后用仿真验证了该方法。     

区分服务网络的协作可调整RED算法

文章描述了一种适用于区分服务网络的新的RED改进算法协作可调整RED算法(CARED)。算法的思想是基于令牌桶标记器的标记概率建立业务实际速率与承诺速率间的定量关系,并以这个关系动态调整下游节点中CARED的丢包概率。为了实现CARED,文章还提出了一种标记概率的计算方法基于包数的滑动窗口测量算法(NSWM)。文中通过详实的仿真实验证明了定量关系的正确性,同时验证了CARED算法不仅为区分服务提供了有效的拥塞控制功能,也保障了区分服务的不同性能要求。     

一种改进的主动队列调度RED算法

本文在分析RED利用EWMA形式计算平均队列长度的局限性的基础上,提出了一种改进的RED算法。该改进RED算法在计算平均队列长度时考虑了当前队列长度的真实情况,并将两者结合起来进行丢弃决策。仿真结果表明改进RED算法在分组丢弃比例和链路利用率上部优于RED。     

基于链路资源改进RED算法研究

Internet上随着网络的不断发展和用户的要求越来越高,以为用户提供更高服务质量（Quality of Service,QoS）为目标的研究越来越重要。而强壮的网络拥塞控制是保障用户服务质量的基础。“随机早期检测算法”（Random Early Detection,RED）[1],由于其在实际应用的广泛,成为研究热点。文章首先探讨远端流和近端流在在传输过程中的不公平性和链路资源利用情况,在此基础上提出路由器将网络中的流分为远近两种区分对待,改进RED算法以提高链路资源的利用率和流在传输过程中的公平性。     

基于平均队列长度和滞留时间的RED算法

在无线通信网络环境下，提出了一种改进的基于平均队列长度和滞留时间的RED算法。这种算法根据平均队列长度和滞留时间计算数据包的丢弃概率，比传统的单纯基于平均队列长度的RED算法相比较，能更有效地实现无线网络中的拥塞控制。     

串联排队RED、ERED网络分析模型

基于串联排队网络理论,建立了包括多个网络节点的串联排队RED和ERED分析模型。推导了具有指数分布分组丢失函数的RED算法(ERED)。在概率生成函数域,推导了节点的离去过程,并将其拟合为后续节点的到达过程,从而,刻画了多节点级联拥塞控制的通信场景。通过求解串联排队系统,得出串联排队RED、ERED以及DT 3种AQM机制的网络性能指标,分析了AQM参数设置对其性能的影响。针对不同突发度的业务源,给出每个节点和系统端到端性能指标。     

拥塞控制在路帱器中的实现

ＲＥＤ算法是路帐器中通常「采用的拥塞控制算法，它采用低通滤顺模型计算平均队长，不仅能很好地检测并避免网络拥塞。耐用能避免全局同步，并有很好的防突发能力。本文介绍了ＲＥＤ算法的原理，参数设置的规则及其实现的方法。     

一种基于速率和队列长度的主动队列管理机制

本文研究了拥塞产生的原因及其表现形式,提出了一种基于报文到达速率和队列长度的随机早丢弃算法(RQ).根据拥塞的严重程度和变化趋势将拥塞划分为六个级别,每个级别采用合适的丢弃概率,从而将队列控制在理想工作点附近.利用经典控制理论,分析了系统的稳定性,给出了参数配置的原则.最后,用NS网络仿真器对算法性能进行了验证.     

RED认证要点

RED指令已正式取代R&TTE。给出了RED认证测试项目和认证需提供的资料清单。对于RED的转证问题、制造商/进口商/欧盟代表/分销商的职责要求、风险评估、用户手册、标签等要点进行了说明。     

QL-CSFQ:一种结合队列长度的CSFQ算法

当存在适应流时,网络流量的测量值与实际值存在比较大的误差,这将严重影响CSFQ算法中公平共享速率α值的计算准确性,从而大大降低网络流之间的公平性.针对这一问题,本文提出了一种结合队列长度的CSFQ算法—QL-CSFQ.在QL-CSFQ算法中,采用结合队列长度状态信息来提高公平共享速率α值的计算准确性,从而有效提高了存在适应流情况下的网络流之间的公平性.大量模拟实验结果表明在存在适应流情况下QL-CSFQ有效消除了测量误差对算法公平性的影响,提高了网络流之间的公平性.     

基于RED的AODV路由协议改进

基于当前使用非常广泛的随机早期检测（RED）拥塞控制机制，提出了一种改进方案：在AODV的路由发现阶段，根据当前平均队列长度来确定以不同的方式响应路由请求，从而使建立起来的路由比较可靠，使数据流可以绕过负担比较重的节点，实现一定程度的拥塞控制和流量均衡，最后对改进前后的协议进行了性能仿真和分析。结果表明，改进后的协议与原协议相比在复杂度没有明显增加的前提下，降低了平均延时和丢包率，提高了网络性能。     

A framework to determine the optimal weight parameter of RED in next‐generation Internet routers

Active buffer management can improve the performance of Transmission Control Protocol/Internet Protocol‐based networks. Random early detection (RED), an active queue management scheme, has been proposed by the Internet Engineering Task Force for next‐generation Internet routers. RED uses a number of parameters, such as buffer thresholds, a packet drop probability and a weight parameter. RED suffers from low throughput and large delay/jitter and induces instability in networks. Previous attempts to improve the performance of RED were based on modifying the thresholds and drop probabilities. In this paper, we show that an optimal value of the weight parameter can improve the performance of RED, and then develop a framework to determine the optimal value of the weight parameter. We show that the optimal weight parameter obtained from our framework improves the performance of RED. Copyright © 2008 John Wiley & Sons, Ltd.     

基于参数估计的随机早期探测改进算法

流量控制是互联网有效运行的必要条件.随机早期探测RED(Random Early Detection)是一种基于整条链路的积极队列管理技术,它能显著地提高网络的利用率、稳定性和公平性,是IETF倡导使用的流量控制技术.它由Hashem首先提出雏型,Floyd对之作出了重大改进,形成了较为完善、实用的RED技术,为了在应用中增大RED的适应范围,Fang又提出基于参数自适应的RED.由于现有方法未能稳定、准确地刻画网络流量特征,在流量特性差异较大时,以上方法均难以有效控制网络流量.在本文中,作者首先给出了估计流量特性的解析方法,该方法较参数自适应具有更好的稳定性和准确性;并据此提出了一种RED改进算法,最后通过动态仿真,证明该改进算法提高了RED的适应性,降低了网络时延,减小了时延抖动.     

一种新颖的数据估计技术--dEWMA及其在RED中的应用

该文首先提出了一种新颖的估计技术——dEWMA(doubleExponentialWeightedMovingAverage)。该技术采用两个EWMA估计器进行数据测量,克服了传统EWMA估计技术不能同时获得良好的时效性和平稳性的缺点,能够为相关应用提供更加准确、灵活的数据估计。其次通过在RED网关中应用dEWMA技术,对其有效性进行验证。仿真结果表明采用dEWMA的RED性能较之未采用dEWMA的RED有明显的改善。     

TCP Unfairness in Ad Hoc Wireless Networks and a Neighborhood RED Solution

Significant TCP unfairness in ad hoc wireless networks has been reported during the past several years. This unfairness results from the nature of the shared wireless medium and location dependency. If we view a node and its interfering nodes to form a “neighborhood”, the aggregate of local queues at these nodes represents the distributed queue for this neighborhood. However, this queue is not a FIFO queue. Flows sharing the queue have different, dynamically changing priorities determined by the topology and traffic patterns. Thus, they get different feedback in terms of packet loss rate and packet delay when congestion occurs. In wired networks, the Randomly Early Detection (RED) scheme was found to improve TCP fairness. In this paper, we show that the RED scheme does not work when running on individual queues in wireless nodes. We then propose a Neighborhood RED (NRED) scheme, which extends the RED concept to the distributed neighborhood queue. Simulation studies confirm that the NRED scheme can improve TCP unfairness substantially in ad hoc networks. Moreover, the NRED scheme acts at the network level, without MAC protocol modifications. This considerably simplifies its deployment.Kaixin Xu is a Ph.D student of the computer science department at UCLA. He joined the Network Research Lab. (NRL) of UCLA at 2000. His research focuses on the ad hoc wireless networking especially protocols at MAC, Network and Transport layers. His recently work includes enhancing TCP performance in multihop ad hoc networks, TCP performance in IEEE 802.11 MAC based ad hoc networks, as well as MAC protocols for utilizing directional antennas and mobility track. He’s also working on network protocols for building hierarchical ad hoc networks. E-mail: xkx@cs.ucla.eduMario Gerla was born in Milan, Italy. He received a graduate degree in engineering from the Politecnico di Milano, in1966, and the M.S. and Ph.D. degrees in engineering from UCLA in 1970 and 1973, respectively. He joined the Faculty of the UCLA Computer Science Department in 1977. His research interests cover the performance evaluation, design and control of distributed computer communication systems; high speed computer networks; wireless LANs (Bluetooth); ad hoc wireless networks. He has been involved in the design, implementation and testing of wireless ad hoc network protocols (channel access, clustering, routing and transport) within the DARPA WAMIS, GloMo projects and most recently the ONR MINUTEMAN project. He has also carried out design and implementation of QoS routing, multicasting protocols and TCP transport for the Next Generation Internet. He is currently an associate editor for the IEEE Transactions on Networking. E-mail: gerla@cs.ucla.eduLantao Qi received her B.E. and M.S. from the department of computer science at Tianjin University China in 2003. From 2000 to 2003 she engaged in research programs in the field of computer networks. Her major research focuses on buffer management, DIFFServ networks as well as mobile ad hoc networking. She has published many technical papers in this field. He recently joined the Agricultural Bank of China. E-mail: ltqi@tju.edu.cnYantai Shu is a professor of the computer science department at Tianjin University, China. He received his B.E., M.S., and Ph. D. degree in the electronics engineering department at Tianjin University. From 1974 to 1991, he was working as a researcher in the Institute of Plasma Physics, Academia Sinica. He has been serving as the vice president of the university from 1993 to 1997. His current research interests are focused on computer communication networks, wireless networks, real-time systems, modeling and simulation. He is a member of the IEEE and the ACM. He has published more than 120 papers and contributed to one book. E-mail:ytshu@tju.edu.cn     

RED theory for quality of service provisioning in wireless communications

In this paper, we study the performance limit of a wireless communication system over a fading channel. The system under study consists of 1) a finite‐buffer discrete‐time queueing system on the link layer; and 2) a rate‐adaptive channel coding system on the physical layer. The objective of this paper is to analyze the relationship among data rate (R), packet error probability (E), and delay bound (D) under the interaction between the link layer and the physical layer. In our analysis, we consider three types of packet errors; that is, 1) packet drop due to full buffer; 2) packet drop due to delay bound violation; and 3) packet decoding error due to channel noise. We obtain an upper bound on the packet error probability. Furthermore, by minimizing the packet error probability over the transmission rate, we obtain an optimal rate control policy that guarantees the user‐specified data rate and delay bound. In the case of constant arrival, the optimal rate control policy results in a rate‐error‐delay triplet; then, by varying data rate and delay bound, we obtain rate‐error‐delay Pareto‐optimal surface, which serves as the performance limit of the system under study. Copyright © 2012 John Wiley & Sons, Ltd.     

An Improved Algorithm of Nonlinear RED Based on Membership Cloud Theory

Concerning on the shortcoming and com-plexity of Random early detection (RED) algorithm in net-work congestion control, a new RED algorithm based on the Hemi-Rise Cloud model (CRED) was proposed, non-linear packet loss strategy was used, and sensitivity and un-certainty of parameters were improved. As a result, queue length could be kept stable in the neighborhood of ref-erence value. Network congestion was well controlled and network resource was used effectively. The stability of the algorithm was studied and the experimental results showed that the proposed algorithm could improve the stability, and had better performance than the RED and Adaptive RED (ARED) algorithms.