Reducing Queue Oscillation at a Congested Link

Queue length oscillation at a congested link causes many undesirable properties such as large delay jitter, underutilization of the link and packet drops in burst. The main reason of this oscillation is that most queue management schemes determine the drop probability based on the current traffic without consideration on the impact of that drop probability on the future traffic. In this paper, we propose a new active queue (AQM) scheme to reduce queue oscillation and realize stable queue length. The proposed scheme measures the current arrival and drop rates, and uses them to estimate the next arrival rate. Based on this estimation, the scheme calculates the drop probability which is expected to realize stable queue length. We present extensive simulation with various topologies and offered traffic to evaluate performance of the proposed scheme. The results show that the proposed scheme remarkably reduces queue length oscillation compared to other well-known AQMs. It is also shown that the proposed scheme improves fairness among TCP flows due to the stable drop probability, and maintains high utilization with small queue length.     

几种主动队列管理算法的仿真及性能评估

采用仿真分析的方法,系统地研究了当前提出的用于Internet路由器缓冲管理的一系列主动队列管理(AQM)算法的性能。根据AQM的设计原理,将当前的AQM算法划分为3类：基于队列长度,基于网络负载和同时基于队列长度和网络负载的AQM算法。仿真研究和分析表明,现有的AQM算法不能适应网络流量的动态变化。     

Stochastic controller as an active queue management based on B-spline kernel observer via particle swarm optimization

Given the fact that the current Internet is getting more difficult in handling the traffic congestion control, the proposed method is compatible with the stochastic nature of network dynamics. Most conventional active queue management is based on the first stochastic moment. In stochastic theory, the first moment is not efficient for non-Gaussian systems that are the same as the network queue size. We propose a new stochastic active queue management technique, based on stochastic control and B-spline window observer, called intelligent probability density function AQM (IPDF-AQM). The IPDF-AQM is based on a PDF control and particle swarm optimization, which not only considers the average queue length at the current time slot, but also takes into consideration the PDF of queue lengths within a round-trip time. We provide a guideline for the selection of the probability of dropping as control input for TCP/AQM system to make the PDF of queue length converge at a certain PDF target based on B-spline approximation and improve the network performance. Simulation results show that the proposed stochastic AQM scheme does improve the end-to-end performance.     

基于模型算法控制的主动队列管理算法

通过深入分析TCP/AQM系统的动态特性,提出一个新的预测模型。基于该模型,结合模型算法控制(MAC)提出一种新的主动队列管理算法(MACAQM),并给出了MACAQM的详细设计过程和参数选取的原则。大量不同网络环境的仿真实验表明了MACAQM算法的有效性。与PI, RaQ和REM等算法相比较, MACAQM具有收敛速度快、队列抖动小的优点。同时, MACAQM的采样间隔相对较大,算法实现简单,所以计算量较小,占用的路由器资源也较少。     

不确定TCP流模型的离散H∞鲁棒主动队列管理算法

针对TCP/IP网络存在参数时变和不确定性下的拥塞控制问题,提出一种新的基于H∞状态反馈控制的离散鲁棒主动列队管理算法(AQM).该方法针对不确定TCP流模型,将短期突发流所占据的带宽作为系统的外部干扰,同时考虑时滞和参数不确定性因素,基于Lyapunov稳定性理论和线性矩阵不等式技术,设计了离散鲁棒状态反馈控制器以保证路由器队列响应的稳定性和鲁棒性.最后,通过NS-2仿真验证了本文方法的有效性.     

Design and analysis of a self-tuning feedback controller for the Internet

Active queue management (AQM) is an effective method used in Internet routers for congestion avoidance, and to achieve a tradeoff between link utilization and delay. The de facto standard, the random early detection (RED) AQM scheme, and most of its variants use average queue length as a congestion indicator to trigger packet dropping. This paper proposes a novel packet dropping scheme, called self-tuning proportional and integral RED (SPI-RED), as an extension of RED. SPI-RED is based on a self-tuning proportional and Integral feedback controller, which considers not only the average queue length at the current time point, but also the past queue lengths during a round-trip time to smooth the impact caused by short-lived traffic dynamics. Furthermore, we give theoretical analysis of the system stability and give guidelines for selection of feedback gains for the TCP/RED system to stabilize the average queue length at a desirable level. The proposed method can also be applied to the other variants of RED. Extensive simulations have been conducted with ns2. The simulation results have demonstrated that the proposed SPI-RED algorithm outperforms the existing AQM schemes in terms of drop probability and stability.     

Multiclass AQM on a TCP/IP router: A control theory approach

Active queue management (AQM) is a well‐known technique to improve routing performance under congested traffic conditions. It is often deployed to regulate queue sizes, thus aiming for constant transmission delay. This work addresses AQM using an approach based on control theory ideas. Compared with previous results in the literature, the novelty is the consideration of heterogeneous traffic, ie, multiclass traffic. Thus, each traffic class may have different discarding policies, queue sizes, and bandwidth share. This feature brings the proposal nearer to real network management demands than previous approaches in the literature. The proposed technique assumes that each class already has a simple controller, designed a priori, and focuses on designing a static state‐feedback controller for the multiclass system, where the design is based on using LMIs for the calculations. For this, optimization problems with LMI constraints are proposed to compute the state‐feedback gains that ensure stability for a large set of admissible initial conditions. These conditions ensure not only closed‐loop stability but also some level of performance. As far as we know, this is the first control theory based approach for the AQM problem on TCP/IP routers that allows a multiclass AQM while also considering time‐varying delays and input saturation. This is an important step to frame AQM in a more formal, yet realistic context, enabling it to address important service level agreement (SLA) directives. The proposal is tested on a simulated system at the end of this paper, showing the feasibility and performance of the approach in the presence of multiclass traffic.     

Managing Internet routers congested links with a Kohonen-RED queue

The behaviour of the TCP AIMD algorithm is known to cause queue length oscillations when congestion occurs at a router output link. Indeed, due to these queueing variations, end-to-end applications experience large delay jitter. Many studies have proposed efficient active queue management (AQM) mechanisms in order to reduce queue oscillations and stabilize the queue length. These AQM attempt to improve the random early detection (RED) model. Unfortunately, these enhancements do not react in a similar manner for various network conditions and are strongly sensitive to their initial setting parameters. Although this paper proposes a solution to overcome the difficulties of configuring the RED parameters by using a Kohonen neural network model; another goal of this study is to investigate whether cognitive intelligence could be placed in the core network to solve such stability problem. In our context, we use results from the neural network area to demonstrate that our proposal, named Kohonen-RED (KRED), enables a stable queue length without complex parameters setting or passive measurements to obtain a correct configuration.     

一种基于双模控制的主动队列管理新算法

随机指数标记算法(REM)是一种有效的主动队列管理算法,但由于TCP/IP网络流量模型呈现非线性特性,故而其控制效果不佳,存在队列稳定性差,对动态流量响应慢等问题.为了解决上述问题,提出了基于双模控制的主动队列管理算法(Fuzzy-REM).算法采用分段控制策略,在瞬时队列偏差大于阈值时,采用模糊控制,反之采用REM控制,从而将模糊控制的快速响应和REM稳态性能好的优点结合起来.NS2中的仿真实验表明,相对于REM算法,Fuzzy-REM提高了队列稳定性,加快了收敛速度,增强了算法对网络环境变化的适应性.     

Anti-windup compensator for active queue management in TCP networks

In this paper, we apply a dynamic anti-windup scheme for improving the performance of a conventional proportional–integral (PI) controller for active queue management (AQM) supporting TCP flows. When a PI controller is used for AQM, the windup phenomenon of the integral action can cause performance degradation because the packet drop probability is limited between 0 and 1. Therefore we suggest a TCP/AQM model with a saturating actuator and apply a dynamic anti-windup method for improving the performance of the conventional PI AQM scheme. The proposed scheme not only provides graceful performance degradation, but also guarantees the stability of the overall system with the linearized TCP model. We verify the performance of the proposed scheme through ns-2 simulations. The simulation results show that our scheme outperforms the conventional PI controller when the traffic load is not stationary, which is always the case in real network environment.     

基于传输控制协议的无线自组网主动队列管理建模

主动队列管理（AQM）通常研究队列控制器的设计．作为被控对象,传输控制协议（TCP）往往利用网络仿真器（NS）的仿真实现,因此有必要研究无线自组网的TCP及AQM特性．基于TCP窗口加性增一乘性减算法及排队原理,推导了TCP窗口及队列的微分方程,再基于比例积分AQM控制,推导了拥塞丢弃概率的微分方程,通过建立联立微分方程组,提出了AdHoc网络TCP／AQM微分模型．对比仿真显示,新模型能较好地估计无线白组网的性能．模型研究也表明,网络跳数,无线丢失和过小的队列成为AQM性能瓶颈,队列信息则有助于TCP区分无线自组网的拥塞丢弃与无线丢失．     

LRED: A Robust and Responsive AQM Algorithm Using Packet Loss Ratio Measurement

Active queue management (AQM) is an effective means to enhance congestion control, and to achieve trade-off between link utilization and delay. The de facto standard, random early detection (RED), and many of its variants employ queue length as a congestion indicator to trigger packet dropping. Despite their simplicity, these approaches often suffer from unstable behaviors in a dynamic network. Adaptive parameter settings, though might solve the problem, remain difficult in such a complex system. Recent proposals based on analytical TCP control and AQM models suggest the use of both queue length and traffic input rate as congestion indicators, which effectively enhances stability. Their response time generally increases however, leading to frequent buffer overflow and emptiness. In this paper, we propose a novel AQM algorithm that achieves fast response time and yet good robustness. The algorithm, called Loss Ratio-based RED (LRED), measures the latest packet loss ratio, and uses it as a complement to queue length for adaptively adjusting the packet drop probability. We develop an analytical model for LRED, which demonstrates that LRED is responsive even if the number of TCP flows and their persisting times vary significantly. It also provides a general guideline for the parameter settings in LRED. The performance of LRED is further examined under various simulated network environments, and compared to existing AQM algorithms. Our simulation results show that, with comparable complexities, LRED achieves shorter response time and higher robustness. More importantly, it trades off the goodput with queue length better than existing algorithms, enabling flexible system configurations     

自适应卡尔曼滤波的主动队列管理算法

PID控制器通过微分环节加快了控制器的调节速度,但PID的参数是固定的,不能根据动态的网络自调整参数,故不能有效控制队列的稳定性。由于神经元网络有自适应性,提出了一种自适应卡尔曼滤波的主动队列管理算法(adaptive-KF-AQM)。它结合卡尔曼滤波和神经元网络方法,根据队列长度及其变化率来估计下一时刻的队列长度,使队列长度在期望值附近波动。仿真结果表明,该算法在队列稳定性、收敛速度、延时和链路利用率等方面都明显优于传统的PID算法。     

基于时滞补偿的网络拥塞控制策略*

中间节点上的主动队列管理策略在保证较高吞吐量的基础上能有效控制队列长度和端到端时延,利用频率域模型降阶拟合方式建立了TCP流量控制中主动队列管理系统的等效模型,应用控制理论中的内模补偿原理设计鲁棒的延时补偿主动队列管理控制算法,克服了大时滞给对队列稳定造成的不利影响.仿真结果表明,该补偿方法在长时滞小期望队列综合性能明显优于已有的RED,REM,PI等算法,链路利用率大大提高.     

应用于Ad Hoc网络TCP/AQM的神经网络PID控制器优化设计

主动队列管理(AQM)PID(Proportional integral derivative)算法的控制效果取决于比例、积分及微分系数的整定,但传统整定往往基于试凑方法和经验知识。根据Ad Hoc网络参量时变的特点,推导Ad Hoc网络的TCP/AQM模型,利用遗传算法动态调整RBF(Radial Basis Function)神经网络PID控制器系数,提出基于遗传算法的RBF神经网络PID-AQM。仿真表明,相较RBF-PID,新算法在信道状态复杂的Ad Hoc网络健壮性更好,并具有较好的队列控制效果。     

一种链路负载自适应的主动队列管理算法

随机早检测(random early detection,简称RED)是IETF推荐部署的主动队列管理(active queue management,简称AQM)算法.RED存在参数难以配置、无法得到与流量无关的平均队长等问题.ARED(adaptive RED)是RED的自适应版本,它根据平均队长动态调节最大标记概率参数,从而得到稳定的平均队长.但ARED没有克服瞬时队列长度振荡问题,且在动态流量环境下性能明显降低.分析了ARED性能问题的原因,并提出了一种链路负载自适应的主动队列管理算法LARED(load adaptiveRED).LARED具有两个特点:自适应链路负载、快速响应队长变化.分析和仿真实验表明,与ARED等其他AQM算法相比,LARED在保持高链路利用率和低时延的同时可以得到稳定的瞬时队长,并且具有良好的响应性和鲁棒性.     

TCP/AQM系统大时间尺度模型

网络传输连接的往返时延(round-trip times,RTT)大小各不相同,因此TCP/AQM系统本质为一多时滞回路耦合系统.由于RTT分布范围远大于控制量调节周期,这给准确评估控制效果带来很大困难.已有基于控制理论的主动队列管理(active queue management,AQM)算法多以流体流模型为基础进行设计,没有充分考虑RTT和采样周期对系统性能的影响.对于TCP/AQM系统,合理的评价方法是对调节过程进行评价,而非仅评价单个采样周期内的控制量是否合适.本文结合数据驱动控制思想和系统自身特征,统一从路由视角对TCP与AQM之间的交互进行抽象,通过时间扩展从更大的时间尺度去评价控制量调节过程,然后基于此模型设计自适应AQM算法–—大时间尺度AQM算法(large time scale AQM,LTSAQM).仿真结果表明,该算法收敛速度快,排队时延抖动小,特别是在长时滞网络环境下,性能明显改善.     

Design of Adaptive PI Rate Controller for Best-Effort Traffic in the Internet Based on Phase Margin

In this paper, we propose an adaptive PI (proportional-integral) rate controller for the AQM (active queue management) router that would support best-effort traffic in the Internet. Unlike most window-based controllers, our rate-based controller design is derived from the classical control theory and it would allow the users to achieve good stability robustness of the AQM control system by specifying a proper phase margin. We also make our controller adaptive by selecting a simple heuristic parameter to monitor the network environment real-time so that the controller would self-tune only when a dramatic change of the network traffic has drifted the monitoring parameter outside its specified interval. Located in the router, the adaptive PI rate controller calculates desirable source window sizes (i.e., source sending rates) based on the instantaneous queue length of the buffer and advertises it to the sources. Our simulations demonstrate that our AQM control system can adapt very well to sudden changes in network environment, thus providing the network with good transient behavior. By making the source sending rate relatively smooth, our adaptive PI rate controller becomes quite suitable for streaming media traffic control in the Internet     

随机早期检测主动队列管理算法的改进研究

主动队列管理（Active Queue Management）算法是近几年网络研究的重点。为了改进和完善现有的AQM算法和设计更好的新算法,需要对主动队列管理机制的实现方法和性能特性进行深入研究,详细讨论了随机早期检测（Random Early Detection）主动队列管理机制的关键技术问题,研究了近年来对RED算法的主要改进算法,总结了这几种算法的优缺点及其有待改进之处,并应用网络仿真器（Network Simulator）对实际网络模型进行了仿真实验,检验了改进算法对网络性能的影响。     

基于PID神经元网络和内模控制的拥塞控制算法*

针对网络系统的大时滞和非线性特性,设计了一种新的拥塞控制算法,将PID神经元网络与内模控制相结合应用于主动队列管理中,并使用Lyapunov理论证明了此算法的稳定性。NS仿真结果表明,这种算法的稳态和瞬态性能都优于PID算法,并且在参数变化和负载扰动时具有很强的鲁棒性。