On designing self-tuning controllers for AQM routers supporting TCP flows based on pole placement

This paper revisits the simple pole placement technique in the classical control theory, and exploits this technique to propose two kinds of controllers for active queue management (AQM) in Internet protocol (IP) routers: the self-tuning proportional controller based on pole placement (ST/spl I.bar/P/spl I.bar/PP) and the self-tuning proportional-plus-integral controller based on pole placement (ST/spl I.bar/PI/spl I.bar/PP). The damping ratio /spl xi/ and undamped natural frequency /spl omega//sub n/ can be appropriately chosen such that: 1) the transient response performance of the system is satisfied and 2) all the poles would lie in the left-half s-plane to guarantee the stability of the control system. The self-tuning controllers can assign proper intervals of /spl xi/ and /spl omega//sub n/ to achieve good AQM performance and thereby adapting the system to significant load changes very well. Furthermore, the ST/spl I.bar/PI/spl I.bar/PP controller can regulate the packet drop probability based on the knowledge of the instantaneous queue size, and clamp the steady value of the queue length to a specified reference value. We verify the effectiveness of these two controllers via OPNET simulation. Our simulation results show the following: 1) choosing appropriate /spl xi/ and /spl omega//sub n/ can successfully satisfy the transient response of the system and 2) when the network load changes, the ST/spl I.bar/P/spl I.bar/PP controller and the ST/spl I.bar/PI/spl I.bar/PP controller exhibit extremely short settling time.     

Using interval phase margin assignment to self-tune a PI AQM controller for TCP traffic

We propose a self-tuning PI (Proportional-Integral) controller for an AQM (Active Queue Management) router supporting TCP traffic in the Internet. Classical control theory is applied in the controller design to meet the phase margin specification in the frequency domain. By assigning a proper interval of the phase margin, we can achieve good AQM performance by making the control system adapt to dramatic load changes. Our self-tuning PI controller self-tunes only when there is a great change in the network environment that would cause the phase margin of the AQM control system to drift outside the specified interval. Based on the knowledge of the queue size, our PI controller can regulate the TCP source window size by adjusting the packet drop probability, thus clamping the steady queue size around a desirable target buffer occupancy. We demonstrate by OPNET^® simulations that with our self-tuning PI controller applied, the network exhibits a good transient behavior. A simple PID (Proportional-Integral-Derivative) controller design method is also provided.     

A genetic algorithm for the design of a fuzzy controller for active queue management

Active queue management (AQM) policies are those policies of router queue management that allow for the detection of network congestion, the notification of such occurrences to the hosts on the network borders, and the adoption of a suitable control policy. This paper proposes the adoption of a fuzzy proportional integral (FPI) controller as an active queue manager for Internet routers. The analytical design of the proposed FPI controller is carried out in analogy with a proportional integral (PI) controller, which recently has been proposed for AQM. A genetic algorithm is proposed for tuning of the FPI controller parameters with respect to optimal disturbance rejection. In the paper the FPI controller design methodology is described and the results of the comparison with random early detection (RED), tail drop, and PI controller are presented.     

基于粒子群优化的网络拥塞控制新算法

PI控制器常用于主动队列管理中,但参数整定上的试凑法具有盲目性,算法的瞬态性能也不够理想.本文推导了基于流体流理论的网络简化模型,基于该模型将集群智能中的改进粒子群优化算法(PSO)应用于PID控制器参数优化,定义了一个综合调节时间、上升时间、超调量、系统静态误差、正弦跟踪误差等动静态性能指标函数,在给定的参数空间进行组合优化搜索,迅速求得获取使性能指标优化函数极小化的一组PID控制器参数,将PID控制器应用于网络主动队列管理系统中.仿真结果表明,在大时滞和突发业务流的冲击两种情况下,该方法设计的控制器的动静态性能优于RED、PI算法,超调量均小于5%,调节时间分别小于5秒、4秒,稳态误差分别小于两个数据包和3个数据包.     

PAQM: an adaptive and proactive queue management for end‐to‐end TCP congestion control

Two functions, the congestion indicator (i.e. how to detect congestion) and the congestion control function (i.e. how to avoid and control congestion), are used at a router to support end‐to‐end congestion control in the Internet. Random early detection (RED) (IEEE/ACM Trans. Networking 1993; 1 (4):397–413) enhanced the two functions by introducing queue length averaging and probabilistic early packet dropping. In particular, RED uses an exponentially weighted moving average (EWMA) queue length not only to detect incipient congestion but also to smooth the bursty incoming traffic and its resulting transient congestion. Following RED, many active queue management (AQM)‐based extensions have been proposed. However, many AQM proposals have shown severe problems with detection and control of the incipient congestion adaptively to the dynamically changing network situations. In this paper, we introduce and analyse a feedback control model of TCP/AQM dynamics. Then, we propose the Pro‐active Queue Management (PAQM) mechanism, which is able to provide proactive congestion avoidance and control using an adaptive congestion indicator and a control function under a wide range of traffic environments. The PAQM stabilizes the queue length around the desired level while giving smooth and low packet loss rates and high network resource utilization. Copyright © 2004 John Wiley & Sons, Ltd.     

主动队列管理中的PID控制器

作为对终端系统上拥塞控制的一种补充,中间节点上的主动队列管理(AQM)策略在保证较高吞吐量的基础上有效地控制队列长度,从而实现了控制端到端的时延,保证QoS的目的。C.Hollot等人(2001)用经典控制理论中频域校正的方法设计了用于AQM的PI控制器,但参数整定上的试凑方法不免代有盲目性;算法的瞬态性能指标也不够理想。为此,该文引入了微分环节来增强系统的响应能力,同时给出了基于稳定裕度的参数整定方法,使PID控制器的稳定性有了绝对保障。仿真试验表明PID算法的调节时间远远短于PI控制器,从而为在负载瞬息万变的网络环境中实现控制分组排队等待时间的目标提供了有力的技术保障。     

基于模糊神经网络的拥塞控制算法研究

本文介绍一种基于模糊神经网络的主动队列管理(AQM)算法,实现网络拥塞控制。利用神经网络来实现模糊推理,可自适应修正隶属函数的参数和加权系数,优化模糊逻辑控制器,从而达到某种性能指标的最优化。仿真结果表明,采用模糊神经网络进行流量速率预测的拥塞控制策略能够使缓冲器队列长度快速收敛到目标值,并且维持小的队列震荡。结果也表明该方法与传统的PD控制器相比具有更好的性能和鲁棒性。     

主动队列管理算法性能评价：比较研究

对当前的主动队列管理算法进行了分类，然后基于NS2网络仿真器比较了它们在不同流量和网络拓扑条件下的多种性能指标。主要结论包括：ARED综合表现最好；PI和REM的瞬态性能较差，对动态流量响应较慢：REM和没有ECN支持的AVQ可以得到较小的时延，但同时牺牲了链路的带宽利用率并造成大量分组丢失；在多瓶颈网络拓扑中，大部分算法的性能有所提高。     

A compensated PID active queue management controller using an improved queue dynamic model

Beside the major objective of providing congestion control, achieving predictable queuing delay, maximizing link utilization, and robustness are the main objectives of an active queue management (AQM) controller. This paper proposes an improved queue dynamic model while incorporating the packet drop probability as well. By applying the improved model, a new compensated PID AQM controller is developed for Transmission Control Protocol/Internet Protocol (TCP/IP) networks. The non‐minimum phase characteristic caused by Padé approximation of the network delay restricts the direct application of control methods because of the unstable internal dynamics. In this paper, a parameter‐varying dynamic compensator, which operates on tracking error and internal dynamics, is proposed to not only capture the unstable internal dynamics but also reduce the effect of uncertainties by unresponsive flows. The proposed dynamic compensator is then used to design a PID AQM controller whose gains are obtained directly from the state‐space representation of the system with no further gain tuning requirements. The packet‐level simulations using network simulator (ns2) show the outperformance of the developed controller for both queuing delay stability and resource utilization. The improved underlying model leads also to the faster response of the controller. Copyright © 2013 John Wiley & Sons, Ltd.     

基于动态矩阵控制的主动队列管理算法

针对Internet系统,通过对流体流模型的分析,提出了一个新的预测模型.该模型形式简单,参数的计算相对容易,并且能根据当前的网络情况有效的预测拥塞窗口的变化.结合动态矩阵控制 (Dynamic Matrix Control,DMC) 理论,提出了一种新的主动队列管理算法——DMCAQM 算法,给出了DMCAQM 的详细设计过程,稳定性分析和参数选取原则.大量不同网络环境的仿真实验表明DMCAQM 算法是有效的.与PI、RaQ 和REM 等算法相比较,DMCAQM 有收敛速度快、队列抖动小的优点.同时,由于DMCAQM 的采样间隔相对较大,而算法实现简单,所以计算量小,占用的路由器资源更少.     

Nonlinear performance evaluation model for throughput of AQM scheme using full factorial design approach

The performance of active queue management (AQM) is measured in terms of throughput, delay, queue size, and loss rate. We have carried out the optimized performance measure of throughput for AQM scheme random early detection (RED) using full factorial design (FDD) technique that is a new approach of performance analysis particularly for congestion control algorithms. We have considered the input factors, viz, buffer size, maximum threshold, and the number of file transfer protocol (FTP) sources for the evaluation of RED that can be used for other AQM schemes, viz, adaptive RED, three‐section RED (TRED), and adaptive queue management with random dropping (AQMRD). The effect of each input factor as well as their interactions are evaluated using factorial design technique that results to obtain the nonlinear equation for performance measure in terms of input factors buffer size, maximum threshold, and the number of FTP sources. Finally, we show the contour plots for variation of performance measure throughput (steady state) from minimum to maximum values with respect to the different setting of input parameters.     

Relative loss rate differentiation: performance of short‐lived TCP flows

The relative differentiated service model provides assurances for the relative quality ordering between service classes, rather than for the actual service level in each class. In this paper, we describe a relative loss rate differentiation scheme where packet drop probabilities are determined according to an active queue management (AQM) mechanism based on random early detection (RED) in a first‐in first‐out (FIFO) queue, are weighted in inverse proportion to the price that the network operator assigns to each service class. Basically, we describe a scheme where relative loss rate differentiation is incorporated directly into AQM. Most TCP flows today, particularly Web flows, can be characterized as short‐lived flows. Using simulations with short‐lived TCP flows, we show that the scheme is very effective in ensuring relative loss rate differentiation between service classes during times of network congestion. Copyright © 2004 John Wiley & Sons, Ltd.     

Rate‐based proportional–integral control scheme for active queue management

Most high‐speed links do not have adequate buffering and as a result Active Queue Management (AQM) schemes that utilize queue size information for congestion control cannot be effectively applied on these links. A high‐speed link will, typically, have small buffers in relation to the bandwidth‐delay product of the link. In this paper we argue that rate‐based AQM schemes be used for such links. The goal here is to match the aggregate rate of the active TCP connections to the available capacity while maintaining minimal queue size and high link utilization. The AQM scheme described here employs a Proportional–Integral (PI) control strategy and explicitly takes into account the time delay in the control process. Copyright © 2006 John Wiley & Sons, Ltd.     

A novel AQM algorithm based on feedforward model predictive control

Developing feedforward model predictive controller as an active queue management (AQM) scheme is studied in this paper. MPC is an advanced control strategy for AQM. However, the conventional MPC is usually an implementable form of feedback MPC. In this paper, a feedforward and feedback optimal control law is presented. It is a clean, easily implementable, version of model predictive control that incorporates feedforward. Firstly, we use the nominal fluid model to design the feedforward control input so that the output tracks the given queue length with small error. Furthermore, in order to achieve robust performance and to reject the (unmeasured) disturbance, the feedback component is designed. In particular, a disturbance observer is incorporated into the prediction output in standard feedback MPC. This framework can significantly improve performance in the presence of measurement noise and certain types of model uncertainty. Finally, the simulation results show the effectiveness of FF‐AQM algorithm.     

Advances in Active Queue Management (AQM) Based TCP Congestion Control

Current end-to-end Internet congestion control under tail-drop (TD) queue management experiences performance degradations such as multiple packet losses, high queueing delay and low link utilization. In this paper, we review recently proposed active queue management (AQM) algorithms for supporting end-to-end transmission control protocol (TCP) congestion control. We focus recently developed control theoretic design and analysis method for the AQM based TCP congestion control dynamics. In this context, we analyze the problems of existing AQM proposals in which congestion is detected and controlled reactively based on current and/or past congestion. Then we argue that AQM based TCP congestion control should be adaptive to the dynamically changing traffic situation in order to detect, control and avoid the current and the incipient congestion proactively. Finally, we survey two adaptive and proactive AQM algorithms, PID-controller and Pro-Active Queue Management (PAQM), designed using classical proportional-integral–derivative (PID) feedback control to overcome the reactive congestion control dynamics of existing AQM algorithms. A comparative study of these AQM algorithms with existing AQM algorithms is given. A simulation study under a wide range of realistic traffic conditions suggests that PID-controller and PAQM outperform other AQM algorithms such as random early detection (RED) [Floyd and Jacobson, 18] and proportional-integral (PI) controller [Hollot et al., 24].     

A stochastic approximation approach to active queue management

Recently, a dynamic adaptive queue management with random dropping (AQMRD) scheme has been developed to capture the time-dependent variation of average queue size by incorporating the rate of change of average queue size as a parameter. A major issue with AQMRD is the choice of parameters. In this paper, a novel online stochastic approximation based optimization scheme is proposed to dynamically tune the parameters of AQMRD and which is also applicable for other active queue management (AQM) algorithms. Our optimization scheme significantly improves the throughput, average queue size, and loss-rate in relation to other AQM schemes.     

PD-RED: to improve the performance of RED

We propose a new active queue management (AQM) scheme to improve the performance of the well-known random early detection (RED) AQM. The new AQM is based on the proportional derivative (PD) control principle, and we call it PD-RED. In PD-RED we introduce minimal changes to RED. We demonstrate the improvement in performance of PD-RED over adaptive RED AQM by simulations.     

Proxy‐RED: an AQM scheme for wireless local area networks

Wireless access points (APs) act as bridges between wired and wireless networks. Since the actually available bandwidth in wireless networks is much smaller than the bandwidth in wired networks, there is a disparity in channel capacity which makes the access point a significant network congestion point in the downstream direction. A current architectural trend in wireless local area networks (WLAN) is to move functionality from APs to a centralized gateway in order to reduce cost and improve features. In this paper, we study the use of RED, a well known active queue management (AQM) scheme, and explicit congestion notification (ECN) to handle bandwidth disparity between the wired and the wireless interface of an access point. Then, we propose the Proxy‐RED scheme, as a solution for reducing the AQM overhead from the access point. Simulations‐based performance analysis indicates that the proposed Proxy‐RED scheme improves the overall performance of a network. In particular, the Proxy‐RED scheme significantly reduces packet loss rate and improves goodput for a small buffer, and minimizes delay for a large buffer size. Copyright © 2006 John Wiley & Sons, Ltd.     

Adaptive fuzzy sliding mode active queue management algorithms

Active queue management (AQM) is aimed at achieving the tradeoff between link utilization and queuing delay to enhance TCP congestion control and is expected to perform well for a wider-range of network conditions. Static AQM schemes despite their simplicity, often suffer from long response time due to conservative parameter setting to ensure stability. Adaptive parameter settings, which might solve this problem, remain difficult from implementation point of view. In this paper, we propose an adaptive fuzzy sliding mode (AFSM) AQM algorithm to achieve fast response and yet good robustness. The AFSM algorithm uses the queue length and its differential as the input of AQM and adjusts fuzzy rules by the measurement of packet loss ratio dynamically. The stability analysis under heterogeneous round trip times provides guidelines for parameter settings in AFSM and guarantees that the stability of AFSM is independent of the active TCP flows. This merit as well as other performances is examined under various network environments. Compared to some typical AQMs, the AFSM algorithm trades off the throughput with queuing delay better and achieves a higher per-flow throughput. Finally, AFSM can be executed at a scale of seconds with the least fuzzy rules.     

一种新的快速响应的AQM算法

该文应用控制理论,设计了具有快速响应特征的比例积分(Proportional-Integral, PI)控制器,并以此提出了一个新的主动队列管理(Active Queue Management, AQM)算法,称为Active PI算法(简称A-PI)。A-PI用一组参数实现快速响应,用另一组参数保证队长稳定,使用门限值实现两组参数间平滑切换,算法实现简单。理论分析和仿真实验表明,A-PI算法在保证队长稳定的同时实现了对不同网络环境的快速响应。