Elsevier

Automatica

Volume 47, Issue 1, January 2011, Pages 26-38
Automatica

A robust adaptive congestion control strategy for large scale networks with differentiated services traffic

https://doi.org/10.1016/j.automatica.2010.08.019Get rights and content

Abstract

In this paper, a robust decentralized congestion control strategy is developed for a large scale network with Differentiated Services (Diff-Serv) traffic. The network is modeled by a nonlinear fluid flow model corresponding to two classes of traffic, namely the premium traffic and the ordinary traffic. The proposed congestion controller does take into account the associated physical network resource limitations and is shown to be robust to the unknown and time-varying delays. Our proposed decentralized congestion control strategy is developed on the basis of Diff-Serv architecture by utilizing a robust adaptive technique. A Linear Matrix Inequality (LMI) condition is obtained to guarantee the ultimate boundedness of the closed-loop system. Numerical simulation implementations are presented by utilizing the QualNet and Matlab software tools to illustrate the effectiveness and capabilities of our proposed decentralized congestion control strategy.

Introduction

The congestion control problem is of paramount importance in communication networks specially given the growing need for speed (bandwidth), size, load, and connectivity of the increasingly integrated services. This fact has necessitated the design and utilization of new network architectures by including more effective congestion control algorithms in addition to the standard TCP-based congestion control schemes. Specifically, the Internet Engineering Task Force (IETF) has proposed the Differentiated Services (Diff-Serv) architecture (Diff-Serv, 2009) to deliver aggregated quality of service (QoS) in IP networks. In the Diff-Serv architecture the traffic is aggregated into different classes of flows and the bandwidth allocation and the packet dropping rules are applied to the traffic classes according to their QoS requirements and specifications.

For the TCP/IP networks, a number of congestion control design techniques have been proposed in the literature (Baines et al., 2000, Floyd and Jacobson, 1993, Lee et al., 2006), which have shown excellent performance and were demonstrated to be robust in a variety of scenarios. However, the current TCP-based congestion control mechanisms cannot adequately address simultaneously the congestion problem and achieve fairness among traffic aggregates within the Diff-Serv networks (Floyd and Jacobson, 1993, Lee et al., 2006). It has been recognized that generally scaling up the existing congestion control approaches that use ad hoc techniques and intuitive methods are not formal in nature and are indeed problematic even with a number of proposed tuning solutions. Furthermore, by simply relying on the TCP congestion control algorithms, the service QoS requirements that are expected from the differentiated services traffic cannot be fully realized (Baines et al., 2000). This problem is more challenging for large scale networks that need to operate under constraints such as bandwidth limitations, real-time requirements, latency management, unknown and time-varying delays, and specially when the resources are not effectively controlled. Therefore, development of new congestion control schemes for large scale Diff-Serv networks is critically needed.

An “ideal” congestion control mechanism must be able to simultaneously satisfy the QoS specifications of the aggregate traffics in addition to congestion avoidance. A large body of work is available on the performance metrics of congestion control algorithms utilized in the literature (Dukkipati and McKeown, 2006, Floyd, 2008, Jain et al., 1984). To model and analyze these performance metrics, such as network throughput, queueing delay, and packet loss rate in a formal, quantitative, and analytical manner is not an easy task, since their effects on the congestion are significantly nonlinear. For this reason, the congestion control problem may become unmanageable unless effective, robust, and decentralized methods are developed. The development of such effective congestion control protocols will require integration of advanced networking and control techniques.

The control systems community has shown a growing interest in addressing the challenges in the area of congestion control. Since the congestion control concept that was introduced in Jacobson (1988), several attempts at control theoretic-based schemes have been made in the literature by using approaches such as optimal control (Segall, 1997), linear control (Kolarov & Ramamurthy, 1997), fuzzy and neural control (Chrysostomou et al., 2003, Ha and Hoang, 2005), predictive adaptive control (Pitsillides & Lambert, 1997), and nonlinear control techniques (Su et al., 2000, Wu et al., 2001). Despite these efforts, formal, quantitative, and analytical investigation of performance of large scale networks with Diff-Serv traffics have not been fully addressed and resolved.

Several new congestion control schemes for Diff-Serv networks whose performance can be analytically established have been presented in the literature by using sliding mode control (Zhang et al., 2008, Zhang et al., 2009) and robust adaptive control (Pitsillides, Ioannou, Lestas, & Rossides, 2005) techniques. The results developed in these works are quite interesting. However, the above solutions have also serious drawbacks. First, the nature of discontinuities of the sliding mode controller may result and introduce unavoidable and undesirable oscillations in the closed-loop system (Utkin & Hoon, 2006), and therefore reduce the effectiveness of the developed congestion control solutions. On the other hand, the approach in Pitsillides et al. (2005) is designed for only a cascade network of switches and considered the bottleneck switch as a single node. Consequently, the presence of unknown and time-varying delays and latencies are not considered in the design of the congestion control scheme. The lack of explicit consideration of the delays will yield a critical challenge and even an instability when the approach is applied to a large scale network consisting of many nodes structured in arbitrary configurations containing feedback (Bouyoucef and Khorasani, 2009, Bouyoucef and Khorasani, 2007, Chen and Khorasani, 2007, Chen and Khorasani, 2010).

The main objective of this paper is to extend the robust congestion control strategy that was proposed in Pitsillides et al. (2005) corresponding to a cascade network of nodes with differentiated services traffic to a large scale network of fully connected nodes in arbitrary configurations. Our proposed congestion control strategy is designed based on nonlinear and adaptive control methodologies. A fluid flow model is developed where the controller is designed in a decentralized manner. This will ensure that the proposed congestion control solutions are feasible to be implemented and scaled up to large scale networks. The justifications and rationale for selecting the fluid flow model and the extension of it to a large scale Diff-Serv network is given in the next section.

The main contributions of this paper can be summarized as follows.

  • (1)

    Extension of the model of a single node system to a model of large scale networks where the inter-node traffic is considered explicitly with unknown and time-varying delays and latencies.

  • (2)

    Proposed a novel decentralized congestion control scheme for a large scale Diff-Serv network. The stability of the closed-loop system that is obtained by applying our proposed controller is guaranteed formally in Theorem 1 and Lemma 1, Lemma 2. We have shown through simulations that improved performance is obtained when compared to the existing approaches in the literature. Our proposed congestion control scheme is quite general and can be applied to a wide range of communication networks with Diff-Serv traffic.

  • (3)

    Unlike other work in the literature, we allow and consider traffic data compressions in designing the congestion control scheme. Optimal values of the traffic compression rates are obtained (or equivalently the minimum allowable compression rates) in order to maintain and guarantee that the overall network congestion control specifications and requirements are satisfied.

The remainder of this paper is organized as follows. In Section 2, the representation of a large scale network that is embedded with Diff-Serv traffic flows is introduced. The formulation of our proposed congestion control methodology as applied to these networks is then presented. Before introducing our proposed decentralized robust congestion control scheme, necessary preliminary results are provided in Section 3. Our novel congestion control strategies for the premium and the ordinary traffic flows are proposed and developed in Section 4, and the stability and convergence properties of the closed-loop system subject to unknown and time-varying delays are investigated. In order to evaluate the performance and robustness of our proposed decentralized control strategies, comprehensive simulations are performed in Section 5 by using the commercially available software tools QualNet (Little, 2009) and Matlab. Finally, conclusions are stated in Section 6.

Section snippets

Fluid flow model of a large scale diff-serv network

As stated above, an “ideal” congestion control mechanism should be able to simultaneously satisfy the QoS specifications of the aggregate traffics in addition to congestion avoidance. The main QoS specifications of the aggregate flows and the performance metrics of any congestion control scheme should include both the node throughput as well as the delay and the packet loss rates (Dukkipati and McKeown, 2006, Floyd, 2008). Given the trade-offs among the performance metrics it is clearly

Preliminary results

Before formally presenting our congestion control strategies a couple of preliminaries that are needed for our subsequent discussions are briefly introduced below. The first background information concerns the definition of and the notion of a uniformly ultimately bounded nonlinear system. The second preliminary result concerns the derivation of stability conditions of switched time-delay systems. As shown in the next section, the closed-loop congestion controlled system belongs to this class

Proposed decentralized robust congestion control strategy

Consider a large scale network with N nodes. Suppose each node has three queues corresponding to the premium, the ordinary and the best-effort traffic. The congestion controller is implemented at the output port of each node, as shown in Fig. 1. The control strategy adopts the Diff-Serv framework that was originally introduced in Diff-Serv (2009). The control objective pursued for the premium traffic is to allocate the output capacity, that is denoted by Cp,i(t), by incorporating an adaptive

Performance evaluations and simulation results

In order to evaluate and quantify the performance of our proposed control strategies, a number of simulations are performed and comparative results are provided in this section. We use a network that consists of a number of randomly distributed nodes with more than one bottleneck link. The details on the simulation model and the selected control parameters are described next.

(1) Simulation model: Our simulation model is shown in Fig. 2. This network consists of 3 clusters where each cluster has

Conclusions

In this paper, a decentralized robust adaptive congestion control strategy for differentiated services (Diff-Serv) traffic in large scale network is proposed. A dynamic queueing model corresponding to the premium and the ordinary traffic are obtained by using a fluid flow model and conservation principles. The LMI conditions that facilitate design of the controller parameters as well as the network traffic compression/transmission gains are derived. These conditions are shown formally to

R.R. Chen received the B.S. and M.S. degrees in electrical engineering from the Beijing Institute of Technology, Beijing, China, in 2000 and 2003, respectively. She is currently working towards her Ph.D. degree in the Department of Electrical and Computer Engineering, Concordia University, Montreal, Canada. Her research interests are in congestion control, robust adaptive control, networked multi-agent systems, time-delay system, and switching systems.

References (44)

  • Bouyoucef, K., & Khorasani, K. (2007). A sliding mode-based congestion control for time delayed differentiated-services...
  • Chen, R.R., & Khorasani, K. (2007). An adaptive congestion control technique for networked control systems. In Proc. of...
  • Chen, R.R., & Khorasani, K. (2010). A Markovian jump congestion control strategy for mobile ad-hoc networks with...
  • C. Chrysostomou et al.

    Fuzzy logic controlled RED: congestion control in TCP/IP differentiated services networks

    Journal Soft Computing

    (2003)
  • S. Deb et al.

    Rate-based versus queue-based models of congestion control

    ACM Transactions on Modeling and Computer Simulation

    (2006)
  • http://www.ietf.org/html.charters/OLD/diffserv-charter.html (accessed Feb....
  • Dukkipati, N., & McKeown, N. (2006). Why flow-completion time is the right metric for congestion control. ACM...
  • J. Filipiak

    Modeling and control of dynamic flows in communication networks

    (1988)
  • Floyd, S. (2008). Metrics for the evaluation of congestion control mechanism. RFC 5166. Informational, Networking...
  • S. Floyd et al.

    Random early detection gateways for congestion avoidance

    IEEE/ACM Transaction on Networking

    (1993)
  • E. Fridman

    Descriptor discretized Lyapunov functional method: analysis and design

    IEEE Transactions on Automatic control

    (2006)
  • E. Fridman et al.

    An improved stabilization method for linear time-delay systems

    IEEE Transactions on Automatic Control

    (2002)
  • Cited by (25)

    • Robust congestion control in cognitive radio network using event-triggered sliding mode based on reaching laws

      2020, Journal of the Franklin Institute
      Citation Excerpt :

      To obtain a better result, second-order sliding mode control has been used in [18] which resulted in an increase in system complexity. Although various attempts are made in [7,13,18] for control congestion yet controller updates are continuous that leads to indispensable over taxing of computation and actuation requirements. In the context of resources limitation in wireless network, an adaptive algorithm to improve QoS has been presented in [19] without considering heterogeneous users in the dynamic network.

    • Congestion control in cognitive radio networks with event-triggered sliding mode

      2018, AEU - International Journal of Electronics and Communications
    • Markovian jump guaranteed cost congestion control strategies for large scale mobile networks with differentiated services traffic

      2014, Automatica
      Citation Excerpt :

      It should also be noted that the results in Chen and Khorasani (2011b) are applicable to only fixed topology networks, whereas the results in the present work are developed and generalized to switching and mobile large scale networks. This is of the main contributions and extensions of our work as compared to those that we have already reported in Chen and Khorasani (2010, 2011a,b). The changes and switchings in the network topologies are modeled by a Markov chain and the dynamics of the mobile network are represented by a nonlinear time-delay system with Markovian jump parameters.

    • Hopf bifurcation control in the XCP for the Internet congestion control system

      2012, Nonlinear Analysis: Real World Applications
      Citation Excerpt :

      The nonlinear dynamics of congestion control system motivates researchers to improve the performance of AQM schemes by existing bifurcation control methods. Subsequently, there have been many papers which have addressed nonlinear behavior such as bifurcation and chaos in models of network systems [12–44]. The rest of this paper is organized as follows.

    View all citing articles on Scopus

    R.R. Chen received the B.S. and M.S. degrees in electrical engineering from the Beijing Institute of Technology, Beijing, China, in 2000 and 2003, respectively. She is currently working towards her Ph.D. degree in the Department of Electrical and Computer Engineering, Concordia University, Montreal, Canada. Her research interests are in congestion control, robust adaptive control, networked multi-agent systems, time-delay system, and switching systems.

    K. Khorasani received the B.S., M.S., and Ph.D. degrees in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign in 1981, 1982 and 1985, respectively. From 1985 to 1988 he was an Assistant Professor at the University of Michigan at Dearborn and since 1988, he has been at Concordia University, Montreal, Canada, where he is currently a Professor and Concordia University Tier I Research Chair in the Department of Electrical and Computer Engineering. His main areas of research are in nonlinear and adaptive control, intelligent and autonomous control of networked unmanned systems, fault diagnosis, isolation and recovery (FDIR), neural network applications to pattern recognition, robotics and control, adaptive structure neural networks, and modeling and control of flexible link/joint manipulators. He has authored/co-authored over 300 publications in these areas.

    This research is supported in part by the Discovery Grants program of the Natural Sciences and Engineering Research Council of Canada (NSERC). This paper was not presented at any IFAC meeting. This paper was recommended for publication in revised form by Associate Editor Hideaki Ishii under the direction of Editor Ian R. Petersen.

    View full text