Internet Traffic Control Using Dynamic Neural Networks

Active Queue Management (AQM) has been widely used for congestion avoidance in Transmission Control Protocol (TCP) networks. Although numerous AQM schemes have been proposed to regulate a queue size close to a reference level, most of them are incapable of adequately adapting to TCP network dynamics due to TCP's non-linearity and time-varying stochastic properties. To alleviate these problems, we introduce an AQM technique based on a dynamic neural network using the Back-Propagation (BP) algorithm. The dynamic neural network is designed to perform as a robust adaptive feedback controller for TCP dynamics after an adequate training period. We evaluate the performances of the proposed neural network AQM approach using simulation experiments. The proposed approach yields superior performance with faster transient time, larger throughput, and higher link utilization compared to two existing schemes: Random Early Detection (RED) and Proportional-Integral (PI)-based AQM. The neural AQM outperformed PI control and RED, especially in transient state and TCP dynamics variation.

[1]  Hitay Özbay,et al.  On the design of AQM supporting TCP flows using robust control theory , 2004, IEEE Transactions on Automatic Control.

[2]  A. Guez,et al.  Neural network architecture for control , 1988, IEEE Control Systems Magazine.

[3]  Jing Peng,et al.  An Efficient Gradient-Based Algorithm for On-Line Training of Recurrent Network Trajectories , 1990, Neural Computation.

[4]  Vishal Misra,et al.  Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED , 2000, SIGCOMM 2000.

[5]  V. Jacobson,et al.  Congestion avoidance and control , 1988, CCRV.

[6]  Donald F. Towsley,et al.  Analysis and design of controllers for AQM routers supporting TCP flows , 2002, IEEE Trans. Autom. Control..

[7]  Yong-Wan Park,et al.  Approximate ML Detection with the Best Channel Matrix Selection for MIMO Systems , 2008 .

[8]  Van Jacobson,et al.  Random early detection gateways for congestion avoidance , 1993, TNET.

[9]  James Aweya,et al.  A control theoretic approach to active queue management , 2001, Comput. Networks.

[10]  Sally Floyd,et al.  Adaptive RED: An Algorithm for Increasing the Robustness of RED's Active Queue Management , 2001 .

[11]  S. Floyd,et al.  A report on recent developments in TCP congestion control , 2001, IEEE Commun. Mag..

[12]  Kang G. Shin,et al.  A self-configuring RED gateway , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[13]  T. V. Lakshman,et al.  SRED: stabilized RED , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[14]  Paul J. Werbos,et al.  Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[15]  Martin May,et al.  Analytic evaluation of RED performance , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[16]  Steven H. Low,et al.  Analysis and design of AQM based on state-space models for stabilizing TCP , 2003, Proceedings of the 2003 American Control Conference, 2003..

[17]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[18]  Kang G. Shin,et al.  Stochastic fair blue: a queue management algorithm for enforcing fairness , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[19]  Lakhmi C. Jain,et al.  Recurrent Neural Networks: Design and Applications , 1999 .

[20]  Steven H. Low,et al.  REM: active queue management , 2001, IEEE Netw..