A class of Active Queue Management algorithm based on BP neural network

As an end-to-end congestion control mechanism, Active Queue Management (AQM) technology maintains smaller queue length and higher link utilization through discarding packets in intermediate network nodes. This paper discussed some previous AQM algorithms, RED, BLUE and RLGD, and found out shortcomings in which by comparing with them. On the Basis of Artificial Intelligence (AI) theory and technology, a new AQM algorithm based on BP neural network is proposed. In the end, the implement of the new Active Queue Management algorithm is presented, and the convergence is proved.

[1]  G. Pal,et al.  Congestion control , 1995 .

[2]  Zhang Yan.  A Robust Active Queue Management Algorithm Based on Reinforcement Learning , 2004 .

[3]  D. Lapsley,et al.  Random early marking for Internet congestion control , 1999, Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042).

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

[5]  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).

[6]  David Ott,et al.  Tuning RED for Web traffic , 2001, TNET.

[7]  Liu Wei-dong.  A Classifier Implementation for Active Queue Management Algorithm , 2004 .

[8]  R. La,et al.  Nonlinear instabilities in TCP-RED , 2004, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[9]  B. Barden.  Recommendations on queue management and congestion avoidance in the Internet , 1998 .