Wuhan University Journal of Natural Sciences

, Volume 6, Issue 3, pp 694–698 | Cite as

Analysis and improvement of TCP congestion control mechanism based on global optimization model

  • Pan Li
  • Li Sheng-hong
  • Gu Shang-jie


Network flow control is formulated as a global optimization problem of user profit. A general global optimization flow control model is established. This model combined with the stochastic model of TCP is used to study the global rate allocation characteristic of TCP. Analysis shows when active queue management is used in network TCP rates tend to be allocated to maximize the aggregate of a user utility functionU s (called,U s fairness). The TCP throughput formula is derived An improved TCP congestion control mechanism is proposed. Simulations show its throughput is TCP friendly when competing with existing TCP and its rate change is smoother. Therefore, it is suitable to carry multimedia applications.

Key words

global optimization additive increase multiplicative decrease congestion control 

CLC number

TP 393 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Paxson V. End-to-End Internet Packet Dynamics.IEEE/ACM Transaction On Networking, 1999,7 (3): 277–292.CrossRefGoogle Scholar
  2. [2]
    Jacobson V. Congestion Avoidance and Control.Proceedings of ACM Sigcomm'88. Palo Alto: CA. 1988. 314–329.Google Scholar
  3. [3]
    Chiu D, Jain R. Analysis of the Increase Decrease Algorithms for Congestion Avoidance.Journal of Computer Networks and ISDN, 1989,17: 1–14.MATHCrossRefGoogle Scholar
  4. [4]
    Kelly F P, Mauloo A K, Tan D K H. Rate Control in Communication Networks: Shadow Prices, Proportional Fairness and Stability.Journal of the Operational Research Society, 1998,49: 237–252.MATHCrossRefGoogle Scholar
  5. [5]
    Hurley P, Boudec L J, Thiran P. A Note on the Fairness of Additive Increase and Multiplicative Decrease.Proceedings of ITC-16, UK: Edinburgh, June 1999.Google Scholar
  6. [6]
    Shenker S. Fundamental Design Issues for the Future Internet.IEEE JSAC, 1995,13: 1176–1188.Google Scholar
  7. [7]
    Steven L, David L. Optimization Flow Control, I: Basic Algorithm and Convergence.IEEE/ACM Transactions on Networking, 1999,7(12): 861–874.Google Scholar
  8. [8]
    Bertsekas D P, John T N.Parallel and Distributed Computation. New Jersey: Prentice-Hall, 1989.MATHGoogle Scholar
  9. [9]
    Ramakrishnan K K, Floyd S. A Proposal to Add Explicit Congestion Notification (ECN) to IP.RFC 2481, January 1999.Google Scholar
  10. [10]
    Braden B, Clark D, Crowcrof J,et al. Recommendations on Queue Management and Congestion Avoidance in the Internet.RFC 2309, April 1998.Google Scholar
  11. [11]
    Brakmo L S, Peterson L L. TCP Vegas: End-to-End Congestion Avoidance on a Global Internet.IEEE JSAC, 1995,13(8): 1465–1480.Google Scholar
  12. [12]
    UCB/LBNL/VINT. Network Simulator-Ns2. http://www-mash.CS.Berkeley.EDU/ns/.Google Scholar

Copyright information

© Springer 2001

Authors and Affiliations

  • Pan Li
    • 1
  • Li Sheng-hong
    • 1
  • Gu Shang-jie
    • 1
  1. 1.Department of Electronic EngineeringShanghai Jiaotong UniversityShanghaiChina

Personalised recommendations