Journal of Communications and Networks

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지

Internet Roundtrip Delay Prediction Using the Maximum Entropy Principle

  • Liu, Peter Xiaoping (Department of Systems and Computer Engineering, Carleton University) ;
  • Meng, Max Q-H (Department of Electronic Engineering, Chinese University of Hong Kong) ;
  • Gu, Jason (Department of Electrical and Computer Engineering, Dalhousie University)
  • Published : 2003.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

Internet roundtrip delay/time (RTT) prediction plays an important role in detecting packet losses in reliable transport protocols for traditional web applications and determining proper transmission rates in many rate-based TCP-friendly protocols for Internet-based real-time applications. The widely adopted autoregressive and moving average (ARMA) model with fixed-parameters is shown to be insufficient for all scenarios due to its intrinsic limitation that it filters out all high-frequency components of RTT dynamics. In this paper, we introduce a novel parameter-varying RTT model for Internet roundtrip time prediction based on the information theory and the maximum entropy principle (MEP). Since the coefficients of the proposed RTT model are updated dynamically, the model is adaptive and it tracks RTT dynamics rapidly. The results of our experiments show that the MEP algorithm works better than the ARMA method in both RTT prediction and RTO estimation.

Keywords

References

  1. J. Postel, RFC 793: Transmission controt protocot, available athttp://www.ibiblio.org/pub/docs/rfc/rfc793.txt, Sept. 1981
  2. V. Jacobson, 'Congestion avoidance and control,' in Proc. ACM SIG-COMM, Stanford, CA, Aug. 1988, pp. 314-329
  3. L. Zhang, 'Why TCP timers don't work well,' in Proc. ACM SIGCOMMSymp. Comm., Architectures, and Protocots, Stowe, Vermont, Aug. 1986,pp. 397-405
  4. J. Padhye et al., A TCP-friendty rate adjustment protocot for continuousmedia fIows over best effort network, UMass-CMPSCI Technical ReportTR 98-04, Oct. 1998
  5. R. Rejaie, M. Handley, and D. Estrin, 'RAP: An end-to-end rate-basedcongesdon control mechanism for realtime streams in the Internet,' inProc. IEEEINF0C0M, New York, NY, Mar. 1999, PP. 1337-1345
  6. P. Liu et al., 'A novel data transport protocol for Intemet robots,' inPmc. WorId Consress Inteltigent Controt and Automation (WCfCA'02),June 2002, pp. 59-65
  7. S. Ployd and K. Fall, 'Promoting the use ofend-to-end congestion controlin the Intemet,' IEEE/ACM Trans. Networking, 7(4): pp. 458-472, 1999 https://doi.org/10.1109/90.793002
  8. K. Goldberg and R. Siegwart (editors), Beyond Webcams: An introductionto ontine robots, The MIT Press, Cambridge, Massachusetts, USA, 2002
  9. M. S. Borella et at., 'Self-similarity of Intemet packet delay,' in Proc.[EEEICC'97, Aug. 1997, pp. 513-517
  10. Q. Li and D. L. Mills, 'Jitter-based delay-boundary prediction of wide-area networks,' IEEE/ACM Trans. Networking, 9(5): pp. 578-590, 2001 https://doi.org/10.1109/90.958327
  11. D. Mills, RFC889: Intemet Delay Experiments, Dec. 1983
  12. R. Jain, 'Divergence of timeout algorithms for packet retransmissions,' inProc. Fifth Phoenix Conf. Comp. and Comm., Scottsdale, Arizona, Mar.1986,PP.174-179
  13. P. Kam and C. Partridge, 'Improving round-trip time estimates in reliabletransport protocols,' ACM Trans. Computer Systems, 9(4): pp. 365-373,1991
  14. J.-C. Bolot, 'Characterizina end-to-end packet delay and loss in the Inter-net,' J. High Speed Networks, 2(3): pp. 305-323, 1993
  15. A. Acharya and J. Saltz, 'A study of Intemet round-trip delay,' TechnicatReport CS-TR-3736, University ofMaryland, Dec. 1996
  16. M. Allman and V. Paxson, 'On estimating end-to-end network path prop-erties,' in Proc. ACM SIGCOM Comput. Commun. Review, 1999, pp. 263-274
  17. D. D. Clark, RFC 813: Window and acknowtedgement strategy in TCP,July 1988
  18. P. Klan and G. Darbellay, 'An information-theoretic adaptive method fortime series forecasting,' Neurat Network WorId, 7(2): pp. 227-238, 1997
  19. D. Kugiumtzis and M. A. Boudourides, 'Chaotic analysis of Internet pingdata,' S0E1S Meeting at Bietefeld, Mar. 27-28, 1998
  20. S. Guiasu and A. Shenitzer, 'Maximum entropy principle,' Mathematicat Intelligencer, 7(1): pp. 42-48, 1985 https://doi.org/10.1007/BF03023004
  21. E. T. Jaynes, 'Information theory and statistical mechanics I,' Physical Review, 106(4): pp. 620-630, 1957 https://doi.org/10.1103/PhysRev.106.620
  22. J. UfRnk, 'Can the maximum entropy principle be explained as a con-sistency requirement?,' Studies in History and Philosophy of Modem Physics, 26(3): pp. 223-261, 1995 https://doi.org/10.1016/1355-2198(95)00015-1
  23. Available at http://www.topjimmy.net