Evaluation of an adaptive packet loss prediction mechanism applied to Voice over IP transmission

• Fabrício Murai UFRJ
• Hugo H. Costa Sato UFRJ
• Edmundo de Souza e Silva UFRJ
• Daniel R. Figueiredo UFRJ

Abstract

Packet loss directly affects the quality of service provided by multimedia applications in computer networks. These negative effects are even more harmful when the loss process shows great variability. These losses can be mascared by using redundancy, however, the quality and efficiency of this mechanism depends on loss process. Thus, an accurate forecast for the loss process in near future can help to improve the performance of redundancy-based mechanisms. and efficiency of this mechanism. In this article, we evaluate different prediction models, such as hidden Markov models (HMM) and autoregressive models (AR), with respect to their capability of forecasting the loss process. We also evaluate how this prediction affects the recovery of loss packets, in terms of recovery rate and overhead. Our results show that recovery performance is directly related to forecast performance.

References

Azevedo, J. A., Netto, B. C. M., de Souza e Silva, E., and Leão, R. M. M. (2006). Freemeeting: um ambiente para trabalho cooperativo e ensino a distância. In Anais do 7 Fórum Internacional de Software Livre, pages 319–323.

Bilmes, J. (1997). A Gentle Tutorial on the EM Algorithm and its Application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models. Technical Report, University of Berkeley, ICSI-TR-97-021.

Bolot, J. C., Parisis, S. F., and Towsley, D. F. (1999). Adaptive FEC-based error control for Internet telephony. In Proceedings of the IEEE INFOCOM, pages 1453–1460.

Brakmo, L. S. and Peterson, L. L. (1995). TCP Vegas: End to end congestion avoidance on a global Internet. IEEE Journal on Selected Areas in Communications, 13(8):1465–1480.

Brockwell, P. J. and Davis, R. A. (2002). Introduction to Time Series and Forecasting. Springer.

de Souza e Silva, E., da Silva, A. P., de A. Rocha, A. A., Leão, R. M. M., Duarte, F. P., Filho, F. J. S., Jaime, G. D. G., and Muntz, R. R. (2006). Modeling, analysis, measurement and experimentation with the Tangram-II integrated environment. In VALUETOOLS, page 7.

Duarte, F. P. (2003). Algoritmo adaptativo para previsão e recuperação de perda de pacotes em aplicações multimídias usando cadeias de Markov ocultas. Master’s thesis, Universidade Federal do Rio de Janeiro - COPPE - Programa de Engenharia de Sistemas e Computação.

Duarte, F. P., de Souza e Silva, E., and Towsley, D. (2003). An adaptive FEC algorithm using hidden Markov chains. SIGMETRICS Performance Evaluation Review, 31(2):11–13.

Elliot, R. J., Aggoun, L., and Moore, J. B. (1995). Hidden Markov Models: Estimation and Control. Springer-Verlag.

Filho, F. J. S. (2006). Previsão de estatísticas de perdas de pacotes usando modelos de markov ocultos. Master’s thesis, Universidade Federal do Rio de Janeiro - COPPE - Programa de Engenharia de Sistemas e Computação.

Filho, F. J. S. and de Souza e Silva, E. A. (2006). Modelling the short-term dynamics of packet losses. In Performance Evaluation Review.

Filho, F. J. S., Watanabe, E. H., and de Souza e Silva, E. A. (2006). Adaptative forward error correction for interactive streaming over the Internet. In Proceedings of the IEEE Globecom.

Mascolo, S., Casetti, C., Gerla, M., Sanadidi, M. Y., and Wang, R. (2001). TCP Westwood: Bandwidth estimation for enhanced transport over wireless links. In Mobile Computing and Networking, pages 287–297.

Paxson, V. (1997). End-to-end Internet packet dynamics. In SIGCOMM ’97: Conference proceedings on Applications, technologies, architectures, and protocols for computer communication, pages 139–152.

Rabiner, L. R. (1989). A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 77(2):257–285.

Robert, S. and Le Boudec, J. Y. (1997). New Models for Pseudo Self-Similar Traffic. Performance Evaluation, 30(1-2):57–68.

Sahinoglu, Z. and Tekinay, S. (1999). On multimedia networks: self-similar traffic and network performance. Communications Magazine, IEEE, 37(1):48–52.

Salamatian, K. and Vaton, S. (2001). Hidden Markov modeling for network communication channels. In Proceedings of the ACM SIGMETRICS, pages 92–101.

Su, Y. C., Yang, C. S., and Lee, C. W. (2004). The analysis of packet loss prediction for Gilbert-model with loss rate uplink. Information Processing Letters, 90:155–159.

Yajnik, M., Moon, S. B., Kurose, J. F., and Towsley, D. F. (1999). Measurement and modeling of the temporal dependence in packet loss. In Proceedings of the IEEE INFOCOM, pages 345–352.