Statistical Analysis and Modeling of Peer-to-Peer Multimedia Traffic

We study peer-to-peer packet traffic arising from passive VoIP and video measurements that are generated by Skype and IPTV clients. We provide a common methodology for the statistical characterization of the packet flows, discuss the user's satisfaction and load estimation. Two main ideas are used in our analysis. Due to the dependence of the data we first partition the observations into independent blocks and deal further with these block-wise independent data. Secondly, loss is generated by packet lengths which exceed the channel capacity in a time unit if the inter-arrival times coincide with this time unit. If the inter-arrival times are random, loss is generated by the lengths of those packets corresponding to transmission rates that exceed the channel capacity. Our methodology is demonstrated by individual Skype flows and the aggregated flow of video packets exchanged with a mobile peer of a SopCast session.

[1]  W. Willinger,et al.  ESTIMATORS FOR LONG-RANGE DEPENDENCE: AN EMPIRICAL STUDY , 1995 .

[2]  Feng Liu,et al.  A Measurement and Modeling Study of P2P IPTV Applications , 2008, 2008 International Conference on Computational Intelligence and Security.

[3]  J. Teugels,et al.  Statistics of Extremes , 2004 .

[4]  N. Markovich.  Nonparametric analysis of univariate heavy-tailed data , 2007 .

[5]  A. Philippe,et al.  A TEST FOR STATIONARITY VERSUS TRENDS AND UNIT ROOTS FOR A WIDE CLASS OF DEPENDENT ERRORS , 2006, Econometric Theory.

[6]  R. Bhattacharya,et al.  THE HURST EFFECT UNDER TRENDS , 1983 .

[7]  Udo R. Krieger,et al.  Statistical Characterization of QoS Aspects Arising from the Transport of Skype VoIP Flows , 2009, 2009 First International Conference on Evolving Internet.

[8]  Marco Mellia,et al.  Revealing skype traffic: when randomness plays with you , 2007, SIGCOMM 2007.

[9]  J. Bendat,et al.  Random Data: Analysis and Measurement Procedures , 1971 .

[10]  Ralf Runde,et al.  The asymptotic null distribution of the Box-Pierce Q-statistic for random variables with infinite variance An application to German stock returns , 1997 .

[11]  J. Teugels.  Renewal Theorems When the First or the Second Moment is Infinite , 1968 .

[12]  G. Box,et al.  On a measure of lack of fit in time series models , 1978 .

[13]  Richard A. Davis,et al.  Introduction to time series and forecasting , 1998 .

[14]  W. G. Cochran.  The distribution of the largest of a set of estimated variances as a fraction of their total , 1941 .

[15]  S. Resnick.  Heavy-Tail Phenomena: Probabilistic and Statistical Modeling , 2006 .

[16]  Miguel A. Labrador,et al.  On the applicability of available bandwidth estimation techniques and tools , 2010, Comput. Commun..

[17]  Udo R. Krieger,et al.  Statistical analysis and modeling of Skype VoIP flows , 2010, Comput. Commun..

[18]  Kishor S. Trivedi.  Probability and Statistics with Reliability, Queuing, and Computer Science Applications , 1984 .

[19]  Patrice Abry,et al.  Wavelet Analysis of Long-Range-Dependent Traffic , 1998, IEEE Trans. Inf. Theory.

[20]  Antonio Pescapè,et al.  Traffic analysis of peer-to-peer IPTV communities , 2009, Comput. Networks.

[21]  Chun-Ying Huang,et al.  Quantifying Skype user satisfaction , 2006, SIGCOMM 2006.

[22]  T. Higuchi.  Approach to an irregular time series on the basis of the fractal theory , 1988 .

[23]  Jan Beran,et al.  Statistics for long-memory processes , 1994 .