Internet Teletraffic Modeling and Estimation.

Cover -- Contents -- List of Tables -- List of Figures -- Preface -- List of acronyms and symbols -- 1 Introduction -- 1.1 Objectives of telecommunications carriers -- 1.2 Traffic characteristics -- 1.3 Questions and contributions -- 1.4 Time series basic concepts -- 1.4.1 Time series examples -- 1.4.2 Operators notation -- 1.4.3 Stochastic processes -- 1.4.4 Time seriesmodeling -- 2 The fractal nature of network traffic -- 2.1 Fractals and self-similarity examples -- 2.1.1 The Hurst exponent -- 2.1.2 Samplemean variance -- 2.2 Long range dependence -- 2.2.1 Aggregate process -- 2.3 Self-similarity -- 2.3.1 Exact second order self-similarity -- 2.3.2 Impulsiveness -- 2.4 Final remarks: why is the data networks traffic fractal? -- 3 Modeling of long-range dependent teletraffic -- 3.1 Classes of modeling -- 3.1.1 Non-parametric modeling -- 3.2 Wavelet transform -- 3.2.1 Multiresolution analysis and the discrete wavelet transform -- 3.3 ModelMWM -- 3.4 Parametric modeling -- 3.4.1 ARFIMAmodel -- 3.4.2 ARFIMA models prediction - optimum estimation -- 3.4.3 Formsof prediction -- 3.4.4 Confidence interval -- 3.4.5 ARFIMAprediction -- 3.5 Longmemorystatistical tests -- 3.5.1 R/Sstatistics -- 3.5.2 GPHtest -- 3.6 Some H and d estimation methods -- 3.6.1 R/Sstatistics -- 3.6.2 Variance plot -- 3.6.3 Periodogram method -- 3.6.4 Whittle's method -- 3.6.5 Haslett and Raftery's MV approximate estimator -- 3.6.6 Abry andVeitch'swavelet estimator -- 3.7 Bi-spectrum and linearity test -- 3.8 KPSS stationarity test -- 4 State-space modeling -- 4.1 Introduction -- 4.2 TARFIMAmodel -- 4.2.1 Multistep prediction with the Kalman filter -- 4.2.2 The prediction power of the TARFIMA model -- 4.3 Series exploratory analysis -- 4.3.1 ARFIMA(0 -- 0.4 -- 0) series -- 4.3.2 MWM series with H = 0.9 -- 4.3.3 Nile river series.

4.4 Prediction empirical studywith theTARFIMAmodel -- 4.4.1 ARFIMA(0, d, 0) series -- 4.4.2 MWMseries -- 4.4.3 Nile river series between years 1007 and 1206 -- 4.4.4 Conclusions -- 5 Modeling of Internet traffic -- 5.1 Introduction -- 5.2 Modeling of the UNC02 trace -- 5.2.1 Exploratory analysis -- 5.2.2 Long memory local analysis of the UNC02 trace -- 5.2.3 Empirical prediction with the TARFIMA model -- 6 Conclusions -- Bibliography -- Index -- About the Authors.

Description based on publisher supplied metadata and other sources.

Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2024. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries.