Advanced Models and Algorithms for Self-Similar IP Network Traffic Simulation and Performance Analysis

Abstract

The paper examines self-similar (or fractal) properties of real communication network traffic data over a wide range of time scales. These self-similar properties are very different from the properties of traditional models based on Poisson and Markov-modulated Poisson processes. Advanced fractal models of sequentional generators and fixed-length sequence generators, and efficient algorithms that are used to simulate self-similar behavior of IP network traffic data are developed and applied. Numerical examples are provided; and simulation results are obtained and analyzed.K e y w o r d s: communication networks, IP network traffic, long-range dependent self-similar processes, advanced generators of self-similar teletraffic

Study about effects of self-similar IP network traffic on queuing and network performance

“…Markovian models for self-similar traffic require including several control parameters with a wide range of input values. As a result, controlling these values in sequential generators is much more complex than in generators of fixed-length sequences of self-similar processes with a given Hurst parameter (Radev and Lokshina, 2010).…”

“…which is continuous for all t, producing smoother spectral density function (Radev and Lokshina, 2010). A method based on the FBNDP adds M independent and identically distributed (iid) alternating FRPs to generate a fractal binomial noise process that serves as the rate function for a Poisson process.…”

“…The algorithm, based on the F-ARIMA method, is too rigorous to be used to generate long sample sequences (Radev and Lokshina, 2010).…”

“…In this work the exact or asymptotic self-similar processes are used in an interchangeable manner, which refers to the tail behaviour of the autocorrelations (Radev and Lokshina, 2010). Modelling and simulation of self-similar IP network traffic is performed with advanced generators of synthetic self-similar sequences, divided into two practical classes: the sequential generators and the fixed-length sequence generators.…”

“…Self-similar video traffic provides possibility for developing models for variable bit rate (VBR) video traffic using heavy-tailed distributions (Radev and Lokshina, 2010). The impact of self-similar models on queuing performance is significant and the main trends in such findings are correspondent to efficient modelling and with 0.5 < H < 1, if for each k large enough ρ k (m) → ρ k , as m → ∞, where ρ k = E[(X i -μ)(X i+k -μ)] / σ 2 .…”