Estimating the Intensity of Long-Range Dependence in Real and Synthetic Traffic Traces

Domańska, Joanna; Domański, Adam; Czachórski, Tadeusz

doi:10.1007/978-3-319-19419-6_2

Cited by 13 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figs. 2 (a-d) show the traffic generated after the simulation using different Pareto parameters, for example, (10, 0.8), (10, 1.2), (10,16) and (10, 1.8). Fig.…”

Section: Hurst Index and Fractal Dimension Calculationmentioning

confidence: 99%

“…5 (a -d) show how Hurst index is estimated using Eq. (13,15,16). The Average of the several methods are considered because each method has its accuracy and error.…”

Section: Hurst Index and Fractal Dimension Calculationmentioning

confidence: 99%

“…Different mathematical models have been developed to investigate various properties and characteristics of information flow across telecommunication networks using Poison, Lognormal or Weibull distribution [11][12][13][14][15][16]. However, to date, there is no systematic method that has addressed traffic irregularity on telecommunication networks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Traffic Modeling Using Raw Packet Generator on Corporate Computer Network

Ali

Vasira

2018

AJRCoS

View full text Add to dashboard Cite

Fractal dimension is mathematically defined as a ratio of statistical complexity of network traffic; its significant manifestation can affect the network performance. In this work, two models of corporate computer networks have been developed using optimized network engineering tool (OPNET) technology. Raw packet generator (RPG) traffic was imposed on the corporate networks and modeled using H = 0.7 and D = 1.3, under the influence of Pareto distribution. Autocorrelation function and power law were used to confirm the presence of fractal traffic on the networks. Average Hurst index (H) of 50 and 100 workstations were estimated using aggregate of variance, absolute moment, periodogram and R/S methods as 0.627, 0.608 and its corresponding fractal dimensions (D) were obtained as 1.371 and 1.391 respectively. These results obtained mean, there is a manifestation of fractal traffic and delay is minimised on the network.

show abstract

“…Figs. 2 (a-d) show the traffic generated after the simulation using different Pareto parameters, for example, (10, 0.8), (10, 1.2), (10,16) and (10, 1.8). Fig.…”

Section: Hurst Index and Fractal Dimension Calculationmentioning

confidence: 99%

“…5 (a -d) show how Hurst index is estimated using Eq. (13,15,16). The Average of the several methods are considered because each method has its accuracy and error.…”

Section: Hurst Index and Fractal Dimension Calculationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Traffic Modeling Using Raw Packet Generator on Corporate Computer Network

Ali

Vasira

2018

AJRCoS

View full text Add to dashboard Cite

show abstract

“…The studies have proven that ignoring these phenomena has a negative effect on the estimation of performance measurements. This can result in mean queue length enlargement at buffers and packet loss probability increase [ 10 ]. Because of this negative impact on the network performance [ 4 ], these phenomena should be taken into consideration in the network management process.…”

Section: Introductionmentioning

confidence: 99%

“…Fractional Gaussian Noise (FGN) is a stationary Gaussian process that is exactly self-similar. For that reason, it has become common to use FGN in the network traffic modeling [ 10 ]. An alternative is to use the Fractional Auto-regressive Integrated Moving Average (FARIMA).…”

Section: Introductionmentioning

confidence: 99%

Long-Range Dependent Traffic Classification with Convolutional Neural Networks Based on Hurst Exponent Analysis

Filus

Domański

Domańska

et al. 2020

Entropy

Self Cite

View full text Add to dashboard Cite

The paper examines the ability of neural networks to classify Internet traffic data in terms of self-similarity expressed by the Hurst exponent. Fractional Gaussian noise is used for the generation of synthetic data for modeling the genuine ones. It is presented that the trained model is capable of classifying the synthetic data obtained from the Pareto distribution and the real traffic data. We present the results of training for different optimizers of the cost function and a different number of convolutional layers in the neural network.

show abstract