Octave fractals

Potortì F.

fractional Gaussian  fractal traffic  traffic generator  hurst parameter  random midpoint disp 

These are functions for Octave, a high-level language which uses a language very similar to that of Matlab. hurstIDC.m is a function for estimating the Hurst parameter of Hurst noise, commonly known as fractional Brownian noise, or fBn. Network traffic generated as an aggregate of many sources has been shown to behave like fBn. A brief introduction to this topic, with references, can be found in Section 3 of Modeling Ka band scintillation as a fractal process. hurstVTP.m is another function with the same purpose, which uses a different algorithm. In fact, estimating the Hurst parameter is not an easy task, and requires some knowledge of the possible pitfalls. A related function, RMDtraffic.m, is conversely used to generate Gaussian Hurst noise, also called fractional Gaussian noise (fGn for short). fGn can be used to produce synthetic traces representing the traffic generated by the aggregation of multiplicity of traffic sources. The product of this third function is only an approximation of real fGn, for performance reasons, and it only contains nonnegative samples, making the results immediately usable as synthetic traffic traces for simulation purposes.