FRACTAL ANALYSIS OF THE INTERNET OF COMPLEX ORGANIZATIONS

Abstract

This paper presents research aimed at understanding the dynamics of the Internet network in organizations with high intensity of Internet network traffic. The case of the National Polytechnic Institute (IPN) is analyzed, applying techniques and tools from statistical mechanics. First, some concepts of complex systems and complex networks are mentioned, which have been applied in the analysis of the Internet network. Next, a statistical analysis and a fractal (scaling) analysis are presented, which describes how these are used to characterize the statistical parameters that govern the dynamics of Internet traffic. In the statistical analysis, it was found that the time series used fit better to heavy-tailed distributions (log-logistic), in different time horizons, this indicates a power law behavior in the series. For the fractal analysis, calculations were generated to determine the local scaling exponents (Hurst exponent), using five autoaffine tracking methods. Finally, based on the analysis of network traffic results and the theory of the small world model of Watts and Strogatz, the model that represents the dynamics of the Internet and the IPN mail server is developed.