Computing, Telecommunication and Control Информатика, телекоммуникации и управление 2687-0517 21 A modelling of intensity of self-similary traffic of telecommunication network with packet transmission of data Нейросетевая прогнозная модель интенсивностей самоподобного трафика телекоммуникационной сети с пакетной передачей данных Gabdrakhmanov Artur a.gabdrahmanov@samara.ttk.ru Gabdrakhmanova Nailya Kwon@ufamail.ru 10 12 2010 6 113 117 121

In this paper the intensity of self-similar traffic of Gigabit Ethernet was researched. For this we use the measurement of real network. The research proves the self-similar nature of real traffic. Opportunity to use neural network to indicate the traffic of Gigabit Ethernet was shown. Our findings are, that our mathematical model is able to predict the intensity of traffic for some steps to forth.

 nonlinear dynamics Takens theorem neural networks multi-layer perceptron time rows self-similar traffic of Gigabit Ethernet