Mathematical Simulation of Infocommunication Networks Applying Chaos Theory

The advent and wide usage of computer networks as well as an increasing number of various network services have resulted in the fact that the network traffic has become more complex and unpredictable. These properties have become especially apparent with the appearance of high-speed data transmission technology. It is connected with the fact that one of the main quality indeces (QoS) of operation of packet transmission networks is the number of lost packets. The loss of packets results to additional network traffic and, finally, “congestions”. At high speeds of data transmission packet losses, expressed in portions of a per-cent, lead to considerable information losses. It has been shown in numerous papers devoted to the research of network traffic that the abovementioned phenomena are related to the properties of the traffic self–similarity which is mainly caused by the TCP protocol behaviour. However to date models have not been offered which adequately describe the behaviour of the communication networks of information systems and which allow to apply the whole arsenal of classic methods of analyzing nonlinear dynamic systems.