One Problem of Structural Optimization of Data Transmission Networks

Abstract

The article discusses issues of analysis and structural optimization of multi-level data transmission networks using the random hypernetwork model. It is argued that the main level subjected to destruction under the influence of the external environment and affecting the performance of the network as a whole is the physical level, but in structural terms physical destruction primarily affects the channel level. In turn, failures at the link level affect the organization of work of the network and higher levels; in particular, they require the redistribution of flows including the rebuilding of virtual channels.

A brief overview of multi-level network models used for the analysis and structural optimization of data transmission networks for various purposes (in particular sensor monitoring networks) is presented, a description of a random hypernetwork is given and some problems of increasing the reliability of the functioning of networks at the network level in the event of possible failures of channels and network nodes are discussed, that is, disruption of its operation at the link level. As indicators of the reliability of the network layer of the network, the possibility (probability) of data transmission between a pair of nodes in principle or via established virtual channels and the mathematical expectation of the number of nodes capable of transmitting data to the central node are considered.

The goals of structural optimization are the best placement of virtual channels in the known structure of the channel network and the placement of base stations in network nodes which connection with the central node is considered to be reliable.