Search academic texts

Information × Registration Number 0221U100688, 0120U102622 , R & D reports Title Research of routing in computer networks of railway transport with use of methods of artificial intelligence popup.stage_title Head Pakhomova Victoria M., Кандидат технічних наук Registration Date 08-01-2021 Organization Dnipro National University of Railway Transport named after Academician V. Lazaryan popup.description2  Classical algorithms for finding the shortest path on the graph, which are the basis of existing routing protocols, which are now used in the ITS of railway transport, in a constantly changing network load can not lead to an optimal solution in real time. But on the one hand there is a well-developed device of artificial intelligence, and on the other hand - MPLS technology, which could significantly improve the quality of ITS rail transport networks, create virtual channels between network nodes, encapsulate different data protocols to transmit a variety of traffic. The structure of the integrated routing system in the ITS of railway transport based on the use of fuzzy networks, which will allow in real time to determine the optimal route in the network for the same number of routers that make up the path of the packet. A study of the average learning error of a fuzzy network with different membership functions and different optimization methods for the computer network of the ITS of the Dnieper Railway. The organization of distribution of traffic flows in the perspective MPLS network for a fragment of ITS of railway transport on the basis of the created software model "CoSThDist" on which input data on traffic are submitted is offered: class of service; sender and recipient identifiers; the vector of average flow rates and data on communication channels that determine the corresponding tunnels in the MPLS network. The optimal parameters of the neural network are determined, which is the basis of the software model, the use of which will allow in real time to obtain the distribution of flows by tunnels according to the class of maintenance of traffic flows and quality parameters. Product Description popup.authors Бондарева Валентина Сергіївна Dоmаns'kа Galina A. Pakhomova Victoria M. Skaballanovych Tetiana I popup.nrat_date 2021-03-17 Close