Open AccessASSESSMENT OF ELEMENTS OF DATA-TRANSFER SYSTEMS BY USING FUZZY NEURAL NETWORKS
Author(s) -
А. А. Олейников,
Ilya Aleksandrovich Beresnev
Publication year - 2020
Publication title -
vestnik astrahanskogo gosudarstvennogo tehničeskogo universiteta. seriâ: upravlenie, vyčislitelʹnaâ tehnika i informatika
Language(s) - English
Resource type - Journals
eISSN - 2224-9761
pISSN - 2072-9502
DOI - 10.24143/2072-9502-2020-4-121-131
Subject(s) - operability , artificial neural network , computer science , data transmission , automation , process (computing) , data mining , telecommunications network , reliability engineering , fuzzy logic , artificial intelligence , machine learning , real time computing , engineering , computer network , operating system , mechanical engineering
The article considers using direct distribution neural networks and fuzzy neural networks for assessing the operational state of data transmission system elements. In order to select the type of artificial neural network that most fully meets the task of redefining data for predicting the operational state of communication network elements, factors presented in quantitative form are taken into account. For that purpose, the amount of data transmitted through active equipment was selected as the most significant factor having a high level of uncertainty in networks with packet data transmission. The predicted values and changes in traffic levels resulting from the operation of a neural network allow to make the predictive analysis of the operability of the communication networks equipment. Automation of the process and analysis of the equipment operability imply commissioning this function to the assessment system for typical elements of data networks with similar operational conditions. This helps to reduce the number of poor-quality decisions on modernization and increase the speed of response to emergency situations.