Maestría en Telecomunicaciones
Permanent URI for this collectionhttp://repositorio.uta.edu.ec/handle/123456789/32901
Browse
Item Ruta óptima de tráfico de una red virtual basada en análisis de datos y algoritmo de Machine Learning(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Telecomunicaciones, 2021) Chávez Fuentes, Carla Patricia; Ríos Villacorta, Percy AlbertoThe present research topic aims to determine the óptimal traffic route of a virtual network based on data analysis and machine learning algorithms through traffic prediction. As in physical networks, in virtual networks, the óptimal design of a communication network is sought. Optimizing a data network is not an easy task, due to the complexity and the number of factors necessary to evaluate to obtain a solution that satisfies the end users [1] [2]. Therefore, in this proposal it is proposed to determine an óptimal traffic route of a virtual network, in order to reduce the unnecessary consumption of network bandwidth and other resources. Traffic analysis of a network is a key point in determining its architecture, which is why it has become a topic of great interest in recent years [3]. The great technological advance that has occurred in recent years, the development of artificial intelligence and the use of machine learning algorithms known by its acronym in English as machine learning (ML), have made it possible to solve several problems in the area of engineering and computer science [4]. Under this first, it is proposed to make use of ML algorithms to determine the óptimal route of traffic in a virtual network through its analysis of traffic and its status. To determine the óptimal route for network traffic, an objective function is intended to minimize the traffic in each node, making the most of the determined bandwidth. The analysis of the topology and architecture of the network will allow to determine the óptimal route for the network traffic so that the design meets the needs of the end users.