Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/87345
Title: Predicción de tráfico en redes móviles mediante Deep Learning
Author: Gran Josa, José Manuel
Director: Vilajosana, Xavier  
Tutor: Lopez Vicario, Jose  
Abstract: The goal of this paper is to present a study for network traffic prediction based on Deep Learnig techniques. With the predictions obtained it is pretended to optimize the virtualized elements of the Cloud RAN mobile access network architecture. Firstly, concepts related to the subject of study will be presented, such as virtualization, 5G and Cloud RAN architecture. They can be seen in detail, as they are part of the context of this study on data prediction. In order to reach this data prediction target, we also be presented tools used in the development of predictive models. We talk about Python, Pandas, Numpy, Matplotlib and Tensorflow. The last one, will allow to develop the models based on Deep Learning. The main algorithms that can be implemented to make the predictions will be analyzed in detail. It will also be considered the preprocessing of the data that can be implemented in order to make those predictions. We will introduce training, validation and test sets and concepts such as underfitting and overrfitting. The results obtained with the various methods of predicted data implemented will be shown. We will see how we get more accurate results using neural networks. The purpose of this prediction work is successfully concluded for its possible application on virtualized elements of the Cloud RAN to optimize its resources and ensure its capabilities.
Keywords: deep learning
C-RAN
virtualization
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jan-2019
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
jgrajos_TFM_01_2019_Scripts.zipScripts243,51 kBUnknownView/Open
jgrajosTFM0119memoria.pdfMemoria del TFM3,65 MBAdobe PDFThumbnail
View/Open
jgrajosTFM0119presentación.pdfPresentación del TFM2,23 MBAdobe PDFThumbnail
View/Open