Time Series Analysis of Telecommunications Data Using Neural Networks

Main Article Content

Jaime G. Alvarado, Alexis D. Alvarado, Edgar V. Lema


This paper shows a model of artificial neural networks (ANN) for the access to the advanced mobile service (AMS)
of the mobile telephony of Ecuador to predict the demand of active lines by technology of mobile telephony of the
data obtained by the Agency of Regulation and Control of Telecommunications (ARCOTEL) in the period December
2008 to August 2020, model that allows technically a better management of the social media that use the frequencies
of the radioelectric spectrum as are the carriers CONECEL S.A., OTECEL S.A. and CNT EP. The methodology
used is based on the analysis of the time series by means of processes of multilayer neural networks of the statistical
package SPSS. For this purpose, 70% of the data was used for training of the neural network and the remaining 30%
as test data of the network already trained and later to make the predictions of the application of the ANN model.
This allows carriers to know the demand and make the best decisions for the management of new technology in the
field of telecommunications.

Article Details