Educational bandwidth traffic prediction using non-linear autoregressive neural networks

##plugins.themes.bootstrap3.article.main##

Shwan Dyllon
Timothy Hong
Ousmane Abdoulaye Oumar
Perry Xiao
Enviado: Nov 29, 2018
Publicado: Sep 30, 2018

Resumen

Time series network traffic analysis and forecasting are important for fundamental to many decision-making processes, also to understand network performance, reliability and security, as well as to identify potential problems. This paper provides the latest work on London South Bank University (LSBU) network data traffic analysis by adapting nonlinear autoregressive exogenous model (NARX) based on Levenberg-Marquardt backpropagation algorithm. This technique can analyse and predict data usage in its current and future states, as well as visualise the hourly, daily, weekly, monthly, and quarterly activities with less computation requirement. Results and analysis proved the accuracy of the prediction techniques.

Palabras clave

Educational bandwidth, traffic prediction

Descargas

La descarga de datos todavía no está disponible.

##plugins.themes.bootstrap3.article.details##

Cómo citar
Dyllon, S., Hong, T., Oumar, O., & Xiao, P. (2018). Educational bandwidth traffic prediction using non-linear autoregressive neural networks. Memorias De Congresos UTP, 1(1), 251-261. Recuperado a partir de http://revistas.utp.ac.pa/index.php/memoutp/article/view/1919