The today networks evolution requires the prediction of traffic demand in order to efficiently use the available resources. The traffic load prediction can be exploited to dynamically allocate the network resources among the different users that, in the 5G world, can be the different verticals. In this scenario, we analyse the application of classical time series predictors to the mobile network traffic in order to evaluate the performance of the considered approaches in terms of complexity and prediction accuracy. Furthermore, we propose an enhancement to the classical Normalized Least Mean Square (NMLS) in order to increase its prediction accuracy, with a negligible complexity increase. The enhancement is based on the application of the Chebyshev's inequality to estimate the prediction error bound. This statistical bound is used to correct the prediction error. The simulation analysis shows the performance improvements given by the proposed scheme.

Prediction of mobile networks traffic: Enhancement of the NMLS technique

Garroppo R. G.;Callegari C.
2020-01-01

Abstract

The today networks evolution requires the prediction of traffic demand in order to efficiently use the available resources. The traffic load prediction can be exploited to dynamically allocate the network resources among the different users that, in the 5G world, can be the different verticals. In this scenario, we analyse the application of classical time series predictors to the mobile network traffic in order to evaluate the performance of the considered approaches in terms of complexity and prediction accuracy. Furthermore, we propose an enhancement to the classical Normalized Least Mean Square (NMLS) in order to increase its prediction accuracy, with a negligible complexity increase. The enhancement is based on the application of the Chebyshev's inequality to estimate the prediction error bound. This statistical bound is used to correct the prediction error. The simulation analysis shows the performance improvements given by the proposed scheme.
978-1-7281-6339-0
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/1056339
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
social impact