5G is expected to bring forth disruptive indus-trial-societal transformation by enabling a broad catalog of (radically new, highly heterogeneous) applications and services. This scenario has called for zero-touch network and service management (ZSM). With the recent advancements in artificial intelligence, key ZSM capabilities such as the runtime prediction of user demands can be facilitated by data-driven and machine learning methods. In this respect, the article proposes a runtime prediction approach that transforms time series forecasting into a simpler multivariate regression problem with artificial neural networks (ANNs), structurally optimized with a genetic algorithm (GA) metaheuristic. Leveraging on a novel set of input features that capture seasonality and calendar effects, the proposed approach removes the prediction accuracy's dependence on the temporal succession of input data and the forecast horizon. Evaluation results based on real telecommunications data show that the GA-optimized ANN regressor has better prediction performance. It achieved average improvements of 59 percent and 86 percent compared to 1-day and 1-hour ahead forecasts obtained with state-of-the-art multi-seasonal time series and long short-term memory forecasting models, respectively. Furthermore, despite its longer training times compared to the baseline models, the proposed ANN regressor relaxes the monitoring requirements in 5G dynamic management systems by allowing less frequent retraining offline.

ANNs Going beyond Time Series Forecasting: An Urban Network Perspective

Pajo J. F.;Bruschi R.;Davoli F.
2021-01-01

Abstract

5G is expected to bring forth disruptive indus-trial-societal transformation by enabling a broad catalog of (radically new, highly heterogeneous) applications and services. This scenario has called for zero-touch network and service management (ZSM). With the recent advancements in artificial intelligence, key ZSM capabilities such as the runtime prediction of user demands can be facilitated by data-driven and machine learning methods. In this respect, the article proposes a runtime prediction approach that transforms time series forecasting into a simpler multivariate regression problem with artificial neural networks (ANNs), structurally optimized with a genetic algorithm (GA) metaheuristic. Leveraging on a novel set of input features that capture seasonality and calendar effects, the proposed approach removes the prediction accuracy's dependence on the temporal succession of input data and the forecast horizon. Evaluation results based on real telecommunications data show that the GA-optimized ANN regressor has better prediction performance. It achieved average improvements of 59 percent and 86 percent compared to 1-day and 1-hour ahead forecasts obtained with state-of-the-art multi-seasonal time series and long short-term memory forecasting models, respectively. Furthermore, despite its longer training times compared to the baseline models, the proposed ANN regressor relaxes the monitoring requirements in 5G dynamic management systems by allowing less frequent retraining offline.
File in questo prodotto:
File Dimensione Formato  
ANNs_Going_Beyond_Time_Series_Forecasting_An_Urban_Network_Perspective.pdf

accesso aperto

Descrizione: Articolo su rivista
Tipologia: Documento in versione editoriale
Dimensione 965.15 kB
Formato Adobe PDF
965.15 kB Adobe PDF Visualizza/Apri

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/11567/1058147
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 2
social impact