Forecasting Busy-Hour Downlink Traffic in Cellular Networks

TL;DR

This paper presents methods for long-term forecasting of busy-hour cellular network traffic, demonstrating that accurate predictions can be made up to two months ahead with improved accuracy using clustering techniques.

Contribution

It introduces a novel approach for long-term busy-hour traffic forecasting, including cluster-based models, and compares multiple algorithms on real network data.

Findings

Forecasting error below 10% for 2-month ahead predictions.

Cluster-based models improve accuracy by up to 8%.

Long-term traffic forecasting is feasible with high accuracy.

Abstract

The dramatic growth in cellular traffic volume requires cellular network operators to develop strategies to carefully dimension and manage the available network resources. Forecasting traffic volumes is a fundamental building block for any proactive management strategy and is therefore of great interest in such a context. Differently from what found in the literature, where network traffic is generally predicted in the short-term, in this work we tackle the problem of forecasting busy hour traffic, i.e., the time series of observed daily maxima traffic volumes. We tackle specifically forecasting in the long term (one, two months ahead) and we compare different approaches for the task at hand, considering different forecasting algorithms as well as relying or not on a cluster-based approach which first groups network cells with similar busy hour traffic profiles and then fits per-cluster forecasting models to predict the traffic loads. Results on a real cellular network dataset show that busy hour traffic can be forecasted with errors below 10% for look-ahead periods up to 2 months in the future. Moreover, when clusters are available, we improve forecasting accuracy up to 8% and 5% for look-ahead of 1 and 2 months, respectively.