Forecasting Internet Traffic by Neural Networks under Univariate and Multivariate Strategies

Paulo Cortez***, Miguel Rio**, Pedro Sousa*, Miguel Rocha*

Universidade do Minho
Departamento de Informatica*
Departamento de Sistemas de Informacao***
P-4710-057 Braga, Portugal

Tel.: +351 253 604430
Fax.: +351 253 604471
E-mail: {{pns,mrocha} (at) di, pcortez (at) dsi}

      UCL (University College London)**
Department of Computer Science
London WC1E 6BT
United Kingdom

Tel: +44 20 7679 7214
Fax: +44 20 7387 1397
E-mail: M.Rio(at)


By improving Internet traffic forecasting, more efficient TCP/IP traffic control and anomaly detection tools can be developed, leading to economic gains due to better resource management. In this paper, Neural Networks (NNs) are used to predict TCP/IP traffic for 39 links of the UK education and research network, under univariate and multivariate strategies. The former uses only past values of the forecasted link, while the latter also uses the traffic from neighbor links of the network topology. Several experiments were held by considering hourly real-world data. The Holt-Winters method was also tested in the comparison.

Overall, the univariate NN approach produces the best forecasts for the backbone links, while a Dijkstra based NN multivariate strategy is the best option for the core to subnetwork links.

Workshop on Artificial Neural Networks and Intelligent Information Processing (ANNIIP), Funchal, Portugal, May 2008