Group for Research in Decision Analysis

Traffic Prediction techniques and their application in telecommunication networks

Zhani Mohamed Faten

With the boom of the metrology, traffic prediction has constituted a new hot research topic of metrology. It can help to improve quality of Network protocols and can help for planning future needs of any complex system in order to assess future capacity requirements and to plan for changes.

However, tuning the prediction model parameters is very crucial to achieve accurate prediction. This work focuses on the design, the empirical evaluation and the analysis of the behavior of models for predicting the throughput of a single link. We also propose « \(\alpha_\text{SNFAQM}\) », a new active queue management mechanism that uses a neurofuzzy prediction method (SNF) to capture traffic variation and accurately detect the future congestion in order to avoid routers congestion.