Internet and Network Traffic Forecasting Tool
A novel tool that predicts future network traffic with significantly improved performance than current techniques.
A new hybrid forecast model
A novel hybrid tool that uses historical ISP network traffic data to predict future network traffic better than currently available forecasting techniques.
Supports external factors
The model also considers external factors or events that can influence network traffic forecasts.
Aids ISP's network planning and traffic engineering
Offers both short and long term forecast
Effectively predicting future network traffic is valuable for ISPs for ensuring efficient utilization of resources and preventing outages. Although there are several tools available today, there is a clear lack of comparative studies among the many forecasting techniques to determine the most effective ones.
We developed a novel hybrid tool that uses historical ISP network traffic data to predict future network traffic both in the short and long-term. Through a comparative analysis, our technique is a significant improvement on currently available traffic forecasting studies. Furthermore, our model can be extended and customized to include various external factors or events that can influence network traffic forecasts.
-
S. Saha, A. Haque and G. Sidebottom, "Analyzing the Impact of Outlier Data Points on Multi-Step Internet Traffic Prediction Using Deep Sequence Models," in IEEE Transactions on Network and Service Management, vol. 20, no. 2, pp. 1345-1362, June 2023, doi: 10.1109/TNSM.2023.3262406.
-
S. Saha and A. Haque, "Out-of-distribution Internet Traffic Prediction Generalization Using Deep Sequence Model", to appear in the proceedings of the 56th 2022 IEEE International Conference on Communications (ICC 2023), Rome, Italy, May 28-June 1, 2023
Gallery
0 / 3