Leveraging Multi-Step Traffic Forecasts for Multi-Period Planning Optical Networks
2026-05-25 • Networking and Internet Architecture
Networking and Internet Architecture
AI summaryⓘ
The authors use multi-step traffic predictions to help plan changes in optical networks ahead of time. They apply a deep learning model to forecast future traffic and then use mathematical methods to decide the best network settings that save spectrum while avoiding service problems. Their methods show a trade-off between how far ahead they predict and the chances of disruptions, meaning the prediction length should be chosen based on what the network operators want. Overall, their approach helps balance efficient network use with maintaining good service.
Optical networksTraffic predictionDeep learningEncoder-decoder modelInteger Linear ProgrammingHeuristic algorithmsSpectrum efficiencyQuality-of-Service (QoS)Prediction horizonService disruption
Authors
Giannis Savva, Hafsa Maryam, Venkatesh Chebolu, Tania Panayiotou, Georgios Ellinas
Abstract
In this work, multi-step traffic predictions are leveraged to enable multi-period planning in reconfigurable optical networks. The proposed framework aims to achieve spectrum savings by adapting the network to predicted time-varying conditions while ensuring the necessary quality-of-service (QoS) levels. Since frequent network (re)configurations may lead to undesired service disruptions, traffic predictions spanning various prediction horizons are exploited to balance the trade-off between spectrum savings and service disruptions. For multi-step-ahead prediction, an encoder-decoder deep learning model is employed to analyze real traffic traces. Subsequently, an Integer Linear Programming (ILP) formulation and heuristic algorithms are developed that use the predictions to proactively (re)optimize future network configurations, enhancing spectrum efficiency while minimizing service disruptions. The approaches are utilized under different scenarios, with the ILP achieving better solutions overall, and the heuristics achieving solutions close to the ILP at significantly lower running times. Further, the results present the effect of the prediction horizon on disruptions and over- and under- provisioning, showcasing that the prediction horizon selection greatly depends on the network operator targets in both network performance and predefined service level agreements.