Machine Learning Algorithms Improve Routing in Optical Networks

Strategic Insights -

A recent study has demonstrated the effectiveness of machine learning algorithms in solving routing, modulation, and spectrum allocation (RMSA) problems in optical networks. The research, conducted at the Federal University Pernambuco, utilized a classification model to propose an algorithm that predicts routes according to call request information and network link states. The algorithm was evaluated on three network topologies and compared to two other routing algorithms, with the classification model achieving superior results.

Key Takeaways:

  • The research introduced a machine learning-based algorithm to solve RMSA problems in optical networks, leveraging historical data to find optimal solutions.
  • The proposed algorithm outperformed Yen's algorithm (k shortest routes) in all three network topologies in terms of blocking probability.
  • When compared to the spectrum continuity based shortest path (SCSP) algorithm, the classification model achieved an average performance gain of 15% and 25% in the six-node and NSFNET network topologies, respectively.
  • The algorithm reduced the time spent to find the RMSA solution compared to the SCSP algorithm in all network topologies considered.
  • The study used a dynamic routing algorithm to train the dataset, which consisted of six-node, NSFNET, and European optical network topologies.
  • The proposed algorithm was evaluated on three network topologies: six-node, NSFNET, and European optical network.
  • The research was funded by the National Natural Science Foundation of China (NSFC) and the Sichuan Science and Technology Program.

Statistics:

  • The proposed algorithm achieved an average performance gain of 15% and 25% in the six-node and NSFNET network topologies, respectively, compared to the SCSP algorithm.
  • The algorithm reduced the time spent to find the RMSA solution compared to the SCSP algorithm by an average of 23.19% in the European network topology at the lowest network loads.
  • The presented research utilized a classification model to solve the routing problem in elastic optical networks.
  • The study used three network topologies to evaluate the routing algorithms: six-node, NSFNET, and European optical network.

Sources:

  • Classification-model Applied To Routing Problem In Flexible-grid Optical Networks. Ieee Transactions On Network and Service Management, 2025;22(3):2747-2763.
  • IEEE Transactions On Network and Service Management can be contacted at: Ieee-inst Electrical Electronics Engineers Inc, 445 Hoes Lane, Piscataway, NJ 08855-4141, USA.
  • Institute of Electrical and Electronics Engineers - www.ieee.org/.
  • Ieee Transactions On Network and Service Management - ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=4275028.