Doctoral Thesis Defense Session No. 53 of PPGEEC - Application of machine learning techniques for planning and management of elastic optical networks

STUDENT: TALISON AUGUSTO CORREIA DE MELO

DATE: 12/09/2024

TIME: 2:00 PM

LOCATION: https://conferenciaweb.rnp.br/sala/karcius-day-rosario-assis

TITLE: Application of machine learning techniques for planning and management of elastic optical networks

KEYWORDS: Elastic optical networks; RSA; Protection; Virtualization; Machine Learning;

ABSTRACT: Elastic Optical Networks (EONs) have emerged as an innovative response to traditional optical networks, bringing new operational concepts that improve flexibility and efficiency in the use of resources. A recurring problem in EONs is Routing and Spectrum Allocation (RSA), which seeks to define a route for each request and allocate an adequate number of slots according to the required demand, using the smallest possible amount of spectrum. This work presents supervised machine learning techniques for the design of virtualization with protection in EONs, with the objective of predicting the total number of spectrum slots required to support all traffic demands. Focusing on Virtual Optical Networks (VONs) subject to specific protection, we investigate the application of Machine Learning (ML) techniques, specifically Multilayer Perceptron (MLP) and Support Vector Regression (SVR), to solve the link capacity problem of EONs with virtualization faster than traditional Integer Linear Programming (ILP) formulations, maintaining results close to the optimal ones. The performance of the models was evaluated through statistical metrics, training time and inference. The results showed that the proposed method is effective in predicting the number of slots required in the physical substrate subject to multiple VONs.

BOARD MEMBERS:

KARCIUS DAY ROSARIO ASSIS (ADVISOR)  UFBA

MARCELA SILVA NOVO                              UFBA

VITALY FELIX RODRIGUEZ ESQUERRE         UFBA

JOSÉ VALENTIM DOS SANTOS FILHO          UFRB

WASLON TERLLIZZIE ARAÚJO LOPES          UFPB

Em 06/12/2024