Banca de DEFESA: VALENTINA ALESSANDRA CARVALHO DO VALE

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : VALENTINA ALESSANDRA CARVALHO DO VALE
DATE: 20/12/2021
TIME: 14:00
LOCAL: Sessão por Videoconferência - Sala: https://meet.google.com/wpf-earg-rpf
TITLE:

RMLSA PROBLEM WITH INCLUSION OF POWER ASSIGNMENT IN TRANSPARENT ELASTIC OPTICAL NETWORKS AND THEIR PHYSICAL LAYER PENALTIES.

KEY WORDS:

Optical Networks, non-linear noises, PRMLSA, power assignment, adaptive models, evolutionary models.

PAGES: 213
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUMMARY:

Elastic Optical Networks (EON) have been chosen as the prominent solution for supporting the growing data traffic with heterogeneous bandwidth requirements, caused by the diversification of offered services. In this context, this research contributed to the mitigation of gaps in the inherent problem of Power, Routing, Modulation Level and Spectrum Assignment (PRMLSA) in EONs, especially the launch power assignment. PRMLSA was addressed in this work by the evaluation of transparent EONs under dynamic traffic and fidelity in the reproduction of physical layer effects through mathematical models of non-linear propagation with low computational complexity. Thus, initially, a study of the importance of appropriate launch power in Quality of Transmission (QoT) was undertaken, followed by the development of launch power assignment techniques: (i) maximum; (ii) minimum; (iii) added margin; (iv) adaptive; (v) evolutionary. The techniques of maximum, minimum and added margin form the basis of adaptive and evolutionary techniques, in which the launching power function was defined as the minimum power value plus a percentage of the difference between the maximum power and the minimum power (power margin). The adaptive technique used an algorithm that starts from a fixed percentage for the power margin and, if necessary, adjusts the power value by adding and/or subtracting a percentage to meet the QoT of the presumed connection and any other active in the network. On the other hand, the evolutionary technique used the Power Series technique through the use of the Particle Swarm Optimization (PSO) meta-heuristic to find adequate values to be assigned to the power margin. Significant reductions in the path request blocking probability (BP), network performance index considered in this research, are observed when the proposed launching power assignment techniques, specifically adaptive and evolutionary ones, are compared to disseminated techniques in the literature.

BANKING MEMBERS:
Interno - 026.367.184-43 - CARMELO JOSÉ ALBANEZ BASTOS FILHO - UFPE
Externo à Instituição - DANIEL AUGUSTO RIBEIRO CHAVES - UPE
Externo à Instituição - HELDER ALVES PEREIRA - UFCG
Externo ao Programa - 2158364 - LEONARDO DIDIER COELHO
Presidente - 1963122 - RAUL CAMELO DE ANDRADE ALMEIDA JUNIOR