Gaussian noise model for multiband optical networks over the C and L bands

• Autor(a) principal: Venda, Pedro Daniel Santo
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/23875
• topic: Gaussian noise model; Multiband transmission; Nonlinear interference; Stimulated Raman scattering; Difusão estimulada de Raman; Interferência não-linear; Modelo de ruído Gaussiano; Rede ótica -- Optical network; Sinais multibanda

The Gaussiannoise (GN) model is an efficient tool to estimate the nonlinear interference (NLI), simplifying current wavelength-division multiplexing (WDM) systems design and analysis. Recently, the generalized Gaussian noise (GGN) model has been proposed for assessing the performance of multiband C+L transmission in WDM optical systems, where the interaction between NLI and stimulated Raman scattering (SRS) must be accurately characterized. In a network scenario, the WDM channels are added and dropped, leading to dynamic traffic variations in the fiber spans (here referred as network utilization). For efficient network planning, the combined SRS and NLI impact on the optical signal-to-noise ratio (OSNR) must be properly evaluated for the whole C+L band and different network utilizations. In this work, an exhaustive study of the performance of the GN and GGN models for point-to-point and network transmission scenarios is performed. For 0 dBm channel launch power and network utilizations from 20% to 100%, the power transfer between the outer channels increases about 5.3 dB due to SRS. At optimal OSNR, the maximum power transfer due to SRS lies between 4.4 dB and 6.4 dB and the maximum OSNR variation along the WDM channels bandwidth is only 0.7 dB. Comparing the OSNR predictions of GGN and GN models closed form approximations (which are more suitable for the performance assessment in network scenarios), it is demonstrated that the maximum difference between methods is below 0.7 dB at optimum OSNR and for maximum C+L band occupancy.