CDC ROADM design tradeoffs due to physical layer impairments in optical networks

Sequeira, Diogo; Cancela, Luís; Rebola, Joao L.

doi:10.1016/j.yofte.2021.102461

Cited by 6 publications

(4 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ROADMs architecture considered for the optical node is the route-and-select with maximum losses of 18 dB [17]. The filtering effects and crosstalk due to ROADM components imperfections [18] are not taken into account in this work. The post-amplifiers are designed to perfectly compensate the ROADM losses.…”

Section: Network Simulationmentioning

confidence: 99%

Impact of Traffic Load and Spectral Occupancy on Gaussian Noise Models Performance for Multiband Networks

Venda

Rebol

Cancela

2022

2022 13th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP)

View full text Add to dashboard Cite

In a network scenario, wavelength divisionmultiplexing channels are added and dropped leading to fluctuations on the network traffic loads along the optical path. In this work, a comparison between the optical signal-to-noise ratio (OSNR) predictions of the recently proposed closed-form generalized Gaussian noise (GGN) model and a closed-form Gaussian noise (GN) model that does not take into account the stimulated Raman scattering (SRS) is performed, for different network traffic loads and spectral occupancy over the entire C+L band. In all results obtained, the maximum difference between the OSNR predictions of GN (without SRS) and GGN models closed forms is below 0.7 dB at optimum OSNR and maximum C+L band occupancy, indicating that the GN-model can also be used in C+L band transmission. For channel launch powers higher than the optimum, the OSNR differences increase up to 3 dB, being the GN-model (without SRS) unsuitable to assess the network performance in such situations.

show abstract

Section: Network Simulationmentioning

confidence: 99%

Impact of Traffic Load and Spectral Occupancy on Gaussian Noise Models Performance for Multiband Networks

Venda

Rebol

Cancela

2022

2022 13th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP)

View full text Add to dashboard Cite

show abstract

“…The PLIs analytical and simulated analysis is done by assessing the optical signal-to-noise ratio (OSNR) obtained for a network with the following parameters: optimum signal power 𝑝 , bit rate 𝑅 =100 Gbit/s, line bit rate 𝑅 , = 120 Gbit/s, quaternary phase-shift-keying (QPSK) modulation, optical bandwidth 𝐵 =30 GHz, amplifier figure noise 𝐹 = 6.9 dB, frequency of the central channel 𝜈 =193.8 THz, fiber section length 𝐿 = 80 km, attenuation coefficient α=0.2 dB/km, pre-amplifier gain 𝐺 = 16 dB, post amplifier gain 𝐺 =19.2 dB for the architecture with spatial granularity or 20.4 dB for the architecture with spatialwavelength granularity, symbol rate 𝑅 = 30 GBaud, 80 wavelength channels per spatial channel, fiber dispersion coefficient 𝐷 = 18 ps/nm/km, nonlinear fiber coefficient 𝛾 = 1.1 𝑊 , channel spacing of 50 GHz, Super-Gaussian passband and stopband filters of 4 th order with -3 dB bandwidth of 45 GHz and different isolations [6].…”

Section: A Roadm Cascade Without In-band Crosstalkmentioning

confidence: 99%

“…Each interfering signal is generated also with a phase difference and a time misalignment, in relation to the original signal. The time misalignment and phase difference are modelled as a uniformly distributed random variables, respectively, between [0, 𝑇 ] and between [0, 2π] as in [6]. A first order interfering term must go through one blocking filter that simulates a 1 st order leakage in WSSs, whereas second order interfering terms must go through two blocking filters.…”

Section: B Roadm Cascade With In-band Crosstalkmentioning

confidence: 99%

Cost, power consumption and performance analysis in SDM ROADM architectures for uncoupled spatial channels

Silva

Cancela

Rebola

2022

2022 13th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP)

View full text Add to dashboard Cite

Currently optical networks are reaching their maximum transport capacity. Different solutions can be used to overcome this capacity limit, but we are going to study, in particular, space division multiplexing (SDM). To use SDM, the reconfigurable optical add/drop multiplexers (ROADMs) need to be adapted to support this multiplexing. In this paper we analyze four switching strategies used in SDM ROADMs and the respective SDM ROADM architectures. These strategies are explained and analyzed, for uncoupled scenarios, in terms of cost and power consumption. The impact of the physical layer impairments (PLIs) amplified spontaneous emission noise, non-linear interference, passband narrowing due to optical filtering and in-band crosstalk is also assessed, considering a cascade of SDM ROADMs with spatial and spatial-wavelength switching granularities. The PLI inband crosstalk, has insignificant impact, in networks with a single spatial channel or in SDM networks with spatial granularity ROADM architecture, but in a SDM network with spatial wavelength granularity this PLI can lead to an OSNR penalty of around 2 dB, when the number of spatial channels is high. The other PLIs have similar impacts in all networks studied.

show abstract

“…The ROADMs architecture considered is the route-and-select with maximum losses of 18 dB [18]. The filtering effects and crosstalk due to ROADM components imperfections [19] are not taken into account in this work. The post-amplifiers are designed to perfectly compensate the ROADM maximum losses.…”

Section: Model For Assessing the Lightpaths Qotmentioning

confidence: 99%

Assessing the Quality of Transmission of Lightpaths in Multiband C+L Networks Through Gaussian Noise Models

Venda¹,

Rebola²,

Cancela³

2022

SSRN Journal

View full text Add to dashboard Cite

CDC ROADM design tradeoffs due to physical layer impairments in optical networks

Cited by 6 publications

References 34 publications

Impact of Traffic Load and Spectral Occupancy on Gaussian Noise Models Performance for Multiband Networks

Impact of Traffic Load and Spectral Occupancy on Gaussian Noise Models Performance for Multiband Networks

Cost, power consumption and performance analysis in SDM ROADM architectures for uncoupled spatial channels

Assessing the Quality of Transmission of Lightpaths in Multiband C+L Networks Through Gaussian Noise Models

Contact Info

Product

Resources

About