Design Optimisation of Power-Efficient Submarine Line through Machine Learning

Abstract

An optimised subsea system design for energy-efficient SDM operation is demonstrated using machine learning. The removal of gain-flattening filters employed in submarine optical amplifiers can result in capacity gains at no additional overall repeater cost.

Power Evolution Modeling and Optimization of Fiber Optic Communication Systems With EDFA Repeaters

“…5). For an intra-EDFA root MSE of ≈ 0.16dB was achieved in [3], and intra-EDFA root MSE of ≈ 0.5dB was achieved in [4]. The MSE is increased to up to ≈ 0.045dB 2 of inter-MSE, with performance relatively flat across average gains and total output power, worst for EDFA A2 operated at high gain or low output power.…”

“…Some solutions exist for countering the SRS effect by launch PSD optimization [7], [8], [9], which do not take into account the gain tilt of practical EDFAs without a GFF. Alternatively, ML can be applied for the end-to-end system [3], [10], [11], which at least in the linear region of transmission can be employed for PSD optimization. Such solutions require a lot of training data to be generated for each link in the network, which is time consuming, and is furthermore susceptible to even minor changes in the link.…”

SNR Optimization of Multi-Span Fiber Optic Communication Systems Employing EDFAs With Non-Flat Gain And Noise Figure

“…Such links are also typically parts of mesh networks, for which equalizing the performance per channel might be preferred to maximizing the total throughput in order to maintain a given quality of service regardless of the allocated frequency channel per user. Furthermore, the throughput is targeted for optimization only by proxy of the optical SNR (OSNR) in [6] [7] [8], which neglects potential frequency dependent transmitter and receiver penalties, as well as the nonlinear distortions accumulated during transmission. This paper is a direct extension of [9], where the EDFA modeling and the system optimization were restricted to the power evolution.…”

“…This comes at the cost of energy consumption and further enhances the non-flatness of the accumulated noise profile, and potentially degrades the overall performance. More recently a few studies have started considering the EDFA response [6][7] [8], but limit the analysis to systems operating in the linear region of the fiber. This is reasonable for transoceanic systems with electrical power constraint, but is sub-optimal for regional and terrestrial networks where maximizing the throughput by operating at the optimum launch power is essential.…”

A State-Variable Approach to Submarine Links Capacity Optimization

“…The quest for energy efficiency has become critical, and new industrial cable designs have arisen, leveraging space division multiplexing (SDM) with either fiber bundles combined with pump-farming solutions at the industrial level [2], [3], or investigations of multi-core fiber solutions at the research level [4]. For each spatial-mode, line optimization has also been rethought, with suggestions to optimize amplifiers bandwidth [1], gain shaping filters [5], [6], or the wavelength division multiplexing (WDM) input power spectral density (PSD) [1], [7], also known as pre-emphasis, most often based on direct numerical or machine-learning optimization.…”