Modeling and mitigation of fiber nonlinearity in wideband optical signal transmission [Invited]

Abstract

The adoption of open optical networks (OONs) requires the development of open and effective network planning tools, enabling the use of multi-vendor or white box transport solutions. Such tools for studying and planning optical networks must be able to take into account the physical layer impairments, including fiber nonlinearity. The use of wideband wavelength division multiplexing in OONs, with channel frequencies extending across the short-, conventional-and long-bands (S, C and L-bands) and beyond, offers a pathway to increasing data rates through the installed fiber infrastructure. However, achievable information rates are limited by the resulting signal distortion due to fiber nonlinearity as signal bandwidths are increased, in particular inter-channel stimulated Raman scattering (ISRS). In this paper, we describe the nonlinear effects observed in wideband transmission systems, and review recently-developed analytical tools, based on the Gaussian noise (GN) model of nonlinear interference with the inclusion of ISRS. Using the ISRS GN model, we assess the impact of fiber nonlinearity on the achievable information rates in transmission systems with bandwidths of up to 12 THz. We demonstrate the use of the model in the optimization of launch power spectral profiles for a variety of dynamic gain equalizer arrangements in a 1000 km standard single-mode fiber link, using particle swarm optimization and the steepest descent algorithm. Such nonlinear models and optimization methods could be applied in OON planning tools, for example in optical link emulators to estimate quality-of-transmission (QoT) and data throughput, and in impairment-aware software-defined network control and management.

Challenges in Extending Optical Fibre Transmission Bandwidth Beyond C+L Band and How to Get There

“…The NLI coefficient η n (f i ) experienced by the channel of frequency f i is calculated using the closed-form ISRS GN model proposed in [12]. This model is well suitable for modelling ultra-wideband transmission systems, and its accuracy compared to nonlinear Schrodinger equation (NLSE) simulations has been demonstrated at [12], [14].…”

“…This creates to a N ch -dimensional optimisation problem. Additionally, in the presence of ISRS, this optimisation is non-convex, leading to multiple local solutions [14].…”

Universal Virtual Lab: A Fast and Accurate Simulation Tool for Wideband Nonlinear DWDM Systems

“…D. Dahan is also with the Department of Electrical Engineering, Holon Institute of Technology, 52 Golomb street, Holon 58102, Israel (email: david.dahan@huawei.com; michael.zarubinsky@huawei.com; liangyunhua1@huawei.com ) penalties as some portions of the channel optical power are transferred from higher to lower optical frequencies via nonlinear inelastic scattering of the signal photons by the fiber silica medium [8]. Accurate and reliable simulations of nonlinear DWDM systems with a large number of interfering channels (IC) are critical for designing and predicting the performance of fiber-communications systems [9][10][11], for the development of nonlinearity tolerant modulation formats and for devising effective nonlinearity equalization techniques [12][13][14], as well as for optical networking related applications such as optical channel monitoring [15], fast and accurate network planning tools [16] and means against physical layer related security threats [17]. Simulation of large channel-count systems are extremely demanding in terms of computational resources and often require unacceptably long computation time when relying on the conventional split-step Fourier-transform method (SSFTM) [18].…”

“…They include the Gaussian noise (GN) model [20], the enhanced Gaussian noise model (EGN) [21], both developed in the frequency domain and the time-varying intersymbol interference (ISI) model of the XPM induced NLIN [22][23]. Some of these models have also been extended to include the ISRS impact [11,24]. These models lead to semi-analytical solutions of the NLIN and in some contexts, they offer an acceptable alternative to SSFTM simulations.…”

Accuracy of EGN model in ultra-wideband optical fiber communication systems

“…It is particularly important to accurately and quickly evaluate the signal-to-noise ratio (SNR) of high-speed and ultra-wideband optical fiber communication systems. 1 The process of optical signal transmission in fiber is complex and unresolvable, but fortunately, the Enhanced GN-model (EGN model), [2][3][4][5][6][7][8][9][10][11] arising from the Gaussian noise perturbation analysis, can evaluate the system SNR accurately. To simplify the calculation, a closed-form formula is proposed using some mathematical approximation.…”