Feedback-Based Channel Frequency Optimization in Superchannels
Abstract
Superchannels leverage the flexibility of elastic optical networks and pave the way to higher capacity channels in space division multiplexing (SDM) networks. A superchannel consists of subchannels to which continuous spectral grid slots are assigned. To guarantee superchannel operation, we need to account for soft failures, e.g., laser drifts causing interference between subchannels, wavelength-dependent performance variations, and filter misalignments affecting the edge subchannels. This is achieved by reserving spectral guardband between subchannels or by employing a lower modulation format. We propose a process that dynamically retunes the subchannel transmitter (TX) lasers to compensate for soft failures during operation and optimizes the total capacity or the minimum subchannel quality of transmission (QoT) performance. We use an iterative stochastic subgradient method that at each iteration probes the network and leverages monitoring information, particularly subchannels signal-to-noise ratio (SNR) values, to optimize the TX frequencies. Our results indicate that our proposed method always approaches the optima found with an exhaustive search technique, unsuitable for operating networks, irrespective of the subchannel number, modulation format, roll-off factor, filters bandwidth, and starting frequencies. Considering a four-subchannel superchannel, the proposed method achieves 2.47 dB and 3.73 dB improvements for a typical soft failure of +/- 2 GHz subchannel frequency drifts around the optimum, for the two examined objectives.
Cite
@article{arxiv.2305.17961,
title = {Feedback-Based Channel Frequency Optimization in Superchannels},
author = {Fabiano Locatelli and Konstantinos Christodoulopoulos and Camille Delezoide and Josep M. Fàbrega and Michela Svaluto Moreolo and Laia Nadal and Ankush Mahajan and Salvatore Spadaro},
journal= {arXiv preprint arXiv:2305.17961},
year = {2023}
}