Machine learning analysis of extreme events in optical fibre modulation instability

Abstract

A central research area in nonlinear science is the study of instabilities that drive extreme events. Unfortunately, techniques for measuring such phenomena often provide only partial characterisation. For example, real-time studies of instabilities in nonlinear optics frequently use only spectral data, limiting knowledge of associated temporal properties. Here, we show how machine learning can overcome this restriction to study time-domain properties of optical fibre modulation instability based only on spectral intensity measurements. Specifically, a supervised neural network is trained to correlate the spectral and temporal properties of modulation instability using simulations, and then applied to analyse high dynamic range experimental spectra to yield the probability distribution for the highest temporal peaks in the instability field. We also use unsupervised learning to classify noisy modulation instability spectra into subsets associated with distinct temporal dynamic structures. These results open novel perspectives in all systems exhibiting instability where direct time-domain observations are difficult.

Performance analysis of dual-pump nonlinear amplifying loop mirror mode-locked all-fibre laser

“…Our approach opens new opportunities for exploring nonlinear cavity dynamics in laser systems where the vast parameter space makes systematic exploration impracticable, yet ideally suited to optimisation by a NN algorithm. Therefore, it confirms the recent interest in the use of NN algorithms for handling complex optical structure designs [24], ultrafast optics processes [25] or complex nonlinear problems [26]. Future work may consider the upgrade of the NN model for handling multi-pulse and multiperiod mode-locking operation.…”

Recent progress on optical rogue waves in fiber lasers: status, challenges, and perspectives

“…Furthermore, research groups have employed machine learning to analyze the generation of extreme events in optical fiber modulation instability. 177 So far, the investigation of RWs in algorithm-controlled fiber lasers has not been yet demonstrated. We believe that the generating mechanism of rouge waves will be effectively studied in pulse fiber lasers with different SAs through human-like intelligent methods.…”

Artificial neural networks for nonlinear pulse shaping in optical fibers

“…Owing to its power of extracting essential information from large amounts of data, machine learning is bringing a revolutionary reform to research in the physical sciences [8]. In the field of photonics, a number of studies have been recently reported in laser design and optimization [9][10][11], complex nonlinear dynamics [12], design of photonic crystal fibers and optical components [13,14], pulse characterization [15], and optical communications [16,17]. In [18], we have shown that the combination of a graphical approach with the machine-learning method of neural networks (NNs) can provide a rapid and precise identification of the parameters of nonlinear pulse shaping systems based on pulse propagation in a normally dispersive fiber that are required to generate pulses with preset temporal features.…”