Deep learning enabled superfast and accurate M²evaluation for fiber beams

Abstract

We introduce deep learning technique to predict the beam propagation factor M 2 of the laser beams emitting from few-mode fiber for the first time, to the best of our knowledge. The deep convolutional neural network (CNN) is trained with paired data of simulated nearfield beam patterns and their calculated M 2 value, aiming at learning a fast and accurate mapping from the former to the latter. The trained deep CNN can then be utilized to evaluate M 2 of the fiber beams from single beam patterns. The results of simulated testing samples have shown that our scheme can achieve an averaged prediction error smaller than 2% even when up to 10 eigenmodes are involved in the fiber. The error becomes slightly larger when heavy noises are added into the input beam patterns but still smaller than 2.5%, which further proves the accuracy and robustness of our method. Furthermore, the M 2 estimation takes only about 5 ms for a prepared beam pattern with one forward pass, which can be adopted for realtime M 2 determination with only one supporting Charge-Coupled Device (CCD). The experimental results further prove the feasibility of our scheme. Moreover, the method we proposed can be confidently extended to other kinds of beams provided that adequate training samples are accessible. Deep learning paves the way to superfast and accurate M 2 evaluation with very low experimental efforts.

Photonic lantern based real-time modal instability diagnostic tool for high power lasers

“…Many approaches utilizing deep learning and neural networks have been proposed recently [7][8][9] while some incorporate physics inspired constraints and modeling to achieve more robust results 10,11 while others have proposed machine learning approaches operating in near real time. [12][13][14][15] The general trend in this category of modal decomposition methods rely on machine learning to shortcut computational steps in optimization routines or in interferometric beam analysis previously presented. The reader is encourage to look in to the more detailed works highlighting works employing optimization methods, 16,17 fractional Fourier transforms, 18 and digital holography.…”

Fiber laser development enabled by machine learning: review and prospect

“…In 2019, Yi An et al utilized a trained CNN to achieve M 2 determination of the fiber beams in about 5 ms with only one near-field beam pattern from the CCD, which is highly competitive in real-time measurement for time-varying beams [131]. This method also shows excellent robustness for imperfect beam patterns, such as noisy patterns and patterns from the CCD with vertical blooming [83] (Fig.…”

Functional Fibers and Functional Fiber-Based Components for High-Power Lasers

“…The mode information carried by laser beam is the key to understand intrinsic properties and transmission characteristics of HPFLs, which are challenging to be completely analyzed even with complicated measurement methods [489,492,493]. However, it should be noted that the emerging machine learning technology is expected to provide an effective and accurate technical scheme for online analysis [488][489][490][491][492][493][494][495][496].…”