High-performance mode decomposition using physics- and data-driven deep learning

Abstract

A novel physics- and data-driven deep-learning (PDDL) method is proposed to execute complete mode decomposition (MD) for few-mode fibers (FMFs). The PDDL scheme underlies using the embedded beam propagation model of FMF to guide the neural network (NN) to learn the essential physical features and eliminate unexpected features that conflict with the physical laws. It can greatly enhance the NN’s robustness, adaptability, and generalization ability in MD. In the case of obtaining the real modal weights (ρ2) and relative phases (θ), the PDDL method is investigated both in theory and experiment. Numerical results show that the PDDL scheme eliminates the generalization defect of traditional DL-based MD and the error fluctuation is alleviated. Compared with the DL-based MD, in the 8-mode case, the errors of ρ2 and θ can be reduced by 12 times and 100 times for beam patterns that differ greatly from the training dataset. Moreover, the PDDL maintains high accuracy even in the 8-mode MD case with a practical maximum noise factor of 0.12. In terms of adaptation, with a large variation of the core radius and NA of the FMF, the error keeps lower than 0.43% and 2.08% for ρ2 and θ, respectively without regenerating new dataset and retraining NN. The experimental configuration is set up and verifies the accuracy of the PDDL-based MD. Results show that the correlation factor of the real and reconstructed beam patterns is higher than 98%. The proposed MD-scheme shows much potential in the application of practical modal coupling characterization and laser beam quality analysis.

General error analysis of matrix-operation-mode decomposition technique in few-mode fiber laser

“…Generally, direct measurement methods require a broadband light source or reference beam, which may limit its application in some scenarios with dynamic changes in linewidth, e.g., in characterizing the spatial characteristics between linewidth and transversal mode instability (TMI) threshold [28].The second kind of MD methods i.e., numerical analysis MD methods are based on computer algorithms, in which only the near-field or far-field beam intensity distribution of the fiber output beams are required. Commonly, the MD algorithms contain the Gerchberg-Saxton (GS) algorithm (a phase-retrieval algorithm) [29], line search algorithm [30], stochastic parallel gradient descent (SPGD) algorithm [31][32][33][34][35], and neural networks [36][37][38][39][40][41][42]. The first three algorithms (i.e., GS, line search and SPGD algorithms) require iterative calculation, resulting in time-consumption (e.g., the decomposition speed of MD technology based on the SPGD algorithm is ~10Hz), and the MD accuracy is also affected by the initial value selection of the algorithm.…”

Simultaneous determination of polarization states and mode coefficients of fiber LP modes with four-channel off-axis holography

“…So far, a variety of mode decomposition (MD) techniques which calculates the mode coefficients have been proposed, such as the spatial spectroscopy method [6][7][8], correlation analysis method [9][10][11], numerical analysis method [12][13][14][15][16], digital holography method [17,18] and the matrix analysis method [29] etc. Both mode weights and phase differences can be determined simultaneously using these methods.…”

Statistical learning method for modal analysis of optical fibers

“…8 There are also approaches combining physical and data driven models. 7 Here, a statistical approach utilizing a fully interpretable complex valued model is proposed. With this, learning the fiber modes based on the light field before and after propagation through the fiber is possible.…”