CC BY
0Artificial intelligence for fiber lasers and sensors
A. Kokhanovskiy1
1- School of Physics and Engineering, ITMO University, St. Petersburg 197101, Russia
In recent years, fiber lasers and optical sensors have become an integral part of numerous high-tech applications, ranging from industrial monitoring and medical diagnostics to telecommunications [1]. These photonics devices offer unique advantages, such as high precision, a wide measurement range, and the ability to operate in extreme conditions. However, as demands for measurement accuracy, speed, and reliability increase, along with the growing volumes of data that need to be processed, there arises a need for new methods of data analysis and processing.
Artificial intelligence (AI) algorithms, particularly machine learning (ML), open up new horizons for improving the performance of optical sensors. With their ability to analyze large volumes of data and uncover hidden patterns, AI can significantly enhance the sensitivity, accuracy, and reliability of these devices. The application of ML allows for the automation of sensor calibration, prediction of sensor failures, and improvement of real-time signal processing.
This article explores the application of artificial intelligence algorithms in both fiber lases and fiber sensors. Special attention will be given to analyzing current achievements and development prospects in this rapidly evolving area. Also, the recent progress of our group will be demonstrated. We demonstrate that the Reinforcement learning algorithm utilizes sophisticated strategies to ensure a guaranteed harmonic mode-locked regime of the highest order by efficiently managing the laser system's pumping power and the nonlinear transmission of a nanotube absorber. Also, Using the example of a fiber sensor based on a multicore optical fiber with densely written fiber Bragg gratings, the possibility of increasing spatial resolution by a factor of five is demonstrated through the interpretation of complex reflection spectra by deep learning algorithms.
[1] G. Genty, et al, Machine learning and applications in ultrafast photonics, Nature Photonics 15(2), 91-101 (2021).
[2] A. Kokhanovskiy, et al, Highly dense FBG temperature sensor assisted with deep learning algorithms, Sensors 21(18) 6188 (2021).
[3] A. Kokhanovskiy, et al, Multistability manipulation by reinforcement learning algorithm inside mode-locked fiber laser, Nanophotonics (2024).
