Classifying beams carrying orbital angular momentum with machine learning: tutorial

Abstract

This tutorial discusses optical communication systems that propagate light carrying orbital angular momentum through random media and use machine learning (aka artificial intelligence) to classify the distorted images of the received alphabet symbols. We assume the reader is familiar with either optics or machine learning but is likely not an expert in both. We review select works on machine learning applications in various optics areas with a focus on beams that carry orbital angular momentum. We then discuss optical experimental design, including generating Laguerre–Gaussian beams, creating and characterizing optical turbulence, and engineering considerations when capturing the images at the receiver. We then provide an accessible primer on convolutional neural networks, a machine learning technique that has proved effective at image classification. We conclude with a set of best practices for the field and provide an example code and a benchmark dataset for researchers looking to try out these techniques.

Machine learning meets singular optics: speckle-based structured light demultiplexing

“…The accuracy of detection using these traditional methods is limited by alignment issues, noise, and beam wandering. Almost a decade ago, artificial intelligencebased structured light recognition models were proposed and used in optical communication links 4,5 . These models uplifted the constraints of traditional detection methods by reducing the complexity, boosting the speed of detection, and automating the process.…”

Generating terahertz multiple vortex beams using graphene metasurfaces

“…Several technologies have been developed to improve electromagnetic spectrum (OAM) at visible wavelengths through a plasmonic metasurface was introduced (Li et al 2021). Additionally, the investigation involved the observation of a surface plasmon vortex that carries OAM within a gold metasurface when subjected to linearly polarized optical excitation (Avramov-Zamurovic et al 2023). Reconfigurable metasurface can change their electromagnetic properties dynamically.…”

Machine-learning-assisted orbital angular momentum recognition using nanostructures

“…To address these issues, AI-powered OAM demultiplexing has been introduced [13][14][15][16], including speckle-based OAM recognition [17] harnessing the advanced computational capabilities of AI, essential for effectively identifying and extracting information from diverse intensity modes of OAM. This ability makes the intricate task of separating and decoding the interwoven OAM channels manageable.…”