Speckle-driven single-shot orbital angular momentum recognition with ultra-low sampling density

Area of Science:

• Optics and Photonics
• Information Processing

Background:

• Orbital angular momentum (OAM) recognition is vital for optical communications and quantum technologies.
• Conventional methods fail in scattering media (e.g., multimode fibers) and require high-resolution sensors.

Purpose of the Study:

• To introduce a novel, efficient OAM recognition technique robust to scattering media.
• To reduce the reliance on high-resolution sensors and complex data acquisition.

Main Methods:

• Developed spatially multiplexed points detection (SMPD), a speckle-driven OAM recognition technique.
• Utilized scattering media as intrinsic encoders by extracting intensity from few spatially distributed points in a speckle plane.
• Reduced sampling requirements significantly compared to conventional imaging.

Main Results:

• Achieved over 99% retrieval accuracy for OAM recognition using only 16 sampling points (0.024% sampling density).
• Demonstrated versatility through spatiotemporally interleaved vortex beams decoding, high-capacity OAM-multiplexed communication, and machine learning tasks (MNIST, Fashion-MNIST).

Conclusions:

• SMPD transforms scattering from a hindrance into an efficient encoding mechanism.
• Establishes a scalable strategy for optical information processing and fiber-based sensing in complex environments.