A team led by Fu Shiyao and Gao Chunqing from the School of Optoelectronics at Beijing Institute of Technology has unveiled an innovative, compact method for the intelligent sensing of photon angular momentum spectra. Leveraging their self-developed periodically gradient metasurfaces, this method effectively extracts the angular momentum characteristics of light beams. Coupled with intelligent analysis via the SADT-Net neural network, it achieves precise measurement of photon angular momentum spectra, boasting a mean square error of 10-6 and supporting up to 34 total angular momentum modes. This groundbreaking work has been published in the prestigious SCI top-tier journal, ACS Photonics.
