Physics-driven untrained neural network for vortex beam compensation in adaptive optics aided underwater wireless optical communications.

Orbital angular momentum (OAM) multiplexing is emerging as a critical technique for achieving high data capacity in underwater wireless optical communications (UWOC). Nonetheless, wavefront distortions induced by underwater turbulence compromise the orthogonality of OAM modes. In this paper, we introduce a physics-driven untrained learning approach for adaptive optics that operates independently of extensive amplitude datasets.

All-optical image transmission through a dynamically perturbed multimode fiber and a ring-core fiber using diffractive deep neural networks.

In this study, we present an all-optical image reconstruction technique leveraging a diffractive deep neural network (D2NN) within a ring-core fiber (RCF) architecture. Orbital angular momentum (OAM) modes are employed to facilitate imaging transmission. We experimentally validate the efficacy of our approach for complex field diffractive image reconstruction through a multimode fiber (MMF) and RCF at a 1550 nm operating wavelength.

Probabilistic shaping and forward error correction scheme employing uneven segmentation mapping for data center optical communication.

The combination of probabilistic shaping (PS) technology and forward error correction (FEC) technology can significantly boost the performance of a transmission system. In this paper, we propose a probabilistic shaping distribution matching algorithm employing uneven segmentation for data center optical networks, while keeping extremely low computational complexity for both encoding and decoding. Based on the proposed probabilistic shaping distribution matching algorithm, we develop a novel integrated scheme of PS and FEC coding that lifts the restrictions on the use of FEC technology and increases the use of interleaver.