We propose an improved hologram with both phase and amplitude modulation to generate superimposed fractional optical vortices (SFOVs). The modulation of the optical field’s amplitude and phase is achieved through the utilization of controllable diffraction efficiency of the transmission function. The resulting interference fringes of an SFOV with four orbital angular momentum (OAM) modes exhibit a distinctive double-petal-like structure, serving as a distinguishable feature for the beam’s topological charges. Accurate demodulation of the multiplexed OAM modes of 256-ary SFOV is achieved using a residual next neural network based on machine learning. To showcase its practical utility, we employ the coherent OAM multiplexing system to transmit a Newton portrait with 0.01% error rate. Furthermore, the system robustly identifies beams propagating through computer-simulated oceanic turbulence channels to aid in the development of underwater optical communication. These promising results demonstrate the potential to further expand the range of modes and enhance the information processing capabilities in optical communication.
Read full abstract