High‐Fidelity Information Transmission Through the Turbulent Atmosphere Utilizing Partially Coherent Cylindrical Vector Beams

As the demand for high‐capacity and high‐fidelity communication systems continues to increase, addressing the challenges posed by noise and atmospheric turbulence disturbances is imperative. This study introduces and experimentally implements a novel free‐space optical communication protocol. This protocol combines the advantages of reducing the spatial coherence of light at the source with the capabilities of convolutional neural networks at the receiver to encode and transmit optical images through a noisy link. Light beams that are robust against noise are generated and atmospheric turbulence is modeled in a laboratory setting by decreasing the degree of spatial coherence of the source. Eight orbital angular momentum states, four polarizations, and eight coherence states of a light source that generates partially coherent cylindrical vector beams are utilized. These elements are employed to achieve a 256‐ary encoding/decoding data transmission within our protocol. This study is expected to catalyze further research into the utilization of partially coherent light and neural networks in the realm of free‐space optical communications.