Deep-Learning-Assisted Detection For Topological Charges of Vortex Beams Through Strong Scattering Medium

Vortex beams have special spiral phase factors, and the communication capacity can be greatly improved by using vortex beams for communication coding. The atmospheric turbulence and haze in the actual communication environment will lead to the scattering of vortex beams and form speckles, which increases the difficulty of information decoding in the vortex optical communication. Therefore, it is of great significance to accurately and efficiently measure the topological charges of vortex beams from the speckles for their application in vortex optical communication. The characteristics of the speckle field formed after vortex beams passing through scattering medium are closely related to the topological charges. Based on the efficient feature extraction of depth neural network, the measurement of topological charges of scattered vortex beams is realized by using classified neural network, and the measurement accuracy is up to 100%.