Abstract
The atmospheric turbulence disturbs the phase fronts of orbital angular momentum (OAM) beams, which significantly affects the detection of beam modes. In this paper, we propose a method based on the convolutional neural network (CNN) to detect OAM modes in atmospheric turbulence. This method does not require a separate system to suppress the influence of atmospheric turbulence. The propagation of multimode OAM beams in atmospheric turbulence is simulated by setting several random phase screens in the transmission channel. We select three levels of atmospheric turbulence. In all these cases, the predicted error of the trained CNN is lower than 2 ×10<sup>-5</sup> , which indicates that our network can detect the mode distribution in multimode OAM beams efficiently and accurately. We believe that this approach for detecting OAM modes holds great promise for potential applications and will provide widespread benefits for many optical fields.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
