Abstract
The helical phasefront and orbital angular momentum (OAM) of vortex electromagnetic waves (VEMW) have attracted extensive attention in expanding communication capacity and enhancing radar imaging resolution. However, the VEMW is sensitive to atmospheric turbulence (AT), which will cause helical phasefront distortion and OAM mode diffusion. Hence, a convolutional neural network (CNN) based method for AT compensation is proposed. Specifically, the developed CNN automatically learns the AT characteristic parameters from the distorted VEMW phasefront and reconstructs the equivalent AT phase screen, and then, compensation is achieved by loading the inverse of the predicted phase on the distorted VEMW. The simulations show that the trained CNN has strong generalization capability and can perform high-quality AT prediction for different turbulence strengths and OAM modes, after compensation, the distorted phase is repaired and the mode purity is significantly improved. Moreover, comparing with the conventional AT compensation method, the proposed method shows smaller computational burdens and higher compensation accuracy.
Published Version
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