Transmission and information encryption of windmill array beams in oceanic turbulence

Oceanic turbulence-induced beam wandering, scintillation and phase distortion are the main factors limiting underwater optical communication performance. To address this issue, this study investigates the propagation of windmill array beams through oceanic turbulence and evaluates their potential application as binary information carriers. A cross-spectral density evolution model based on the extended Huygens-Fresnel principle and Nikishov's temperature-salinity spectrum is employed to analyze the propagation characteristics of windmill beams in oceanic turbulence. Numerical simulation results suggest that windmill beams can preserve their basic structure during propagation in oceanic turbulence. Furthermore, laboratory experiments conducted in a 1.2 m turbulence simulator demonstrate that windmill beams preserve stable and distinguishable leaf structures under pump-induced turbulence and varying salinity levels. Since the number of leaves remains constant at different propagation distances, we further propose an effective encoding scheme mapping leaf number 1-4 to 2-bit symbols, combined with convolutional neural network (CNN) decoding. Experimental results demonstrate that the system, integrated with CNN-based decoding, maintains high recognition accuracy despite turbulence-induced distortions. The CNN-based decoder consistently achieves classification confidence above 98.0% under turbulent conditions and salinity levels ranging from 0 to 3.3%. These findings suggest windmill beams as a promising approach to enhance reliability and capacity in underwater optical communication.