We design an optical feedback network making use of machine learning (ML) techniques and demonstrate via simulations its ability to correct for the effects of turbulent propagation on optical modes. This artificial neural network scheme relies only on measuring the intensity profile of the distorted modes, making the approach simple and robust. The network results in the generation of various mode profiles at the transmitter that, after propagation through turbulence, closely resemble the desired target mode. The corrected optical mode profiles at the receiver are found to be nearly identical to the desired profiles, with near-zero mean square error indices. We are hopeful that the present results combining the fields of ML and optical communications will greatly enhance the robustness of free-space optical links.

中文翻译：

---

用人工神经网络进行湍流校正

我们设计了一种利用机器学习（ML）技术的光反馈网络，并通过仿真展示了其校正湍流传播对光模影响的能力。这种人工神经网络方案仅依赖于测量变形模式的强度分布，从而使该方法既简单又健壮。网络导致在发射机处生成各种模式轮廓，这些模式轮廓在通过湍流传播之后非常类似于所需的目标模式。发现接收器处的校正后的光学模式轮廓与所需轮廓几乎相同，均方误差指数接近零。我们希望将ML和光通信领域结合起来的当前结果将大大增强自由空间光链路的鲁棒性。