Mitigating the ambiguity problem in the CNN-based wavefront correction.

In this work, we propose an attention-based adaptive optics method that uses a non-local block to integrate phase diversity with a convolutional neural network (CNN). The simulation results showcase the effectiveness of the proposed method to mitigate the ambiguity problem of phase retrieval and better performance than traditional CNN-based wavefront correction.

Jointly recognizing OAM mode and compensating wavefront distortion using one convolutional neural network.

In this work, a new recognition method of orbital angular momentum (OAM) is proposed. The method combines mode recognition and the wavefront sensor-less (WFS-less) adaptive optics (AO) by utilizing a jointly trained convolutional neural network (CNN) with the shared model backbone. The CNN-based AO method is implicitly applied in the system by providing additional mode information in the offline training process and accordingly the system structure is rather concise with no extra AO components needed. The numerical simulation result shows that the proposed method can improve the recognition accuracy significantly in different conditions of turbulence and can achieve similar performance compared with AO-combined methods.

DNN-based aberration correction in a wavefront sensorless adaptive optics system.

Existing wavefront sensorless (WFS-less) adaptive optics (AO) generally require a search algorithm that takes lots of iterations and measurements to get optimal results. So the latency is a serious problem in the current WFS-less AO system, especially in applications to free-space optics communication. To solve this issue, we propose a deep neural network (DNN)-based aberration correction method. The DNN model can detect the wavefront distortion directly from the intensity images, thereby avoiding time-consuming iterative processes. Since the tip-and-tilt mode of Zernike coefficients are considered, the tip-tilt correction system is not necessarily required in the proposed method. From our simulation results, the proposed method can effectively reduce the computation time and has an impressive improvement of root mean square (RMS) in different turbulence conditions.