Speckle-based structured light shift-keying for non-line-of-sight optical communication

Abstract

We report an experimental proof of concept for speckle-based one-to-three non-line-of-sight (NLOS) free space optical (FSO) communication channels employing structured light shift-keying. A 3-bit gray image of resolution 100×100 pixels is encoded in Laguerre–Gaussian or Hermite–Gaussian beams and decoded using their respective intensity speckle patterns via trained 1D convolutional neural network. We have achieved an average classification accuracy of 96% and 93% using LGml and HGpq beams, respectively, among all three channels. It demonstrates the directional independence and broadcasting capability of speckle-based decoding (SBD) in FSO communication using structured light. Further, we have extended the study from 2D to 1D SBD in one-to-three NLOS FSO communication channels to decrease the computational cost and to emphasize the importance of the 1D SBD approach.

Machine learning meets singular optics: speckle-based structured light demultiplexing

“…In this case, we first extract the features from speckle images of various modes using a wavelet scattering network (WSN). The corresponding features given by scattering coefficients are fed to 1D CNN for the training classification model of LG and HG modes 10,11 . The same Alexnet can also be trained for these data but extracting the feature beforehand the model training accelerates the training process and improves accuracy especially when the dataset is small.…”

“…Therefore, any small portion of the field containing sufficient speckles is adequate for beam recognition. This opens the possibility of non-line-of-sight optical communication where the detectors can be placed at any off-axis position which is not possible with direct beam detection methods 10 . Also, these methods are robust to noises such as atmospheric turbulence noise and Gaussian white noise.…”