The ability to tailor a specific electromagnetic field pattern along an arbitrary selected surface is interesting and of substantial importance considering its numerous immediate applications. It belongs to a class of inverse source problems, and as such it is challenging when only partial data is given. Here, a deep learning-based method that can map the electromagnetic field from an arbitrarily selected surface to a flat surface is presented. In addition, phase retrieval capability is demonstrated for finding both the phase and amplitude on an input flat surface from knowing only the amplitude on an arbitrarily selected surface.
The possibility of employing the spatial degree of photons for communications is gaining interest in recent years due to its unbounded dimensionality. A natural basis to span the transverse profile of photons is comprised of Laguerre-Gaussian (LG) modes which are characterized with two topological numbers: l, the orbital index, describing the orbital-angular-momentum (OAM) in units of “h-bar” per photon in the beam and p, which is the radial index or radial quantum number. One of the main challenges for utilizing LG modes in communications is the ability to perform mode-sorting and demultiplexing of the incoming physical data-flow. Nowadays, there are two leading approaches to mode demultiplexing. The first approach uses intricate optical setups in which the l and p degrees of freedom are coupled to other degrees of freedom such as the angle of propagation or the polarization of the beam. Most of these methods address either the OAM or the radial index degrees of freedom. The second approach, which emerged recently, suggests using just a camera to detect the intensity of the incoming light beam and to utilize a deep neural network (DNN) to classify the beam. To date, demonstrated DNN-based demultiplexers addressed solely the OAM degree of light. We report on an experimental demonstration of state-of-the-art mode demultiplexing of Laguerre-Gaussian beams according to both their orbital angular momentum and radial topological numbers using a flow of two concatenated deep neural networks. The first network serves as a transfer function from experimentally-generated to ideal numerically-generated data, while using a unique "Histogram Weighted Loss" function that solves the problem of images with limited significant information. The second network acts as a spatial-modes classifier. Our method uses only the intensity profile of modes or their superposition with no need for any phase information.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.