Learning to Extract Distributed Polarization Sensing Data From Noisy Jones Matrices (Th2A.14)
Presenter: Mohammad Farsi, Chalmers University of Technology
We consider the problem of recovering spatially resolved polarization information from receiver Jones matrices. We introduce a physics-based learning approach, improving noise resilience compared to previous inverse scattering methods, while highlighting challenges related to model overparameterization.
Authors:Mohammad Farsi, Chalmers University of Technology / Christian Häger, Chalmers University of Technology / Magnus Karlsson, Chalmers University of Technology / Erik Agrell, Chalmers University of Technology