I worked on something similar years ago. As you know, Bi Refringence is the underlying mechanism behind faraday rotation. I was observing several radio galaxies at 8 mm (36 ghz). I developed a process I called "Polarization diversity interferometry. I detected the signal with four separate receiver horns mounted at the phase center of the dish. The horns were configured to detect the signal in both RCP and LCP polarization and vertical and horizontal polarization. I then arbitrarily selected one of the signals as the "Early wave" and introduced it to a variable delay. I then looked at the correlation function between the "early wave " and the "late wave" as function of delay time. This has the effect of "unwinding" a signal that has undergone Faraday rotation. Because birefringence is determined by a combination of the index of refraction of the intervening medium and the local magnetic field the correlation function contains signatures revealing the properties of both.

While this work was done years ago, in the 1970's, It is still a useful observing strategy. I see no reason that a similar approach in optical astronomy could not be useful.

I hope this helps, good luck.

Charles Ivie