The lasers generating the signal are not perfect and generate a Gaussian-like profile of wavelengths, albeit narrow.
In this experiment[1] they sent single photons alongside the classical data:
Alice’s qubit is then encoded onto the horizontal (H) and vertical (V ) components of the heralded photon’s polarization |ψ〉A = α|H〉 + β|V 〉using polarization waveplates.
This photon is then multiplexed using an O-band/C-band WDM to co-propagate with the 400-Gbps C-band classical signal over 15.2 km of spooled optical fiber (SMF-28(R) ULL) to Charlie.
To prevent the detection of C-band photons due to insufficient filter isolation, we cascade two cascaded O-band/C-band WDMs before the FBGs and a 1290-nm DWDM to achieve >190 dB rejection of C-band light.
The lasers are not Gaussian but Lorenztian. This is a detail that’s not terribly important.
The lasers generate frequencies in C-band at ~194THz. The O-band photon source is ~229THz. Classically these fields do not interact.
For passive beam splitting, there is also no quantum interference. The photo detectors are broadly sensitive to both frequency bands. So if the C-band light hits the photodiode you won’t be able to count photons. The filters are there to reduce the background “noise” of the C-band light hitting the photodiode that’s being used for photon counting. The use of filters for this purpose is very common and really not that interesting.
In this experiment[1] they sent single photons alongside the classical data:
Alice’s qubit is then encoded onto the horizontal (H) and vertical (V ) components of the heralded photon’s polarization |ψ〉A = α|H〉 + β|V 〉using polarization waveplates.
This photon is then multiplexed using an O-band/C-band WDM to co-propagate with the 400-Gbps C-band classical signal over 15.2 km of spooled optical fiber (SMF-28(R) ULL) to Charlie.
To prevent the detection of C-band photons due to insufficient filter isolation, we cascade two cascaded O-band/C-band WDMs before the FBGs and a 1290-nm DWDM to achieve >190 dB rejection of C-band light.
WDM = wavelength division multiplexer, FBG = fiber Bragg grating[2].
If it was just a distraction, why did they bother adding 190+ dB of filter rejection?
[1]: https://opg.optica.org/optica/fulltext.cfm?uri=optica-11-12-...
[2]: https://en.wikipedia.org/wiki/Fiber_Bragg_grating