Experimental validation of XPM mitigation using a generalizable multi-task learning neural network.

We address the development of efficient neural network (NN)-based post-equalizers in long-haul coherent-detection dense wavelength-division multiplexing (DWDM) optical transmission systems. To achieve a high level of generalization of the NN-based equalizers, we propose to employ multi-task learning (MTL). MTL refers to a single shared machine learning (NN) model that can perform multiple different (albeit related) tasks.

Demanding response time requirements on coherent receivers due to fast polarization rotations caused by lightning events.

Lightning events can cause fast polarization rotations and phase changes in optical transmission fibers due to strong electrical currents and magnetic fields. Whereas these are unlikely to affect legacy transmission systems with direct detection, different mechanisms have to be considered in systems with local oscillator based coherent receivers and digital signal processing. A theoretical analysis reveals that lightning events can result in polarization rotations with speeds as fast as a few hundred kRad/s.