Stratified-Sampling Estimation of PDL-Induced Outage Probability in Nonlinear Coherent Systems

We present a novel adaptive stratified-sampling (SS) algorithm to efficiently assess the polarization-dependent loss (PDL) induced outage probability (OP) in presence of fiber nonlinearity and polarization mode dispersion. We show that, at standard OP values, the SS algorithm more than quadruples the efficiency in simulation time with respect to the standard Monte Carlo method at equal accuracy. We applied the SS algorithm to polarization-division multiplexing quadrature phase-shift keying transmissions in both dispersion-managed and dispersion-unmanaged links, showing a non-negligible impact of PDL using typical ITU-T recommendations. Finally, we show that estimating the OP by neglecting the interaction of nonlinearity and PDL can be largely optimistic.