In wavelength-division-multiplexed (WDM) systems, performance is affected by the bit patterns on the various channels and, in order to account for this pattern dependence, one should perform a large number of simulations, each by using different patterns of suitable length in all channels and then average all results. However, since the number of patterns to be considered for averaging could be considerable, this approach would be very demanding from a computational point of view, even using multicanonical Monte Carlo (MMC) simulations. In single-channel systems, pattern dependence can be accounted for through the pattern perturbation method, a computationally efficient way for performing a random walk over an extended state space, which includes, in addition to noise components, also the transmitted bit pattern. In this way, not only the most relevant noise configurations but also the most relevant bit patterns (those having a major impact on the error events) are sampled more frequently by the MMC simulation, leading to enhanced efficiency in estimating the error probability. Here, the pattern perturbation method is extended to the WDM case, and simulations results are reported to show the effectiveness of the proposed approach.
© 2010 IEEEPDF Article