We present numerical evidence that maximum-likelihood sequence estimation (MLSE)-based receivers, assuming ideal implementation, could operate over as much as 11 000 ps/nm of chromatic dispersion (700 km of G.652 fiber), keeping the total penalty with respect to back-to-back within approximately 3 dB. We argue that these results suggest that the penalty of an MLSE-based receiver versus total chromatic dispersion could be bounded to an asymptotic value, provided that enough trellis states are used to properly deal with the channel memory.
No takes yet. Share an insight, caveat, or question.
Bosco et al. (2006) studied this question.