Abstract

In this paper, closed-form analytical bounds on PMD-induced symbol error rate and outage probability in high-speed long-haul optical orthogonal frequency division multiplexed (OFDM) systems are derived, evaluated, and verified experimentally in order to assess the PMD tolerance of this modulation format. To obtain the analytical results, the PMD channel is modeled as a linear time-invariant system, whose end-to-end transfer function is used to upper-bound symbol error and outage probabilities in IM/DD optical OFDM transmission. The symbol error rate bounds predicted by the general analytical model are verified experimentally on a 10-Gb/s OFDM system with instantaneous DGD ranging between $\Delta\tau = 0$ and 120 ps. The outage probability bounds indicate that, if no RF guard bands are required, OFDM enables high-speed transmission with at least twice the PMD tolerance provided by an equivalent uncompensated OOK-based system at system outage probabilities $P_{{\rm out,\ sys}}<10^{-5}$. If RF guard bands are required to mitigate effects of other distortions, it is shown that a penalty in the system PMD tolerance proportional to the intermediate RF subcarrier frequency, $f_{{\rm RF}}$, is exerted. Consequently, a tolerance tradeoff exists in IM/DD OFDM systems, wherein $f_{\rm RF}$ as well as baseband constellation size may be viewed as design parameters that can be optimized depending on specific system requirements.

© 2008 IEEE

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription