Estimating Outages Due to Polarization Mode Dispersion Using Extreme Value Statistics

Abstract
Polarization mode dispersion (PMD) is a time varying phenomenon, which in a network can result in system outages. In this paper, it is demonstrated that the method of extreme value statistics can be used to model the occurrence of these rare events. The extreme value behavior of the bit error rate (BER) due to PMD is analyzed both theoretically and experimentally, revealing that the logarithm of the BER follows a Gumbel distribution