PMD emulation

Abstract

As PMD has become an increasingly significant issue in high-bit-rate fiber optic systems, a need has developed for laboratory instruments and software tools capable of rapidly exploring the effects of PMD on various test items. The random nature of PMD dictates that to characterize its effects on transmitter/receiver pairs and on PMD compensation systems, one must repeatedly measure the system performance over a wide sample space of PMD states. This need has spurred the development of several methods for accurately and rapidly emulating the random variations of PMD in real fibers, as well as techniques for generating specific components and combinations of first-and higher-order PMD in a predictable and repeatable way. This chapter reviews several of these methods for both statistical and deterministic PMD emulation. The underlying concepts and rationales for various design architectures are discussed. A common analytical model for describing multisection all-order emulators is presented and a simple design example is used to further illustrate the concepts.

All-Fiber PMD Emulator with Reduced Number of Polarization Controllers between Sections

“…Following established PMD theory [1][2]9], we use computer simulation to evaluate the emulator design and compare it with other schemes. As an example, we use six off-the-shelf polarization controllers and seven sections of PMF.…”

“…The need for polarization-mode-dispersion (PMD) emulation has become well-recognized as a major challenge in combating PMD effects in high performance optical communication systems [1][2]. Ideal features of practical PMD emulators include: (i) low loss and polarization-dependent-loss (PDL), (ii) high stability (or even repeatability), (iii) fast control speed, (iv) accurate first-and higher-order statistical distributions, (v) quadratic fall-off of the autocorrelation function outside a limited frequency range (an essential requirement for multi-channel systems), etc.…”

Emulator with inverse trend in first-order and second-order polarisation mode dispersion

“…SO-PMD results in the ACF being asymmetric although it does not have influence on the BACF [23]. d) The emulator should have low insertion loss and exhibit negligible polarisation dependent loss (PDL) [17]. PDL influences both the FO-and SO-PMD statistics [24].…”

“…b) The emulator should be able to reach any combination of FO-and SO-PMD values [17]. c) When the autocorrelation function (ACF) is averaged over an ensemble of fibre realisation, the background autocorrelation function (BACF) should tend towards zero outside the limited frequency range.…”

Optical Fiber Measurement

“…Interested readers on the statistics of DGD generated by different types of PMD emulators can find details at [38,39,40,41]. Different control mechanisms and algorithms, such as the control of SOP or optical phase retardation between birefringence sections or the PM fiber sections control of the DGD provided by each section, may exhibit different PMD statistics.…”