存在偏振相关损失条件下偏振模色散的特点

偏振模色散和偏振相关损失的联合作用，会产生不规则的色散现象，这一点不能直接通过琼斯距阵本征分析法的运用进行描述。由此引出了对偏振模色散和偏振相关损失共存条件下的研究。通过对琼斯本征分析法进行修正，来分析存在偏振相关损失条件下偏振模色散的特征距阵，理论分析表明，由于偏振相关损失的影响，即使在忽略差分损耗频率相关性的条件下，偏振模色散的特征距阵也会产生根本性的变化。     

基于PLC的PMD补偿器中PDL效应的分析

应用穆勒（Müller）矩阵对硅基平面光波导线路（Planar lightwave circuit PLC）补偿器中PDL对差分群时延(Differential group delay DGD)分布的影响进行了分析，在不同的PDL值下对各种不同级次的PLC单元级联情况的DGD值分布进行了讨论，并与Maxwell分布进行了比较，得到了在特定PDL值情况下最为优化的PLC级次.     

Combined effect of PMD and PDL on four-wave-mixing-induced crosstalk in WDM system

We derive an explicit autocorrelation function (ACF) formula of state of polarization for a fiber transmission system with polarization mode dispersion (PMD) and polarization dependent loss (PDL), which is found to agree well with Monte Carlo simulation. Then we use the new ACF to investigate the combined effect of PMD and PDL on the polarization multiplexed scheme. We find the performance of the polarization multiplexed scheme can be deteriorated more severely than the case without PDL.     

Power penalty of cascaded Bragg gratings in the presence of PMD and PDL

Considering the presence of birefringence in Bragg gratings, the spectral polarization mode dispersion (PMD) and polarization dependent loss (PDL) induced power penalty of eye opening is examined for a network with cascaded Bragg gratings. It is shown that when the birefringence in the grating is less than 10⁻⁵, the induced power penalty is less than 0.1 dB for a network with up to 30 cascaded Bragg gratings. However, when the grating birefringence is on the order of 10⁻⁴, the induced power penalty can be as high as 5.6 dB for a network with 14 cascaded Bragg gratings.     

Statistical PMD and PDL effects emulator based on polarization maintaining optical fiber segments

Low cost statistical PMD/PDL emulator is presented. The emulator was initially designed as a classical PMD emulator. But PDL measurements of the designed emulator show that it can be also used as statistical PDL emulator. The emulator was constructed by concatenating 15 highly birefringence optical fiber segments and randomly varying the mode coupling between them, by rotating the polarization state between them. The achieved numerical and experimental results show that the designed emulator can mimic the polarization effects (PMD and PDL) of a real modern optical fiber link.     

PMD emulation

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.     

A polarization mode dispersion (PMD) emulator with tuneable first-order PMD and constant second-order PMD

We design a polarization mode dispersion (PMD) emulator with fixed second-order polarization mode dispersion (SO-PMD) but varying first-order PMD (FO-PMD). The emulator constitutes of an optical delay line (ODL), a polarization controller (PC) and a fixed number of randomly concatenated polarization maintaining fibre (PMF) segments. An understanding of the SO-PMD equation is the first vital step to consider before designing such an emulator. The control of the differential group delay (DGD) statistics with wavelength proves to be the key measure for this design. Results show that the mean DGD (or the mean magnitude of the FO-PMD vector ()) of the emulator is biased towards the dominant wavelength-independent of the ODL. This is provided the dominant is by far greater than FO-PMD contributions from the other cascaded sections. Experimentally it is shown that when the DGD (Δτ) is wavelength-independent due to the absence of mode coupling, or when the wavelength-dependent DGD spectra do not change with time due to fixed mode coupling, there is negligible influence on the SO-PMD. The PC angle is controlled at an angle θ to ensure that the sub-emulator is always parallel to the ODL . Thus by rotating the mode coupling angle θ, we change the wavelength-dependent DGD spectra thereby ensuring SO-PMD variation.     

PMD compensation techniques

Mitigating PMD is still a challenge in todays optical communications. After giving the key performance indicators dictated by the statistical nature of PMD, we introduce the classification of PMD compensation schemes into 2 categories: optical and electrical PMD compensators. In a first part, we explain the operation principle of a PMD compensator by taking a detailed look at the basic optical PMD compensator and appropriate feedback signals. More complex multistage structures and a feed-forward adaptation approach are also discussed. This first part is closed by results from a one-year field trial confirming the behavior and performance of a prototype compensator. In the second part of this article, electronic equalization for PMD mitigation is explained. Starting with a discussion on performance and adaptation of linear equalizers suitable for analog electronic signal processing, finally also the Viterbi equalizer basing on digital signal processing is analyzed. A comparative review of mitigation by optical or electronic means concludes the discussion.     

PMD compensation based on a new type dynamic first-order PMD compensator

A dynamic first-order polarization mode dispersion (PMD) compensator based on garnet and yttrium vanadate crystal has been proposed and implemented. Consisting of a differential group delay (DGD) generator and a Faraday rotator (FR), this PMD compensator has only two degrees of freedom. Feedback control and compensation algorithm are both very simple. Experimental results reveal the compensator behaviors to be excellent for PMD compensation in 40-Gb/s optical time domain multiplexing (OTDM) system.     

PMD compensation based on a new type dynamic first-order PMD compensator

A dynamic first-order polarization mode dispersion (PMD) compensator based on garnet and yttrium vanadate crystal has been proposed and implemented. Consisting of a differential group delay (DGD) generator and a Faraday rotator (FR), this PMD compensator has only two degrees of freedom. Feedback control and compensation algorithm are both very simple. Experimental results reveal the compensator behaviors to be excellent for PMD compensation in 40-Gb/s optical time domain multiplexing (OTDM)system.     

Numerical Modeling of PMD

A procedure for evaluating system performance considering all orders of polarization mode dispersion (PMD) is presented. The method provides engineering rules for computing system penalty introduced by PMD. Computer simulation of PMD induced penalty indicates that for generating a realistic system budget, effect of higher order PMD in presence of chromatic dispersion is essential. Considering all of these penalties, in order to obtain a system-reach of 3200 km at 40 Gb/s, a fiber PMD of less than 0.04 ps/km^1/2 is required when at the signal wavelength chromatic dispersion is 0. In presence of 18 ps/nm chromatic dispersion for the same system reach this PMD value reduces to about 0.02 ps/km^1/2.     

The inverse PMD problem

When a fiber is characterized by measured polarization mode dispersion (PMD) vector data, inversion of these data is required to determine the frequency dependence of the fibers Jones matrix and, thereby, its pulse response. We briefly review approaches to PMD inversion and discuss three second-order models used for this purpose. We report extension of inversion to fourth-order PMD using higher-order concatenation rules, rotations of higher power designating higher rates of acceleration with frequency, and representation of these rotations by Stokes vectors.     

Mueller Matrix Analysis of PDL Components

A detailed Mueller-Stokes analysis of an arbitrary elliptical dichroic polarizer is presented. Explicit expressions for the Jones matrix, the Stokes parameters, and the Mueller matrix for a distributed and localized dichroic polarizer acting as a polarization-dependent loss (PDL) element are derived. Application to wavelength-dependent PDL elements is discussed. The Poincare sphere representation and the application of the Mueller matrices for the study of randomly oriented concatenated PDL elements and PDL vector measurement are addressed.     

双折射光纤中偏振模色散的抑制

利用双折射光纤中孤子自捕获现象可以抑制偏振模色散,但这种抑制技术对传输参量有严格的限制.本文提出在双折射光纤中周期性地改变快慢轴的方法来抑制偏振模色散. 研究发现,这种方法在一定程度上能有效地抑制偏振模色散,并且光纤间偏振轴的夹角偏差有利于对偏振模色散的进一步抑制.     

GVD compensation schemes with considering PMD

Three group velocity dispersion (GVD) compensation schemes, i.e., the post-compensation, pre-compensation and hybrid-compensation schemes, are discussed with considering polarization mode dispersion (PMD). In the 10- and 40-Gbit/s non-return-zero (NRZ) on-off-key (OOK) systems, three physical factors, Kerr effect, GVD and PMD are considered. The numerical results show that, when the impact of PMD is taken into account, the GVD pre-compensation scheme performs best with more than 1 dB better of average eye-opening penalty (EOP) when input power is up to 10 dBm in the 10-Gbit/s system. However the GVD post-compensation scheme performs best for the case of 40 Gbit/s with input power less than 13 dBm, and GVD pre-compensation will be better if the input power increased beyond this range. The results are different from those already reported under the assumption that the impact of PMD is neglected. Therefore, the research in this paper provide a different insight into the system optimization when PMD, Kerr e     

PMD统计模型的改进

对于PMD的研究,分段模型是较优的选择.原始模型经过数学推导, 可转化为迭代模型.转化后的模型物理意义明晰,在直观上便于看出PMD随光纤分段增加的累积过程,具有较强的移植性.理论分析指明了数值方法产生误差的缘由,给出了实用数值计算的迭代模式,对于PMD补偿具有指导意义,同时使计算变得简洁.     

一种可编程PMD模拟器

为了研究光纤通信中的偏振模色散(PMD)及其对高速光通信系统的影响和补偿技术,研制了一种可编程控制的PMD模拟器,它由6组双折射晶体和磁光晶体组成,群时延差(DGD)变化范围为±(0.3~35) ps,可设为不同平均值的Maxwell统计分布随机变化,也可以设为其它要求的数学分布.实验证明,该模拟器测量值与理论符合得很好,且具有较高的重复性和稳定性,响应时间小于1 ms.这种模拟器可用于10 Gb/s和40 Gb/s光传输系统PMD问题的研究.     

高阶PMD统计特性研究

讨论了三阶偏振模色散(PMD)的统计特性及其对于高阶PMD补偿系统的影响.数值模拟结果表明,随着即时差分群时延(DGD)的增加,二阶PMD单位矢量旋转速率的统计覆盖范围的变化趋势与主偏振态(PSP)旋转斜率的相应变化正好相反.此外,随着即时DGD的增加,二阶PMD模值的统计覆盖范围不断增加,使得高速光通信系统中高阶PMD补偿的实现将更加困难.     

二阶偏振模色散测量分析

在庞加莱球的基础上,对原偏振模色散的测量方法进行了理论分析.研究表明,原方法属于间接测量,在测量二阶偏振模色散中存在明显不足.一阶偏振模色散测量的误差影响二阶偏振模色散的测量,间接测量降低了二阶偏振模色散的测量准确度.提出了一种二阶偏振模色散测量的新方法,保证了测量准确度,并通过实验中的统计结果得到证实.     

DIRECT算法在PMD补偿中的应用

第一次把DIRECT算法用于偏振模色散补偿中作为反馈控制算法的第一部分,来寻找初始的全局最优解。用著名的Hartman6函数的返回值模拟偏振度的测量值,给出了DIRECT算法对Hartman6函数全局最优解的寻找过程,通过与其它几种全局最优化算法的比较,对DIRECT算法进行了初步的评价。