WO2001028134A1

WO2001028134A1 - Method and apparatus for measuring pmd of a dispersion compensation grating

Info

Publication number: WO2001028134A1
Application number: PCT/US2000/027729
Authority: WO
Inventors: David Way; Tiejun Xia
Original assignee: MCI Worldcom Inc
Current assignee: Verizon Business Global LLC
Priority date: 1999-10-08
Filing date: 2000-10-06
Publication date: 2001-04-19
Anticipated expiration: 2002-04-08
Also published as: JP2003511681A; EP1222758A1; AU7871100A; EP1222758A4; US6229606B1

Abstract

A method, a system and a computer program product configured to determine a differential group delay of two sequential optical beams based on a frequency dependence of a length of an optical path traveled by the two sequential optical beams in a dispersion compensating grating (60). The determined differential group delay provides a measure of polarization mode dispersion in the dispersion compensation grating (60).

Description

METHOD AND APPARATUS FOR MEASURING PMD OF A DISPERSION

COMPENSATION ORATING

The present invention relates to electronic instruments for testing optical devices, and in particular to a method and apparatus for measuring polarization mode dispersion (PMD) of a dispersion grating, such as a dispersion compensation grating (DCG).

As a light pulse, with a finite spectrum of optical frequencies, travels tlirough a medium, e.g., on optical fiber, two dispersion phenomena occur. First, because light travels tlirough the medium at different velocities depending on its frequency, the pulse of multiple frequency will experience a chromatic dispersion and will be spread out in time during its propagation through the medium. Second, a polarization mode dispersion (PMD) occurs because light travels through the medium at different velocities depending on its polarization. It is well known that PMD can limit the available transmission bandwidth in fiber optics transmission links, hi other words, as the bit rate of a system mcreases, the PMD tolerance of that system decreases.

To minimize the chromatic dispersion in high bit rate systems (for example, a OC-48 system having a 2.488 Gbps bit rate) it has been suggested to connect the optical fiber to a dispersion compensation grating (DCG). Such DCG is specifically designed to compensate for chromatic dispersion. However, a DCG may potentially increase PMD in the optical fiber/DCG system, and thus, may deteriorate transmission hi high bit rate systems. There is, therefore, a need for a method and apparatus capable of accurately measuring PMD of a DCG so as to determine whether a particular DCG can be incorporated into a high bit rate optical liber system without deterioration of the optical signal.

Several conventional methods exist to measure PMD in optical fibers, hi particular, a well-established method for measuring PMD hi optical fibers uses the so-called "Jones Matrix Eigenaπalysis" (JME). A description of this method is provided for example in U.S. Patent Nos. 5,227,623 and 5,298,972 to Hefner and in B. L. Heffher, "Automated Measurement of Polarization Mode Dispersion Using Jones Matrix Eigenanalysis", IEEE Photonics Technology Letters, Vol. 4, No. 9, September 1 92, the entire contents of which are hereby incorporated by reference.

An example of an apparatus used in the conventional JME method to measure PMD in an optical fiber is shown in Fig. 1. An optical source 1 , which may be a conventional tunable laser source, generates sequential optical beams with different frequencies/wavelengths, for example 1300 nm and 1302 nm. The polarization synthesizer 20 receives the optical beams from the optical source 10 and produces three sequential states of polarization for each optical beam. Preferably, the three states of polarization are separated by 60°. The polarized optical beams travel sequentially though the optical fiber 30 under test, which produces the PMD being measured. The optical beams, having experienced PMD through tile optical fiber, then enter an analyzer 40, such as an optical polarization meter. Optical polarization meter 40 measures the intensity of the received optical beams with the different polarizations and generates the so called "Stokes parameters," which are well known hi the art and defined in Principles of Optics, by M. Bom and E. Wolf, Pergamon Press, 4* Edition, London, 1970, pages 30-32, the content of which is hereby incorporated by reference. For each optical beam with a particular wavelength, the Jones matrices are then computed in processor 50 from the measured responses, i.e., from the Stokes parameters, according to a known algoridim. One such algorithm is described in U.S. Patent No. 5,298,972, the entire content of which is hereby incorporated by reference.

Processor 50 may be a conventional personal computer and can be connected to the

optical source 10 to read and set the frequency of the optical beams. Processor 50 is also

coimected to polarization synthesizer 20 to read and set the polarization of the optical beams. Once the Jones matrices, which capture die characteristics of the optical fiber under ttst, are determined, then PMD can be determined using a conventional Jones Matrix Eigcnanalysis, wlu^'cli is well known to persons of ordinary skill in the art, described for example in U.S. Patent No. 5,227,623 at column 12, line 21 to column 16, line 21, the entire content of which is incorporated herewith by reference.

Briefly summarizing, the conventional Jones Matrix Eigenanalysis method for measuring PMD hi an optical fiber, defines die following quantities:

Eι(ω) is the input field of an optical beam of frequency ω before entering an optical

fiber. E|(ω) has a unit vector e₍.

E₂(ω) is the output field of an optical beam of frequency ω after exiting die optical i fiber. E₂(ω) has a unit vector έj.

T(ω) is the Jones matrix, characteristic of the optical fiber, so that E₂(ω) - T(ω) • e,. (1)

The output field E₂(ω) can be expressed in terms of a magnitude term σ(ω) and a phase couponent φ(ω):

E₂(ω) = σ(ω) e^{l (0),}e2. (2) Taking the partial derivative of E₂(ω) with respect to ω, and using die prime symbol to denote differentiation with respect to ω, one obtains:

E'₂(ω) - T'(ω) • e, - (σ'(ω)/σ(ω) + iφ'(ω))E₂(ω), (3) where δβi /δω = 0 because 4ι is fixed at input and δέyδω is 0 to first order over ω.

From Equation (1) and ωsuining that the optical fiber is not perfectly polarizing so that its

Jones matrix is non-singular, one obtains: έi - r'tø) • &(»). (4)

Inserting (4) into (3) and using the approximation T'(ω) - [T(ω+Δω) - T(ω)]/Δω, one obtains:

[T(ω+Δω) - T(ω)]/Δω • T (ω) * E₂(ω) - (σ'(ω)/σ(ω) + iφ'(ω))E₂(ω), which simplifies to:

T(α>+Δω) • T '(ω) - (1 + iφ'(co)Δω)I - 0, (5) where I is the unit matrix, and where it is assumed that Δω • σ'(ω)/σ(ω) is 0 for small Δω because losses are essentially die same at ω and at ω + Δω.

For an optical fiber, it is assumed that the phase component of the output field, φ(ω), is the product of a propagation constant, β, which is a function of frequency, times the d stance z traveled by the light beam along the axis of the optical fiber. For an optical fiber, it is assumed that z is independent of frequency, so that: φ(ω) - β(ω) • z, and φ'(ω) - δβ(ω)/δω • z. δβ(ω)/δω is the reciprocal of the group velocity v_g, so that ψ'(ω) ^» z /vg = τ_g, (6) where τ_g is the group delay of ihe optical beam through the optical fiber. Substituting (6) into (5) and solving (5) for the eigenvalues gives:

Δt ^•= τ,a - τ_glU - Arg( pi / ρ_π ) / Δω, (7) where 1 and JJ denote the two principal states of polarization; pi and p» arc the eigenvalues of

T(ω+Δω) • T'^l(ω); Arg denotes the argument function; and Δτ, wlύch is called the differential

group delay, corresponds to the PMD in the optical fiber. The computation of Δτ is performed by processor 50, which is configured to carry out the above analysis based on the data received from optical polarization meter 40.

As noted above, the conventional JME method asssumcs that die length of the

traveled optical path by light, z, is independent of frequency, so that Δτ corresponds exclusively to the effect of PMD. This is an accurate assumption in optical fibers, and JME is thus an accurate method for measuring PMD in optical fibers.

However, in a DCG, by definition, die length of die optical path traveled by light, z, is dependent of frequency, i.e., z = z(ω). Specifically, light is reflected off a DCG at a depth that is dependent of its frequency. The DCG is designed so that tiie difference in travel time for different wavelengths caused by the difference in length of the optical path traveled by the light of different wavelength offsets, or compensates for, the difference in traveled time caused by the cliromatic dispersion. Consequently, in a DCG, light of different wavelengths will have different lengths of traveled optical path and the JME assumption noted above is no longer correct. Thus, if the conventional JME method were to be used to measure PMD of a

I

DCG, the computed differential group delay, Δτ, would not correspond to the true PMD of

the ΩCG, but to the sum of the delay caused by PMD plus any delay caused by the difference in length of the traveled optical path hi die DCG. Therefore, using the conventional JME method for measuring PMD in a DCG does not give an accurate result. Accordingly, an object of the present invention is to provide a method and apparatus to accurately measure PMD in a DCG.

A second object of tiie present invention is to provide a modified JME method and an apparatus to implement such a modified JME method, which takes into consideration the frequency dependence of the length of the traveled optical path in the DCG when determining die PMD of a DCG. h a first embodiment, the present invention provides a method for determining polarization mode dispersion (PMD) of a DCG, including generating two sequential optical beams of different frequencies; exposing tiie grating to the two sequential optical beams; determining a differential group delay of the two sequential optical beams based on a frequency dependence of a length of an optical path traveled by the two sequential optical beams hi the DCG.

A preferred emodiment of the present invention includes determining the Jones matrix of the DCG using the Stokes parameters of the DCG.

Another preferred embodiment of the present invention includes determining ti e eigenvalues of the equation:

T(ω ι-Δω) ♦ T-'(ω) - (1 + iφ'(ω)Δω)I « 0, where T is the Jones matrix of the DCG, ω is an optical beam frequency variable, Δ(ω) is a frequency difference between the two sequential optical beams traveling tlirough the DCG, I is the unit matrix, and φ'(ω) is the partial derivative, with respect to ω, of a phase component of a field output from the DCG, wherein ψ'(ω) - τ, + β(ω) • o z(ω)/δω, where τ_t is a group delay due to PMD, β(ω) is die propagation constant of the DCG and z is a

variable representing the traveled length of the optical path.

Another preferred embodiment of the present invention includes deterniining the

differential group delay, Δτ, by calculating:

Δτ - Arg( pi / p» ) / Δω - (βi - βn) * δz(ω)/δω,

where Arg denotes the argument function, pi and pu are the eigenvalues of T(ω+Δω) • T (ω),

and βi and βn are die two propagation constants on the two principal axes of PMD.

Another preferred embodiment of the present invention includes dctcπrtining the differential

group delay, Δτ, by calculating:

Δτ - (Arg( p, / pu ) / Δω} / (1 - λD c / 2L_λ), where λ is die average optical beam wavelength between the two scqucncial optical beams, c

is die speed of light, D is the dispersion value of the DCG, and L\ is a distance from the

entrance of die DCG to a plane reflecting an optical beam of wavelength λ.

In accordance with another embodiment, the present invention provides a system to implement Uie above method. hi yet another embodiment, the present invention provides a computer program

product to implement the above method. i

A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as die same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:

FIG. 1 is a block diagram of a conventional JME apparatus; FIG. 2 is a block diagram of a modified JME apparatus to measure MPD in a DCG according to the present invention.

FIG. 3 is a schematic representation of a DCG for wliich PMD is measured by the

apparatus and method of the present invention.

FIG. 4 is a schematic representation of a computer implementing the present

invention.

DESCRIPTION OF THE PRPFF.RR KD EMBODIMENTS Referring now to the drawings, wherein like reference numerals designate identical or corresponding parts throughout the several views, a first embodiment of the present invention in ttie form of an apparatus for measuring the PMD of a DCG (as shown in the block diagram of Fig. 2) is described.

In Fig.2, optical source 10 generates at least two sequential optical beams with different frequencies. Polarization synthesizer 20 sets the polarization of the beams; in an exemplary embodiment, the separation is by 45°. The polarized optical beams travel tlirough grating 60 and then enter polarization meter 40, which detects the beams' polarization. Data from polarization meter 40 is then read by processor 70, which controls and reads both optical

I source 1 and polarization synthesizer 20. To minimize introduction of any PMD from the hardware components other than the grating 60, polarization-mamtaining fibers 25 are used to connect the grating 60 to the rest of the hardware.

Some of the components used in the present invention for measuring the PMD of a

DCG can be the same as the ones used for measuring the PMD of an optical fiber. In particular, optical source 10, polarization synthesizer 20 and polarization meter 40, of the conventional system as in Fig. 1 can be employed in an embodiment of the present invention.

The commonality of components is advantageous because it allows persons of ordinary skill in the art, who ore already familiar with these components, to implement the present invention without having to purchase and familiarize themselves with many new components.

The present invention is therefore practical and cost-effective to persons working in die field.

Alternatively, optical source 10 can be a tunable laser source, such as HP8167A manufactured by Hewlett-Packard™, polarization synthesizer 20 and optical polarization meter 40 can be part of a single unit polarization analyzer, such as HP8509B, manufactured by Hewlett-Packard™.

The device being tested in Fig. 2 is grating 60, and the data from optical polarization meter 40 is received by processor 70, which is configured to perform the analysis described below In order to deteπnine die PMD of grating 60. Grathlg 60 can be any grating. However, when the present invention is used to determine the PMD of gratings used with optical fibers to compensate tiie chromatic dispersion that occurs in optical fibers, a DCG is used. Therefore, grating 60 is preferably a DCG.

An example of a DCG 60, which may be used in the present invention, is schematically represented in Fig. 3. As mentioned above, a DCG 60 is designed to compensate for chromatic dispersion. Threfore, a DCG 60 is designed so that the traveled length of tiie optical path depends on the frequency of the optical beam traveling through the grating. For example, in Fig. 3, optical beam 81 having a frequency α>ι travels a longer optical paUi tiian optical beam 82, having frequency ω₂. The difference in traveled length Δz is therefore frequency dependent, i.e., Δz - Δz(ω).

DCG 60 is typically made by exposing an optical fiber with an intensity modulated X- ray beam. The intensity modulation generates ripples in the optical fiber and a grating is thus created. The intensity modulation can be controlled to obtain different grating characteristics, such, s a dispersion compensation grating . Processor 70 (Fig. 2) may be implemented by a personal computer as well known in die art or any computing device including a memory device coimected to a processing unit. Processor 70 receives data from optical polarization meter 40 and then computes Stokes parameters and die Jones matrix for die DCG 60 in a similar manner as described above for a conventional PMD measurement for an optic fiber. However, processor 70, according lo the

present invention, is configured to compute die PMD of grating 60 based on die analysis below.

As noted above, in a DCG 60, the lengtii of the traveled optical padi depends on the frequency of tiio optical beam. In other words, Equation (6) is not correct for a DCG 60. Instead, one has:

φ(ω) ^« β(ω) • z(ω), so that φ'(ω) =» δβ(ω)/δω • z(ω) + β(ω) • δz(ω)/δω, or

φ'(ω) = τ, + β(ω) • δz(ω)/δω (8)

where τ_κ Is the group delay due lo PMD, and the second term is related to the "designed"

delay caused by the DCG 60. Substituting Equation (8) into Equation (5), and solving Equation (5) for tiie eigenvalues gives: (τ,.ι + βi • δz(ω)/δω) - (τ,,_π + β„ • δz(ω)/δω) = Arg( pi / pu ) / Δω, (9)

where i and pu are tiie eigenvalues of T(ω+Δco) • T^*'(ω), T representing the Jones matrix for

I die DCG, βi and βu are die two propagation constants on the two principal axes of PMD, Δω is die frequency difference between two sequential optical beams, and Λrg denotes the argument function.

Solving Equation (9) for Tgj - τ^u = Δτ, the differential group delay, corresponding to die

PMD in die DCG 60, gives:

Δτ - t,.i - τ_fcU - Arg( p, / p_u ) / Δω - (βi - β„) • δz(ω)/δω. (10) where βi and βu can be expressed as:

to βi = 2 π ni / anci βi, = 2 π n„ / λ, (1 1) where nι and nι₍ arc die total index of refraction for the grating material for the first and second principle states of polarization, respectively, and λ is the average optical beam wavelength for two sequential measurements.

In most cases, as shown in G. Agraval, Nonlinear Fiber Optics. Academic Press, pp 8- 9, the content of which is hereby incorporated by reference, it can be assumed that: ιi| * n_u ~ 11, - c _j Lλ, (12) where n_g is die group velocity index, c is the speed of light, τ_g is die group delay through the DCG, and is the distance from the entrance of the grating to the equivalent plane reflecting .(lie optical beam of wavelength λ (see Fig. 3). In practice, L(λ), and thus λ, Is provided by the grating manufacturer.

Also, δz(ω)/δω can be expressed as: δz(ω)/δω = - λ^J D / 2 π, (13) where λ is the average optical beam wavelength between two sequenciai optical measurements, and D is the dispersion value of the grating, which in practice is also provided by the grathlg manufacturer.

Replacing (12) into (11), then (11) and (13) into (10) gives: Δτ = {Arg( p, / p„ ) / Δω) / (1 - λD c / 2L_λ). (14)

Equation (14) therefore gives a practical solution to measuring the differential group delay, i.e., PMD, of a DCG.

As shown in Equation (10), when compared to Equation (7), the present invention provides a method for accurately measuring the PMD of a DCG by taking into consideration die frequency dependence of the length of die traveled optical path in the DCG. This frequency dependence leads to an additional term hi the expression of die differential group delay, Δτ, when compared to die expression used in the conventional JME method. The present invention titus provides a modified JME mediod, wliich can be easily implemented by persons of ordinary skill in the art to accurately deteπnine die PMD of a DCG.

Advantageously, once die PMD of a DCG is determined to be satisfactory, the DCG can be inlcuded in a high bit rate network with confidence. The present invention therefore provides a useful, concrete and tangible result.

All or a portion of tiie invention may be conveniently implemented using conventional general purpose computers or microprocessors programmed according to die teachings of the present invention, as will be apparent to tiioso skilled hi the computer art. Appropriate sσfiware can be readily prepared by programmers of ordinary skill based on the teachings of the present disclosure, as will be apparent to those skilled in the software art.

Figure 4 is a schematic illustration of a computer system 100 for implementing the method of die present invention. The computer system 100 includes a computer housing 102 for housing a mother board 104, which contains a CPU 106, a memory 108 (e.g., random access memory (RAM), dynamic RAM (DRAM), statio RAM (SRAM), synchronous DRAM (SDRAM), flash RAM, read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), and electrically erasable PROM (EEPROM)), and other optional special purpose logic devices (e.g., application specific integrated circuits (ASICs)) or configurable logic devices (e.g., generic array of logic (GAL) or reprogrammable field programmable gate arrays (FPGΛs)). The computer system 100 also includes plural input devices, such as a keyboard 122 and a mouse 124, and a display card 110 for controlling a monitor 120. In addition, the computer system 100 further includes a floppy disk drive 114; other removable media devices (e.g., a compact disc 119, a tape, and a removable magneto-optical media); and a hard disk 112, or other fixed, high density media drives, connected using an appropriate device bus (e.g., a small computer system interface (SCSI) bus, and erdianced integrated device electronics (IDE) bus, or an ultra-direct memory access (DMA) bus). The computer system 100 may additionally include a compact disc reader 1 18, a compact disc reader-writer unit, or a compact disc juke box, each of which may be connected to the same device bus or another device bus. Λldiough the compact disc 119 is shown in a CD caddy, the compact disc 1 19 can be inserted directly into CD-ROM drives wliich do not require caddies, hi addition, a printer may provide printed listings of the data structure stored and/or generated by die computer system 100.

As stated above, die system includeu at least one computer readable medium or memory programmed according to the teachings of the invention and for containing data structures, tables, records, or other data described herein. Examples of computer readable media are compact discs, hard disks, floppy disks, tape, magneto-optical disks, PROMs (EPROM, EEPROM, Flash EPROM), DRAM, SRAM, SDRAM, etc. Stored on any one or on a combination of computer readable media, die present invention includes software for controlling bodi the hardware of the computer 100 and for enabling die computer 100 to interact witii a human user (e.g., α consumer). Such software may include, but is not limited lo, device drivers, operating systems and user apphcations, such as development tools. Such computer readable media further includes die computer program product of die present invention for performing all or a portion (if processing is distributed) of the processing perfoπned in implementing die invention. The computer code devices of the present invention can be any interpreted or executable code mechanism, including but not limited to scripts, interpreters, dynamic link libraries, Java classes, and complete executable programs. Moreover, parts of the processing of the present invention may be distributed for better performance, reliability, and or cost. The invention may also be implemented by the preparation of application specific integrated circuits or by interconnecting an appropriate network of conventional component circuits, as will be readily apparent to those skilled in the art.

Obviously, numerous modifications and variations of die present invention are possible in liglit of die above teachings. It is dierefore to be understood that widύn the scope ^of the appended claύns, the invention may be practiced otherwise than as specifically described herein.

u

Claims

WHAT IS CLAIMED AS NEW AND DESIRED TO BE SECURED BY LETTERS PATENT OF THE UNITED STATES IS:

1. A mediod for determining polarization mode dispersion (PMD) of a grating, comprising: generating two sequential optical beams of different frequencies; exposing said grating to said two sequential optical beams; and deteπnining a differential group delay of said two sequential optical beams based on a frequency dependence of a length of an optical path traveled by said two sequential optical beams in said grathlg.

2. The method of Claim 1, wherein detcπnining said differential group delay is based on a frequency dependence of the length of the optical path traveled in a dispersion coπpensation grating (DCG).

3. The method of Claim 2, wherein deteππining said differential group delay comprises determining a Jones matrix of the DCG.

4. The method of Claim 3, wherein deteπnining the Jones matrix of the DCG comprises determining Stokes parameters of the DCG.

5. The mediod of Claim 3, wherein deteπnining said differential group delay comprises deteπnining the eigenvalues of die equation:

T(ω+Δω) ^■ T^"'(ω) * (1 + iφ'(ω)Δω)I - 0, where T is die Jones matrix of die DCG, ω is an optical beam frequency variable, Δ(ω) is a frequency difference between said two sequential optical beams traveling dirough said DCG, I is the unit matrix, and φ'(ω) is the partial derivative, with respect to ω, of a pliasc component of a field output from the DCG, wherein φ'(ω) = τ_g + β(co) • δ z(ω)/δω, die DCG where τ_g is a group delay due to PMD, β(ω) is die propagation constant of and z is a variable representing the traveled IeπgUi of the optical path.

6. The method of Claim 5, wherein determining said differential group delay, Δτ, comprises calculating:

Δτ = Arg( pi / p_u ) / Δω - (βi - β_u) • δz(ω)/δω, where Arg denotes die argument function, i and p» are the eigenvalues of T(ω+Δω) ♦ ^•T^" (ω), and βi and βu are two propagation constants on die two principal axes of PMD.

7. The mediod of Claim 6, wherein determining said differential group delay, Δτ, comprises calculating:

Δτ = { Arg( p, / p_u ) / Δω) / (1 - λD c / 2L_λ), where λ is an average optical beam wavelength, c is die speed of light, D is die dispersion value of die DCG, and Li is a distance from the entrance of the DCG to a plane reflecting an optical beam of wavelength λ.

8. A system for determining the polarization mode dispersion (PMD) of a grating, comprising: an optical source configured to generate two sequential optical beams having different frequencies; a grating configured to receive said two sequential opliual beams; a detection mechanism configured to detect the polarization of said two sequential optical beams exiting said grating; and a processor configured to deteπnine a differential group delay of said two sequential optical beams based on a frequency dependence of a length of an optical path traveled by said iwo sequential optical beams in said grating.

9. The system of Claim 8, wherein said grating comprises a dispersion compensation grating (DCG).

10. The system of Claim 9, wherein said processor is configured to determine a Jones matrix of the DCG.

11. The system of Claim 10, wherein said processor is configured to determine the eigenvalues of die equation:

T(ω+Δω) • r'(co) . (i + iψ'(ω)Δω)I - 0, where T is the Jones matrix of the DCG, ω is an optical beam frequency variable, Δ(ω) is a frequency difference between said two sequential optical beams traveling through said grating, I is die unit matrix, and ψ'(ω) is the partial derivative, widi respect to ω, of a phase component of a field output from tiie DCG, wherein φ'(ω) - x_a + β(ω) • δ z(ω)/δω, where τ_g is a group delay due to PMD and β(ω) is die propagation constant of the DCG and z is a variable representing tiie length of the traveled optical path.

12. The system of Claim 11, wherein said processor is configured to determine said differential group delay, Δτ, by calculating:

Δτ = Arg( p, / p„ ) / Δω - (βj - β_u) . δz(ω)/δω, where Arg denotes the argument function, p, and u are die eigenvalues of T(ω+Δω) • T^" (ω), and βi and βu arc two propagation constants on the two principal axes of PMD.

13. The system according to Claim 12, wherein said processor is configured to determine said differential group delay, Δτ, by calculating: Δτ = (Λrg( Pi / p„ ) / Δω) / (1 - λD c / 2L_λ), where λ is an average optical beam wavelength, c is the speed of light, D is the dispersion value of die DCG, and L . is a distance from the entrance of die DCG to a plane reflecting an optical beam of wavelengtii λ.

14. A computer program product, comprising: a computer storage medium and a computer program code mechanism embedded in the computer storage medium configured to cause a computer to determine a differential group delay of two sequential optical beams based on a frequency dependence of a length of an optical path traveled by said two sequential optical beams in a dispersion compensation .grating (DCG).

15. The computer program product of Claim 14, wherein said computer program code mechanism comprises a first computer code device configured to determine a Jones matrix of said DCG.

16. The computer program product of Claim 15, wherein said computer program code mechanism further comprises a second computer code device configured to determine the eigenvalues of the equation:

T(ω+Δω) ^• T'(ω) - (1 + iψ'(ω)Δω)I - 0, where T is tiie Jones matrix of die DCG, ω is an optical beam frequency variable, Δ(ω) is a frequency difference between said two sequential optical beams traveling through said DCG, I is the unit matrix, and ψ'(ω) is the partial derivative, with respect to ω, of a phase component of a field output from the DCG, wherein ψ'(ω) - τ, + β(ω) • δ z(ω)/δω, where τ_g is a group delay due lo PMD, β(ω) is the propagation constant of die DCG and z is a variable representing die traveled length of tiie optical path.

IB

17. The computer program product of Clahn 16, wherein said computer program code mechanism further comprises a third computer code device configured to calculate:

Δτ - Arg( i / p_π ) / Δω - (β, - β„) • δz(ω)/δω, where Δτ is the differential group delay of said two sequential optical beams, Arg denotes die argument function, i and pu are the eigenvalues of T(ω+Δω) • T^"'(ω), and βi and βn are two propagation constants on die two principal axes of PMD.

18. The computer program product of Claim 17, wherein said computer program code mechanism further comprises a fourth computer code device configured to calculate:

Δτ = (Arg( p, / p„ ) / Δω} / (1 - λD c / 2L_λ), where λ is die average optical beam wavelength, c is die speed of light, D is the dispersion value of die DCG, and LA is a distance from the entrance of die DCG to a plane reflecting an optical beam of wavelength λ.