JP2002243585A - Method for determining properties of optical device and inspection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and device for determining elements or the like of the Jones matrix of an optical device without requiring an interferometer. SOLUTION: This method comprises a process for generating an incoming light beam; a process for dividing the incoming light beam to a first light beam and a second light beam; a process for making the first light beam having two mutually delayed parts into a given initial polarized light and connecting it to the optical device to be tested; a process for advancing the second light beam to a route different from the first light beam; a process for superposing the first light beam to the second light beam and performing an interference between the first and second light beams in the superimposed light beam; a process for separating the superimposed light beam to depending third light beam and fourth light beam; a process for detecting the powers of the third and fourth light beams as the function of frequency when synchronizing the frequency of the incoming light beam over a given frequency range; and a process for deriving the properties of the optical device to be tested from the frequency dependency of the detected powers.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a method for inspecting an optical device and an apparatus used for the method. In particular,
The present invention provides a method for determining characteristics of an optical device to be inspected,
For example, determining the elements of a Jones matrix of an optical device.

[0002]

BACKGROUND OF THE INVENTION A Jones matrix contains information about the optical properties of the device under test, which can be a fiber or an optical component. Knowing the Jones matrix of optical components
Insertion loss, reflectance or transmittance, polarization dependent loss (P
DL) and many important optical properties of components such as polarization mode dispersion (PMD).

[0003] PMD is a fundamental property of single-mode optical fibers and components such that signal energy at a given wavelength is resolved into two orthogonal polarization modes with slightly different velocities of propagation. The resulting difference in propagation between polarization modes is called differential group delay, or DGD. The term PMD generally refers to physical phenomena, in particular, DGD
Used to mean the average or expected value of. PM
The characteristic that defines D is DGD, which is the basic state of polarization (PSP). Both are usually a function of wavelength in a single mode fiber system. Long fibers generally exhibit random polarization connections, and consequently P
MD can be evaluated as the square root of fiber length for fibers longer than a few kilometers. Conventionally known fibers are limited to tens of picoseconds of DGD per route kilometer. Furthermore, previously known components for such fiber communication systems can exhibit only one tenth of a picosecond of DGD.

[0004] PMD causes a number of significant capacity impairments, including pulse broadening. In this regard, it has an effect similar to chromatic dispersion damage, but with significant differences.
Chromatic dispersion is a relatively stable phenomenon because it results from the change in propagation delay with wavelength caused by the interaction and size of the fiber material. The overall chromatic dispersion in the communication system can be calculated from the sum of the parts, but the position and value of the dispersion compensator can be set in advance. On the other hand, the PMD of a single mode fiber at any given signal wavelength is not stable and requires the communication system designer to make statistical predictions of the effects of the PMD and to make corrections. Make it impossible. In addition, PMD
Even if the chromatic dispersion is made sufficiently small, it becomes a limiting factor.

This means that the average value of the channel PMD, ie the DGD for the fiber over wavelength and time, is 20
This is because ps can often be exceeded. This value is in the range of the bit resolution of the 40 Gbit / s communication system, so that the communication system will be adversely affected by PMD. Furthermore, in the prior art, communication system components are often guided in cascades, for example, in multiple cascades of Bragg gratings in fibers. Although a single member of such a cascade shows only about one-tenth of a picosecond as the DGD, the whole cascade can exhibit a DGD that reaches the resolution of transmittance. Therefore, there is a great need to be able to obtain accurate information about PMD in each single member.

[0006] The above-mentioned problems have affected the development of many measurement methods for PMD measurement. The following describes some of the well-known methods.

In the fixed analyzer PMD measurement method, P
MD is determined statistically from the number of peaks and valleys in optical power transmission through the polarizer as the wavelength is scanned. The polarizer placed immediately before the detector is called an analyzer so that the method can be understood by name. The response of the fixed analyzer can be Fourier changed, yielding a spectrum indicating the degree of mode coupling, allowing the PMD to be calculated from a Gaussian fit or from a second moment algorithm.
A problem with the fixed analyzer method is that it is not possible to measure PMD components that exhibit a bandwidth that is less than the change in optical power transmission over wavelength.

Another method is an interferometric method, in which the PMD is determined from an electric field autocorrelation function using a broadband source. The PMD value is calculated using an algorithm based on the second moment. The problem with this method is that
When PMD is caused by pure birefringence, P
It only produces the correct value of MD. However, this method is incapable of generating a valid PMD value when the PMD is wavelength-dependent.

Another method is to use Poincare Arc or SO.
It is called P (state of polarization) and uses a polarimeter to capture the arc that has traveled the Poincare sphere by the output polarization of the test equipment over a series of wavelength increments. However, PMD cannot be measured when the polarization accidentally couples with the main polarization state of the test device. Another problem is that a high resolution polarimeter, a type of polarimeter that tends to be very effective, is needed. In addition, chromatic dispersion cannot be measured with this method.

Another method is the so-called Jones matrix specific analysis or JME method. This method determines DGD and PSP as a function of wavelength from measurements of the transmission matrix at a range of wavelengths. This method also uses an expensive polarimeter. This method does not give information about chromatic dispersion.

Finally, there are well-known methods of measuring PMD more or less directly. These methods are, for example,
A modulation phase method and a pulse delay method,
Measurement of the change in modulation phase from between the main states of polarization, and
The PMD is measured from the change in the pulse arrival time. The disadvantage of these methods lies in the dependence of the obtained results on the pulse shape.

Accordingly, it is an object of the present invention to improve the determination of characteristics in an optical device.

[0013]

The present invention is solved by the invention described in the claims. An advantage of the present invention is that it can derive transmission characteristics and reflection characteristics,
For example, deriving the PMD of a DUT by just determining the elements of the Jones matrix of the device under test (DUT), which is described in more detail below, without requiring the use of expensive polarimeters And to be able to measure the chromatic dispersion of the DUT simultaneously. To that end, all of the above-mentioned disadvantages of the prior art are solved by the present invention. Moreover, it is possible to derive additional information from the retrieved Jones matrix of the DUT. From the measured Jones matrix, the insertion loss, reflectance and transmittance, loss of dependence (PDL), group delay, chromatic dispersion, differential group delay, key states of polarization, and higher order PMD parameters Can be determined.

Preferably, a Jones matrix is used. The information in the Jones matrix can be represented in a number of different ways, one option being to carry almost the same information as a Jones matrix containing all the actual elements as a 4x4 matrix. This is called the Mueller matrix. However, the absolute phase characteristics, and consequently,
Chromatic dispersion cannot be described by a Mueller matrix. There are several ways to achieve the same goal. One subject of the present invention is PMD, PDL, PSP, DGD,
It is a method of measuring group delay, chromatic dispersion, and the like with the described interferometric method and apparatus. However, it is possible to develop creatively with tensors other than the Jones matrix.

The term "coherent" in this application
Means that the coherence length of the incident light beam is greater than the different lengths of the paths of the superimposed light beams.

In a preferred embodiment of the invention, the device comprises a first Mach-Zehnder interferometer, whereby a polarization setting tool is arranged in the measuring arm,
A laser beam connects to a DUT having a predetermined polarization. This direction of polarization is then defined as the x-axis in the coordinate system of the Jones matrix calculation. Thus, the first two elements in the Jones matrix can be easily removed. In this creative second implementation, the other two elements of the Jones matrix are DU
It is extracted from the same interferometer by changing the direction of polarization of the light beam incident on T. Preferably, the polarization is changed to a polarization orthogonal to the previous polarization to facilitate evaluation of the result. In this regard, it is further preferred that the initial polarization is linear and the changed polarization is changed by 90 ° with respect to this initial polarization.

In a further preferred embodiment, a second implementation of the creative method for determining the other two elements of the Jones matrix is not necessary. The Jones matrix can be completely determined in just one run. For complete capture of the Jones matrix, the DUT needs to be tested with two orthogonal input polarizations. Therefore, in this preferred embodiment, the concept of a so-called "single scan" measurement is proposed. In this concept, a polarization delay process generates two, preferably orthogonal, polarization signals that are delayed with respect to each other. The polarization delay process provides two polarization states that are simultaneously guided to the DUT and “coded” by different propagation delays. Preferably, in order to delay the two signals with respect to each other, the first signal beam is split into two undelayed beams and a first delayed beam in this polarization delay process. Subsequently, the polarization, or a part thereof, of the first beam is combined with the light of the local oscillator, ie, the second beam.
Superimposed with the luminous flux. Preferably, the two parts of the first beam are polarized orthogonally with respect to each other and do not interfere when they are superimposed. However, 2
The two polarizations create an interference pattern when superimposed on the second beam. Due to the delay introduced by the polarization delay process, the detected interference signal has two different frequencies. These interference frequencies can be individually separated and analyzed by using electronic or digital filters. The analysis performed allows the determination of the elements of the Jones matrix in the optical device. Numerous advantages are obtained in connection with the above-described embodiment of the single scan concept. First, in an embodiment, two scans are performed so that the DUT can change its optical properties during the time between scans. Therefore, the stability required for the DUT is very high. Furthermore, the measurement speed can be increased dramatically according to the single scan concept, since performing two scans takes a lot of time. In addition, the measurement accuracy is impaired by vibration noise, phase noise during the measurement or disturbance of the light source. This can also be avoided by this embodiment. As a further advantage of the present invention, the required accuracy for the frequency to which each of the data is assigned is such that in this embodiment two scans are performed simultaneously and both relate to a certain frequency of the laser light source, Less than the accuracy required in other embodiments that perform two scans.

It will be appreciated by those skilled in the art that numerous other ways of splitting the first beam, delaying one portion relative to the other, and recombining them into different, preferably orthogonal, polarizations can be created. It is clear.

Preferably, in each of the polarization delay units, two polarization beam splitters are used to perform the polarization delay process, one of which converts the first light beam into a first delayed light beam and a first non-delayed light beam. And the other is recombined into the first delayed light beam and the first non-delayed light beam. This two with two mouths each
The two polarizing beam splitters are connected to each other by a polarization maintaining fiber. One of these two fibers delays one portion of the split first light beam longer than the other, as described above. Alternatively, the recombining beam splitter may be replaced by a polarization maintaining coupler.

In another preferred embodiment, which is either a dual or single scan method, a second parallel to the first is used.
Mach-Zehnder interferometer is used. In this second interferometer, the same coherent laser beam of the laser light source is connected by a beam splitter in front of these interferometers. With the help of this second interferometer, which has no optics on its measuring arm, when scanning the laser, the resulting light flux of the first interferometer caused by the non-linearity in the scanning speed was detected. Errors in power can be eliminated.

Instead of using a second interferometer, use a second port of a second polarizing beam splitter that recombines the two parts of the first beam to output a signal used for wavelength measurement. Is possible. This is done by connecting this port directly to a polarizer that creates interference between the two ports of the first light flux, evaluating the scanning speed when the interference pattern is generated and scanning the frequency of the laser light source. . Since the DUT is connected to the other port of the polarization delay unit, this port is not affected by the DUT and can therefore be used as a reference. An advantage of the present invention is that it reduces costs, reduces maintenance and creative equipment floor space required without the need for a second interferometer.

The invention may be implemented by one or more suitable software programs stored on or provided separately by any data carrier and executable on any suitable data processing unit, Or, it is clear that it is partially supported. The software program is preferably used so that optical properties can be evaluated from the measured data.

[0023]

BRIEF DESCRIPTION OF THE DRAWINGS Other objects and numerous attendant advantages of the present invention will become more readily apparent by reference to the detailed description when taken in conjunction with the accompanying drawings, wherein: FIG. It will be well understood. The components in the figures need not have the same scale, but instead are exaggerated for the purpose of clearly illustrating the principles of the invention. Basically or functionally identical or similar configurations are described with the same reference symbols.

In the following, reference is made in detail to the accompanying drawings, in which FIG. 1 shows an interferometric determination of the frequency-dependent Jones matrix of an optical transmission device (DUT) 2 to be tested according to the invention. FIG. 1 is a schematic explanatory view of a first preferred embodiment of the apparatus 1 for the present invention. The device 1 and the respective method according to FIG. 1 described below are one mode of implementing the invention. Other modes are shown in FIGS.
Referring to FIG. With the device shown in FIG. 1, the DUT 2, which is an optical component and can be a fiber, a Bragg grating or some other optical component or even air, is to be characterized by its chromatic dispersion and its PMD. The device 1 comprises a tunable laser 4 as a signal source which is continuously tuned in frequency.
including. The laser 4 emits a coherent laser beam 6. The laser beam 6 is connected to a first beam splitter 8 that splits the coherent laser beam 6 into a first incident beam 10 and a second incident beam 12. The first incident light beam 10 is connected to a second beam splitter 14. The second incident light flux 12 is connected to a third beam splitter 16. The second beam splitter 14 splits the first incident light beam 10 into a third laser beam 18 and a fourth laser beam 20. The third beam splitter 16 splits the second laser beam 12 into a fifth laser beam 22 and a sixth laser beam 24.

The third laser beam 18 comprises a polarization controller 26 (Hewlett-Packard HP8169A) having three subunits 26a, 26b and 26c.
Can be connected). By passing through the polarization controller 26, the laser beam indicated by 19 is adjusted in its polarization and connected to the DUT 2. After passing through the DUT 2, the laser beam 19
Is again coupled to the laser beam 20 of FIG. The fourth laser beam 20 travels a different optical path length (for example, 7 meters) as compared to the third laser beam 18 and the polarized laser beam 19 traveling from the second beam splitter 14 to the fourth beam splitter 28. I do. In the fourth beam splitter 28, the polarized laser beam 19 and the fourth laser beam 20 are superimposed, and the polarized laser beam 19 and the fourth laser beam 20 interfere with each other to perform the first superposition. The resulting laser beam 30 is produced. The first superposed laser beam 30 is then connected to a polarizing beam splitter 32 which splits the beam 30 into a seventh beam 34 and an eighth beam 36. The light beam 34 is then connected to a first photodiode 38. The light flux 36 is connected to the second photodiode 40. The polarization beam splitter 32, the first photodiode 38, and the second photodiode 40 constitute a polarization split receiver. First photodiode 38 and second photodiode 40
Converts their outputs to an analog-to-digital converter (ADC) 42 (N) connected to an evaluation unit (not shown).
AT-MI of National Instruments
O-16DE-10) to allow the detected data to be evaluated.

The second beam splitter 14, the third laser beam 18, the polarized laser beam 19, the fourth laser beam 20, and the fourth beam splitter 28 constitute a Mach-Zehnder interferometer 44. The third laser beam 18 and the polarized laser beam 19 are coupled to the Mach-Zehnder interferometer 4
4 measurement arms. The fourth laser beam 20 is
The reference arm of the Mach-Zehnder interferometer 44 is configured. D
The UT 2 is arranged on a measurement arm of the Mach-Zehnder interferometer 44.

The fifth laser beam 22 and the sixth laser beam 24 have different paths before they are superimposed on the fifth beam splitter 46 because the beam 24 travels a long distance symbolized by the loop 25. To progress. Ejecting the fifth beam splitter 46 is a second superimposed light beam 28 detected by a third photodiode 50. Third photodiode 50
Outputs each signal to an analog / digital converter (ADC) 42. Third beam splitter 1
The sixth and fifth laser beams 22, the sixth laser beam 24 and the fifth beam splitter 46 constitute a reference interferometer 52 for the measurement interferometer 44. This reference interferometer 52
Support is provided as part of the apparatus 1 to tune the tuning speed of the laser 4 to avoid possible non-linearities. For this purpose, the output of the photodiode 50 is
42 inputs. The ADC 42 thus obtains information regarding the occurrence of scanning speed non-linearities in the laser 4. Based on this information, this non-linearity is subtracted from the measurement results of the measurement interferometer 44 by the evaluation unit.

The tunable laser 4 is an input to the ADC 42 for triggering the ADC 42.

The creative process according to the invention proceeds as follows.

The third laser beam 18 obtains a predetermined polarization by the polarization controller 26 and becomes a polarized laser beam 19. The laser beam 19 polarized by the predetermined polarization is connected to the DUT 2. After passing through the DUT 2, the polarized laser beam 19 is superimposed on the fourth laser beam 20, ie, constitutes the reference arm of the Mach-Zehnder interferometer 44. The resulting first superimposed beam 30 is then combined with a seventh laser beam 34, an orthogonally polarized component of beam 30, and an eighth beam
Is connected to the polarization beam splitter 32, which produces the laser light flux 36. These orthogonal light beams 34 and 36 are detected by photodiodes 38 and 40, and the output signals of the photodiodes 38 and 40 are output from an ADC 42.
Received by When the signal is received by ADC 42, the evaluation unit can calculate and determine the following two (composite) elements of the Jones matrix in DUT2. The other two elements of the Jones matrix in DUT 2 are the polarization controller 2
6 by changing the polarization of the polarized laser beam 19 using the method described above and performing the above-described steps of the second embodiment of the method (or by using the polarization retardation steps described below). The altered polarization (not shown) of the polarized laser beam thus produced is preferably orthogonal to the polarization of the polarized laser beam 19 in the first implementation of the method. Therefore,
Missing 2 of Jones matrix in DUT2
It is possible to calculate one element. Having a complete Jones matrix makes it very easy to extract the transmittance or reflectance, DGD, PMD, PSP, PDL or chromatic dispersion of DUT2.

To illustrate the determination of the differential group delay (DGD) by the Jones matrix, the following description shows the calculations used to make this determination.

The Jones matrix U is what gives the relationship between the Jones vector at the output _{E b} and the input _{E a} of DUT 2.

[0033]

(Equation 1)

Here, ω is the optical frequency of the input light. From the Jones matrix itself, the active group delay (DGD), the two main states of polarization (PSP),
The reflectivity and the refractive index and the polarization dependent loss (PDL) can be determined. Using a conventionally known polarimeter, the phase relationship between the left and right terms of the above equation is ambiguous, which hampers determination of group delay, chromatic dispersion, and the like. The present invention
However, as with the Jones matrix, it allows the determination of the correct phase relationship between both terms of this equation. In the following, this information is used to calculate the group delay associated with each of the two PSPs of the device.

[0035] When the E _{a ±} and E _{b ±} is the primary condition in the output and the input of DUT2 each (unknown), according to the present invention, it is possible to establish the following relationship.

[0036]

(Equation 2)

Ea _± and _{Eb ±} are normalized such that their average phase disappears. Im {E _x · E _y } = 0 Here, the main state is a first approximation that is not related to frequency. Therefore, the following can be applied.

[0038]

(Equation 3)

The group delay of the device, τ _g, is defined as:

(Equation 4)

Thus, the above expression can be rewritten as a generalized eigenvalue problem.

[0041]

(Equation 5)

The eigenvalue λ is as follows.

[0043]

(Equation 6)

Thus, the two group delays are exactly the imaginary parts of the eigenvalues. Differential group delay (DGD)
Can be calculated directly according to the following equation:

[0045]

(Equation 7)

To be able to determine the Jones matrix U, it is necessary to perform two partial calculations with the orthogonal input polarizations E _a1 and E _{a2 of the} polarized laser beam 19 respectively. If those two input polarizations are used as the fundamental vectors of the Jones representation, this would correspond to the following input vectors.

[0047]

(Equation 8)

The corresponding vector can be read as follows:

[0049]

(Equation 9)

The symbol U _mn in this case indicates four elements in the Jones matrix. Light 20 from the reference arm can be described by the Jones vector as follows:

[0051]

(Equation 10)

In this case, τ _r indicates the group delay of the reference arm 20. For the sake of simplicity, it is assumed in the following that the power is distributed uniformly for the two basic states, and that there is no relative phase difference between them (linearly polarized light on the polarizing beam splitter 32). Incident at an angle of 45 °).

[0053]

[Equation 11]

The light incident on the detector then gives the following relationship by superposition with the measurement signal:

[0055]

(Equation 12)

The following applies with respect to detector power.

[0057]

(Equation 13)

The size U _mn is the term cos (φ _mn + τ
Assuming that the frequency changes much more slowly than _r ω), both U _mn (ω) and φ _mn (ω)
8, 40 can be determined from the reference signal. The derivative is numerically calculated from the matrix U (ω),
It can be taken out from the GD. In the case of real measurements, errors arise, especially in determining the absolute phase term of U (ω). In this regard, the light of the reference arm 20 may encounter the polarizing beam splitter 32 elliptically rather than linearly (φ ≠ 0). Furthermore, it is questionable whether the measurement consists of two wavelength scans of the laser light source 4. This problem can be solved by the embodiment of FIGS. Due to the above uncertainty,
There is _a phase error φ _a in the first column and φ _{b in} the second column of the Jones matrix U. Instead of matrix U, the following matrix is measured.

[0059]

[Equation 14]

This is equivalent to a polarization controller 26 connected upstream or downstream, so that no change occurs in the extracted DGD.
The main conditions that have arisen, however, are different. Therefore, no further correction is needed for DGD determination.

In a first attempt, the above method was used as described above using the apparatus of FIG. 1 to measure a high birefringent fiber (HiBi-fiber) as DUT2.
The results are evident in FIG.

In FIG. 2, the upper plot shows the group delay of the two main axes. The abscissa shows the wavelength in nm and the ordinate shows the group delay in ps. The lower plot shows the difference between the two group delays, showing the DGD in ps over the nm wavelength. DGD, while its absolute value fluctuates greatly across wavelengths, is clearly very good at 10 ps. One reason is that D
There would be Fabry-Perot interference in the optical polarization controller 26 of the free beam present in the measurement arm in addition to the UT (see FIG. 1). Under some circumstances, the polarization controller 26 can be located upstream of the second beam splitter 14 so that the variation in group delay does not play any role. This, however, also affects the polarization in the reference arm 20, which must be taken into account in the evaluation procedure.

FIG. 3 shows a measurement using the above method without a DUT in the apparatus of FIG.

In FIG. 3, the upper plot shows the group delay of the two main axes. The abscissa shows the wavelength in nm and the ordinate shows the group delay in ps. The lower plot shows the difference between the two group delays, showing the DGD in ps over the nm wavelength. As expected, DGD
Close to zero. Although a large deviation from the ideal value is recognized, it is possible to evaluate the measurement accuracy of the device 1 according to the present invention, which is several picoseconds. For convenience of description of these parts, reference is made to the description of Example 1 in FIG.

FIG. 4 shows a second embodiment 10 of the present invention.
FIG. 4 is a schematic explanatory diagram of a zero. The main difference between the embodiment 100 of FIG. 4 and the embodiment 1 of FIG. 1 lies in the introduction of a polarization delay unit (PDU) 102 into the path of the third laser beam 18.
The PDU 102 includes the second beam splitter 14 and the DUT
It is arranged on the path of the third laser beam 18 between the two. PDU
102 is a first polarizing beam splitter (PBS) 10
4 and a second PBS 106. First PBS1
04 divides the third laser beam 18 into a first portion 18a and a second portion 18b. Two parts 18a, 18b
Are provided by polarization maintaining fibers (PMF) that maintain the polarization of the respective portions 18a, 18b, respectively. Subsequently, the first portion 18a and the second portion 18b
Are recombined by the second PBS 106. PB
Both S104, 106 are considered to be made with two ports of PMF and two ports of single mode fiber (SMF). When the devices 104, 106 are connected using a standard matching connector, the light is completely directed to one port of the second PBS 106 and no light exits from the other port of the second PBS 106. Accordingly, the PMF of each light flux 18a, 18b is rotated by 45 ° to generate output at both ports of the second PBS 106.

However, the first portion 18a and the second portion 18b are separated by the first PBS 104 and the second portion 18b because the light beam 18b travels a long distance shown by the loop 108.
Travel different optical distances between the PBSs 106. This means that the second part 18b is delayed with respect to the first part 18a. Besides, both parts 1
Since 8a and 18b are orthogonal to each other, the second PBS
They do not interfere when recombined by.

The second PBS 106 has two outlets 112.
And 114 are provided. At the two outlets 112 and 114, both parts 18a, 18b are present because they do not interfere with each other. Light beams 18a, 18 exiting the exit 112
On the path b, the beam splitter 116 and the shutter 11
8, a PBS 32 connected to the DUT 2, the beam splitter 28 and finally to the detectors 38 and 40 is arranged. The optical characteristics of the DUT 2 are measured by transmittance.
In the beam splitter 28, the two parts 18a, 1
8b is the local oscillation light 2 as in the embodiment of FIG.
0 is superimposed. The local oscillation light 20 reaches the beam splitter 28 via the beam splitter 120. Thus, the two polarizations of both parts 18 a, 18 b and thus of the light bundle 18 are superimposed on the local oscillator light 20. Thus, both polarizations create an interference pattern at polarization splitter detectors 38 and 40 at the same time. Because of the different propagation delays of the two input polarizations, the two interference patterns have different electrical spectra. For this purpose, a digital filter (not shown) in the evaluation unit
Can split the information belonging to each input polarization. As a result, the second scan that was required in Example 1 of FIG. 1 for orthogonally polarized beams 18 is advantageously not required in Example 100.

In addition, DUT 2 can also be measured in reflectivity. To achieve this purpose, a beam splitter 116 is provided in the path of the reflected light beam 122. Reflected light flux 12
2 includes those reflective portions of portions 18a and 18b. The light bundle 122, including those reflections at the portions 18 a and 18 b, is superimposed on the local oscillator light 20 at the beam splitter 116. This superimposed light 124 is then directed to a PBS 126 that splits the light into a light flux 128 and a light flux 130. Luminous flux 128
Is connected to a detector 132, and the light flux 130 is connected to a detector 134. Detectors 132 and 134
Thus, the optical characteristics of the DUT 2 can be measured by the reflectance. In addition, based on the introduction of the PDU 102, this superimposed light 124 is
And a portion where the local oscillation light 20 and the portion 18b which is delayed with respect to the first portion 18a are superposed. Thus, the optical characteristics of the DUT 2 can be measured. To that end, the digital filter can split the information belonging to the respective input polarizations of the respective portions 18a and 18b in the same way as done with the detectors 38 and 40 for the transmittance measurement.

Instead of the reference interferometer 52, the second port 114 of the PBS 106 is connected to the reference port so that the portions 18a and 18b interfere with each other to detect the light 18a and 18b emitted through the PMF-connected polarizer 136. Detector 1 used as an interferometer and connected to polarizer 136
A superimposed light beam 138 is generated that exhibits an interference pattern detectable by 40. Therefore, the embodiment of FIG. 1 can be dramatically simplified.

In the embodiment 100, the input polarization of the system severely affects the measurement performance. Preferably, the input polarization is such that the light in the local oscillator 6 is a four detector 3
8, 40, 132, 134 should be selected to be evenly divided. Therefore, each of PBS10
Light incident on 4, 32, 126 is substantially polarized and should preferably achieve a 50% split ratio. The effective polarization states are located on a large circle on the Poincare sphere. Usually, the position of this circle is the position of each PBS 104,
32, 126 are different. 50 on the corresponding PBS
It is guaranteed that there are two intersections for each two of these circles that produce a% split ratio. In most cases,
All three circles do not intersect. Therefore, PBS10
A split ratio of 50% for all 4, 32, 126 cannot be guaranteed. In other words, in the worst case, 50
That is, a compromise that is not a percentage can be compromised. This worst case is when the input polarization of PDU 102 is 2
This corresponds to a linear polarization state with a misalignment of 2.5 °. In this worst case, sin ² (22.5 °) = 15
A minimum split ratio of% results, which leads to a sufficiently acceptable contrast in the interference pattern. PDU1
The 02 optimal input polarization is to be found during the initialization procedure (see below).

FIG. 5 shows a third embodiment 300 of the present invention.

The difference between the embodiment 300 of FIG. 5 and the embodiment 100 of FIG. 4 is that the PBS 106 is replaced by a polarization maintaining coupler 302. Thus, both output ports 112 and 114 emit the same signal. The use of output ports 112, 114 output signals is the same as in embodiment 100 in FIG. In the PDU 102, only the long path 18 b is used as shown by the symbol 304 because of the use of the polarization maintaining coupler 302.
Must be rotated by °.

FIG. 6 shows a fourth embodiment 400 of the present invention. In FIG. 6, the setting of the polarization delay unit is as follows.
This is the same as that of the embodiment 100 in FIG. Unlike the embodiment 100 of FIG.
Is not used for local oscillator wavelength measurements.
Instead, the aperture 402 of the beam splitter 120 not used in the embodiment 100 of FIG.
00 is used to split the input light of DUT2. Mouth 4
Since the fiber 404 used to transmit the signal from 02 is a single mode fiber (SMF), the polarization is not maintained because the polarizer 136 cannot be properly positioned in either case. Thus, a second polarizer 406 rotated by 45 ° with respect to the first polarizer 136 provides an interference pattern in each case. A beam splitter 408 is disposed on the fiber 404 to enable use of the second polarizer 406. The signal transmitted by the second polarizer 406 is detected by the second detector 410.

FIG. 7 shows a fifth embodiment 500 according to the present invention.
FIG. Embodiment 500 is similar to embodiment 400 of FIG. However, in the embodiment 500 of FIG. 7, the PDU 102 shows different settings. Instead of the second PBS 106, two Faraday reflecting mirrors 502 and 504 are provided. The use of Faraday mirrors 502, 504 may cause problems when the polarization is not properly aligned with the axis of the PMF.
This is to avoid a long member of the PMF inside the DU 102. The input beam 18 is split by the first PBS 104 into two linear polarization states (SOPs) and travels along the SMF before being reflected by the Faraday mirrors 502 and 504, respectively. Unlike ordinary reflectors,
The Faraday reflecting mirrors 502 and 504
Convert each of the P to orthogonal reflected polarization states. In this manner, light exits through fourth port 506 of PBS 104 without the need for a circulator. The use of PBS 104 to recombine the reflected light reflected by Faraday mirrors 502, 504 also ensures that the two delayed components are orthogonally polarized but do not interfere. When the Faraday mirrors 502, 504 do not produce completely orthogonal polarization states, a small portion of the light
It is not reflected back to the laser light source 4 and disturbs the signal path.

The output port 506 of the PBS 104 is connected to the PM
F508 and connect the portions 18a, 18b to the DUT as in the embodiment 100 shown in FIG.
2

The PMC 510 is used to transmit a small amount of light guided through the polarizer 136 to generate a wavelength reference signal. Two PMFs for both 45 ° and 90 ° connections
This setting makes use of a commonly matched polarization maintaining connector, as there is no need in between.

Further, the power is applied to each passage 18a, 1
8b, it is possible to direct the PDU 102 of the embodiment 500 of FIG. 7 to the power detectors in the respective passages 18a, 18b for measurement (not shown in FIG. 7). .

Before the DUT 2 is connected to the device 500 of FIG. 7, two photodetectors 38, 40 are used to provide a feedback circuit (not shown) to ensure that each of the local oscillating lights 6 is split by the PBS 32. ) Are connected together. This is done by opening the optical shutter 118 and using the control circuit to properly adjust the polarization controller 26.

To measure the power of each of the polarizations 18a, 18b incident on DUT 2, DUT 2 is replaced with a standard single mode fiber and the experiment is performed as described above. The total power in each of the polarizations 18a, 18b is two photodiodes 38, 40
Can be determined by adding these signals. Since standard single mode fibers are not considered to be involved in polarization dependent loss (PDL), the power of each frequency in the interfering signal indicates the power in one of the polarizations of portions 18a, 18b.

FIG. 8 shows a sixth embodiment 600 according to the present invention.
FIG. 2 is a schematic explanatory view of the present invention. The difference between the embodiment 600 and the embodiment 500 is that all the reference paths and the PDU 102
Is set by the PMF. Moreover, the local oscillating light 20 does not overlap with the reflected portions 18a, 18b at the beam splitter 116. In this embodiment 600, the local oscillating light 20 is superimposed on the portions 18a, 18b reflected by the beam splitter 602 before the light enters the beam splitter 126. In the embodiment 600, the polarization splitting receivers 604, 606 are preferably implemented with bulk optics to avoid problems with PMF in the DUT path.

FIG. 9 shows a seventh embodiment 700 of the present invention.
FIG. 2 is a schematic explanatory view of the present invention. FIG. 4 of the embodiment 700
Is different from that of Example 100 in the following.
The PDU 102 does not include the PBS 104. Instead, the PDU 102 is stored in the path 18a, 1
8b includes polarization setting tools 702 and 704. The path 18 is divided into the paths 18a, 1a by the beam splitter 706.
8b. Switches 708 and 710 are provided on the respective paths 18a and 18b. Road 18
a, 18b are joined by PBS 106.

Further, the embodiment 700 has a phase calibration unit (PCU) 712. PCU 712 includes portions 18a, 1a provided by beam splitter 714.
8b is superimposed with the reference signal 20 provided by the third beam splitter 716 with the help of the beam splitter 718. This superimposed signal is again split by beam splitter 718 and detected by photodiodes 720 and 722 of PCU 712.

The beam splitter 28, the path 30, the beam splitter 32, and the detectors 38 and 40 constitute a transmission receiver 734. Beam splitter 736, road 73
7, beam splitter 126 and detectors 132 and 1
34 constitutes a reflection receiver 738.

A switch 724 is provided on the reference arm 20 to cut off the reference signal. Switches 708, 710 and 724 are not required to perform this creative method.

One advantage of using PCU 712 is described below. Using the "single scan" technique, multiple interference frequencies (the number depends on which architecture was chosen) can be found in the power spectrum of the received data. By individually banding around each of the peaks, information about the Jones matrix element associated with the interference peak can be retrieved. For example, the filtered interference peak associated with the first Jones matrix element has an amplitude and phase given by:

[0086]

(Equation 15)

However, the filtered interference peaks associated with the second Jones matrix element have different amplitudes and phases given by:

[0088]

(Equation 16)

The difference occurs because the delays τ _xx and τ _yx are different.

The Jones matrix must have the following structure, where τ is any delay.

[0091]

[Equation 17]

However, to achieve this, one must know the difference ω (τ _xx −τ _yx ), among other things. This must be known with an accuracy of a few femtoseconds in order to maintain the maximum accuracy of the measurement. The best way to accomplish this is to measure the best frequency of these two delayed signals using LO before the two delayed signals pass through DUT2. This is PCU71
2 is performed. In that case (for example, FIGS. 9 and 10
(As in the case of architectures 700 and 800 in FIG. 3), the interfering signals corresponding to those embodiments described above, respectively, are cos (ωτ _xx ), assuming no large PMD.
And cos (ωτ _yx ) will be at least approximately proportional. From these signals, the difference ω (τ _xx −τ _yx ) can be determined. This allows

[0093]

(Equation 18) Can be replaced by the following modified signal:

[0094]

[Equation 19]

By performing the same procedure on all the interfering signals, it is possible to create a Jones matrix in the manner shown above.

[0096]

(Equation 20)

Preferably, the difference ω (τ _xx −τ _yx ) is measured by the PCU 712 at the same time as the measurement being performed, for at least the following reasons. The reasons are: 1) Due to temperature or stress deformation, the value measured the first time is different at other times. In fact, even if any inaccuracy is relatively small,
2) Many of the errors that can occur in these systems can confuse the interferometer and give noisy phase measurements, which can lead to poor PMD measurements. This is due to vibrational noise. Since the light travels a very similar path for both phase calibration by the PCU 712 and transmission receiver measurements by the detectors 38, 40, the acoustic noise measured at both locations can be similar. . As a result, the DU using the phase calibration receiver 712 is
Vibration noise and acoustic noise in the measurement of T2 can be corrected.

FIG. 10 shows an eighth embodiment 80 according to the present invention.
FIG. 2 is a schematic explanatory diagram of a zero. Example 800 is
The second PDU 102-2 is used for the reference arm 20. This second PDU 102-2 generates two parts 20a, 20b of the reference signal 20, in which the parts 20a, 20b are delayed with respect to each other in the same way as the PDU 102. However, PDU 102-2 includes a different loop 108-2 that doubles delayed portion 20b with undelayed portion 20a. Second P on LO road
The advantages of providing the DU 102-2 are as follows.
In the double sweep technique in FIG. 1 for determining the Jones matrix, the polarization beam splitter 3
2 is used in front of the transmissive receiver 38 to define the output x, y axes. In other words, the x component of the output light from DUT 2 is directed to receiver 38, where the interference is recorded and used to determine the Jones matrix element. Similarly, the y component of the output light from the DUT 2 is guided to another receiver 40 and measured. x
When the axially polarized light is the first input to the DUT 2, the x-axis polarized output detector 3
8 will provide information about the _Uxx Jones matrix. Similarly, output detector 40 polarized in the y-axis direction will provide information about element U _xy .

The second PDU as in embodiment 800
, Into the LO path, two orthogonally advanced polarizations can be used to define the output x, y axes. In other words, light passing through the short LO PDU 102-2 path 20a is defined to be polarized in the x-axis direction at the receiver, and light passing through the long path 20b is defined to have y-axis polarization at the output. Can be D
Both of the two orthogonal lightwaves coming from UT2 PDU 102 interfere with each of the two lightwaves from LOPDU 102-2. When properly implemented, this is 4
Produces two distinct interference frequencies. Each of these interferences contains information about one element in the Jones matrix.

Further, the embodiment 800 in FIG. 10 does not include the PBSs 32 and 126 because they no longer require the second PDUs 102-2 as described above.

In addition, the embodiment 800 shown in FIG.
Uses two photodiodes for each receiver. This is the transmission receiver 802 of the embodiment 800.
Means that the beam splitter 28, the obtained path 30 and the second obtained path 804 are set and both paths 30, 804 are detected by the two diodes 806 and 808. Similarly, the reflection receiver 810 set by the beam splitter 812 is separated by the photodiodes 818 and 820 into the beam splitter 8.
12 to detect both paths 814, 816.

The two photodiodes 720, 722, 806, 80 in the receivers 712, 802, 810
One of the advantages of 8,818,820 is as follows. In most measurements, the laser power generated from the tunable laser 4 has a relative intensity noise (RI
N). These noise fluctuations can have a wide frequency content that can be extended into the range of frequencies where the measurements are attempted. The end result can be simple noise fluctuations in measurements such as group delay, insertion loss, PDL, etc. This is a rather small effect, but it is desirable that it be measured very precisely because this small effect can be significant. This noise is 2
Photodiodes 720, 722, 806, 80
8, 818, 820 can be eliminated by using creative solutions. This creative solution consists of two light waves E ₁ = E ₀₁ e ^jωx and E ₂ = E ₀₂ e
^{jω (t + τ) work} together in the coupler to obtain an intensity I ₁ (t) at the _first output port, where
The intensity is expressed as follows.

[0103]

(Equation 21)

The strength at the other arm would be as follows:

(Equation 22)

Here, RIN is N₁Relative to light wave 1
With strong intensity noise E₁= (E₀₁+ N₁) e ^jωtAs and N₂
Where E is the relative intensity noise of lightwave 2._Two= (E₀₂+ N_Two) e
^jωtBy rewriting the two light waves as
You. When two light waves are added in the coupler, one of the
The intensity measured by the
become that way.

[0106]

(Equation 23)

Due to the action of the coupler on the light, the light intensity at the other arm is as follows:

[0108]

(Equation 24)

Obviously, the noise can be removed by a simple subtraction of the measured intensities at the two output arms of the coupler. The result is as follows.

[0110]

(Equation 25)

In these interferometric techniques, subtraction is a conventional way to eliminate RIN, since the interference term is of interest.

Both embodiments 700 and 800 include a wavelength reference unit (WRU) 730 located upstream of the beam splitter 14 and connected to the path of the light beam 6 by the beam splitter 732. WRU730
Measures the wavelength of the light beam 6 as a criterion in the evaluation process in the present invention for further improving the accuracy of the evaluation result. However, a WRU is not essential to the practice of the present invention.

The present invention will be described with reference to the above-described embodiment. The present invention relates to a method for determining the characteristics of an optical device (2) to be tested. 19) and a step of splitting into a second light beam (20), and the first light beam (18, 19) having a predetermined polarization.
Connecting the second light beam (20) to the first light beam (18,
19) a step of traveling on a different path from the first light beam (18, 19) and the second light beam (20), and the first light beam (18) in the resulting superimposed light beam. , 19) and the second light beam (20), and when tuning the frequency of the incident light beam over a predetermined frequency range, the superimposed light beam (30) as a function of frequency and polarization. A) detecting the power of the optical device under test from the frequency dependence of the detected power.

Preferably, the method according to the invention further comprises:
Splitting the superimposed light beam (30) into a third light beam (34) and a fourth light beam (36) in a polarization-dependent state, and tuning the frequency of the incident light beam over a predetermined frequency range; Detecting the power of the third light beam (34) and the fourth light beam (36) as a function of frequency.

Preferably, the method according to the invention further comprises:
Extracting the elements of the Jones matrix in the optical device under test (2) from the frequency dependence of the detected power.

Preferably, the method according to the invention further comprises:
The first light flux (18, 19) for the predetermined polarization;
Changing the polarized light to the changed polarized light, and performing the process according to claim 1 a second time with respect to the changed polarized light.

Preferably, the method according to the invention further comprises:
After splitting the incident light beam (6), the first light beam (1
8, 19) or setting the polarization of the incident light beam (6) before the division.

Preferably, the method according to the invention further comprises:
Changing the polarization of the first light flux (18, 19) to orthogonal polarization.

Preferably, the method according to the invention further comprises:
Converting the predetermined polarized light into linearly polarized light.

Preferably, the method according to the invention further comprises:
Splitting an incident light beam (6) into a first initial light beam (10) and a second initial light beam (12); and performing the process according to claim 1 using the first initial light beam (10). Process and
Splitting the second initial light beam (12) into a fifth light beam (22) and a sixth light beam (24);
After each of 24) travels on a different path, the fifth light flux (22) and the sixth light flux (24) are superimposed,
The fifth in the resulting superimposed beam (48)
Causing interference between the light beam (22) and the sixth light beam (24) and tuning the frequency of the incident light beam (6) over a predetermined frequency range. Detecting the power of the light beam (48), detecting the non-linearity of the tuning rate of the incident light beam (6) over a predetermined frequency range, and detecting the detected non-linearity when detecting the non-linearity. Correcting the effect of the third light beam and the fourth light beam on the detected power using the information on the linearity.

Preferably, the method according to the present invention further comprises:
The first light beam (18, 19) is transferred to a first portion (18a).
And a second part (18b);
Delaying the portion (18b) with respect to the first portion (18a); recombining the first portion (18a) and the second portion (18b); Connecting the re-coupled portion with the optical device under test.

Preferably, the method according to the invention is such that the polarization of said portions (18a, 18b) is at least approximately orthogonal to each other.

Preferably, the method according to the invention further comprises:
Each of the recombined portions (18a, 18b) has at least about 5 powers of coherent beam power.
To have 0%.

Preferably, the method according to the invention further comprises:
Filtering the detected peak of the power spectrum; locating the peak in the spectrum with respect to each of the portions (18a, 18b); and detecting the frequency and polarization dependence of the detected power. Taking out the optical characteristics of the optical device to be subjected to the test from the viewpoint of performance.

Preferably, the method according to the invention further comprises:
Creating interference between the recombined portions (18a, 18b) of the resulting superimposed beam (138); and tuning the frequency of the incident beam (6) over a predetermined frequency range. Continuously detecting the power of the resulting superimposed light flux as a function; and detecting non-linearity in the tuning gradient frequency when tuning the frequency of the incident light flux over a predetermined frequency range; When detecting the linearity, the detected powers of the third light beam (34) and the fourth light beam (36) generated by the non-linearity using the information of the detected non-linearity. And correcting the effect on.

Preferably, the method according to the invention further comprises:
The polarization of the recombined portions (18a, 18b) allows the recombined portions (18a, 18b) to be polarized.
b) overlapping to cause interference.

Preferably, the method according to the invention further comprises:
Extracting the polarization mode dispersion of the device under test (2), preferably from information obtained by measurements expressed as Jones matrix elements of the device under test (2), of the device under test (2) Deriving the chromatic dispersion of the device under test (2) from the Jones matrix element, and changing the basic state in the polarization of the device under test (2) from the Jones matrix element of the device under test (2); Removing the polarization dependent loss of the device under test (2) from the Jones matrix element of the device under test (2); removing the polarization dependent loss of the device under test (2) from the Jones matrix element of the device under test (2) Short and long radicals in the polarization of the device under test (2) Retrieving the short and long group delays associated with dynamic conditions; retrieving the insertion loss of the device under test from the Jones matrix element of the device under test (2); Extracting the transmittance and reflectivity of the device under test (2) from the Jones matrix element, or a high rate of change of the group delay with respect to frequency from the Jones matrix element of the device under test (2). At least one of the steps of extracting the next polarization mode dispersion is provided.

Preferably, the method according to the invention further comprises:
Splitting the first light beam (18, 19) into a first part (18a) and a second part (18b) in a polarization-dependent manner.

Further, according to the present invention, there is provided an apparatus (2) for determining an optical characteristic of an apparatus to be tested, wherein a path of an incident light beam (6) is changed to a first light beam (18, 19) traveling on a first path. A first beam splitter (14) for splitting a second light beam (20) traveling on the second path, wherein the optical device (2) to be tested is connected to the first path and A first beam splitter (14) adapted to be connected to the first light beam (18, 19) having the polarization of the second light beam, the second light beam (20) being connected to the first light beam (18, 19). After traveling on different paths, the first luminous flux (18, 19)
Between the first light beam (18, 19) and the second light beam (20) in the resulting superimposed light beam traveling on a path obtained by superposing the second light beam (20) with the second light beam (20). And a second beam splitter (28) disposed on the first path and the second path, which cause interference with the first path and the second path, when the frequency of the incident light beam (6) is tuned over a predetermined frequency range, The resulting superimposed luminous flux (30,80) traveling in the obtained path as a function of polarization
4,814,816), a detector unit arranged in the obtained path for detecting the power, and an evaluation unit for extracting optical characteristics of the optical device (2) to be tested from the detected power. (38, 40,
806, 808, 818, 820).

Preferably, the superimposed light beam (30) is combined with the third light beam (34) traveling on the third path and the fourth light beam (36) traveling on the fourth path in a polarization-dependent manner. )
And the third beam (34) as a function of frequency when tuning the frequency of the incident beam (6) over a predetermined frequency range with the polarization beam splitter (32) in the obtained path for splitting into A first power detector in said third path for detecting said power of said first power detector;
And tuning the frequency of the incident light beam (6) over a predetermined frequency range as a function of frequency.
The fourth power for detecting the power of the light beam (36)
And a second power detector (40) in the path of (1).

Preferably, an evaluation unit is provided for extracting the Jones matrix element of the optical device (2) under test from the frequency dependence of the detected power.

Preferably, the first beam splitter (14) and the second beam splitter include a Mach-Zehnder interferometer and / or a Michelson interferometer.

Preferably, the first light flux (1
8, 19) comprising a polarization setting tool in the first path for adjusting the polarization of the first beam splitter (26) to the predetermined polarization. It is arranged on the path of the incident light beam (6) before and after 14).

Preferably, the polarization setting tool (26)
Linearly adjusts the polarization of each of the light beams (6, 18).

Preferably, the incident light beam (6) further comprises a first initial light beam (10) traveling on a first initial path and a second initial light beam (10).
A third beam splitter (8) disposed on the incident light beam (6) for splitting into a second initial light beam (12) traveling on the initial path, and the second initial light beam (12). 5
A fourth beam splitter (16) disposed on the second initial path for splitting into a fifth light flux (22) traveling on a path of the second and a sixth light flux (24) traveling on the sixth path; After each of the light beams (22, 24) travels on a different path, the fifth light beam (22) and the sixth light beam (24) are superimposed on each other and travel on a second obtained path. The fifth beam (2) in the resulting superimposed beam (48)
A fifth beam splitter (46) disposed on the fifth path and the sixth path so as to cause interference between 2) and the sixth light beam (24); and the incident light beam (6). Are tuned over a predetermined frequency range, the resulting superimposed flux (4) as a function of frequency.
8) a third power detector (50) disposed in the second obtained path for detecting the power of the power detector, whereby the output of the power detector (50) is Connected to an evaluation unit, detecting any non-linearity in the tuning gradient frequency when tuning the frequency of the coherent light beam over a predetermined frequency range, and when the evaluation unit detects any non-linearity, The evaluation unit uses the information of the detected nonlinearity to generate the third light beam (34) and the fourth light beam (36) generated by the nonlinearity.
Is configured to correct the effect on the detected power.

Preferably, a polarization delay unit (1
02), the first light flux (18, 19) is divided into a first portion (18a) and a second portion (18b), and the second light beam is divided with respect to the first portion (18a). Part (18
b) the first part (18a) and the second part
Is reconfigured to connect the recombined portion with a different polarization to the optical device to be tested (2).

Preferably, the polarization delay unit (10
2) has a first polarizing beam splitter (104) for splitting the first light beam (18, 19) into a first part (18a) and a second part (18b).

Preferably, the polarization delay unit (10
2) has a second polarizing beam splitter (106) that recombines the first part (18a) and the second part (18b).

Preferably, the polarization delay unit (10
2) a first optical path for the first part (18a) and a second optical path for the second part (18b) having a longer optical path length than the first optical path; And the second portion (18) with respect to the first portion (18a).
b) is delayed.

Preferably, the polarization delay unit (10
2) a polarizing device (104, 106, 502, 504)
The optical device (2) to be tested is connected to the recombined parts (18a, 18b) with different polarizations.

Preferably, the polarization delay unit (10
2) having the device (114, 116, 120, 510) and the said recombined part (1
8a, 18b) to the polarizers (136, 406) to cause interference between said paths in the resulting superimposed luminous flux (138) traveling on the paths obtained, said apparatus further comprising Power detector (140, 4
10) detecting the power of the resulting superimposed light beam (138) as a function of frequency when tuning the frequency of the coherent light beam over a predetermined frequency range, thereby detecting the power detector (14).
0,410) is connected to the evaluation unit to detect a non-linearity in the function of the tuning gradient frequency when tuning the frequency of the incident beam (6) over a predetermined frequency range, and the evaluation unit When detecting any non-linearity, the evaluation unit uses the detected non-linearity to calculate the third light flux (34) and the fourth light flux (36) caused by the non-linearity. The effect on the detected power is corrected.

Preferably, according to the invention, the device (11
4, 116, 120, 510) connected to the output port (114) of the second polarizing beam splitter which is not connected to the optical device (2) to be tested or to the optical device (2) to be tested. The output port (11) of the second polarizing beam splitter (106)
A polarization maintaining coupler (510) connected to 4), or
At least one beam splitter (116, 120) connected to the output port (114) of the second polarizing beam splitter (106) connected to the optical device (2) to be tested.

[Brief description of the drawings]

FIG. 1 is a schematic explanatory view of an apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram showing measurement results of advanced birefringent fibers;
(A) a plot showing the group delay of the two spindles, and (b) 2
FIG. 4 shows a plot showing the difference between two group delays.

3A and 3B are measurement results without a DUT, showing (a) a plot showing group delay of two main axes, and (b) a plot showing a difference between two group delays.

FIG. 4 is a schematic explanatory view of a second embodiment according to the present invention.

FIG. 5 is a schematic explanatory view of a third embodiment according to the present invention.

FIG. 6 is a schematic explanatory view of a fourth embodiment according to the present invention.

FIG. 7 is a schematic explanatory view of a fifth embodiment according to the present invention.

FIG. 8 is a schematic explanatory view of a sixth embodiment according to the present invention.

FIG. 9 is a schematic explanatory view of a seventh embodiment according to the present invention.

FIG. 10 is a schematic explanatory diagram of an eighth embodiment according to the present invention.

 ──────────────────────────────────────────────────続 き Continued on the front page (71) Applicant 399117121 395 Page Mill Road Palo Alto, California U.S.A. S. A. (72) Inventor Greg Van Wigaren Los Gatos Dardaneri Lane, California, United States No. 22 114 (72) Inventor Douglas Bunny Los Altos, California, United States 897 F-Term (reference) 2G086 EE06 EE12 KK01 KK07

Claims (30)

[Claims] 1. A method for determining characteristics of an optical device under test, comprising: dividing an incident light beam into a first light beam and a second light beam; and testing the first light beam having a predetermined polarization with the first light beam. Connecting the second light beam to a path different from the first light beam, superimposing the first light beam and the second light beam, and causing the superimposition. Causing interference between the first light beam and the second light beam in the obtained light beam; and when the frequency of the incident light beam is tuned over a predetermined frequency range, the frequencies are superimposed as a function of frequency and polarization. A method comprising: detecting the power of a light beam; and extracting characteristics of the optical device under test from the frequency dependence of the detected power. 2. The method according to claim 1, further comprising the step of dividing the superimposed light beam into a third light beam and a fourth light beam in a polarization-dependent state, and adjusting the frequency of the incident light beam over a predetermined frequency range. Detecting the power of the third light beam and the fourth light beam as a function. 3. The method according to claim 1, further comprising the step of extracting a Jones matrix element in the optical device under test from the frequency dependence of the detected power. 4. The method according to claim 1, further comprising: changing the polarization of the first light beam to the changed polarization with respect to the predetermined polarization; and performing the process according to claim 1 for the second time with respect to the changed polarization. And a step of executing.
The method described in. 5. The method according to claim 1, further comprising: setting the polarization of the first light beam after dividing the incident light beam, or setting the polarization of the incident light beam before the division. 4. The method according to 4. 6. The method according to claim 4, further comprising the step of changing the polarization of the first light beam to orthogonal polarization. 7. The method according to claim 4, further comprising the step of converting said predetermined polarized light into linearly polarized light. 8. The method according to claim 1, further comprising: dividing the incident light beam into a first initial light beam and a second initial light beam; performing the process according to claim 1 using the first initial light beam; Splitting the initial light beam of No. 2 into a fifth light beam and a sixth light beam;
Superposing the light beam and the sixth light beam, and causing interference between the fifth light beam and the sixth light beam in the generated superposed light beam; and adjusting the frequency of the incident light beam to a predetermined frequency range. Detecting the power of the superimposed light flux as a function of frequency when tuning over; detecting the non-linearity of the tuning rate of the incident light flux over a predetermined frequency range; and detecting the non-linearity. And correcting the effect of the third light beam and the fourth light beam on the detected power caused by the non-linearity using the detected non-linearity information. Item 4
8. The method according to any one of claims 1 to 7. 9. The method according to claim 1, wherein the first light beam is divided into a first portion and a second portion.
Splitting the second portion with the first portion; delaying the second portion with respect to the first portion; re-combining the first portion and the second portion; Connecting the re-coupled portion with the optical device under test. 10. The method of claim 9, wherein the polarizations of the portions are at least approximately orthogonal to one another. 11. The apparatus of claim 9 wherein each of said recombined portions has at least about 50% of the power of the coherent light beam.
0. The method according to any one of 0. 12. The method further comprising: filtering the detected peak of the power spectrum; locating the peak in the spectrum with respect to each of the portions; and detecting the frequency of the detected power and the frequency of the power. Extracting the optical characteristics of the optical device to be tested from the polarization dependence. 13. The method further comprising: causing interference between the recombined portions of the resulting superimposed light beam; and tuning the frequency of the incident light beam as a function of frequency when tuning the frequency of the incident light beam over a predetermined frequency range. Continuously detecting the power of the superimposed light beam, detecting the non-linearity at the tuning gradient frequency when tuning the frequency of the incident light beam over a predetermined frequency range, and detecting the non-linearity And correcting the effect of the third light flux and the fourth light flux on the detected power caused by the non-linearity using the detected non-linearity information. Item 13. The method according to any one of Items 9 to 12. 14. The method of claim 13, further comprising the step of superimposing said recombined portions to cause interference by polarizing said recombined portions. 15. The step of extracting the polarization mode dispersion of the device under test, preferably from information obtained by measurement expressed as a Jones matrix element of the device under test, the Jones of the device under test. Extracting the chromatic dispersion of the device under test from a matrix element; extracting the basic state in the polarization of the device under test from the Jones matrix element of the device under test; the Jones of the device under test Removing the polarization dependent loss of the device under test from a matrix element; short and long associated with the short and long basic state of polarization of the device under test from the Jones matrix element of the device under test. Group delay Retrieving the insertion loss of the device under test from the Jones matrix element of the device under test; determining the transmittance and reflectance of the device under test from the Jones matrix element of the device under test. 15. The method according to claim 4, further comprising the step of extracting or extracting higher order polarization mode dispersion, such as different rates of change of group delay with respect to frequency, from the Jones matrix element of the device under test. The method according to any of the above. 16. The method according to claim 9, further comprising the step of splitting said first light beam into a first part and a second part in a polarization-dependent manner. 17. An apparatus for determining optical characteristics of a device under test, comprising: dividing an incident light beam path into a first light beam traveling on a first path and a second light beam traveling on a second path. A first beam splitter, wherein the optical device under test comprises a first beam splitter connected to the first path and connected to the first light beam having a predetermined polarization; After the second light beam travels on a different path from the first light beam, the second light beam travels on a path obtained by superimposing the first light beam and the second light beam. A first beam path and a second beam splitter disposed on the second path for causing interference between the first light beam and the second light beam; and adjusting a frequency of the incident light beam to a predetermined frequency range. Tuning as a function of frequency and polarization A detector unit disposed in the obtained path for detecting the power of the generated superimposed light beam traveling in the obtained path; and an optical device for receiving the test from the detected power. An evaluation unit for extracting optical characteristics. 18. The obtained light beam for splitting the superimposed light beam in a polarization dependent manner into a third light beam traveling on a third path and a fourth light beam traveling on a fourth path. A polarization beam splitter in a path, and a first in the third path for detecting the power of the third light flux as a function of frequency when tuning the frequency of the incident light flux over a predetermined frequency range. A power detector and a second power detector in the fourth path for detecting the power of the fourth light beam as a function of frequency when tuning the frequency of the incident light beam over a predetermined frequency range. 18. The device according to claim 17, wherein the device comprises: 19. The apparatus according to claim 17, further comprising an evaluation unit for extracting the Jones matrix element of the optical device under test from the frequency dependence of the detected power. 20. The first beam splitter and the second beam splitter.
Beam splitter is a Mach-Zehnder interferometer and / or
Alternatively, a Michelson interferometer is included.
20. The apparatus according to any one of claims 19 to 19. 21. A polarization setting tool in the first path for adjusting a polarization of the first light beam with respect to the predetermined polarization, wherein the polarization setting tool includes the polarization setting tool. 21. A method according to any one of claims 17 to 20, wherein the method is arranged in the path of the incident light beam before and after one beam splitter. 22. The method according to claim 17, wherein the polarization setting tool linearly adjusts the polarization of each of the light beams. 23. An apparatus according to claim 15, wherein said incident light beam is disposed on said incident light beam for dividing said incident light beam into a first initial light beam traveling on a first initial path and a second initial light beam traveling on a second initial path. A third beam splitter disposed on the second initial path for splitting the second initial light beam into a fifth light beam traveling on a fifth path and a sixth light beam traveling on a sixth path; A fourth beam splitter and the fifth beam splitter each having traveled a different path.
And causing interference between the fifth light beam and the sixth light beam in the generated superimposed light beam that travels on the second obtained path by superimposing the light beam and the sixth light beam. A fifth beam splitter disposed on a fifth path and the sixth path, wherein the frequency of the incident light beam is tuned over a predetermined frequency range; A third power detector arranged in the second obtained path for detecting power, whereby the output of the power detector is connected to the evaluation unit and the coherent light beam When tuning frequencies over a predetermined frequency range, any non-linearities in the tuning gradient frequency are detected, and the evaluation unit detects any non-linearities When I,
The evaluation unit uses the detected non-linearity information to correct the effect of the third light beam and the fourth light beam caused by the non-linearity on the detected power. A method according to any of claims 17 to 22, wherein the method comprises: 24. Further, a polarization delay unit is provided, the first light beam is divided into a first part and a second part, and the second part is delayed with respect to the first part. 24. Apparatus according to any of claims 17 to 23, wherein the first part and the second part are recombined and the recombined part with different polarization is connected to the optical device under test. . 25. The apparatus according to claim 24, wherein said polarization delay unit comprises a first polarization beam splitter for splitting said first light beam into a first part and a second part. 26. The polarization delay unit according to claim 24, wherein the polarization delay unit has a second polarization beam splitter that recombines the first part and the second part.
An apparatus according to claim 5. 27. A second optical path for the second part, wherein the polarization delay unit has a first optical path for the first part and a longer optical path length than the first optical path. 27. Apparatus according to any one of claims 24 to 26, comprising: delaying the second part with respect to the first part. 28. The polarization delay unit having a polarization device for connecting an optical device under test to the recombined portion with a different polarization.
28. The apparatus according to any one of items 4 to 27. 29. The path in the resulting superimposed light beam which travels a path obtained by providing the polarizer with said recombined part with different polarizations having a device, said polarization delay unit. Wherein the apparatus further comprises a power detector in the resulting path to tune the frequency of the coherent light beam over a predetermined frequency range and the resulting superimposed light beam as a function of frequency. Detecting the nonlinearity in the function of the tuning gradient frequency when the output of the power detector is connected to the evaluation unit and tunes the frequency of the incident light beam over a predetermined frequency range; And when the evaluation unit detects any non-linearity, the evaluation unit uses the detected non-linearity to perform Claim and to correct the effect on the detected power of the third light flux generated by the nonlinearity and the fourth light beam 24
29. The apparatus according to any one of claims to 28. 30. The apparatus of claim 30, wherein the apparatus is not connected to the optical device under test.
A polarization maintaining coupler connected to the output port of the polarizing beam splitter, or the output port of the second polarizing beam splitter connected to the optical device under test, or connected to the optical device under test. 30. The apparatus of claim 29, wherein the device is at least one beam splitter connected to the output port of the second polarizing beam splitter.