Wavelength dispersion measuring device and wavelength dispersion measuring method

Abstract

To provide a wavelength dispersion measuring device that can accurately measure wavelength dispersion even in samples prone to exhibiting nonlinear optical phenomena. [Solution] The wavelength dispersion measuring device 1C comprises a light source 10, an optical pulse train generation unit 20, an optical amplifier 30, a measurement unit 40, a calculation unit 50, a spectrometer 60, and a control unit 70. The optical pulse train generation unit 20 generates an optical pulse train including a first optical pulse and a second optical pulse based on the optical pulses output from the light source 10. The spectrometer 60 measures the spectrum of the optical pulse train after it has been output from the optical pulse train generation unit 20 and passed through the optical amplifier 30. Based on the spectrum measured by the spectrometer 60, the control unit 70 controls the generation of the optical pulse train by the optical pulse train generation unit 20 so that the spectral shapes of the first optical pulse and the second optical pulse are symmetrical.

Description

Technical Field

[0001] The present invention relates to a wavelength dispersion measuring apparatus and a wavelength dispersion measuring method.

Background Art

[0002] As an apparatus for measuring the wavelength dispersion of a sample, there is known an apparatus described in Patent Document 1. The wavelength dispersion measuring apparatus described in this document generates an optical pulse train including a first optical pulse and a second optical pulse having different peak wavelengths, determines the time intervals of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train after propagating through the sample, and determines the wavelength dispersion of the sample based on this time interval or the like.

Prior Art Document

Patent Document

[0003]

Patent Document 1

Non-Patent Document

[0004]

Non-Patent Document 1

[0005] The inventors attempted to measure the wavelength dispersion of various samples using the above-described wavelength dispersion measuring device and found that, depending on the sample, nonlinear optical phenomena easily occur with the measurement probe light pulse, making it impossible to accurately obtain the wavelength dispersion. Therefore, the inventors attempted to implement measures to suppress the occurrence of nonlinear optical phenomena in the sample while performing wavelength dispersion measurement, but simple measures such as reducing the intensity of the probe light were not sufficient to accurately measure the wavelength dispersion.

[0006] The present invention was made to solve the above-mentioned problems, and aims to provide a wavelength dispersion measuring device and a wavelength dispersion measuring method that can accurately measure wavelength dispersion even in samples that are prone to exhibiting nonlinear optical phenomena. [Means for solving the problem]

[0007] The wavelength dispersion measuring device of the present invention is a device that propagates an optical pulse train containing a first optical pulse and a second optical pulse having different peak wavelengths to a sample, and measures the wavelength dispersion of the sample based on the shape of the autocorrelation of the time waveform of the optical pulse train after propagation or the shape of the crosscorrelation between it and the time waveform of a reference optical pulse.

[0008] A first aspect of the wavelength dispersion measuring device of the present invention comprises: (1) a light source that outputs optical pulses having a bandwidth; (2) an optical pulse train generation unit that generates and outputs an optical pulse train including a first optical pulse having a spectral peak at a first wavelength within the bandwidth and a second optical pulse having a spectral peak at a second wavelength within the bandwidth, based on the optical pulses output from the light source; (3) an optical amplifier that amplifies and outputs the optical pulse train output from the optical pulse train generation unit; (4) a spectrometer that measures the spectrum of the optical pulse train output from the optical amplifier; and (5) a control unit that controls the generation of an optical pulse train by the optical pulse train generation unit such that, among the spectra measured by the spectrometer, the spectral shape of the first optical pulse is symmetrical with respect to the first wavelength, and the spectral shape of the second optical pulse is symmetrical with respect to the second wavelength.

[0009] A second aspect of the wavelength dispersion measuring device of the present invention further comprises, in addition to the first aspect, (1) a measuring unit that measures the time interval between the peaks of a first optical pulse and a second optical pulse based on the shape of the autocorrelation of the time waveform of the optical pulse train output from the optical amplifier or the shape of the crosscorrelation between the time waveform of the optical pulse train output from the optical amplifier or the time waveform of a reference optical pulse, with the sample positioned such that the optical pulse train output from the optical pulse train generation unit propagates through the sample and is then input to the optical amplifier; and (2) a calculation unit that determines the wavelength dispersion of the sample based on the time interval measured by the measuring unit.

[0010] In a third aspect of the wavelength dispersion measuring device of the present invention, in addition to the first or second aspect, the control unit controls the generation of the optical pulse train by the optical pulse train generation unit so that the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier is within the range of 1:10 to 10:1.

[0011] In a fourth aspect of the wavelength dispersion measuring device of the present invention, in addition to the first to third aspects, the control unit controls the generation of an optical pulse train by the optical pulse train generation unit so that the spectral shapes of the first optical pulse and the second optical pulse are Gaussian, Sech-type, or hyper-Gaussian.

[0012] In a fifth aspect of the wavelength dispersion measuring device of the present invention, in addition to the first to fourth aspects, the optical amplifier is an optical fiber amplifier or a semiconductor optical amplifier.

[0013] The chromatic dispersion measurement method of the present invention involves propagating a train of optical pulses, including a first optical pulse and a second optical pulse having different peak wavelengths, through a sample, and measuring the chromatic dispersion of the sample based on the shape of the autocorrelation of the time waveform of the propagated optical pulse train or the shape of the crosscorrelation between it and the time waveform of a reference optical pulse.

[0014] A first aspect of the wavelength dispersion measurement method of the present invention comprises: (1) an optical pulse train generation step in which an optical pulse train generation unit generates and outputs an optical pulse train including a first optical pulse having a spectral peak at a first wavelength within the band and a second optical pulse having a spectral peak at a second wavelength within the band, based on an optical pulse having a band output from a light source; (2) an optical amplification step in which the optical pulse train output from the optical pulse train generation unit is amplified and output by an optical amplifier; (3) a spectroscopy step in which the spectrum of the optical pulse train output from the optical amplifier is measured by a spectrometer; and (4) a control step in which the generation of an optical pulse train by the optical pulse train generation unit is controlled such that, among the spectra measured by the spectrometer, the spectral shape of the first optical pulse is symmetrical with respect to the first wavelength, and the spectral shape of the second optical pulse is symmetrical with respect to the second wavelength.

[0015] A second aspect of the wavelength dispersion measurement method of the present invention further comprises, in addition to the first aspect, (1) a measurement step of measuring the time interval between the peaks of the first and second optical pulses based on the shape of the autocorrelation of the time waveform of the optical pulse train output from the optical amplifier or the shape of the crosscorrelation between the time waveform of the optical pulse train output from the optical amplifier or the time waveform of a reference optical pulse, with the sample positioned such that the optical pulse train output from the optical pulse train generator propagates through the sample and is then input to the optical amplifier; and (2) a calculation step of determining the wavelength dispersion of the sample based on the time interval measured in the measurement step.

[0016] In the third aspect of the wavelength dispersion measurement method of the present invention, in addition to the first aspect or the second aspect, the control step controls the generation of the optical pulse train by the optical pulse train generation unit so that the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier is within the range of 1:10 to 10:1.

[0017] In the fourth aspect of the wavelength dispersion measurement method of the present invention, in addition to the first to third aspects, the control step controls the generation of the optical pulse train by the optical pulse train generation unit so that the spectral shape of each of the first optical pulse and the second optical pulse is a Gaussian shape, a Sech-type shape, or a hyper-Gaussian shape.

[0018] In the fifth aspect of the wavelength dispersion measurement method of the present invention, in addition to the first to fourth aspects, in the optical amplification step, an optical fiber amplifier or a semiconductor optical amplifier is used as the optical amplifier.

Effect of the Invention

[0019] According to the present invention, even for a sample in which a non-linear optical phenomenon is likely to occur, the wavelength dispersion can be accurately measured.

Brief Description of the Drawings

[0020] [Figure 1] FIG. 1 is a diagram showing the configuration of the wavelength dispersion measurement device 1A. [Figure 2] FIG. 2 is a diagram showing a configuration example of the optical pulse train generation unit 20. [Figure 3] FIG. 3 is a diagram showing a configuration example of the measurement unit 40. [Figure 4] FIG. 4 is a graph showing an example of the relationship between the input optical intensity and the optical transmission efficiency in an optical waveguide made of InP. [Figure 5] FIG. 5 is a diagram showing the configuration of the wavelength dispersion measurement device 1B. [Figure 6] FIG. 6 is a graph showing the result of wavelength dispersion measurement performed using the wavelength dispersion measurement device 1B. [Figure 7] Figure 7 is a graph showing an example of the wavelength dependence of the gain of an optical amplifier. [Figure 8] Figure 8 is a graph showing the results of a simulation of the effect of changes in spectral shape on wavelength dispersion measurement values ​​when performing wavelength dispersion measurements using the wavelength dispersion measurement device 1B. [Figure 9] Figure 9 shows the configuration of the wavelength dispersion measurement device 1C. [Figure 10] Figure 10 is a flowchart showing an example of how the control unit 70 of the wavelength dispersion measurement device 1C sets the amplitude modulation pattern. [Figure 11] Figure 11 is a schematic graph showing the target spectrum determined in step S2 and the Si spectrum measured in step S5. [Figure 12] Figure 12 is a graph showing the results of a simulation conducted on the relationship between RMS error and wavelength dispersion measurement error. [Figure 13] Figure 13 shows the spectrum of the optical pulse train output from the optical pulse train generation unit 20 (Figure 13(a)) and the spectrum of the optical pulse train output from the optical amplifier 30 (Figure 13(b)) during wavelength dispersion measurement performed using the wavelength dispersion measurement device 1C. [Figure 14] Figure 14 shows the autocorrelation shape of the time waveform of the optical pulse train measured by the measurement unit 40 when no sample S is placed during wavelength dispersion measurement using the wavelength dispersion measurement device 1C (Figure 14(a)), and the autocorrelation shape of the time waveform of the optical pulse train measured by the measurement unit 40 when sample S is placed (Figure 14(b)). [Figure 15] Figure 15 is a graph showing the results of wavelength dispersion measurements performed using the wavelength dispersion measurement device 1C. [Figure 16] Figure 16 illustrates the intensity ratio of the peaks of the first and second optical pulses in the time waveform of the optical pulse train output from the optical amplifier 30. [Figure 17] Figure 17 shows the relationship between the wavelength dependence of the gain of the optical amplifier 30 and the spectra of the first and second optical pulses, respectively. [Figure 18] Figure 18 shows another example of the configuration of the measurement unit 40. [Modes for carrying out the invention]

[0021] Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the attached drawings. In the description of the drawings, the same elements will be denoted by the same reference numerals, and redundant descriptions will be omitted. The present invention is not limited to these examples, but is indicated by the claims, and all modifications within the meaning and scope equivalent to the claims are intended to be included.

[0022] First, the configurations of the comparative examples 1A and 1B of wavelength dispersion measurement devices will be described, and then the configuration of the embodiment 1C of wavelength dispersion measurement device will be described.

[0023] Figure 1 shows the configuration of the wavelength dispersion measurement device 1A. The wavelength dispersion measurement device 1A shown in this figure comprises a light source 10, an optical pulse train generation unit 20, a measurement unit 40, and a calculation unit 50. The wavelength dispersion measurement device 1A measures the wavelength dispersion of a sample S placed on the optical path between the optical pulse train generation unit 20 and the measurement unit 40.

[0024] The light source 10 repeatedly outputs optical pulses with a defined bandwidth. The light source 10 may be, for example, a supercontinium (SC) light source, an ASE (Amplified Spontaneous Emission) light source, or the like.

[0025] The optical pulse train generation unit 20 generates an optical pulse train based on the optical pulses output from the light source 10, including a first optical pulse having a spectral peak at a first wavelength λ1 and a second optical pulse having a spectral peak at a second wavelength λ2, and outputs this optical pulse train. The first wavelength λ1 and the second wavelength λ2 are within the bandwidth of the optical pulses output from the light source 10 and are different wavelengths from each other. There may be a time difference between the first optical pulse and the second optical pulse when they are output from the optical pulse train generation unit 20. A specific example of the configuration of the optical pulse train generation unit 20 will be described later.

[0026] The measurement unit 40 receives the optical pulse train output from the optical pulse train generation unit 20 and after it has propagated through the sample S, and measures the time intervals between the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train. The measurement unit 40 can be configured to include an autocorrelation analyzer to determine the autocorrelation shape of the time waveform of the optical pulse train, and it can also be configured to include a cross-correlation analyzer to determine the cross-correlation shape between the time waveform of the optical pulse train and the time waveform of a reference optical pulse. Specific examples of the configuration of the measurement unit 40 will be described later.

[0027] The calculation unit 50 performs the necessary calculations based on the time interval measured by the measurement unit 40, the difference between the first wavelength λ1 and the second wavelength λ2, and the length of the optical propagation path in the sample S, to determine the chromatic dispersion of the sample S. The chromatic dispersion obtained here is the value at the intermediate wavelength ((λ1+λ2) / 2) between the first wavelength λ1 and the second wavelength λ2.

[0028] Figure 2 shows an example of the configuration of the optical pulse train generation unit 20. The optical pulse train generation unit 20 shown in this figure comprises a diffraction grating 21, a lens 22, a spatial light modulator 23, a lens 24, and a diffraction grating 25.

[0029] The diffraction grating 21 receives an optical pulse Pa output from the light source 10 and diffracts the optical pulse Pa at a diffraction angle corresponding to its wavelength, thereby spatially separating the optical pulse Pa for each wavelength and outputting each wavelength's optical pulse in a different direction. The lens 22 receives the optical pulses of each wavelength output from the diffraction grating 21 in different directions, collimates them, and outputs the collimated light P1 to the spatial light modulator 23.

[0030] The spatial light modulator 23 has a modulation plane that can modulate the amplitude of light at each pixel position. The amplitude modulation pattern on the modulation plane can be set by an external electrical signal. The spatial light modulator 23 inputs the light P1, which has been collimated by the lens 22, to the modulation plane, amplitude modulates it according to the wavelength of the light, and outputs the modulated light P2. The spatial light modulator 23 may also be capable of phase modulation in addition to amplitude modulation of light at each pixel position.

[0031] The spatial light modulator 23 can generate a first optical pulse Pb1 and a second optical pulse Pb2 by setting the amplitude modulation amount according to the wavelength. Furthermore, the spatial light modulator 23 can set a time difference between the first optical pulse Pb1 and the second optical pulse Pb2 by making the phase modulation amounts in the bands centered on wavelengths λ1 and λ2, respectively, different from each other.

[0032] Lens 24 focuses the light P2 (first light pulse Pb1 and second light pulse Pb2) output from the spatial light modulator 23 at a common position on the diffraction grating 25. The diffraction grating 25 diffracts the first light pulse Pb1 and the second light pulse Pb2, combines them, and outputs them as a light pulse train Pb on the same optical path.

[0033] In the configuration shown in Figure 2, a diffraction grating 21 is used as a separation unit to spatially separate the light pulses output from the light source 10 by wavelength, but a prism may be used instead. In the configuration shown in Figure 2, a diffraction grating 25 is used as a multiplexing unit to combine the first light pulse Pb1 and the second light pulse Pb2 output from the spatial light modulator 23 and output them as a light pulse train Pb on the same optical path, but a prism may be used instead. Concave mirrors may be used instead of lenses 22 and 24. Also, in the configuration shown in Figure 2, the spatial light modulator 23 was a transmissive type, but a reflective type spatial light modulator may be used instead, in which case lenses 22 and 24 can be made common, and diffraction gratings 21 and 25 can be made common.

[0034] Figure 3 shows an example of the configuration of the measurement unit 40. The measurement unit 40 shown in this figure has a configuration that includes an autocorrelator and comprises a lens 41, a nonlinear optical element 42, a lens 43, a beam splitter 44, mirrors 45a to 45d, mirrors 46a to 46c, a stage 47, and a photodetector 48.

[0035] The beam splitter 44 receives the optical pulse train Pb output from the optical pulse train generation unit 20 and propagating through the sample S. This optical pulse train Pb includes a first optical pulse Pb1 and a second optical pulse Pb2. The beam splitter 44 splits the input optical pulse train Pb into two optical pulse trains, Pba and Pbb. The beam splitter 44 outputs one optical pulse train Pba to the mirror 45a and the other optical pulse train Pbb to the mirror 46a. The splitting ratio of the beam splitter 44 may be 1:1.

[0036] One optical pulse train Pba is sequentially reflected by mirrors 45a to 45d and input to lens 41. The other optical pulse train Pbb is sequentially reflected by mirrors 46a to 46c and input to lens 41. Mirrors 46a and 46b are mounted on stage 47. Stage 47 is movable in the direction of the double arrows in the figure. Moving stage 47 sets the optical path length difference between the optical path of optical pulse train Pba from beam splitter 44 through mirrors 45a to 45d to lens 41 and the optical path of optical pulse train Pbb from beam splitter 44 through mirrors 46a to 46c to lens 41. In other words, moving stage 47 makes it possible to set the time difference between the timing when optical pulse train Pba reaches nonlinear optical element 42 and the timing when optical pulse train Pbb reaches nonlinear optical element 42 to any value.

[0037] Lens 41 focuses the optical pulse trains Pba and Pbb at a common position on the nonlinear optical element 42. The optical pulse trains Pba and Pbb are incident on the nonlinear optical element 42 from different directions. As a result, a second harmonic is generated in the nonlinear optical element 42. The intensity of this second harmonic depends on the magnitude of the correlation between the time waveform of the optical pulse train Pba and the time waveform of the optical pulse train Pbb. The second harmonic (correlated light Pc) output from the nonlinear optical element 42 is received by the photodetector 48 after passing through lens 43.

[0038] As the nonlinear optical element 42, for example, KTP (KTiOPO4) crystal, LBO (LiB3O5) crystal, BBO (β-BaB2O4) crystal, etc. may be used. Alternatively, instead of the nonlinear optical element 42, a phosphor that emits fluorescence may be used as the correlated light Pc. Examples of phosphors that can be used include coumarin, stilbene, rhodamine, etc.

[0039] The photodetector 48 receives correlated light Pc for each light pulse train Pb and outputs an electrical signal with a value corresponding to the intensity of the correlated light Pc. When the difference in the time when light pulse trains Pba and Pbb reach lens 41 is set to a specific value by moving the stage 47, the value of the electrical signal output from the photodetector 48 that received the correlated light Pc is determined. The relationship between the time difference, which is set to a different value for each light pulse train, and the output electrical signal value from the photodetector 48 represents the autocorrelation of the time waveform of the light pulse train Pb. Based on this autocorrelation shape of the time waveform of the light pulse train Pb, the time intervals between the peaks of the first light pulse Pb1 and the second light pulse Pb2 contained in the light pulse train Pb can be determined.

[0040] Figure 18 shows another example of the configuration of the measurement unit 40. The measurement unit 40A shown in this figure has a configuration that includes a cross-correlator and comprises a lens 41, a nonlinear optical element 42, a lens 43, mirrors 49a to 49d, a stage 47, and a photodetector 48.

[0041] The optical pulse train Pb, output from the optical pulse train generation unit 20 and propagating through the sample S, is input to the lens 41. This optical pulse train Pb includes a first optical pulse Pb1 and a second optical pulse Pb2. The reference optical pulse Pr is sequentially reflected by mirrors 49a to 49d and input to the lens 41. The reference optical pulse Pr is a single pulse and may be an optical pulse branched by a beam splitter provided between the light source 10 and the optical pulse train generation unit 20. Mirrors 49b and 49c are mounted on the stage 47. The stage 47 is movable in the direction of the double arrows in the figure. By moving the stage 47, the time difference between the timing when the optical pulse train Pb reaches the nonlinear optical element 42 and the timing when the reference optical pulse Pr reaches the nonlinear optical element 42 can be set to any desired value.

[0042] Lens 41 focuses the optical pulse train Pb and the reference optical pulse Pr at a common position on the nonlinear optical element 42. The optical pulse train Pb and the reference optical pulse Pr are incident on the nonlinear optical element 42 from different directions. As a result, a second harmonic is generated in the nonlinear optical element 42. The intensity of this second harmonic depends on the magnitude of the cross-correlation between the time waveform of the optical pulse train Pb and the time waveform of the reference optical pulse Pr. The second harmonic (correlated light Pc) output from the nonlinear optical element 42 is received by the photodetector 48 after passing through lens 43.

[0043] The photodetector 48 receives the correlated light Pc for each light pulse train Pb and outputs an electrical signal with a value corresponding to the intensity of the correlated light Pc. When the difference in the time when the light pulse train Pb and the reference light pulse Pr reach the lens 41 is set to a specific value by moving the stage 47, the value of the electrical signal output from the photodetector 48 that received the correlated light Pc is determined. The relationship between the time difference, which is set to a different value for each light pulse train, and the output electrical signal value from the photodetector 48 represents the cross-correlation between the time waveform of the light pulse train Pb and the time waveform of the reference light pulse Pr. Based on this cross-correlation shape, the time intervals between the peaks of the first light pulse Pb1 and the second light pulse Pb2 contained in the light pulse train Pb can be determined.

[0044] The inventors attempted to measure the wavelength dispersion of various samples using such a wavelength dispersion measuring device 1A (Figure 1), and found that in some cases, the wavelength dispersion could not be accurately obtained for certain samples. For example, when an optical waveguide made of SiN was used as sample S, the wavelength dispersion of the optical waveguide could be obtained relatively accurately, whereas when an optical waveguide made of Si or InP was used as sample S, the wavelength dispersion of the optical waveguide could not be accurately obtained. This is thought to be due to the following reasons.

[0045] In other words, in the case of an optical waveguide made of Si or InP, a nonlinear optical phenomenon (e.g., two-photon absorption) occurs when an optical pulse train propagates through the waveguide, causing a significant decrease in the intensity of the optical pulse train output after it has propagated through the waveguide. If the intensity of the optical pulse train input to the measurement unit 40 is low, the measurement of the autocorrelation shape of the time waveform of the optical pulse train becomes inaccurate, the measurement of the time interval between the peaks of the first optical pulse Pb1 and the second optical pulse Pb2 also becomes inaccurate, and the measurement of wavelength dispersion also becomes inaccurate.

[0046] Figure 4 is a graph showing an example of the relationship between input light intensity and light transmission efficiency in an optical waveguide made of InP. Light transmission efficiency is calculated by dividing the output light intensity from the optical waveguide by the input light intensity to the waveguide. A light pulse train with a center wavelength of 1540 nm was input to the optical waveguide. As shown in this graph, the light transmission efficiency decreases as the input light intensity increases. This is thought to be due to the effects of nonlinear optical phenomena (particularly two-photon absorption).

[0047] In order to measure the autocorrelation shape of the time waveform of the optical pulse train in the measurement unit 40, the average output intensity from the sample must be, for example, 500 μW or more. However, in this example, undesirable nonlinear optical phenomena occur, and the average output intensity is only 40 μW for an average input intensity of 1 mW. If the average input intensity is reduced to, for example, 0.1 mW, the nonlinear effect is reduced and the light transmission efficiency improves, but the average output intensity is only about 10 μW, and in this case as well, it is difficult to detect the autocorrelation shape.

[0048] In order to suppress the occurrence of nonlinear optical phenomena in sample S, the inventors conceived of a configuration in which an optical amplifier 30 is placed in the optical path between sample S and measurement unit 40, as shown in Figure 5, in order to sufficiently reduce the intensity of the optical pulse train input to sample S and to accurately measure the autocorrelation shape of the time waveform of the optical pulse train in measurement unit 40.

[0049] Figure 5 shows the configuration of the wavelength dispersion measurement device 1B. The wavelength dispersion measurement device 1B shown in this figure differs from the configuration of the wavelength dispersion measurement device 1A (Figure 1) in that it further includes an optical amplifier 30. The optical amplifier 30 receives the optical pulse train output from the optical pulse train generation unit 20 and propagating through the sample S, amplifies this optical pulse train, and outputs it to the measurement unit 40. Examples of optical amplifiers 30 include optical fiber amplifiers (OFA) and semiconductor optical amplifiers (SOA). Examples of OFAs include rare earth element doped optical fiber amplifiers such as erbium-doped fiber amplifiers (EDFA) and fiber Raman amplifiers (FRA).

[0050] The inventors used such a wavelength dispersion measuring device 1B (Figure 5) to attempt to measure the wavelength dispersion of various samples by sufficiently reducing the intensity of the light pulses output from the light source 10 so that nonlinear optical phenomena would not occur in the sample S, and by setting the gain of the optical amplifier 30 to an appropriate value so that the intensity of the light pulse train input to the measurement unit 40 was sufficiently increased. However, when the optical waveguide composed of Si or InP was used as the sample S, the wavelength dispersion of the optical waveguide could not be accurately obtained even when using the wavelength dispersion measuring device 1B (Figure 5). Moreover, when the optical waveguide composed of SiN was used as the sample S, although the wavelength dispersion of the optical waveguide could be obtained relatively accurately when using the wavelength dispersion measuring device 1A (Figure 1), the wavelength dispersion of the optical waveguide could not be accurately obtained when using the wavelength dispersion measuring device 1B (Figure 5).

[0051] Figure 6 is a graph showing the results of wavelength dispersion measurements performed using wavelength dispersion measurement device 1B. Here, a polarization-maintaining optical fiber from Thorlabs was used as sample S. The optical amplifier 30 used was Thorlabs' erbium-doped optical fiber amplifier, EDFA100P. This graph shows not only the measured data when using EDFA100P, but also data provided by Thorlabs and data described in the paper. As shown in this graph, the measured data when using EDFA100P differed significantly from the data provided by Thorlabs and data described in the paper.

[0052] This phenomenon is thought to be caused by the following: Generally, optical amplifiers have gain in a predetermined wavelength band, and within that wavelength band, the gain is wavelength-dependent. Figure 7 is a graph showing an example of the wavelength dependence of the gain of an optical amplifier. This graph is published in Non-Patent Literature 1, which is the datasheet for Thorlabs' EDFA100P. As shown in this graph, the gain of an optical amplifier is wavelength-dependent.

[0053] Generally, optical amplifiers are configured by combining an optical amplification medium with gain and a gain-flattening filter to flatten the wavelength dependence of the gain overall. However, even with such a configuration, optical amplifiers cannot completely flatten the wavelength dependence of the gain. Furthermore, the wavelength dependence of the gain differs depending on the input light intensity.

[0054] Therefore, even if the spectra of the first and second optical pulses output from the optical pulse train generation unit 20 and input to the sample S have a desirable shape, the spectra of the first and second optical pulses output from the optical amplifier 30 and input to the measurement unit 40 will differ from the desirable shape due to the wavelength dependence of the gain of the optical amplifier 30. It is thought that the change in the spectra of the first and second optical pulses output from the optical amplifier 30 to an undesirable shape is the cause of the inaccurate measurement of the autocorrelation shape of the time waveform of the optical pulse train in the measurement unit 40. The inventors conducted a simulation to confirm this.

[0055] Figure 8 is a graph showing the results of a simulation of the effect of changes in spectral shape on the wavelength dispersion measurement value when performing wavelength dispersion measurement using the wavelength dispersion measurement device 1B. Here, a sample S with a wavelength dispersion value of 17.5 ps / nm / km was assumed. It was assumed that in the optical pulse train input to the measurement unit 40, the spectrum of one optical pulse maintains symmetry around the peak wavelength, while the spectrum of the other optical pulse does not maintain symmetry around the peak wavelength and becomes asymmetrical. The symmetry of the optical pulse spectrum was defined as the ratio (B / A) of the spectral area A on the shorter wavelength side of the peak wavelength to the spectral area B on the longer wavelength side of the peak wavelength. When the spectrum of the optical pulse is symmetrical around the peak wavelength, the value of the ratio (B / A) is 1. The horizontal axis of this graph represents the ratio (B / A) representing the spectral symmetry, and the vertical axis represents the simulated value of wavelength dispersion. As shown in this graph, in order to accurately determine the wavelength dispersion, it is necessary that the spectral symmetry of each optical pulse in the optical pulse train input to the measurement unit 40 is good.

[0056] Based on the above findings, the inventors conceived the configuration of the wavelength dispersion measurement device 1C shown in Figure 9. Figure 9 is a diagram showing the configuration of the wavelength dispersion measurement device 1C. The wavelength dispersion measurement device 1C shown in this figure differs from the configuration of the wavelength dispersion measurement device 1B (Figure 5) in that it further includes a spectrometer 60 and a control unit 70.

[0057] The spectrometer 60 measures the spectrum of the light pulse train output from the optical amplifier 30. A beam splitter may be provided in the optical path between the optical amplifier 30 and the measurement unit 40. In this case, the beam splitter splits the light pulse train output from the optical amplifier 30 into two, outputting one branch to the measurement unit 40 and the other branch to the spectrometer 60. Alternatively, the spectrometer 60 may be placed in place of the measurement unit 40 without providing such a beam splitter.

[0058] The control unit 70 controls the generation of optical pulse trains by the optical pulse train generation unit 20 based on the spectrum measured by the spectrometer 60. Specifically, the control unit 70 controls the generation of the first optical pulse Pb1 and the second optical pulse Pb2 by the spatial light modulator 23 of the optical pulse train generation unit 20 so that the spectral shape of the first optical pulse is symmetrical around the first wavelength λ1, and the spectral shape of the second optical pulse is symmetrical around the second wavelength λ2.

[0059] In this wavelength dispersion measurement device 1C, the generation of optical pulse trains by the optical pulse train generation unit 20 is adjusted by feedback control using the spectrometer 60 and the control unit 70. That is, the control unit 70 sets the amplitude modulation pattern on the modulation plane of the spatial light modulator 23 so that the spectral shape of the first optical pulse is symmetrical around the first wavelength λ1, and the spectral shape of the second optical pulse is symmetrical around the second wavelength λ2, among the spectra measured by the spectrometer 60. Since the wavelength dependence of the gain of the optical amplifier 30 differs depending on the input light intensity, it is preferable to repeat the spectral measurement by the spectrometer 60 and the setting of the amplitude modulation pattern by the control unit 70. At this time, the sample S may or may not be placed on the optical path between the optical pulse train generation unit 20 and the optical amplifier 30. After the amplitude modulation pattern on the modulation plane of the spatial light modulator 23 is appropriately set, the wavelength dispersion of the sample S is measured using the measurement unit 40 and the calculation unit 50 with the sample S placed on the optical path between the optical pulse train generation unit 20 and the optical amplifier 30.

[0060] Figure 10 is a flowchart showing an example of how the control unit 70 of the wavelength dispersion measurement device 1C sets the amplitude modulation pattern.

[0061] In step S1, the spectrum L of the light pulse output from the light source 10 is measured. In step S2, the target spectrum T of the first and second light pulses output from the optical amplifier 30 is determined. The target spectrum T has a symmetrical shape around the peak wavelength and may be, for example, a Gaussian shape, a Sech shape, or a hyper-Gaussian shape.

[0062] In step S3, the amplitude modulation pattern M0 to be initially set on the modulation plane of the spatial light modulator 23 is designed using the spectrum L measured in step S1 and the target spectrum T determined in step S2, according to equation (1) below. In step S4, the designed amplitude modulation pattern M i The modulation plane of the spatial light modulator 23 is set. Initially, the amplitude modulation pattern M0 designed in step S3 is set as the modulation plane of the spatial light modulator 23.

[0063]

number

[0064] In step S5, the spectra S of the first and second optical pulses output from the optical amplifier 30 are analyzed. i The spectrum is measured using the spectrometer 60. The first spectrum measured, S0, is the one obtained when the amplitude modulation pattern M0 designed in step S3 is set on the modulation plane of the spatial light modulator 23.

[0065] In step S6, the spectrum S measured in step S5 is used. i If the result is sufficiently close to the target spectrum T, the process ends; otherwise, proceed to step S7. In step S7, the amplitude modulation pattern M i Based on this, the amplitude modulation pattern M is obtained by equation (2) below. i+1 A new design is created. Then, the process returns to step S4 and continues from step S4 onward. Steps S4 to S7 are repeated until the process is determined to have ended in step S6.

[0066]

number

[0067] The termination determination in step S6 can be performed, for example, as follows. Figure 11 shows the target spectrum determined in step S2 and the spectrum S measured in step S5. i This graph schematically illustrates the difference between the two spectra. The magnitude of the difference between the two spectra can be evaluated using the RMS error (RMS_error) in equation (3) below. The process can be considered complete when this RMS error in equation (3) falls below a predetermined value. In this equation, ω is the angular frequency, p1 is the peak intensity of the target spectrum, and p2 is the measured spectrum S i This is the peak intensity, and N is the number of data points. The sum on the right side is calculated in the range of Sp-2σ to Sp+2σ, where Sp is the peak wavelength of the target spectrum and σ is the standard deviation.

[0068]

number

[0069] Figure 12 is a graph showing the results of a simulation on the relationship between RMS error and wavelength dispersion measurement error. The simulation conditions were as follows: A SiN optical waveguide was assumed as sample S, with a length of 5 mm for the optical propagation path in the waveguide and a wavelength dispersion of 2000 ps / nm / km. The shape of the target spectrum T was set to Gaussian, and the spectral width to 5 nm. The wavelength midway between the first and second wavelengths was set to 1550 nm, and the difference between the first and second wavelengths was set to 20 nm. The time difference between the first and second optical pulses was set to 5 ps. As shown in this graph, the wavelength dispersion measurement error is approximately proportional to the RMS error. For example, if a measurement accuracy of 5% or less for the wavelength dispersion measurement error is required, the spectral shape of the optical pulse after optical amplification can be improved by feedback until the RMS error is 0.2 or less.

[0070] The wavelength dispersion measurement method using the wavelength dispersion measurement device 1C is as follows. The wavelength dispersion measurement method is broadly divided into an adjustment step to adjust the generation of the optical pulse train by the optical pulse train generation unit 20, and a measurement step to measure the wavelength dispersion of the sample S. In the adjustment step, the sample S may be placed in the optical path between the optical pulse train generation unit 20 and the optical amplifier 30, or the sample S may not be placed.

[0071] In the adjustment step, the optical pulse train generation unit 20 generates and outputs an optical pulse train including a first optical pulse having a spectral peak at a first wavelength and a second optical pulse having a spectral peak at a second wavelength, based on the optical pulse with a bandwidth output from the light source 10 (optical pulse train generation step). The optical pulse train output from the optical pulse train generation unit 20 is amplified and output by the optical amplifier 30 (optical amplification step). The spectrum of the optical pulse train output from the optical amplifier 30 is measured by the spectrometer 60 (spectroscopy step). Then, the generation of the optical pulse train by the optical pulse train generation unit 20 is controlled so that the spectral shape of the first optical pulse is symmetrical around the first wavelength, and the spectral shape of the second optical pulse is symmetrical around the second wavelength, based on the spectrum measured by the spectrometer 60 (control step).

[0072] In the measurement step, the optical pulse train generation unit 20 generates and outputs an optical pulse train including a first optical pulse having a spectral peak at a first wavelength and a second optical pulse having a spectral peak at a second wavelength, based on the optical pulse with a bandwidth output from the light source 10 (optical pulse train generation step). The optical pulse train output from the optical pulse train generation unit 20 and propagated through the sample S is amplified and output by the optical amplifier 30 (optical amplification step). The measurement unit 40 measures the time interval between the peaks of the first optical pulse and the second optical pulse based on the autocorrelation shape of the time waveform of the optical pulse train output from the optical amplifier 30 (measurement step). Then, the chromatic dispersion of the sample S is determined based on the time interval measured by the measurement unit 40 (calculation step).

[0073] Figures 13 to 15 show the results of wavelength dispersion measurements performed using the wavelength dispersion measurement device 1C. Here, the center wavelength of the optical pulse train output from the optical pulse train generation unit 20 was set to 1541 nm. The intensity of the optical pulse train output from the optical pulse train generation unit 20 was 5 μW. A 5 m long polarization-maintaining optical fiber from Thorlabs was used as sample S. A Thorlabs EDFA100P was used as the optical amplifier 30. The intensity of the optical pulse train output from the optical amplifier 30 was 4.1 mW.

[0074] Figure 13(a) shows the spectrum of the optical pulse train output from the optical pulse train generation unit 20. The peak wavelength difference between the two optical pulses output from the optical pulse train generation unit 20 was 18.0 nm. Figure 13(b) shows the spectrum of the optical pulse train output from the optical amplifier 30. The peak wavelength difference between the two optical pulses output from the optical amplifier 30 was 17.7 nm.

[0075] Figure 14(a) shows the autocorrelation shape of the time waveform of the optical pulse train measured by the measurement unit 40 when no sample S is placed. Figure 14(b) shows the autocorrelation shape of the time waveform of the optical pulse train measured by the measurement unit 40 when sample S is placed. In these figures, the horizontal axis is the time difference between optical pulse trains Pba and Pbb set by the movement of the stage 47. The vertical axis is the output signal value from the photodetector 48.

[0076] When sample S was not placed (i.e., when the length of the polarization-maintaining optical fiber as sample S was 0), the time difference between the first and second optical pulses output from the optical amplifier 30 was 3.89 ps. In contrast, when a polarization-maintaining optical fiber with a length of 5 m was placed as sample S, the time difference between the first and second optical pulses output from the optical amplifier 30 was 5.47 ps. In other words, the placement of the 5 m polarization-maintaining optical fiber changed the time difference between the first and second optical pulses output from the optical amplifier 30 by only 1.58 ps. From these values, a measured chromatic dispersion of the polarization-maintaining optical fiber at a wavelength of 1541 nm was obtained to be approximately 17.5 ps / nm / km.

[0077] Figure 15 is a graph showing the results of wavelength dispersion measurements performed using the wavelength dispersion measurement device 1C. This graph shows not only the measured data when using EDFA100P as the optical amplifier 30, but also the measured data when not using the optical amplifier 30, the data provided by Thorlabs, and the data published in the paper. As shown in this graph, the measured data of wavelength dispersion measurements performed using EDFA100P as the optical amplifier 30 in the wavelength dispersion measurement device 1C agrees well with the data provided by Thorlabs and the data published in the paper, and also agrees well with the measured data when not using the optical amplifier 30 in the wavelength dispersion measurement device 1C.

[0078] As described above, in this embodiment, in order to measure the wavelength dispersion of sample S, which is prone to exhibiting nonlinear optical phenomena, the intensity of the light pulse train input to sample S is sufficiently reduced, and an optical amplifier 30 is provided to increase the intensity of the light pulse train input to the measurement unit 40. As a result, even if the shape of the light pulse changes due to the wavelength dependence of the gain of the optical amplifier 30, the control unit 70 controls the generation of the light pulse train by the spatial light modulator 23 so that the spectral shape of the first light pulse in the spectrum measured by the spectrometer 60 is symmetrical around the first wavelength λ1, and the spectral shape of the second light pulse is symmetrical around the second wavelength λ2. This makes it possible to accurately measure the wavelength dispersion of sample S.

[0079] Furthermore, regardless of whether or not nonlinear optical phenomena are likely to occur in sample S, it becomes possible to use a broadband light source 10 with low output intensity, thus enabling the measurement of chromatic dispersion over a wider wavelength range.

[0080] When the control unit 70 controls the generation of an optical pulse train by the spatial light modulator 23, it is preferable that the spectral shape of the first optical pulse in the spectrum measured by the spectrometer 60 be symmetrical around the first wavelength λ1, and the spectral shape of the second optical pulse be symmetrical around the second wavelength λ2, and that the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier 30 be within the range of 1:10 to 10:1. Figure 16 is a diagram illustrating the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier 30. Figure 16(a) shows the time waveform of the optical pulse train when the peak intensity ratio is 1:1. Figure 16(b) shows the time waveform of the optical pulse train when the peak intensity ratio is 1:10. By having the peak intensity ratio of the optical pulse train output from the optical amplifier 30 and input to the measurement unit 40 be within this range, the measurement unit 40 can more accurately measure the autocorrelation shape of the time waveform of the optical pulse train. It is more preferable to have a peak intensity ratio within the range of 1:5 to 5:1.

[0081] Furthermore, when the control unit 70 controls the generation of an optical pulse train by the spatial light modulator 23, it is preferable to set the first and second wavelengths in the gain band of the optical amplifier 30 in a band where the wavelength dependence of the gain is relatively small. Figure 17 shows the relationship between the wavelength dependence of the gain of the optical amplifier 30 and the spectra of the first and second optical pulses, respectively. In the example shown in Figure 17(a), the wavelength dependence of the gain of the optical amplifier 30 is relatively small around wavelengths of 1532 nm and 1550 nm, respectively. Therefore, as shown in Figure 17(b), it is preferable to set the peak wavelength of the spectrum of the first optical pulse (first wavelength) around wavelength 1532 nm, and the peak wavelength of the spectrum of the second optical pulse (second wavelength) around wavelength 1550 nm. [Explanation of Symbols]

[0082] 1A~1C...Wavelength dispersion measuring device, 10...Light source, 20...Optical pulse train generation unit, 21...Diffraction grating, 22...Lens, 23...Spatial light modulator, 24...Lens, 25...Diffraction grating, 30...Optical amplifier, 40,40A...Measurement unit, 41...Lens, 42...Nonlinear optical element, 43...Lens, 44...Beam splitter, 45a~45d...Mirror, 46a~46c...Mirror, 47...Stage, 48...Photodetector, 49a~49d...Mirror, 50...Calculation unit, 60...Spectrometer, 70...Control unit.< / url:>

Claims

1. An apparatus for measuring the chromatic dispersion of a sample by propagating a train of optical pulses, including a first optical pulse and a second optical pulse having different peak wavelengths, to a sample, based on the shape of the autocorrelation of the time waveform of the propagated optical pulse train or the shape of the crosscorrelation between it and the time waveform of a reference optical pulse, A light source that outputs optical pulses with a bandwidth, An optical pulse train generation unit generates and outputs an optical pulse train including a first optical pulse having a spectral peak at a first wavelength within the band and a second optical pulse having a spectral peak at a second wavelength within the band, based on the optical pulses output from the light source. An optical amplifier that amplifies and outputs the optical pulse train output from the optical pulse train generation unit, A spectrometer for measuring the spectrum of the light pulse train output from the optical amplifier, A control unit controls the generation of the optical pulse train by the optical pulse train generation unit such that, among the spectra measured by the spectrometer, the spectral shape of the first optical pulse is symmetrical around the first wavelength, and the spectral shape of the second optical pulse is symmetrical around the second wavelength. A wavelength dispersion measuring device equipped with the following features.

2. With the sample arranged such that the optical pulse train output from the optical pulse train generation unit propagates through the sample and is then input to the optical amplifier, a measurement unit measures the time interval between the peaks of the first optical pulse and the second optical pulse based on the shape of the autocorrelation of the time waveform of the optical pulse train output from the optical amplifier or the shape of the crosscorrelation between it and the time waveform of a reference optical pulse, A calculation unit that determines the wavelength dispersion of the sample based on the time interval measured by the measurement unit, The wavelength dispersion measuring device according to claim 1, further comprising the following:

3. The control unit controls the generation of the optical pulse train by the optical pulse train generation unit so that the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier is within the range of 1:10 to 10:

1. The wavelength dispersion measuring apparatus according to claim 1.

4. The control unit controls the generation of the optical pulse train by the optical pulse train generation unit so that the spectral shapes of the first optical pulse and the second optical pulse are Gaussian, Sech, or hypergaussian. The wavelength dispersion measuring apparatus according to claim 1.

5. The optical amplifier is either an optical fiber amplifier or a semiconductor optical amplifier. The wavelength dispersion measuring apparatus according to claim 1.

6. A method for measuring the chromatic dispersion of a sample by propagating a train of optical pulses, which includes a first optical pulse and a second optical pulse having different peak wavelengths, through the sample, and based on the shape of the autocorrelation of the time waveform of the propagated optical pulse train or the shape of the cross-correlation between it and the time waveform of a reference optical pulse, A light pulse train generation step in which, based on a light pulse having a bandwidth output from a light source, a light pulse train generation unit generates and outputs a light pulse train including a first light pulse having a spectral peak at a first wavelength within the bandwidth and a second light pulse having a spectral peak at a second wavelength within the bandwidth, The optical amplification step involves amplifying the optical pulse train output from the optical pulse train generation unit using an optical amplifier and outputting it, A spectroscopic step of measuring the spectrum of the light pulse train output from the optical amplifier using a spectrometer, A control step of controlling the generation of the optical pulse train by the optical pulse train generation unit such that, among the spectra measured by the spectrometer, the spectral shape of the first optical pulse is symmetrical around the first wavelength, and the spectral shape of the second optical pulse is symmetrical around the second wavelength; A wavelength dispersion measurement method comprising the following features.

7. With the sample positioned such that the optical pulse train output from the optical pulse train generation unit propagates through the sample and is then input to the optical amplifier, a measurement step is performed to measure the time interval between the peaks of the first optical pulse and the second optical pulse based on the shape of the autocorrelation of the time waveform of the optical pulse train output from the optical amplifier or the shape of the crosscorrelation between it and the time waveform of a reference optical pulse. A calculation step to determine the wavelength dispersion of the sample based on the time interval measured by the measurement step, The wavelength dispersion measurement method according to claim 6, further comprising the above.

8. The control step controls the generation of the optical pulse train by the optical pulse train generation unit so that the intensity ratio of the peaks of the first optical pulse and the second optical pulse in the time waveform of the optical pulse train output from the optical amplifier is within the range of 1:10 to 10:

1. The wavelength dispersion measurement method according to claim 6.

9. The control step controls the generation of the optical pulse train by the optical pulse train generation unit so that the spectral shapes of the first optical pulse and the second optical pulse are Gaussian, Sech, or hypergaussian. The wavelength dispersion measurement method according to claim 6.

10. In the optical amplification step, an optical fiber amplifier or a semiconductor optical amplifier is used as the optical amplifier. The wavelength dispersion measurement method according to claim 6.

Images