WAVEGUIDE ESTIMATION APPARATUS, WAVEGUIDE ESTIMATION METHOD, AND PROGRAM
The waveguide estimation device enhances the accuracy of group delay difference and polarization-dependent loss estimation in optical signals by using time window processing, addressing the issue of low signal-to-noise ratios without additional hardware.
Patent Information
- Application Number
- JP2023522052
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-05-18
- Publication Date
- 2024-11-07
- Estimated Expiration
- 2041-05-18
AI Technical Summary
Existing digital coherent communication systems face challenges in accurately estimating group delay difference and polarization-dependent loss due to low signal-to-noise ratios, leading to reduced transmission capacity in optical signals.
A waveguide estimation device and method that employs time window processing to multiply adaptive filter tap coefficients by a time window function, enabling accurate estimation of group delay difference and polarization-dependent loss without requiring additional hardware, even at low signal-to-noise ratios.
Improves the estimation accuracy of group delay difference and polarization-dependent loss in optical signals, maintaining transmission quality without increasing computational load or power consumption.
Smart Images

Figure 0007578894000006 
Figure 0007578894000007 
Figure 0007578894000008
Abstract
Description
[Technical field]
[0001] The present invention relates to a waveguide estimation device, a waveguide estimation method, and a program. [Background technology]
[0002] With the development of information technology, the total amount of data traffic is increasing exponentially. In optical communications, optical fibers are used as waveguides in the transmission lines of optical signals and in optical devices. Optical fibers are used for fixed lines, which require a large transmission capacity, because of their low attenuation of optical signals and wide available bandwidth.
[0003] Factors that limit the transmission capacity of optical signals are polarization mode dispersion (PMD) and polarization dependent loss (PDL). Polarization mode dispersion is the group delay difference that occurs between the polarization modes of an optical signal. Polarization dependent loss is the difference between the maximum and minimum values of optical loss (insertion loss of an optical signal) that depends on the polarization mode. The main factors of polarization mode dispersion and polarization dependent loss are the distortion that occurs in the optical fiber during its manufacture and the birefringence that depends on the bending of the optical fiber after manufacture.
[0004] When an optical signal input to an optical fiber is output from the optical fiber, the polarization state of the output optical signal may fluctuate due to polarization mode dispersion. Here, when the polarization state of the optical signal input to the optical fiber is in a specific state, the fluctuation of the polarization state of the optical signal output from the optical fiber is suppressed.
[0005] Such a specific state is called the principal state of polarization (PSP). The amount of distortion caused by polarization mode dispersion in the waveform of an optical signal is expressed using the differential group delay (DGD) between the orthogonal polarization modes in the principal state of polarization.
[0006] The group delay difference and the polarization dependent loss cause distortion in the waveform of an optical signal, and the group delay difference and the polarization dependent loss are the causes of interference between symbols. Of these, the group delay difference fluctuates randomly at high speed in response to temperature changes and mechanical vibrations in the waveguide (optical fiber). For this reason, distortion due to the group delay difference has been considered the biggest factor limiting the transmission capacity of optical signals.
[0007] However, in recent years, the realization of a communication technology called digital coherent communication has made it possible to compensate for distortions caused by polarization mode dispersion (differential group delay) and polarization dependent loss. Digital coherent communication is a communication technology that combines digital signal processing and coherent communication. Coherent communication is a communication technology that derives the electric field information of an optical signal by using the interference waveform between a received optical signal and local light.
[0008] In digital coherent communication, the impulse response of the transmission path between the transmitter and the receiver and the waveguide in the optical device is estimated using an adaptive filter. Based on the estimated impulse response, it is possible to estimate the differential group delay and the polarization dependent loss (see Non-Patent Document 1).
[0009] By performing a convolution operation using the inverse of the impulse response (the inverse matrix of the transfer function) on the optical signal received by the receiver, it is possible to compensate for distortion due to polarization mode dispersion (differential group delay).In contrast, since polarization dependent loss is accompanied by energy loss, it is not possible to completely compensate for polarization dependent loss using digital signal processing.
[0010] When the peak of the impulse response is long in the time direction and the distortion of the optical signal waveform is large, it is necessary to increase the number of taps of the adaptive filter (the number of samples of the input signal used for the convolution operation in the time direction) in advance. However, increasing the number of taps increases the amount of calculation and power consumption of the digital signal processing. Also, when the distortion of the waveform becomes too large, it may not be possible to compensate for the distortion even if the number of taps of the adaptive filter is increased in advance. [Prior art documents] [Non-patent literature]
[0011] [Non-Patent Document 1] FN Hauske, M. Kuschnerov, B. Spinnler, and B. Lankl, "Optical Performance Monitoring in Digital Coherent Receivers," Journal of Lightwave Technology, vol.27, No.16, pp.3623-3631 (2009) Summary of the Invention [Problem to be solved by the invention]
[0012] The tap coefficients of the adaptive filter are multiplied by the received signal (digital signal) corresponding to the optical signal during the convolution operation. The Fourier transform result of the tap coefficients, "W(ω)", is expressed as the inverse matrix (2 × 2 matrix) of the waveguide transfer function, as shown in equation (1).
[0013]
number
[0014] Differential group delay "τ estim (ω)" is expressed as equation (2).
[0015]
number
[0016] Polarization dependent loss "H PDL (ω)" is expressed as equation (3).
[0017]
number
[0018] Here, since dedicated hardware for estimating the differential group delay and the polarization dependent loss is not required, the hardware of the digital coherent receiver does not need to be changed. In addition, it is possible to estimate the differential group delay and the polarization dependent loss in real time during the transmission of the optical signal without degrading the transmission quality of the optical signal.
[0019] However, due to the large number of relays of the optical signal and the high modulation speed, the signal-to-noise ratio of the optical signal may be less than a predetermined value. When the signal-to-noise ratio of the optical signal is less than a predetermined value, the accuracy of estimating the group delay difference and the polarization dependent loss decreases. Thus, when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value, there is a problem in that the accuracy of estimating the group delay difference and the polarization dependent loss cannot be improved.
[0020] In view of the above circumstances, an object of the present invention is to provide a waveguide estimation device, a waveguide estimation method, and a program that are capable of improving the estimation accuracy of at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of an optical signal transmitted through a waveguide is less than a predetermined value. [Means for solving the problem]
[0021] One aspect of the present invention is a waveguide estimation device including: a time window processing unit that multiplies a time series tap coefficient of an adaptive filter by a time window function; and an estimating unit that estimates at least one of a group delay difference of a frequency series or a time series and a polarization dependent loss of a frequency series or a time series based on the time series tap coefficient within the time window.
[0022] One aspect of the present invention is a waveguide estimation method executed by a waveguide estimation device, the waveguide estimation method including: a time window processing step of multiplying a time series tap coefficient of an adaptive filter by a time window function; and an estimation step of estimating at least one of a group delay difference of a frequency series or a time series and a polarization dependent loss of a frequency series or a time series based on the tap coefficient of the time series within the time window.
[0023] One aspect of the present invention is a program for causing a computer to function as the above-described waveguide estimation device. Effect of the Invention
[0024] According to the present invention, even if the signal-to-noise ratio of an optical signal transmitted through a waveguide is less than a predetermined value, it is possible to improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss. [Brief description of the drawings]
[0025] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a waveguide estimation device in a first embodiment. [Diagram 2] 5A to 5C are diagrams illustrating examples of time window processing, differential group delay estimation processing, and polarization dependent loss estimation processing in the first embodiment. [Diagram 3] 5 is a flowchart showing an example of the operation of the waveguide estimation device in the first embodiment. [Figure 4] FIG. 11 is a diagram illustrating an example of threshold processing in a modified example of the first embodiment. [Diagram 5] FIG. 11 is a diagram illustrating an example of the configuration of a waveguide estimation device in a second embodiment. [Figure 6] 13A and 13B are diagrams illustrating an example of time window processing and differential group delay estimation processing in the second embodiment. [Figure 7] 10 is a flowchart showing an example of the operation of the waveguide estimation device in the second embodiment. [Figure 8] FIG. 13 is a diagram illustrating an example of the configuration of a waveguide estimation device in a third embodiment. [Figure 9] FIG. 13 is a diagram illustrating a first example of time window processing and frequency window processing in the third embodiment. [Figure 10] 13 is a flowchart showing an example of the operation of the waveguide estimation device in the third embodiment. [Figure 11] FIG. 13 is a diagram illustrating a second example of time window processing and frequency window processing in the modified example of the third embodiment. [Figure 12] 13 is a flowchart showing an example of the operation of a waveguide estimation device in a modified example of the third embodiment. [Figure 13] FIG. 2 is a diagram illustrating an example of a hardware configuration of a waveguide estimation device in each embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0026] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail with reference to the drawings. (First embodiment) 1 is a diagram showing an example of the configuration of a waveguide estimation device 1a in the first embodiment. The waveguide estimation device 1a is a device that estimates (measures) at least one of a frequency-series or time-series differential group delay (DGD) and a frequency-series or time-series polarization-dependent loss (PDL) for an optical signal transmitted through an optical transmission line and a waveguide in an optical device.
[0027] For example, the waveguide estimation device 1a estimates at least one of a frequency-series or time-series group delay difference and a frequency-series or time-series polarization-dependent loss by offline analysis of a result of an optical signal transmission experiment in a laboratory. For example, the waveguide estimation device 1a may estimate the group delay difference and the polarization-dependent loss in real time while an installed digital coherent receiver and a digital coherent receiver are transmitting an optical signal.
[0028] The waveguide estimation device 1a performs time window processing on the time series tap coefficients of the adaptive filter (tap coefficients for each sample time). That is, the waveguide estimation device 1a multiplies the time series tap coefficients of the adaptive filter by a time window function. The waveguide estimation device 1a estimates the differential group delay and the polarization dependent loss based on the time series tap coefficients within the time window.
[0029] In the following, the tap coefficients of the time series outside the time window may not be used in the estimation process of the differential group delay and the polarization dependent loss (they may simply be ignored), or may be replaced with, for example, 0 and then used in the estimation process of the differential group delay and the polarization dependent loss.
[0030] The waveguide estimation device 1 a includes an acquisition unit 10 , a storage unit 11 , a time window processing unit 12 a , a group delay difference estimation unit 13 a , a polarization dependent loss estimation unit 14 , and an output unit 15 .
[0031] The acquisition unit 10 acquires the time series tap coefficients of the adaptive filter from, for example, a predetermined information processing device (not shown). The acquisition unit 10 outputs the time series tap coefficients of the adaptive filter to the time window processing unit 12a. The storage unit 11 stores the time window function.
[0032] The time window processing unit 12a (calculation unit) multiplies the time series tap coefficients of the adaptive filter by a time window function. The group delay difference estimation unit 13a (estimation unit) estimates the group delay difference as shown in equation (2) based on the time series tap coefficients within the time window. The polarization dependent loss estimation unit 14 (estimation unit) estimates the polarization dependent loss as shown in equation (3) based on the time series tap coefficients within the time window. The output unit 15 outputs each estimation result to a predetermined information processing device (not shown).
[0033] Next, the time window processing unit 12a, the differential group delay estimating unit 13a, and the polarization dependent loss estimating unit 14 will be described. FIG. 2 is a diagram showing an example of a time window process, a differential group delay (DGD) estimation process, and a polarization dependent loss (PDL) estimation process in the first embodiment. The time window processing unit 12a shown in FIG. 2 includes a time series tap coefficient "h xx", "h xy ", "h yx " and "h yy 2 shows an example of the group delay difference of the frequency series (group delay difference for each frequency). The polarization dependent loss estimation unit 14 shown in FIG. 2 shows an example of the polarization dependent loss of the frequency series (polarization dependent loss for each frequency).
[0034] In Figure 2, the time window centered on the peak time of the time series tap coefficients (samples) is the tap coefficient of the adaptive filter, h xx ", "h xy ", "h yx " and "h yy " are defined in the IEEE 802.11b, IEEE 802.11b, and IEEE 802.11b. When the time series tap coefficients within the time window are used for the Fourier transform, the longer the length (width) of the time window, the better the frequency resolution of the differential group delay and the frequency resolution of the polarization dependent loss. In this way, the length of the time window and the frequency resolution are inversely proportional to each other.
[0035] However, if the frequency resolution is higher than necessary, the signal-to-noise ratio per point (sample) of the frequency sequence decreases. In other words, if the noise of the tap coefficients is assumed to be constant in the time direction, the signal-to-noise ratio per point of the frequency sequence decreases in proportion to the number of tap coefficients of the time sequence within the time window.
[0036] Therefore, by appropriately selecting the number of tap coefficients used in the Fourier transform, it is possible to improve the signal-to-noise ratio of the differential group delay and the polarization dependent loss estimated using the frequency series signal. Here, the length of the time window centered on the peak time of the time series tap coefficients may be determined empirically, for example.
[0037] As a result of determining the length of the time window, the position and length of the time window in the time direction may differ for each element (tap coefficient) of the inverse matrix of the transfer function. In this case, the tap coefficients in the time domain having different positions and lengths may be replaced with 0 before being used in the estimation process.
[0038] The time window processing unit 12a may multiply the time series tap coefficients within the time window by a predetermined window function. For example, an apodization function such as a "triangle function" or a "Happ-Genzel function" is used as the predetermined window function. This makes it possible to suppress the occurrence of ripples in the target spectrum.
[0039] Next, an example of the operation of the waveguide estimation device 1a will be described. 3 is a flowchart showing an example of the operation of the waveguide estimation device 1a in the first embodiment. The acquisition unit 10 acquires time-series tap coefficients (step S101). The time window processing unit 12a executes time window processing on the time-series tap coefficients. That is, the time window processing unit 12a multiplies the time-series tap coefficients by a time window function (step S102).
[0040] The group delay difference estimation unit 13a estimates the group delay difference in the waveguide (the system from the transmitter to the receiver) based on the time-windowed time-series tap coefficients (step S103). The polarization dependent loss estimation unit 14 estimates the polarization dependent loss in the waveguide based on the time-windowed time-series tap coefficients (step S104). The output unit 15 outputs each estimation result to a predetermined information processing device (not shown) (step S105).
[0041] As described above, the time window processing unit 12a multiplies the time series tap coefficients of the adaptive filter by the time window function. The group delay difference estimation unit 13a (estimation unit) estimates the group delay difference of the frequency series or the time series based on the tap coefficients of the time series within the time window. The polarization dependent loss estimation unit 14 (estimation unit) may estimate the polarization dependent loss of the frequency series or the time series based on the tap coefficients of the time series within the time window.
[0042] In this way, the group delay difference estimator 13a estimates the group delay difference of the frequency series or the time series based on the tap coefficients of the time series within the time window.The group delay difference estimator 13a estimates the polarization dependent loss of the frequency series or the time series based on the tap coefficients of the time series within the time window.
[0043] This makes it possible to improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value. Here, since dedicated hardware for estimating the differential group delay and the polarization dependent loss is not required, there is no need to change the hardware of the digital coherent receiver.
[0044] (Modification of the first embodiment) The modification of the first embodiment differs from the first embodiment in that threshold processing is performed on time-series tap coefficients. The modification of the first embodiment will be described focusing on the differences from the first embodiment.
[0045] FIG. 4 is a diagram showing an example of threshold processing in a modified example of the first embodiment. In FIG. 4, the time series tap coefficients "h xx ", "h xy ", "h yx " and "h yy 1 shows an example of "A lower threshold 20 and an upper threshold 21 are set for the time series tap coefficients based on a predetermined condition. The time window processing unit 12a performs threshold processing on the time series tap coefficients (e.g., absolute values of the tap coefficients) using the lower threshold 20 and the upper threshold 21.
[0046] The group delay difference estimator 13a performs an estimation process including a Fourier transform on tap coefficients that are above a lower threshold 20 or below an upper threshold 21 among the tap coefficients of the time series within the time window. The group delay difference estimator 13a may perform an estimation process including a Fourier transform on tap coefficients that are within a range from the lower threshold 20 to the upper threshold 21 and within a predetermined margin range among the tap coefficients of the time series within the time window.
[0047] The group delay difference estimating unit 13a may perform an estimation process including a Fourier transform on the tap coefficients of the time series within the time window that are above a lower threshold 20 or below an upper threshold 21. The group delay difference estimating unit 13a estimates the group delay difference based on the result of the Fourier transform as shown in equation (2).
[0048] The polarization dependent loss estimator 14 performs an estimation process including a Fourier transform on the tap coefficients of the time series within the time window that are above the lower threshold 20 or below the upper threshold 21. The polarization dependent loss estimator 14 may perform an estimation process including a Fourier transform on the tap coefficients of the time series within the time window that are within the range from the lower threshold 20 to the upper threshold 21 and within a predetermined margin range.
[0049] The polarization dependent loss estimator 14 may perform an estimation process including a Fourier transform on the tap coefficients of the time series within the time window that are greater than a lower threshold 20 or less than an upper threshold 21. Based on the result of the Fourier transform, the polarization dependent loss estimator 14 estimates the polarization dependent loss as shown in equation (3).
[0050] As described above, the group delay difference estimator 13a estimates the group delay difference of the frequency series or the time series based on the tap coefficients of the time series within the time window that are above the lower threshold or below the upper threshold. The polarization dependent loss estimator 14 may estimate the polarization dependent loss of the frequency series or the time series based on the tap coefficients of the time series within the time window that are above the lower threshold 20 or below the upper threshold 21.
[0051] This makes it possible to improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value.
[0052] Second embodiment The second embodiment differs from the first embodiment in that the length of the time window is updated based on the average value of the group delay difference of the frequency series or the time series. The second embodiment will be described focusing on the differences from the first embodiment.
[0053] 5 is a diagram showing a configuration example of a waveguide estimation device 1b in the second embodiment. The waveguide estimation device 1b is a device that estimates (measures) at least one of a frequency-series or time-series differential group delay (DGD) and a frequency-series or time-series polarization-dependent loss (PDL) for an optical signal transmitted through an optical transmission line and a waveguide in an optical device.
[0054] The waveguide estimation device 1 b includes an acquisition unit 10 , a storage unit 11 , a time window processing unit 12 b , a group delay difference estimation unit 13 b , a polarization dependent loss estimation unit 14 , and an output unit 15 .
[0055] 6 is a diagram showing an example of time window processing and estimation processing of group delay difference in the second embodiment. The time window processing unit 12b derives the optimal length of the time window by updating the length of the time window based on the average value (frequency average) of the group delay difference of the frequency series. The time window processing unit 12b may derive the optimal length of the time window by updating the length of the time window based on the average value (time average) of the group delay difference of the time series. In the second embodiment, the length of the time window is, for example, a constant multiple of the average value of the group delay difference.
[0056] Next, we explain why it is possible to derive the optimal length of the time window based on the average value (frequency average) of the differential group delay. The length of the time window can be derived based on the spread of the peak in the power spectrum of the frequency spectrum of the differential group delay (for example, the full width at half maximum). The length of the time window may also be derived based on the spread of the peak in the power spectrum of the frequency spectrum of the polarization dependent loss. Each power spectrum is the square of the absolute value of the Fourier transform result of the frequency spectrum. Therefore, the unit of the horizontal axis of each power spectrum is time.
[0057] The result of the Fourier transform of the power spectrum coincides with the autocorrelation function. Therefore, the spread of the autocorrelation function and the spread of the peak of the power spectrum are inversely proportional to each other. Therefore, the optimal length of the time window can be derived based on the spread of the autocorrelation function of the frequency spectrum of the differential group delay. The optimal length of the time window may also be derived based on the spread of the autocorrelation function of the frequency spectrum of the polarization dependent loss.
[0058] The differential group delay and the polarization dependent loss are both probabilistic. The ideal probability distribution of the differential group delay and the ideal probability distribution of the polarization dependent loss have each been clarified by previous research. Furthermore, the autocorrelation function of the differential group delay and the autocorrelation function of the polarization dependent loss have each been uniquely determined by previous research.
[0059] Regarding the effects of the first-order group delay difference and the first-order polarization dependent loss, when a waveguide (optical fiber) is considered to be a continuous limit of a minute birefringent material, the autocorrelation function of the square of the group delay difference is expressed as shown in Equation (4) (Reference 1 (M. Shtaif and A. Mecozzi, "Study of the frequency autocorrelation of the differential group delay in fibers with polarization mode dispersion," Optics Letters, Vol. 25, pp. 707-709 (2000))).
[0060]
number
[0061] Here, "τ" represents the differential group delay. "Δω" represents the deviation in the frequency direction from the center of the frequency. The full width at half maximum in the frequency direction is expressed as in equation (5).
[0062]
number
[0063] Here, " represents the ensemble average of the random variable "A". The polarization-dependent loss also has the same probability distribution (Reference 2 (C. Antonelli, A. Mecozzi, LE Nelson, and P. Magill, "Autocorrelation of the polarization-dependent loss in fiber routes," Optics Letters, Vol.36, pp.4005-4007 (2011))).
[0064] Regarding the power spectrum of the frequency spectrum of the group delay difference and the power spectrum of the frequency spectrum of the polarization dependent loss, the full width at half maximum in the frequency direction of each spectrum is determined not according to the polarization dependent loss but according to the average value (square root) of the group delay difference.
[0065] Therefore, the group delay difference estimating unit 13b uses all the taps of the adaptive filter to temporarily estimate the average value (frequency average) of the group delay difference of the frequency sequence. That is, the group delay difference estimating unit 13b temporarily estimates the average value of the group delay difference of the frequency sequence based on all the acquired tap coefficients. The time window processing unit 12b or the group delay difference estimating unit 13b determines the length of the time window to be a constant multiple of the estimated average value. The time window processing unit 12b or the group delay difference estimating unit 13b may add a predetermined margin to the length of the time window.
[0066] In the ensemble average of the group delay difference, the smaller the average value of the group delay difference is, the fewer the samples (points) that are statistically independent in the frequency direction are. Therefore, the average value (frequency average) of the group delay difference estimated at one time may significantly differ from the ensemble average of the original group delay difference. In such a case, the time window processing unit 12b or the group delay difference estimating unit 13b may perform a time averaging process on the time series tap coefficients. This allows the time window processing unit 12b or the group delay difference estimating unit 13b to improve the estimation accuracy of the group delay difference.
[0067] Next, an example of the operation of the waveguide estimation device 1b will be described. Fig. 7 is a flowchart showing an example of the operation of the waveguide estimation device 1b in the second embodiment. Step S201 is the same as step S101 shown in Fig. 3. The time window processing unit 12b acquires the average value of the group delay difference from the group delay difference estimation unit 13a. The time window processing unit 12b multiplies the time-series tap coefficients by a time window function. Here, the time window processing unit 12b adjusts the length of the time window based on the average value of the group delay difference. For example, the time window processing unit 12b adjusts the length of the time window so that it becomes a constant multiple of the average value of the group delay difference (step S202).
[0068] The group delay difference estimator 13b estimates the group delay difference in the waveguide based on the time-series tap coefficients on which the time window processing has been performed. The group delay difference estimator 13b outputs the average value of the group delay difference to the time window processing unit 12b (step S203). The polarization dependent loss estimator 14 estimates the polarization dependent loss in the waveguide based on the time-series tap coefficients on which the time window processing has been performed. The polarization dependent loss estimator 14 may output the average value of the polarization dependent loss to the time window processing unit 12b (step S204).
[0069] The group delay difference estimator 13b or the polarization dependent loss estimator 14 determines whether the estimation results of the group delay difference and the polarization dependent loss can be output. For example, when the number of executions of step S203 reaches a predetermined number or more, the group delay difference estimator 13b determines that the estimation results of the group delay difference and the polarization dependent loss can be output (step S205).
[0070] If it is determined that the estimated results of the group delay difference and the polarization dependent loss cannot be output (step S205: NO), the group delay difference estimator 13b returns the process to step S202. If it is determined that the estimated results of the group delay difference and the polarization dependent loss can be output (step S205: YES), the group delay difference estimator 13b and the polarization dependent loss estimator 14 proceed to step S206. Step S206 is the same as step S105 shown in FIG. 3.
[0071] As described above, the group delay difference estimator 13b estimates the average value of the group delay difference of the frequency series or time series. The time window processor 12b determines the length of the time window based on a constant multiple of the average value.
[0072] This makes it possible to improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value.
[0073] Third embodiment The third embodiment differs from the first embodiment in that frequency window processing is performed on the group delay difference of the frequency sequence. The third embodiment will be described focusing on the differences from the first embodiment.
[0074] 8 is a diagram showing a configuration example of a waveguide estimation apparatus 1c in the third embodiment. The waveguide estimation apparatus 1c is an apparatus that estimates (measures) at least one of a frequency-series or time-series differential group delay (DGD) and a frequency-series or time-series polarization-dependent loss (PDL) for an optical signal transmitted through an optical transmission line and a waveguide in an optical device.
[0075] In digital coherent communication, the tap coefficients are determined empirically, for example, according to the characteristics of the transmission path. For this reason, the peak of the time series tap coefficients may spread over a certain period of time. In this case, the estimation accuracy of the differential group delay and the polarization dependent loss decreases.
[0076] Therefore, the waveguide estimation device 1c first multiplies the time series tap coefficients of the adaptive filter by a time window function. The waveguide estimation device 1c then multiplies the frequency series used in the Fourier transform of the time series tap coefficients by a frequency window function. Furthermore, the waveguide estimation device 1c derives a shorter time window length based on the result of the frequency window processing. This makes it possible to further suppress the deterioration of the estimation accuracy.
[0077] In the following, the frequency sequence outside the frequency window may not be used in the process of deriving the time window length (it may simply be ignored), or may be replaced with, for example, 0 and then used in the process of deriving the time window length.
[0078] The waveguide estimation device 1c includes an acquisition unit 10, a storage unit 11, a time window processing unit 12c, a group delay difference estimating unit 13c, a polarization dependent loss estimating unit 14, an output unit 15, and a frequency window processing unit 16. The storage unit 11 stores a time window function and a frequency window function.
[0079] 9 is a diagram showing a first example of time window processing and frequency window processing in the third embodiment. For example, when higher-order polarization mode dispersion exists in the waveguide (when the change in differential group delay differs depending on the frequency of the optical signal), the peak time of the impulse response in the time-series waveform of the tap coefficient may differ for each frequency of the optical signal.
[0080] In such a case, the frequency window processor 16 multiplies the frequency sequence used in the Fourier transform of the tap coefficients by a frequency window function in the form of strips arranged in the frequency direction. That is, the frequency window processor 16 extracts frequency sequences for each frequency domain from among all frequency sequences. This narrows the time spread of the impulse response, and therefore the length of the time window can be shortened.
[0081] The frequency windowing unit 16 may multiply the frequency sequence extracted so as not to include a frequency component whose frequency is 0 by a sine function in the time direction. In this way, the frequency sequence extracted so as not to include a frequency component whose frequency is 0 is down-converted to a frequency sequence that includes a frequency component whose frequency is 0.
[0082] The time window processing unit 12c multiplies the time series tap coefficients by a time window function for each frequency domain (frequency window) of the frequency window processing. The group delay difference estimation unit 13c estimates the group delay difference in the waveguide for each frequency domain (frequency window) of the frequency window processing based on the time series tap coefficients on which the time window processing has been performed. The polarization dependent loss estimation unit 14 estimates the polarization dependent loss in the waveguide for each frequency domain (frequency window) of the frequency window processing based on the time series tap coefficients on which the time window processing has been performed.
[0083] In this way, by estimating the differential group delay and polarization dependent loss for each frequency region using a time window whose length is shortened by using a frequency window, the differential group delay and polarization dependent loss can be estimated for the entire frequency region of interest.
[0084] Next, an example of the operation of the waveguide estimation device 1c will be described. Fig. 10 is a flowchart showing an example of the operation of the waveguide estimation device 1c in the third embodiment. Step S301 is similar to step S101 shown in Fig. 3. The frequency window processing unit 16 multiplies the frequency sequence used for the Fourier transform of the tap coefficients by a frequency window function in the form of stripes aligned in the frequency direction (step S302).
[0085] The time window processing unit 12c multiplies the time series tap coefficients by a time window function for each frequency domain of the frequency window processing (step S303). The group delay difference estimation unit 13c estimates the group delay difference in the waveguide for each frequency domain of the frequency window processing based on the time series tap coefficients on which the time window processing has been performed (step S304). The polarization dependent loss estimation unit 14 estimates the polarization dependent loss in the waveguide for each frequency domain of the frequency window processing based on the time series tap coefficients on which the time window processing has been performed (step S305). Step S306 is the same as step S106 shown in FIG. 3.
[0086] As described above, the frequency window processing unit 16 extracts the frequency sequence used in the Fourier transform of the tap coefficients for each frequency domain. The group delay difference estimating unit 13c estimates the group delay difference for each extracted frequency sequence based on the tap coefficients of the time series within the time window. The polarization dependent loss estimating unit 14 may estimate the polarization dependent loss for each extracted frequency sequence based on the tap coefficients of the time series within the time window.
[0087] This makes it possible to further improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value.
[0088] (Modification of the third embodiment) The difference between the modified example of the third embodiment and the third embodiment is that frequency window processing is performed on a frequency region with a small absolute value or on a group delay difference with a small amount of variation. The difference between the modified example of the third embodiment and the third embodiment will be mainly described.
[0089] 11 is a diagram showing a second example of time window processing and frequency window processing in a modified example of the third embodiment. The absolute value or variation of the group delay difference may differ for each frequency domain. For example, due to the influence of high-order polarization mode dispersion or as a probabilistic phenomenon, there may be frequency domains in which the absolute value of the group delay difference is small or the variation is small in some frequency domains.
[0090] In such a case, the frequency window processing unit 16 multiplies the frequency sequence used for the Fourier transform of the tap coefficients in the frequency domain where the absolute value of the group delay difference is small or the amount of fluctuation is small by a frequency window function. That is, the frequency window processing unit 16 extracts the frequency sequence in the frequency domain where the absolute value of the group delay difference is small or the amount of fluctuation is small from all the frequency sequences. This narrows the time spread of the impulse response, and therefore the length of the time window can be shortened.
[0091] The time window processing unit 12c multiplies the time series tap coefficients by a time window function without performing frequency window processing. The group delay difference estimation unit 13c estimates the group delay difference in the waveguide based on the time series tap coefficients after the time window processing. The polarization dependent loss estimation unit 14 estimates the polarization dependent loss in the waveguide based on the time series tap coefficients after the time window processing.
[0092] The frequency windowing unit 16 multiplies the frequency sequence used for the Fourier transform of the tap coefficients in the frequency domain where the absolute value of the group delay difference is small or the amount of fluctuation is small by a frequency window function. The frequency windowing unit 16 may multiply the frequency sequence used for the Fourier transform of the tap coefficients in the frequency domain where the absolute value of the polarization dependent loss is small or the amount of fluctuation is small by a frequency window function.
[0093] The time window processing unit 12c multiplies the time series tap coefficients by a time window function after frequency window processing has been performed. The group delay difference estimation unit 13c estimates the group delay difference in the waveguide based on the time series tap coefficients after time window processing. The polarization dependent loss estimation unit 14 estimates the polarization dependent loss in the waveguide based on the time series tap coefficients after time window processing.
[0094] In this way, by using a time window whose length is shortened by using a frequency window to estimate the frequency domains in which the absolute values of the group delay difference and the polarization dependent loss are small or the amount of fluctuation is small for each frequency domain, the group delay difference and the polarization dependent loss can be estimated.
[0095] Next, an example of the operation of the waveguide estimation device 1c will be described. 12 is a flowchart showing an example of the operation of a waveguide estimation device 1c in a modified example of the third embodiment. Each step from step S401 to step S405 is similar to each step from step S201 to step S205 shown in FIG.
[0096] If it is determined that the estimated results of the group delay difference and the polarization dependent loss cannot be output (step S405: NO), the frequency window processing unit 16 detects a frequency region where the absolute value of the group delay difference is small or the amount of fluctuation is small (step S406). The frequency window processing unit 16 performs frequency window processing on the frequency series in the frequency region where the absolute value of the group delay difference is small or the amount of fluctuation is small. That is, the frequency window processing unit 16 multiplies the frequency series in the frequency region where the absolute value of the group delay difference is small or the amount of fluctuation is small by a frequency window function (step S407). The frequency window processing unit 16 returns the process to step S402.
[0097] If it is determined that the estimation results of the group delay difference and the polarization dependent loss can be output (step S405: YES), the group delay difference estimator 13c and the polarization dependent loss estimator 14 proceed to step S408. Step S206 is the same as step S206 shown in FIG.
[0098] As described above, the frequency window processing unit 16 extracts a frequency sequence used in the Fourier transform of the tap coefficients for a frequency region with a small absolute value of the group delay difference or a small amount of fluctuation. The group delay difference estimation unit 13c estimates the group delay difference for the extracted frequency sequence based on the tap coefficients of the time series within the time window. The polarization dependent loss estimation unit 14 may estimate the polarization dependent loss for the extracted frequency sequence based on the tap coefficients of the time series within the time window.
[0099] This makes it possible to further improve the accuracy of estimating at least one of the differential group delay and the polarization dependent loss even when the signal-to-noise ratio of the optical signal transmitted through the waveguide is less than a predetermined value.
[0100] The apparatus of the present invention can also be realized by a computer and a program, and the program can be recorded on a recording medium or provided via a network.
[0101] (Hardware configuration example) FIG. 13 is a diagram showing an example of a hardware configuration of the waveguide estimation device 1 in each embodiment. The waveguide estimation device 1 corresponds to each of the above-mentioned waveguide estimation device 1a, the above-mentioned waveguide estimation device 1b, and the above-mentioned waveguide estimation device 1c. Some or all of the functional units of the waveguide estimation device 1 are realized as software by a processor 100 such as a CPU (Central Processing Unit) executing a program stored in a storage device 101 having a non-volatile recording medium (non-transient recording medium) and a memory 102. The program may be recorded in a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or a non-transient recording medium such as a storage device such as a hard disk built into a computer system. The communication unit 103 acquires information such as a tap coefficient, a time window, and a frequency window. The communication unit 103 transmits an estimation result to a predetermined information processing device (not shown).
[0102] Some or all of the functional units of the waveguide estimation device 1 may be realized using hardware including an electronic circuit (electronic circuit or circuitry) using, for example, an LSI (Large Scale Integrated circuit), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field Programmable Gate Array), or the like.
[0103] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Industrial Applicability]
[0104] The present invention is applicable to digital coherent receivers. [Explanation of symbols]
[0105] 1a, 1b, 1c...waveguide estimation device, 10...acquisition unit, 11...storage unit, 12a, 12b, 12c...time window processing unit, 13a, 13b, 13c...group delay difference estimation unit, 14...polarization dependent loss estimation unit, 15...output unit, 16...frequency window processing unit, 20...lower limit threshold, 21...upper limit threshold, 100...processor, 101...storage device, 102...memory, 103...communication unit
Claims
1. a time window processing unit that multiplies a time series of tap coefficients of the adaptive filter by a time window function; an estimation unit that estimates at least one of a group delay difference of a frequency series or a time series and a polarization dependent loss of a frequency series or a time series based on a tap coefficient of the time series within a time window; A waveguide estimation device comprising:
2. The estimation unit estimates at least one of a group delay difference of the frequency series or the time series and a polarization dependent loss of the frequency series or the time series based on a tap coefficient that is above a lower threshold or below an upper threshold among the tap coefficients of the time series within a time window. The waveguide estimation device according to claim 1 .
3. The estimation unit estimates an average value of a group delay difference of the frequency series or the time series, The time window processing unit determines a length of the time window based on a constant multiple of the average value. The waveguide estimation device according to claim 1 or 2.
4. a frequency window processing unit that cuts out the frequency series used in the Fourier transform of the time series of tap coefficients for each frequency domain, The estimation unit estimates, for each of the extracted frequency series, at least one of a group delay difference of the frequency series or the time series and a polarization dependent loss of the frequency series or the time series based on a tap coefficient of the time series within the time window. The waveguide estimation device according to claim 1 .
5. A waveguide estimation method executed by a waveguide estimation device, comprising: a time window processing step of multiplying a time series of tap coefficients of the adaptive filter by a time window function; an estimation step of estimating at least one of a group delay difference of a frequency series or a time series and a polarization dependent loss of a frequency series or a time series based on tap coefficients of the time series within a time window; The waveguide estimation method includes:
6. The estimating step further includes a step of estimating at least one of a group delay difference of the frequency series or the time series and a polarization dependent loss of the frequency series or the time series based on tap coefficients of the time series within a time window that are above a lower threshold or below an upper threshold. The waveguide estimation method according to claim 5 .
7. The estimating step further includes the step of estimating an average value of a differential group delay of the frequency series or the time series; The time window processing step further includes a step of determining a length of the time window based on a constant multiple of the average value. The waveguide estimation method according to claim 5 or 6.
8. A program for causing a computer to function as the waveguide estimation device according to any one of claims 1 to 4.
Citation Information
Patent Citations
Optical transmission device, optical transmission system, and polarization dependent loss monitor
JP2016152556A
Method And Apparatus For Polarization-Division-Multiplexed Optical Receivers
US20120002979A1
