Optical transmission characteristic estimation device, optical transmission characteristic estimation method, and program

Abstract

This optical transmission characteristic inference device infers a transmission characteristic of an optical transmission path by: using a non-linear Schrodinger equation to model the optical effect of an optical transmission path; dividing the optical transmission path into N segments (where N is an integer greater than 1); and solving the non-linear Schrodinger equation by N step computations using, as the step size, each of the segment lengths of the N segments resulting from the division. The optical transmission characteristic inference device is provided with a step size updating unit that updates each of the N step sizes on the basis of an inferred optical power distribution and an inferred dispersion distribution indicated by the transmission characteristic.

Description

[Technical Field]

[0001] The present invention relates to an optical transmission characteristics estimation device, an optical transmission characteristics estimation method, and a program. [Background technology]

[0002] When operating an optical transmission system, the basic characteristics of the optical fiber that makes up the optical transmission path have a major impact on transmission performance. Here, the basic characteristics of optical fiber include optical power, distribution of loss and dispersion, and location of fault points. For example, if the optical power is too high, the influence of nonlinear optical effects in the optical fiber increases, reducing the signal-to-noise ratio (hereinafter referred to as "SNR"). If the loss is too high, the optical power attenuates accordingly, reducing the SNR.

[0003] Therefore, knowing the characteristics of optical fibers is important for the operation, maintenance, and monitoring of optical transmission systems. Optical transmission paths are composed of various devices other than optical fibers, such as optical amplifiers and optical filters, and knowing the characteristics of these devices is also important for the operation, maintenance, and monitoring of optical transmission systems.

[0004] The characteristics of devices such as optical fibers, optical amplifiers, and optical filters can generally be measured using analog measuring instruments such as OTDRs (Optical Time Domain Reflectometers) and optical spectrum analyzers. However, measurements using analog measuring instruments require direct measurement of each optical node and optical fiber, which poses the problem of high equipment and operating costs.

[0005] To solve this problem, in recent years, digital longitudinal monitoring (DLM) has been proposed, which is a technology that detects the characteristics of various devices in an optical transmission system by digital signal processing on the receiving side of the optical transmission system, instead of measurements using analog measuring instruments (see, for example, Patent Documents 1 and 2 and Non-Patent Documents 1 to 4). DLM is based on a digital coherent optical transmission system, and monitors the optical power and other characteristics of the optical transmission line by performing digital signal processing on the received signal obtained by coherently detecting the optical signal transmitted by the optical transmission line. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2021 / 124415 [Patent Document 2] International Publication No. 2021 / 199317 [Non-patent literature]

[0007] [Non-Patent Document 1] T. Sasai, et al., “Simultaneous detection of anomaly points and fiber types in multi-span transmission links only by receiver-side digital signal processing”, IEEE, 2020 Optical Fiber Communications Conference and Exhibition (OFC), Paper Th1F.1, March 2020. [Non-patent document 2] T. Sasai, et al., “Physics-oriented learning of nonlinear Schrodinger equation: optical fiber loss and dispersion profile identification”, arXiv:2104.05890, Apr 2021. [Non-patent document 3] T. Sasai, et al., “Digital backpropagation for optical path monitoring: loss profile and passband narrowing estimation”, IEEE, 2020 European Conference on Optical Communications (ECOC), Paper Tu2D.1, Dec 2020. [Non-patent document 4] T. Sasai, et al., “Revealing Raman-amplified power profile and Raman gain spectra with digital backpropagation”, IEEE, 2021 Optical Fiber Communications Conference and Exhibition (OFC), Paper M3I.5, June 2021. Summary of the Invention [Problem to be solved by the invention]

[0008] Compared to analog measuring instruments, DLMs are simpler to use because they do not require on-site measurement. However, DLMs have the problem of estimating device characteristics more accurately than analog measuring instruments. More specifically, when using DLMs to estimate the propagation direction distribution of the optical power of an optical signal transmitted through an optical transmission line, if there is a discontinuous power fluctuation in the optical transmission line (e.g., lumped amplification by an optical amplifier or abnormal loss in an optical fiber), the accuracy of estimating the optical power distribution near the discontinuous point is lower than that of analog measuring instruments.

[0009] For example, suppose that measurements using an OTDR and a DLM are performed on an optical transmission line 200 shown in Figure 16(a), which includes a plurality of optical fibers 201-1 to 201-5 and a plurality of optical amplifiers 202-1, 202-2, 202-3, and 202-4 inserted between the plurality of optical fibers 201-1 to 201-5. Figure 16(b) is a graph showing the results of the OTDR measurement and the DLM measurement. The vertical axis represents the magnitude of optical power, and the horizontal axis represents distance in units of km. In the graph of Figure 16(b), the dashed line represents the results of the OTDR measurement, and the solid line represents the results of the DLM measurement. The length of each of the optical fibers 201-2 to 201-5 is 70 km, and the position of 0 km on the graph in FIG. 16(b) is the connection point between the optical fiber 201-1 and the optical amplifier 202-1 in FIG. 16(a), that is, the position of the optical fiber input in the optical amplifier 202-1.

[0010] As can be seen from the graph in FIG. 16(b), the measurement results using the DLM are less accurate than the measurement results using the OTDR in the section around 70 km indicated by reference numeral 301, the section around 140 km indicated by reference numeral 302, and the section around 210 km indicated by reference numeral 303. The sections indicated by reference numerals 301, 302, and 303 all have one thing in common: they are located near the beginning and end of the optical transmission span near the optical amplifiers 202-2 to 202-4. This means that the DLM has measurement dead zones at the beginning and end of the optical transmission span near the optical amplifiers 202-2 to 202-4, where the estimation accuracy decreases. One possible way to improve the estimation accuracy of the optical power distribution near the optical amplifiers 202-2 to 202-4 is to use an optical signal with a higher optical power than that used during normal operation. However, increasing the optical power is avoided because it degrades the quality of the optical signal due to nonlinear optical effects, which degrades communication quality.

[0011] In view of the above circumstances, the present invention aims to provide a technology that makes it possible to obtain estimation results with appropriate accuracy in the measurement dead zone when estimating the characteristics of an optical transmission line using DLM, even when using an optical signal with low optical power output such as that used during normal operation. [Means for solving the problem]

[0012] One aspect of the present invention is an optical transmission characteristic estimation device that models the optical effect of an optical transmission line by a nonlinear Schrödinger equation, divides the optical transmission line into N sections (N is an integer of 2 or more), and finds a solution to the nonlinear Schrödinger equation by performing N step calculations, with each step size being the section length of the N divided sections, to estimate the transmission characteristics of the optical transmission line, and the optical transmission characteristic estimation device includes a step size update unit that updates each of the N step sizes based on an estimated optical power distribution and an estimated dispersion distribution indicated by the transmission characteristics.

[0013] One aspect of the present invention is an optical transmission characteristic estimation method that models the optical effect of an optical transmission line by a nonlinear Schrödinger equation, divides the optical transmission line into N sections (N is an integer of 2 or more), and estimates the transmission characteristics of the optical transmission line by finding a solution to the nonlinear Schrödinger equation through calculations of N steps, with each step size being the section length of the N divided sections, and that updates each of the N step sizes based on an estimated optical power distribution and an estimated dispersion distribution indicated by the transmission characteristics.

[0014] One aspect of the present invention is a program for causing a computer to execute the following steps: modeling the optical effects of an optical transmission line using a nonlinear Schrödinger equation, and dividing the optical transmission line into N sections (N is an integer of 2 or more); finding a solution to the nonlinear Schrödinger equation by performing N step calculations, with each step size being the section length of the N divided sections, to estimate the transmission characteristics of the optical transmission line; and updating each of the N step sizes based on an estimated optical power distribution and an estimated dispersion distribution indicated by the transmission characteristics. [Effects of the Invention]

[0015] According to the present invention, when estimating the characteristics of an optical transmission line using DLM, it is possible to obtain estimation results with appropriate accuracy in the measurement dead zone, even if an optical signal with low optical power output, such as that used during normal operation, is used. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a block diagram showing a configuration of an optical transmission system according to a first embodiment. [Figure 2] FIG. 4 is a diagram illustrating a procedure for estimating optical power in the first embodiment. [Figure 3] FIG. 4 is a diagram illustrating a procedure for estimating a dispersion coefficient in the first embodiment. [Figure 4] FIG. 2 is a block diagram showing the configuration of an SSFM (Split-Step Fourier Method) processing unit according to the first embodiment. [Figure 5] FIG. 4 is a diagram illustrating the relationship between estimated optical power and step size in the first embodiment. [Figure 6] FIG. 4 is a diagram illustrating a processing flow of the optical receiving device according to the first embodiment. [Figure 7] FIG. 4 is a diagram illustrating a processing flow by a step size update unit according to the first embodiment. [Figure 8] FIG. 3 is a diagram illustrating an outline of a step size change performed by a step size update unit according to the first embodiment. [Figure 9] 1 is a graph (part 1) showing the results of an experiment conducted using the optical receiving device of the first embodiment. [Figure 10] 10 is a graph (part 2) showing the results of an experiment conducted using the optical receiving device of the first embodiment. [Figure 11] FIG. 10 is a block diagram showing the configuration of an optical receiving device according to a second embodiment. [Figure 12] FIG. 10 is a diagram illustrating a processing flow by a step size update unit according to the second embodiment. [Figure 13] FIG. 10 is a diagram illustrating the content of processing by a step size update unit according to the second embodiment. [Figure 14] FIG. 10 is a block diagram of another configuration example (part 2) of the optical receiving device. [Figure 15] FIG. 10 is a block diagram of another configuration example (part 3) of the optical receiving device. [Figure 16] 10 is a graph comparing the measurement results using an OTDR with the estimation results using a conventional DLM. DETAILED DESCRIPTION OF THE INVENTION

[0017] (First embodiment) Hereinafter, embodiments of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing the configuration of an optical transmission system 100 according to a first embodiment. The optical transmission system 100 is, for example, a digital coherent optical transmission system, and includes an optical transmitter 3, an optical receiver 1, and an optical transmission path 2 connecting the optical transmitter 3 and the optical receiver 1.

[0018] The optical transmitting device 3 converts a digital transmission signal, for example, consisting of a sequence of 0s and 1s, into a format expressed by four components: an in-phase component (hereinafter referred to as an I (In-phase) component) for X polarization and a quadrature component (hereinafter referred to as a Q (Quadrature) component) for Y polarization. Hereinafter, a transmission signal expressed by these four components will also be referred to as a transmission symbol. The optical transmitting device 3 provides each of the four electrical signals obtained by the conversion to each of four Mach-Zehnder optical modulators, and modulates each of the four light beams obtained by branching laser light emitted by a signal light source provided inside the device to generate an optical signal. The optical transmitting device 3 converts the modulated I and Q component optical signals for X polarization into an X-polarized optical signal, converts the I and Q component optical signals for Y polarization into a Y-polarized optical signal, and polarization-combines the X and Y-polarized optical signals. The optical transmitter 3 transmits the polarization multiplexed QPSK (Quadrature Phase Shift Keying) optical signal generated by polarization combining to the optical transmission line 2.

[0019] The optical transmission line 2 includes an optical fiber and an optical amplifier, and transmits the polarization multiplexed QPSK optical signal sent from the optical transmitter 3 to the optical receiver 1 .

[0020] The optical receiving device 1 includes a coherent receiving unit 11 and an optical transmission characteristic estimating unit 12. The coherent receiving unit 11 is connected to the optical transmission line 2 and receives and coherently detects an optical signal transmitted by the optical transmission line 2. For example, when the optical transmitting device 3 transmits a polarization-multiplexed QPSK optical signal as described above, the coherent receiving unit 11 separates the received optical signal into an X-polarized wave and a Y-polarized wave. The coherent receiving unit 11 detects the I-component and Q-component of each of the X-polarized wave and the Y-polarized wave by causing interference between each of the X-polarized and Y-polarized optical signals after the separation and a laser beam emitted from a local oscillator light source provided inside the coherent receiving unit 11. The coherent receiving unit 11 converts each of the I-component and Q-component optical signals of each of the X-polarized and Y-polarized waves into four-series analog electrical signals, and then converts the converted four-series analog signals into four-series digital signals using four internal analog-to-digital converters and outputs the digital signals. Hereinafter, the four-series digital signals output by the coherent receiver 11 will be referred to as received signals.

[0021] The optical transmission characteristic estimation unit 12 includes a compensation processing unit 13, a coefficient updating unit 14, a transmission characteristic calculation unit 15, a step size updating unit 16, and an estimation result output unit 17. The compensation processing unit 13 performs compensation on the received signal output by the coherent receiving unit 11 to remove various influences that the optical signal has received while propagating through the optical transmission path 2, and restores the transmitted symbol from the received signal. The compensation processing unit 13 includes an SSFM processing unit 31, an adaptive equalization unit 32, a frequency offset compensation unit 33, and a carrier phase noise compensation unit 34. The SSFM processing unit 31 applies SSFM (split-step Fourier transformation), which is one of the numerical analysis methods for finding a solution to the nonlinear Schrödinger equation, to the received signal.

[0022] (Procedure for estimating transmission characteristics using DLM) Here, a procedure for estimating the characteristics of the optical transmission line 2 using DLM will be described with reference to FIGS. 2 and 3. For example, as shown in FIGS. 2 and 3, assume that the optical transmission line 2 includes optical fibers 21-1 and 21-2 and an optical amplifier 22-1 inserted between the optical fibers 21-1 and 21-2. When the optical signal of the transmission symbol 71 propagates through the optical transmission line 2, two optical effects occur. Note that the diagram of the transmission symbol 71 shown in FIGS. 2 and 3 is a diagram showing an example of a constellation of either the X polarization or the Y polarization of a polarization-multiplexed QPSK optical signal, and shows four code points of QPSK. An actual transmission symbol includes two symbols represented by a constellation like the transmission symbol 71, one corresponding to the X polarization and the other corresponding to the Y polarization.

[0023] One of the two optical effects is a nonlinear optical effect in which the phase of an optical signal rotates in proportion to the magnitude of the optical power of the optical signal, as shown in the graphs of FIGS. 2(a) and 2(b). The horizontal axis of the graphs of FIGS. 2(a) and 2(b) represents the distance in the optical transmission line 2, and the origin of the horizontal axis is the connection point between the optical transmitter 3 and the optical transmission line 2 shown in FIG. 1, i.e., the start point of the optical transmission line 2. The end point of the optical transmission line 2 is the connection point between the coherent receiver 11 and the optical transmission line 2. The vertical axis of the graph of FIG. 2(a) represents the magnitude of the optical power of the optical signal propagating through the optical transmission line 2. The vertical axis of the graph of FIG. 2(b) represents the magnitude of the nonlinear phase rotation occurring in the optical signal propagating through the optical transmission line 2. As shown in the graph of FIG. 2(a), the optical power of the optical signal propagating through the optical transmission line 2 attenuates with increasing distance from the origin, and after being amplified by the optical amplifier 22-1, it again attenuates with increasing distance from the origin. In this case, the amount of nonlinear phase rotation occurring in the optical signal propagating through the optical transmission line 2 will change in proportion to the change in optical power shown in the graph of FIG. 2(a), as shown in the graph of FIG. 2(b).

[0024] The other of the two optical effects is a linear optical effect due to chromatic dispersion, as shown in the graph of Fig. 3(a). The horizontal axis of the graph of Fig. 3(a) represents the distance in the optical transmission line 2, as in Figs. 2(a) and 2(b). The vertical axis of the graph of Fig. 3(a) represents the magnitude of the amount of chromatic dispersion that occurs in the optical signal propagating through the optical transmission line 2. As can be seen from the graph of Fig. 3(a), a certain amount of chromatic dispersion occurs in each of the optical fibers 21-1 and 21-2, and the amount of dispersion that occurs differs depending on the type of the optical fibers 21-1 and 21-2.

[0025] Due to these two optical effects and the influence of other noises, etc., the received symbol 72, which is the symbol of the received signal after coherent detection by the coherent receiver 11, has a disturbed constellation compared to the transmitted symbol 71, as shown in Figures 2 and 3, and the positions of the four QPSK code points cannot be clearly identified.

[0026] The two optical effects occurring in the optical transmission line 2 described above can be modeled by the nonlinear Schrodinger equation of the following equation (1).

[0027]

number

[0028] In the above equation (1), E is an optical signal represented by a complex electric field whose power is normalized to 1. z is a variable indicating the position on the optical transmission line 2, and is expressed as the distance from the origin, which is the connection point between the optical transmitter 3 and the optical transmission line 2. β in the first term on the right-hand side is a dispersion coefficient, and t is a variable indicating time. γ in the second term on the right-hand side is a nonlinear constant, and P(z) is a function indicating the optical power distribution in the optical transmission line 2, i.e., the optical power of the optical signal at any position z on the optical transmission line 2. "j" on the right-hand side is an imaginary unit.

[0029] The first term on the right side of equation (1) represents the chromatic dispersion that occurs in the optical signal propagating through the optical transmission line 2, and the second term on the right side represents the nonlinear phase rotation that occurs in the optical signal propagating through the optical transmission line 2. The nonlinear Schrodinger equation shown in equation (1) can be solved by, for example, SSFM, which is a numerical analysis method.

[0030] In the DLM used in this embodiment, a technique called DBP (Digital Back Propagation) is used to restore a transmission symbol from a received signal output by the coherent receiving unit 11. More specifically, SSFM for an equation obtained by multiplying the right side of equation (1) by −1 is applied to the received signal. That is, the optical transmission line 2 is divided into N sections, and nonlinear phase rotation amounts φ1 to φ2, which indicate the magnitude of nonlinear phase rotation, are calculated for each of the N divided sections. N and the dispersion coefficients β1 to β2, which indicate the magnitude of chromatic dispersion. N After appropriately determining N, the received signal is alternately subjected to inverse nonlinear phase rotation and inverse dispersion. Here, N is an integer equal to or greater than 2, and hereinafter, the symbol "n" will be used to indicate any integer between 1 and N.

[0031] The graphs in Figures 2(c) and 3(b) show an example in which the optical transmission line 2 is divided into eight sections, with N=8, so that each section has the same length. The vertical axis of the graph in Figure 2(c) indicates the amount of nonlinear phase rotation, just like the vertical axis of the graph in Figure 2(b), and the vertical axis of the graph in Figure 3(b) indicates the amount of chromatic dispersion, just like the graph in Figure 3(a). The horizontal axes of the graphs in Figures 2(c) and (d) and 3(b) indicate distance, and their direction is opposite to that of the horizontal axes of the graphs in Figures 2(a) and (b) and 3(a). In other words, the connection point between the optical transmission line 2 and the coherent receiver 11, which is the end point of the optical transmission line 2, is the origin of the horizontal axes of the graphs in Figures 2(c) and (d) and 3(b). Furthermore, in other words, any position on the graphs in Fig. 2(c) and (d) and the graph in Fig. 3(b) can be expressed as "z inv ” and the distance from the start point to the end point of the optical transmission line 2 is “L.” In this case, “z” and “z” are arbitrary positions on the horizontal axis of the graphs in FIGS. 2(a) and 2(b) and the graph in FIG. 3(a). inv " means z=Lz inv The graphs in Figure 2(c) and Figure 3(b) are not based on the z coordinate axis, but on the z inv Since the coordinate axes are shown as (a) and (b), the quantities represented on the vertical axes of the graphs in Figures 2(c) and 3(b) are referred to as the estimated inverse nonlinear phase rotation amount and the estimated inverse dispersion amount, respectively.

[0032] The received signal output from the coherent receiver 11 is subjected to N nonlinear phase rotation amounts φ1 to φ shown in the graph of FIG. 2(c). N 3B, the coherent receiving unit 11 outputs a received signal with N dispersion coefficients β1 to β2. N As a result, an effect opposite to that caused by chromatic dispersion occurring in the optical transmission line 2 is exerted.

[0033] Here, N nonlinear phase rotation amounts φ1 to φ determined in SSFM are N and N dispersion coefficients β1 to β Nand chromatic dispersion are respectively able to approximate the nonlinear phase rotation and chromatic dispersion that actually occur in the optical transmission line 2. An approximation state means that, for example, when the graph in Fig. 2(c) is made line-symmetrical about the vertical axis, the shape of the graph nearly matches the shape of the graph in Fig. 2(b), and when the graph in Fig. 3(b) is made line-symmetrical about the vertical axis, the shape of the graph nearly matches the shape of the graph in Fig. 3(a).

[0034] When the approximation is successful, the symbol of the signal that has been inversely affected by SSFM, i.e., the recovered symbol 73 shown in Figures 2 and 3, will nearly match the transmitted symbol 71, and the SNR of the signal that has been inversely affected by SSFM will be high. In contrast, when the approximation is not successful, the SNR of the signal that has been inversely affected by SSFM will be low.

[0035] Therefore, N nonlinear phase rotation amounts φ1 to φ used in SSFM are N and N dispersion coefficients β1 to β N In order to update the SNR so as to increase the SNR, an optimization algorithm such as a gradient method is used. To use the optimization algorithm, for example, a predetermined training signal is transmitted to the optical transmitting device 3 as a transmission signal. As a result, the coherent receiving unit 11 outputs a received signal corresponding to the training signal, and SSFM is applied to the received signal. In the optimization algorithm, based on the signal inversely affected by SSFM and the known training signal, new N nonlinear phase rotation amounts φ1 to φ are calculated so as to increase the SNR of the signal inversely affected by SSFM. N and the new N dispersion coefficients β1 to β N The optimization algorithm calculates N nonlinear phase rotation amounts φ1 to φ N and N dispersion coefficients β1 to β NBy repeatedly updating the above, it is possible to restore the transmitted symbols from the received signals with a high SNR. Note that even without transmitting a training signal, the transmitted signals can be restored by performing normal demodulation using a general optical receiving device. This makes it possible to estimate the transmission characteristics even during operation.

[0036] In the compensation processing unit 13, the SSFM processing unit 31 includes N linear compensation units 40-1 to 40-N, N nonlinear compensation units 50-1 to 50-N, and a control unit 60, as shown in FIG. 4, in order to perform the above-mentioned SSFM processing. In the SSFM processing unit 31, the linear compensation unit 40-1 is connected to the output side of the coherent receiving unit 11, and the nonlinear compensation unit 50-1 is connected to the output side of the linear compensation unit 40-1. The linear compensation unit 40-1 and the nonlinear compensation unit 50-1 perform the first step of SSFM processing. From the second step onwards, that is, when n is an integer between 2 and N, the linear compensation unit 40-n is connected to the output side of the nonlinear compensation unit 50-(n-1), and the nonlinear compensation unit 50-n is connected to the output side of the linear compensation unit 40-n. The linear compensation unit 40-n and the nonlinear compensation unit 50-n perform the nth step of SSFM processing.

[0037] In the following description, n is assumed to be any integer between 1 and N. The linear compensation unit 40-n includes a Fourier transform unit 41-n, a chromatic dispersion compensation unit 42-n, and an inverse Fourier transform unit 43-n. The Fourier transform unit 41-n performs a Fourier transform on the signal provided from the previous stage to convert it into a frequency domain signal. Here, the Fourier transform performed by the Fourier transform unit 41-n is, for example, an FFT (Fast Fourier Transform). The chromatic dispersion compensation unit 42-n performs dispersion compensation by performing the calculation shown in the following equation (2) on the signal after the Fourier transform by the Fourier transform unit 41-n.

[0038]

number

[0039] In the above equation (2), (·) represents the output of the Fourier transform unit 41-n, that is, the signal after the Fourier transform by the Fourier transform unit 41-n. n " is the dispersion coefficient of the nth step, "ω" is the angular frequency, that is, ω=2πf, and f is the frequency in the frequency domain in which the signal after the Fourier transform by the Fourier transform unit 41-n exists. n " is the step size of the nth step, and is the length of the nth section as seen from the end point side of the optical transmission line 2 when the optical transmission line 2 is divided into N sections. n The reason why " is the nth section from the end point side of the optical transmission line 2, not from the start point side, is that the SSFM process is inv This is because the processing is performed in order from the starting point on the coordinate axis. "j" is the imaginary unit.

[0040] The inverse Fourier transform unit 43-n performs an inverse Fourier transform on the signal that has been dispersion compensated by the chromatic dispersion compensator 42-n to convert it into a time domain signal. Here, the inverse Fourier transform performed by the inverse Fourier transform unit 43-n is, for example, an IFFT (Inverse Fast Fourier Transform).

[0041] The n-th step nonlinear compensation section 50-n performs the calculation shown in the following equation (3) on the signal after the inverse Fourier transform by the inverse Fourier transform section 43-n to compensate for the nonlinear phase rotation.

[0042]

number

[0043] In the above equation (3), (·) means the output of the inverse Fourier transform unit 43-n, that is, the signal after the inverse Fourier transform by the inverse Fourier transform unit 43-n. n " is the nonlinear constant for the nth step, and is a predetermined constant. P(z n ) is the position z on the optical transmission line 2 n is the optical power of the optical signal at zn is the distance between the center position of the nth section as seen from the end point side of the optical transmission line 2 when the optical transmission line 2 is divided into N sections, and the start point of the optical transmission line 2. n The reason why "dz" is set to the center position of the nth section when viewed from the end point side of the optical transmission line 2, rather than from the start point side, is that n This is the same reason as in the case of ", and as shown in Figure 2(c), the SSFM process is inv This is because the processing is performed in order from the starting point on the coordinate axis. 2 " means the square of the absolute value of the output of the inverse Fourier transform unit 43-n. As in equation (2), "dz n " is the step size of the nth step, and "j" is the imaginary unit. Note that the nonlinear constant γ n and optical power P(z n ) and the nonlinear phase rotation amount φ n That is, φ n =γ n P(z n ) relationship.

[0044] The control unit 60 is connected to the coherent receiving unit 11, the coefficient updating unit 14, the step size updating unit 16, the chromatic dispersion compensators 42-1 to 42-N, and the nonlinear compensators 50-1 to 50-N. N , and dispersion coefficients β1 to β N and step size dz1~dz N and configure the settings.

[0045] Returning to FIG. 1, the adaptive equalization unit 32 performs, for example, separation of the polarization multiplexed signal and compensation for waveform distortion caused by polarization mode dispersion, etc., on the signal whose chromatic dispersion and nonlinear phase rotation have been compensated for by the SSFM processing unit 31, using a linear filter such as an FIR (Finite Impulse Response) filter.

[0046] The frequency offset compensation unit 33 performs frequency offset compensation on the signal compensated for by the adaptive equalization unit 32, compensating for the frequency difference between the light emitted by the signal light source provided in the optical transmitting device 3 and the light emitted by the local oscillator light source provided in the coherent receiving unit 11.

[0047] The carrier phase noise compensator 34 performs phase offset compensation on the signal that has been frequency offset compensated for by the frequency offset compensator 33, compensating for the phase difference between the light emitted by the signal light source provided in the optical transmitter 3 and the light emitted by the local oscillator light source provided in the coherent receiver 11.

[0048] The coefficient update unit 14 is a functional unit that performs the above-mentioned DBP processing. The coefficient update unit 14 stores in advance in an internal storage area the training signal transmitted by the optical transmitting device 3. The coefficient update unit 14 applies a predetermined evaluation function to the training signal stored in the internal storage area and to the signal obtained after the compensation processing unit 13 has performed compensation on the received signal obtained when the optical transmitting device 3 transmits the training signal, and calculates new N nonlinear phase rotation amounts φ1 to φ according to an optimization algorithm that minimizes the value of the evaluation function. N and the new N dispersion coefficients β1 to β N and calculate.

[0049] Here, the evaluation function is, for example, a function that calculates N new nonlinear phase rotation amounts φ1 to φ so as to increase the SNR of the signal after compensation by the compensation processor 13. N and the new N dispersion coefficients β1 to β N When calculating , a function is applied that calculates a smaller value as the SNR of the signal after compensation by the compensation processing unit 13 increases. However, the evaluation function is not limited to such a function, and may be a function that calculates the squared error between the signal after compensation by the compensation processing unit 13 and the training signal, or may be the evaluation function disclosed in Patent Document 1, or may be any other evaluation function.

[0050] As an optimization algorithm applied to the coefficient update unit 28, the nonlinear phase rotation amounts φ1 to φ N and dispersion coefficients β1 to β N A method of optimizing each of the above points may be applied, or a gradient method such as backpropagation or steepest descent, or other existing learning processing methods used in the field of machine learning may be applied.

[0051] The transmission characteristic calculation unit 15 calculates the nonlinear phase rotation amount φ n and the dispersion coefficient β n The coefficient update unit 14 calculates the transmission characteristics of the optical transmission line 2 based on the above. n and the dispersion coefficient β n are discrete values. Therefore, the transmission characteristic calculation unit 15 calculates the optical power distribution and dispersion distribution that indicate the transmission characteristics of the optical transmission line 2 as follows.

[0052] The transmission characteristic calculation unit 15 calculates the nonlinear phase rotation amounts φ1 to φ N Each of these is expressed as the corresponding nonlinear constant γ1 to γ N Dividing by z as shown in the graph in Figure 2(d) inv The transmission characteristic calculation unit 15 linearly interpolates the calculated N estimated optical powers to obtain z inv The function P(z inv ) is calculated. The transmission characteristic calculation unit 15 calculates the calculated function P(z inv ) is converted into a function of z, and a function P(z) indicating the estimated optical power at an arbitrary position z on the optical transmission line 2 is taken as the estimated optical power distribution of the optical transmission line 2.

[0053] The transmission characteristic calculation unit 15 calculates the dispersion coefficient β output by the coefficient update unit 14. n Linearly interpolate z inv The function β(z inv ) is calculated. The transmission characteristic calculation unit 15 calculates the calculated function β(z inv) is converted into a function of z to obtain a function β(z) that indicates an estimated dispersion coefficient at an arbitrary position z in the optical transmission line 2, which is taken as the estimated dispersion distribution of the optical transmission line 2. The function P(z) calculated by the transmission characteristic calculation unit 15 and the function β(z) indicate the transmission characteristics of the optical transmission line 2.

[0054] The step size update unit 16 updates the new step size dz n Here, referring to FIG. 5, the function P(z) showing the estimated optical power distribution and the step size dz n The relationship between the above and the step sizes z1 to z8 will be explained. The graph in Fig. 5 shows an example in which the optical transmission line 2 is equally divided into eight sections, where N=8, that is, each of the eight step sizes dz1 to dz8 has the same length. As described above, the order of the step sizes dz1 to dz8 and the order of the center positions z1 to z8 of the step sizes dz1 to dz8 are arranged in order from the end point side of the optical transmission line 2. In the graph in Fig. 5, the horizontal axis represents the distance in the optical transmission line 2, as in Figs. 2(a), (b) and 3(a), and the vertical axis represents the magnitude of the estimated optical power.

[0055] The above-mentioned method of estimating transmission characteristics by SSFM and DBP, when considering the estimation of optical power distribution among the transmission characteristics to be estimated, can be said to be a method of estimating γ1P(z1)dz1 to γ8P(z8)dz8, which correspond to the area of ​​the dashed rectangle in each of the eight sections, as shown in the graph of Figure 5, and calculating the estimated optical power P(z1) to P(z8) by dividing by the nonlinear constants γ1 to γ8 and step sizes dz1 to dz8 at each of the eight positions z1 to z8. However, when the step sizes dz1 to dz n 16 occurs. Therefore, the step size update unit 16 updates new step sizes dz1 to dz2 based on the function P(z) indicating the estimated optical power distribution, the function β(z) indicating the estimated dispersion distribution, and the signal bandwidth of the optical transmission line 2. NThe new step sizes dz1 to dz2 applied in each of the N steps of the SSFM processor 31 are determined so that the proportional relationship shown in the following equation (4) holds under the condition that the sum of the new step sizes dz1 to dz2 is the distance "L" between the start point and the end point of the optical transmission line 2. N Detect combinations of

[0056]

number

[0057] In the above equation (4), "p" and "q" are arbitrary real numbers that are predetermined. "Bw" is the signal bandwidth of the optical transmission line 2, and is a predetermined value. dz(z) is a function that indicates the step size at an arbitrary position z in the optical transmission line 2. The step size update unit 16 updates the step size dz in accordance with the magnitude of the estimated optical power P(z) and the dispersion coefficient β(z) so as to satisfy the proportional relationship shown in equation (4). n By changing the length of "p" and "q" in equation (4), the width of the dashed rectangle shown in Figure 5, i.e., the width over which the sectional area is calculated, changes. This makes it possible to increase or decrease the estimated optical power P(z1) to P(z8) for each section. Therefore, by setting appropriate values ​​for "p" and "q" in equation (4), the estimated optical power in the section near the measurement dead zone can be made appropriate, and the accuracy of the estimation can be improved without increasing the optical power of the optical signal transmitted by the optical transmitter 3.

[0058] When the step size update unit 16 determines that the newly detected step size is optimal, the estimation result output unit 17 outputs to the outside the function P(z) and the function β(z) that indicate the transmission characteristics used to calculate the new step size.

[0059] (Processing by the optical receiving device of the first embodiment) Next, the processing by the optical receiving device 1 will be described with reference to Fig. 6 to Fig. 8. Fig. 6 is a flowchart showing the flow of processing by the optical receiving device 1. Before the processing in the flowchart of Fig. 6 starts, the following initial settings are performed in the optical receiving device 1.

[0060] A compensation coefficient calculated in advance is set in each of the adaptive equalizer 32, the frequency offset compensator 33, and the carrier phase noise compensator 34. As a calculation method for the compensation coefficients for each of the adaptive equalizer 32, the frequency offset compensator 33, and the carrier phase noise compensator 34, for example, a calculation method that has been used conventionally can be applied. Predetermined nonlinear constants γ1 to γ N and z inv A value indicating the distance "L" between the start point and the end point of the optical transmission line 2, which is used to convert the coordinate axis of x into the coordinate axis of z, is written in advance.

[0061] The control unit 60 of the SSFM processing unit 31 stores dispersion coefficients β1 to β N and the initial value of the nonlinear phase rotation amount φ1 to φ N The initial value and step size dz1~dz N The initial values ​​of the dispersion coefficients β1 to β N and nonlinear phase rotation amounts φ1 to φ N As the initial value of , any value determined arbitrarily may be applied. For example, the dispersion coefficients β1 to β N and nonlinear phase rotation amounts φ1 to φ N The initial value of each of the dispersion coefficients β1 to β N For step sizes dz1 to dz2, the total dispersion of the optical transmission line 2 estimated by a conventional method may be divided by the number of divisions N. N As the initial value of , the length of one section when the optical transmission line 2 is equally divided into N sections is applied.

[0062] Data of the training signal transmitted by the optical transmitting device 3 is written in advance in a storage area inside the coefficient updating unit 14. Two thresholds, each having appropriate values ​​determined in advance, are written in advance in a storage area inside the coefficient updating unit 14. The threshold for determining convergence of the dispersion coefficient and the threshold for determining convergence of the amount of nonlinear phase rotation are written in advance in a storage area inside the coefficient updating unit 14. The previous dispersion coefficients β1 to β N and the previous nonlinear phase rotation amount φ1 to φ N The control unit 60 has an area for storing the distribution coefficients β1 to β N and the initial value of the nonlinear phase rotation amount φ1 to φ N The initial values ​​of are written in advance.

[0063] A threshold value ε for determining convergence of the step size is written in advance in a storage area inside the step size update unit 16. The value of the threshold value ε is a positive real number, and an appropriate value for determining convergence of the step size is predetermined. Predetermined values ​​of "p" and "q", a predetermined value indicating the signal bandwidth "Bw" of the optical transmission line 2, and a value indicating the distance "L" between the start point and the end point of the optical transmission line 2 are written in advance in a storage area inside the step size update unit 16. Predetermined nonlinear constants γ1 to γ N is written in advance. An area for storing a counter k indicating the number of repetitions is provided in the internal storage area of ​​the step size update unit 16, and is initialized as k=1. The previous P k-1 (z1)~P k-1 (z N ) is stored in the area, and the nonlinear phase rotation amounts φ1 to φ N The initial values ​​of each of the nonlinear constants γ1 to γ N P0(z1) to P0(z N ) is written in advance as an initial value. The subscript under the letter P indicates the number of times the step size update unit 16 repeats the process.

[0064] After the above-described initial setting of the optical receiving device 1 is completed, the optical transmitting device 3 generates a polarization multiplexed QPSK optical signal using a predetermined training signal as a transmission signal, and transmits the generated polarization multiplexed QPSK optical signal to the optical receiving device 1 via the optical transmission line 2. The coherent receiving unit 11 of the optical receiving device 1 receives the polarization multiplexed QPSK optical signal transmitted by the optical transmission line 2. The coherent receiving unit 11 coherently detects the received polarization multiplexed QPSK optical signal to generate a received signal (step S1).

[0065] When the control unit 60 of the SSFM processing unit 31 receives the received signal output from the coherent receiving unit 11, the control unit 60 calculates the dispersion coefficients β1 to β2 stored in the internal storage area. N The initial value and step size dz1~dz N The control unit 60 sets each of the nonlinear phase rotation amounts φ1 to φ in the corresponding chromatic dispersion compensation units 42-1 to 42-N. N The initial value and step size dz1~dz N In the first process of step S2, the control unit 60 sets the step sizes dz1 to dz2 output by the step size update unit 16 to the nonlinear compensation units 50-1 to 50-N corresponding to the step sizes dz1 to dz2. N Since the step sizes dz1 to dz2 are not included, the step sizes dz1 to dz3 are stored in the internal memory. N In the second and subsequent processing of step S2, the control unit 60 sets the initial values ​​of the step sizes dz1 to dz2 output by the step size update unit 16 in the chromatic dispersion compensation units 42-1 to 42-N and the nonlinear compensation units 50-1 to 50-N. N Since the step size dz1 to dz N are set in the corresponding chromatic dispersion compensation units 42-1 to 42-N and nonlinear compensation units 50-1 to 50-N (step S2).

[0066] Thereafter, the SSFM processing unit 31 performs N steps of processing on the received signal output by the coherent receiving unit 11. The processing of the first step will be explained below. The Fourier transform unit 41-1 of the linear compensation unit 40-1 takes in the received signal output by the coherent receiving unit 11. Note that the taking in of the received signal by the Fourier transform unit 41-1 and the taking in of the received signal by the control unit 60 are performed in parallel. The Fourier transform unit 41-1 performs FFT on the taken in received signal and outputs the FFTed signal to the chromatic dispersion compensation unit 42-1. The chromatic dispersion compensation unit 42-1 takes in the FFTed signal output by the Fourier transform unit 41-1. After the control unit 60 has finished setting the dispersion coefficient β1 and the step size dz1, the chromatic dispersion compensation unit 42-1 performs the calculation of equation (2), that is, performs exp(jβ1 / 2ω 2 dz1) to perform dispersion compensation. The chromatic dispersion compensator 42-1 outputs the dispersion-compensated signal to the inverse Fourier transformer 43-1.

[0067] The inverse Fourier transform unit 43-1 accepts the dispersion-compensated signal output by the chromatic dispersion compensation unit 42-1, performs IFFT on the accepted dispersion-compensated signal, and outputs the IFFT-completed signal to the nonlinear compensation unit 50-1. The nonlinear compensation unit 50-1 accepts the IFFT-completed signal output by the inverse Fourier transform unit 43-1, and performs the calculation of equation (3) on the accepted IFFT-completed signal to compensate for nonlinear phase rotation. The nonlinear compensation unit 50-1 outputs the signal compensated for nonlinear phase rotation to the Fourier transform unit 41-2 of the linear compensation unit 40-2.

[0068] Thereafter, similar to the processing of the first step described above, the processing of the second step and thereafter is performed in step order by the linear compensation unit 40-n (where n = 2 to N) and the nonlinear compensation unit 50-n (where n = 2 to N). The nonlinear compensation unit 50-N in the Nth step outputs the signal after compensation for the nonlinear phase rotation to the adaptive equalization unit 32 (step S3).

[0069] The adaptive equalization unit 32 takes in the signal after compensation for nonlinear phase rotation output by the nonlinear compensation unit 50-N, and performs adaptive equalization processing on the taken signal after compensation for nonlinear phase rotation. The adaptive equalization unit 32 outputs the signal after adaptive equalization to the frequency offset compensation unit 33. The frequency offset compensation unit 33 takes in the signal after adaptive equalization output by the adaptive equalization unit 32, and performs frequency offset compensation processing on the taken signal after adaptive equalization. The frequency offset compensation unit 33 outputs the signal after frequency offset compensation to the carrier phase noise compensation unit 34. The carrier phase noise compensation unit 34 performs phase offset compensation processing on the signal after frequency offset compensation output by the frequency offset compensation unit 33. The carrier phase noise compensation unit 34 outputs the signal after phase offset compensation to the coefficient update unit 14 (step S4).

[0070] The coefficient update unit 14 receives the phase offset compensated signal output by the carrier phase noise compensation unit 34, applies an evaluation function to the received phase offset compensated signal and a training signal stored in an internal storage area, and generates new N dispersion coefficients β1 to β2 according to an optimization algorithm that minimizes the value of the evaluation function. N and N new nonlinear phase rotation amounts φ1 to φ N is calculated (step S5).

[0071] The coefficient update unit 14 updates the calculated new dispersion coefficients β1 to β N and the previous dispersion coefficients β1 to β N and the threshold value for determining convergence of the dispersion coefficient and the calculated new nonlinear phase rotation amount φ1 to φ N and the previous nonlinear phase rotation amounts φ1 to φ stored in the internal memory area. N and the threshold value for determining the convergence of the nonlinear phase rotation amount, the dispersion coefficients β1 to β N and the nonlinear phase rotation amount φ1 to φ N The coefficient update unit 14 determines whether or not the new dispersion coefficients β1 to β N Each of these and the corresponding previous dispersion coefficients β1 to β NIf the calculated sum of squared errors is less than a threshold value for determining convergence of the dispersion coefficients, the dispersion coefficients β1 to β N It is determined that the new nonlinear phase rotation amounts φ1 to φ N and the corresponding previous nonlinear phase rotation amounts φ1 to φ N and if the calculated sum of squared errors is less than a threshold value for determining convergence of the nonlinear phase rotation amount, the nonlinear phase rotation amounts φ1 to φ N The coefficient update unit 14 determines that the dispersion coefficients β1 to β N and the nonlinear phase rotation amount φ1 to φ N If both of the dispersion coefficients β1 to β N and the nonlinear phase rotation amount φ1 to φ N It may be determined that the convergence has occurred when either one of the above has converged. Alternatively, a value indicating an error other than the sum of squared errors may be calculated as a comparison target for determining the convergence (step S6).

[0072] When it is determined in step S6 that convergence has not occurred (step S6, No), the coefficient update unit 14 updates new dispersion coefficients β1 to β N The previous distribution coefficients β1 to β N is written into the storage area, and new nonlinear phase rotation amounts φ1 to φ N The previous nonlinear phase rotation amount φ1 to φ in the internal storage area is N The coefficient update unit 14 writes the new distribution coefficients β1 to β N and the new nonlinear phase rotation amounts φ1 to φ N and output to the control unit 60 of the SSFM processing unit 31.

[0073] The control unit 60 updates the new dispersion coefficients β1 to β N and the new nonlinear phase rotation amounts φ1 to φ N The coefficient update unit 14 takes in the dispersion coefficients β1 to β2 set in the chromatic dispersion compensation units 42-1 to 42-N. N Each of the following is the corresponding dispersion coefficient β1 to β NThe coefficient updating unit 14 updates the nonlinear phase rotation amounts φ1 to φ set in the nonlinear compensation units 50-1 to 50-N. N The nonlinear phase rotation amounts φ1 to φ N When updating, the step sizes dz1 to dz2 set in the chromatic dispersion compensators 42-1 to 42-N and the nonlinear compensation units 50-1 to 50-N are rewritten as follows: N is not updated (step S7). After that, the process of step S3 is performed.

[0074] On the other hand, when it is determined in step S6 that convergence has occurred (step S6, Yes), the coefficient update unit 14 updates the previous distribution coefficients β1 to β N The area storing the dispersion coefficients β1 to β N The initial value of φ1 to φ2 is written, and the previous nonlinear phase rotation amount φ1 to φ2 is written in the internal memory area. N The nonlinear phase rotation amounts φ1 to φ are stored in the area N For example, the coefficient update unit 14 performs an initialization process by writing the initial values ​​of the dispersion coefficients β1 to β N and the initial value of the nonlinear phase rotation amount φ1 to φ N The initial values ​​of the dispersion coefficients β1 to β2 are stored in a separate area of ​​the internal storage area. N , and the nonlinear phase rotation amounts φ1 to φ N The coefficient update unit 14 performs initialization using the initial values ​​of the new N dispersion coefficients β1 to β N and N new nonlinear phase rotation amounts φ1 to φ N and are output to the transmission characteristic calculation unit 15.

[0075] The transmission characteristic calculation unit 15 calculates the new N dispersion coefficients β1 to β N and N new nonlinear phase rotation amounts φ1 to φ N The transmission characteristic calculation unit 15 takes in the new N dispersion coefficients β1 to β N The transmission characteristic calculation unit 15 calculates a function β(z) that indicates the estimated dispersion distribution of the optical transmission line 2 from the above-mentioned procedure, and calculates the new N nonlinear phase rotation amounts φ1 to φN and the nonlinear constants γ1 to γ N From this, the step size update unit 16 calculates a function P(z) that indicates the estimated optical power distribution of the optical transmission line 2. The transmission characteristic calculation unit 15 outputs the calculated function β(z) and function P(z) to the step size update unit 16 as the transmission characteristics of the optical transmission line 2 (step S8).

[0076] The step size updater 16 takes in the function β(z) indicating the transmission characteristics output by the transmission characteristic calculator 15 and the function P(z), and starts the step size update process subroutine shown in FIG. 7 (step S9).

[0077] The step size update unit 16 reads out "p", "q", the signal bandwidth "Bw" of the optical transmission line 2, and the distance "L" between the start point and the end point of the optical transmission line 2 from an internal storage area. The step size update unit 16 generates a function shown in the right-hand side of equation (4) using the imported function β(z) and function P(z) and the read out "p", "q", and "Bw". The step size update unit 16 generates new step sizes dz1 to dz N Under the condition that the sum of these is the distance "L" between the start point and the end point of the optical transmission line 2, a new step size dz1 to dz2 that satisfies equation (4) is calculated. N For example, the combination of dz1 to dz N The entire value is detected while increasing or decreasing (step Sa1).

[0078] For example, assume that the function P(z) representing the estimated optical power distribution acquired by the step size update unit 16 is P1(z) shown in Fig. 8(a) and (b). Fig. 8(a) and (b) show the case where N=8, and Fig. 8(a) shows the case where the step sizes dz1 to dz8 are all the same length as the initial values. 01 ~dz 08 This shows the case where step size dz is applied. 01 ~dz 08 The center position z of each 01 ~z 08 The estimated optical power P1(z 01 )~P1(z08 ) and the nonlinear phase rotation amounts φ1 to φ8 acquired by the transmission characteristic calculation unit 15 are expressed as follows: φ1=γ1×P1(z 01 ),…,φ8=γ8×P1(z 08 ) is the relationship.

[0079] Here, the step size update unit 16 sets the step sizes dz1 to dz8 shown in FIG. 8(b). 11 ~dz 18 In this case, the step size dz 11 ~dz 18 The center position of z 11 ~z 18 and the original z 01 ~z 08 Therefore, z 11 ~z 18 The estimated optical power P1(z 11 )~P1(z 18 ) and z 11 ~z 18 z corresponding to 01 ~z 08 The estimated optical power P1(z 01 )~P1(z 08 ) will have different values.

[0080] Returning to FIG. 7, the step size update unit 16 updates the detected new step sizes dz1 to dz N From the new z1~z N is calculated (step Sa2).

[0081] Returning to FIG. 6, the step size update unit 16 retrieves the previous estimated optical power P from the internal storage area. k-1 (z1)~P k-1 (z N ) is read out. In the first case, P0(z1) to P0(z N ) is written as the initial value, the step size update unit 16 updates P0(z1) to P0(z N ) will be read out.

[0082] The step size update unit 16 calculates new z1 to z2 in step Sa2 of the step size update processing subroutine. N Substituting into the function P(z), P k (z1)~P k (z N ) and calculate the calculated P k (z1)~P k (z N ) and the read P k-1 (z1)~P k-1 (z N ) and then determine whether the following equation (5) is satisfied.

[0083]

number

[0084] That is, the step size update unit 16 updates |P k (z1)-P k-1 (z1)|,…,|P k (z N )-P k-1 (z N )|, and determines whether the maximum value among the results of the calculations is less than a threshold ε stored in an internal storage area. If the formula (5) is satisfied, the step size update unit 16 updates the new step sizes dz1 to dz N On the other hand, if the formula (5) is not satisfied, the step size update unit 16 determines that the new step sizes dz1 to dz N has not converged and is not in an optimal state (step S10).

[0085] The step size update unit 16 does not satisfy the equation (5), that is, the new step sizes dz1 to dz N If it is determined that the step size dz1 to dz2 have not converged (step S10, No), the step size update unit 16 adds 1 to the value of the counter k stored in the internal storage area and sets the new value of the counter k as the step size dz1 to dz2 detected in step Sa1 of the subroutine for the step size update process.N to the control unit 60 of the SSFM processing unit 31 (step S11). After that, the process proceeds to step S2.

[0086] On the other hand, in step S10, the step size update unit 16 determines whether the formula (5) is satisfied, that is, the new step sizes dz1 to dz N When it is determined that the step size update unit 16 has converged and reached an optimal state (Yes in step S10), the step size update unit 16 outputs the currently acquired transmission characteristics, i.e., the function P(z) indicating the estimated optical power distribution and the function β(z) indicating the estimated variance distribution, to the estimation result output unit 17. The estimation result output unit 17 acquires the functions P(z) and β(z) output by the step size update unit 16, and outputs the acquired functions P(z) and β(z) to the outside as the estimated results of the transmission characteristics (step S12). This ends the processing.

[0087] (Experimental results) 9 and 10 are graphs showing the experimental results when the optical transmission line 2 includes four 70 km optical fibers and three optical amplifiers inserted between the four optical fibers. In this experiment, the function β(z) indicating the dispersion distribution is set to a constant value, so Equation (4) can be replaced by the following Equation (6):

[0088]

number

[0089] In this experiment, four values ​​of "p" were used: "0," "-0.5," "-1," and "-1.5." The graph in FIG. 9 shows the change in function dz(z), which indicates the step size for each position z in the optical transmission line 2, for each value of p. In the graph in FIG. 9, the horizontal axis represents distance in units of km, and the origin is the start point of the optical transmission line 2. The vertical axis represents the value of function dz(z), which indicates the step size, also in units of km. In FIG. 9, the dashed-dotted line graph corresponds to the case where p=0, the dotted line graph corresponds to the case where p=-0.5, the dashed line graph corresponds to the case where p=-1, and the dashed-dotted line graph corresponds to the case where p=-1.5.

[0090] In the graph of FIG. 9, when p=0, the right side of equation (6) becomes a constant, so dz(z) is a constant value, that is, the step sizes dz1 to dz N 10 shows a case where each of the lines has the same length.

[0091] The graph in Fig. 10 shows the change in absolute power for each position z in the optical transmission line 2 for each value of p. In the graph in Fig. 10, the horizontal axis represents distance in units of [km], as in the graph in Fig. 9, and the origin is the start point of the optical transmission line 2. The vertical axis represents the magnitude of optical power in units of [dBm]. Here, when an OTDR is used, the optical power represents the value obtained by adding the OTDR measurement value to the optical fiber input power, and when the optical receiving device 1 of this embodiment is used, the step sizes dz1 to dz2 output by the estimation result output unit 17 are used. N is the value of the function P(z) of estimated optical power when is optimal. In Fig. 10, the solid line graph shows the OTDR measurement value, and the graphs other than the solid line show changes in the function P(z) of estimated optical power. In Fig. 10, the correspondence between the type of each line (two-dot chain line graph, dotted line graph, dashed line graph, and one-dot chain line graph) and the value of p is the same as in the graphs in Fig. 9.

[0092] As can be seen from the graph in Figure 10, the change in the dashed line graph for p = -1 tracks the change in the OTDR measurement value shown by the solid line, even near the optical amplifier, which is the measurement dead zone. In contrast, for values ​​other than p = -1, it deviates from the change in the OTDR measurement value shown by the solid line. Therefore, when p = -1, the measurement dead zone can be reduced, and an estimated optical power distribution is obtained that closely matches the true optical power distribution measured by OTDR, demonstrating improved estimation accuracy compared to the case of p = 0, which corresponds to the conventional DLM.

[0093] The optical transmission characteristic estimator 12 included in the optical receiving device 1 of the first embodiment models the optical effect of the optical transmission line 2 using a nonlinear Schrödinger equation, divides the optical transmission line 2 into N sections, and estimates the transmission characteristics of the optical transmission line by solving the nonlinear Schrödinger equation through calculations of N steps, with each step size being the length of each of the N divided sections. In the optical transmission characteristic estimator 12, the step size updater 16 updates each of the N step sizes based on the estimated optical power distribution and estimated variance distribution indicated by the transmission characteristics. This makes it possible to obtain estimation results with appropriate accuracy in the measurement dead zone when estimating the characteristics of the optical transmission line using DLM, even when using an optical signal with low optical power output that is used during normal operation.

[0094] This is also supported by the graphs in Figures 9 and 10. That is, the step size updater 16 reduces the step size when the estimated optical power is large, and conversely, increases the step size when the estimated optical power is small. As described above, increasing the step size increases the width of the piecewise quadrature, thereby increasing the value obtained by piecewise quadrature. That is, increasing the step size increases the amount of nonlinear phase rotation. Conversely, decreasing the step size decreases the amount of nonlinear phase rotation. In other words, by adjusting the step size, the step size updater 16 keeps the amount of nonlinear phase rotation occurring in the optical transmission line 2 close to a constant value. As a result, for example, if the received signal contains noise and the optical power is low, the amount of nonlinear phase rotation is likely to be buried in the noise. However, by changing the step size and increasing the amount of nonlinear phase rotation as described above, the amount of nonlinear phase rotation is less likely to be buried in the noise, thereby improving the estimation accuracy.

[0095] (Second embodiment) 11 is a block diagram showing the configuration of an optical receiving device 1a according to the second embodiment. The optical receiving device 1a is a device used in place of the optical receiving device 1 included in the optical transmission system 100 of the first embodiment, and is connected to an optical transmission line 2 to receive an optical signal transmitted by an optical transmitting device 3 via the optical transmission line 2. In the optical receiving device 1a, the same components as those in the optical receiving device 1 of the first embodiment are denoted by the same reference numerals, and only the different components will be described below.

[0096] The optical receiving device 1a includes a compensation processing unit 13, a coefficient updating unit 14, a transmission characteristics calculation unit 15, a step size updating unit 16a, and an estimation result output unit 17. The optical receiving device 1a is intended to be used when it is known that there is no abnormal loss in the optical transmission line 2. Here, the abnormal loss refers to a loss that causes a sudden drop in optical power at a certain position in the optical transmission line 2 due to an optical fiber in the optical transmission line 2 being bent midway or on the verge of breaking.

[0097] The step size updater 16a has the same configuration as the step size updater 16 of the first embodiment, except for the configuration described below. When there is no abnormal loss in the optical transmission line 2, the optical power attenuates as the transmission distance increases due to loss in the optical fiber. Therefore, the step size updater 16a calculates the change in loss in the optical transmission line 2, i.e., the function α(z) indicating the loss distribution, from the remaining distribution obtained by removing the measurement dead zone from the estimated optical power distribution indicated by the function P(z) calculated by the transmission characteristics calculation unit 15.

[0098] The step size update unit 16a uses the following equation (7) to which the calculated function α(z) is applied instead of equation (4), and calculates new step sizes dz1 to dz N The step size dz1 to dz2 is set under the condition that the sum of these is the distance "L" between the start point and the end point of the optical transmission line 2. N Detect combinations of

[0099]

number

[0100] (Processing by the optical receiving device of the second embodiment) The processing by the optical receiving device 1a of the second embodiment is processing in which the subroutine for step size update processing in step S9 in the flowchart of Fig. 6 showing the processing flow of the optical receiving device 1 of the first embodiment is replaced with processing of the subroutine shown in Fig. 12. Therefore, in Fig. 6, the processing other than step S9 is the same as that in the first embodiment.

[0101] The process of step S9 in the second embodiment will be described below with reference to the flowchart in Fig. 12. The step size update unit 16a takes in the function P(z) and the function β(z) that indicate the transmission characteristics output by the transmission characteristic calculation unit 15. The step size update unit 16a removes the measurement dead zone portion from the taken-in function P(z), for example, by the following procedure.

[0102] The measurement dead zone is a portion where the value of the function P(z) increases suddenly, as shown in the graph of FIG. 16(b). The step size update unit 16a stores in advance in an internal storage area a predetermined threshold value for the increase rate, which is determined based on the increase rate in the measurement dead zone. Based on the threshold value for the increase rate stored in the internal storage area, the step size update unit 16a removes from the function P(z) a portion where an increase that exceeds the threshold value for the increase rate occurs.

[0103] For example, assume that the function P(z) exhibits a distribution as shown in the graph of FIG. 13. The values ​​indicated by the vertical and horizontal axes in the graph of FIG. 13 are the same as those in the graph of FIG. 2(a). In this case, by using the above-described threshold value for the increase rate, the step-size update unit 16a removes the portion indicated by the dashed line 81, which is included in the measurement dead zone portion 80, from the function P(z). However, simply removing the portion indicated by the dashed line 81 from the function P(z) makes the function P(z) discontinuous. Therefore, the step-size update unit 16a performs an interpolation as shown by the dotted line 82 by extending the remaining portion of the function P(z) according to the slope of the function P(z), thereby achieving a continuous, uninterrupted change with respect to the distance z. After performing this interpolation, the step-size update unit 16a calculates the function α(z) indicating the loss distribution (step Sb1).

[0104] The step size update unit 16a reads out "p", "q", the signal bandwidth "Bw" of the optical transmission line 2, and the distance "L" between the start point and the end point of the optical transmission line 2 from an internal storage area. The step size update unit 16a generates a function represented by the right-hand side of equation (7) using the calculated function α(z), the imported function β(z), and the read "p", "q", and "Bw". The step size update unit 16a generates new step sizes dz1 to dz N Under the condition that the sum of these is the distance "L" between the start point and the end point of the optical transmission line 2, a new step size dz1 to dz2 that satisfies equation (7) is calculated. N For example, the combination of dz1 to dz N The entire value is detected while increasing or decreasing (step Sb2).

[0105] Thereafter, the step size update unit 16a performs the same process as that of step Sa2 in FIG. 7 (step Sb3).

[0106] As described above, the function α(z) showing the loss distribution obtained by removing the measurement dead zone from the function P(z) and performing interpolation shows a change that is close to the true loss distribution. -α(z)z / 10 " is a function that shows the rate of change in optical power expressed using the function α(z) that shows the loss distribution. The more accurately the function α(z) approximates the true loss distribution, the more accurately the "10 -α(z)z / 10 " approximates the rate of change in the true optical power. Therefore, the step size updater 16a of the second embodiment can set more appropriate step sizes dz1 to dz2 than the step size updater 16 of the first embodiment. N This allows for more rapid detection.

[0107] In the second embodiment, the step-size update unit 16a calculates the function α(z) indicating the loss distribution by removing the measurement dead zone from the function P(z) and performing interpolation. In contrast, the function α(z) is disclosed in advance as a specification for each type of optical fiber. For example, Non-Patent Document 2 discloses examples of the function α(z) as 0.199 dB / km, 0.230 dB / km, and 0.225 dB / km for SSMF (Standard Single Mode Fiber), DSF (Dispersion Shifted Fiber), and NZ-DSF (Non-Zero Dispersion Shifted Fiber), respectively. In this way, if the typical value of the function α(z) for each of the multiple optical fibers included in the optical transmission line 2 is known, the typical values ​​of the function α(z) for each of the multiple optical fibers may be concatenated without any discontinuity at the distance z to generate the function α(z) indicating the overall loss distribution of the optical transmission line 2 in advance, and the generated function α(z) may be applied to the step-size update unit 16a.

[0108] In the first and second embodiments, P(z) in equation (4) and α(z) in equation (7) are unknown in the initial processing, so the step sizes dz1 to dz2 having the same length are used. N is used as an initial value, and in the initial processing of step S8, a function P(z) is acquired as a rough profile of the estimated optical power distribution of the optical transmission line 2. In contrast to this, a case is assumed where a function α(z) indicating the overall loss distribution of the optical transmission line 2 is used, which is generated by concatenating functions α(z) previously disclosed as specifications for each type of optical fiber as described above. In this case, the step sizes dz1 to dz N As the initial value of the step size dz1 to dz2, which have different lengths detected in advance according to the overall loss distribution of the optical transmission line 2, N can be applied, it is possible to reduce the time required to estimate the transmission characteristics.

[0109] (Another example of the optical receiving device (part 1)) 1, the order of the SSFM processing unit 31, adaptive equalization unit 32, frequency offset compensation unit 33, and carrier phase noise compensation unit 34 in the compensation processing unit 13 may be changed as follows. For example, the adaptive equalization unit 32 may be connected to the output side of the coherent receiving unit 11, the frequency offset compensation unit 33 may be connected to the output side of the adaptive equalization unit 32, the carrier phase noise compensation unit 34 may be connected to the output side of the frequency offset compensation unit 33, and the SSFM processing unit 31 may be connected to the output side of the carrier phase noise compensation unit 34. In other words, the location of the SSFM processing unit 31 is arbitrary. In a state where the coherent receiving unit 11, adaptive equalization unit 32, frequency offset compensation unit 33, and carrier phase noise compensation unit 34 are connected in this order, the SSFM processing unit 31 may be inserted between the adaptive equalization unit 32 and the frequency offset compensation unit 33, or between the frequency offset compensation unit 33 and the carrier phase noise compensation unit 34. However, when the connection order is changed as described above, the coefficient update unit 14 takes in the signal output by the last functional unit of the rearranged compensation processing unit 13. When the order of the functional units is changed in this way, the order of the processes shown in steps S3 and S4 in the flowchart of Fig. 6 is changed in accordance with the order of the functional units.

[0110] (Another example of the optical receiving device (part 2)) An optical receiving device 1b shown in Fig. 14 may be used instead of the optical receiving device 1 shown in Fig. 1. In the optical receiving device 1b, the same components as those in the optical receiving device 1 are assigned the same reference numerals, and the different components will be described below. The optical receiving device 1b includes a compensation processing unit 13b instead of the compensation processing unit 13, a coefficient updating unit 14b instead of the coefficient updating unit 14, and further includes a demodulation unit 18. Like the compensation processing unit 13, the compensation processing unit 13b includes an SSFM processing unit 31, an adaptive equalization unit 32, a frequency offset compensation unit 33, and a carrier phase noise compensation unit 34, but the adaptive equalization unit 32 is not connected to the SSFM processing unit 31, and instead the coefficient updating unit 14b is connected to the output side of the SSFM processing unit 31, and the adaptive equalization unit 32 is connected to the output side of the coefficient updating unit 14b.

[0111] In the case of the optical receiving device 1b, the signal output by the SSFM processing unit 31 cannot be compared with the training signal. Therefore, it is necessary to generate a signal for comparison in advance by giving the training signal characteristics and inverse characteristics of the compensation by the adaptive equalization unit 32, frequency offset compensation unit 33, and carrier phase noise compensation unit 34, and store the previously generated comparison signal in an internal storage area of ​​the coefficient update unit 14b. When the coefficient update unit 14b takes in the signal output by the SSFM processing unit 31, it calculates new dispersion coefficients β1 to β2 by an optimization algorithm that minimizes the value of the evaluation function based on the taken-in signal and the comparison signal stored in the internal storage area. N and the new nonlinear phase rotation amounts φ1 to φ N The following calculation is made:

[0112] When the coefficient update unit 14b receives a signal output by the SSFM processing unit 31, it outputs the received signal as is to the adaptive equalization unit 32. Therefore, it is sufficient that the processing of step S5 is performed after the processing of step S3 shown in Fig. 6, and the processing of step S4 is no longer essential from the viewpoint of DLM processing.

[0113] The demodulator 18 demodulates the transmission signal from the transmission symbols included in the phase-offset-compensated signal output by the carrier phase noise compensator 34. Note that the demodulator 18 is a functional unit that demodulates the transmission signal from the transmission symbols restored from the received signal, and therefore may be provided in the optical receiving device 1 of the first embodiment. In that case, the demodulator 18 is connected to the output side of the carrier phase noise compensator 34.

[0114] (Another example of the optical receiving device (part 3)) Instead of the optical receiving device 1 in Fig. 1, an optical receiving device 1c shown in Fig. 15 may be used, and an optical transmission characteristic estimation device 12c may be connected to the optical receiving device 1c. The optical transmission characteristic estimation device 12c is, for example, a device provided as one functional unit of a network control device, and is connected to the optical receiving device 1c via a communication network. In the optical receiving device 1c and the optical transmission characteristic estimation device 12c, the same components as those in the optical receiving device 1 and the optical receiving device 1b are denoted by the same reference numerals, and only the different components will be described below.

[0115] The optical receiving device 1c includes a coherent receiving unit 11 connected to the optical transmission line 2, a compensation processing unit 13, and a demodulation unit 18. The optical transmission characteristic estimation device 12c includes a compensation processing unit 13, a coefficient updating unit 14, a transmission characteristic calculation unit 15, a step size updating unit 16c, and an estimation result output unit 17. The SSFM processing unit 31 of the compensation processing unit 13 of the optical transmission characteristic estimation device 12c is connected to the output side of the coherent receiving unit 11 of the optical receiving device 1c, and takes in the received signal output by the coherent receiving unit 11. The step size updating unit 16c has the same configuration as the step size updating unit 16 of the first embodiment, and further, in the processing of step S12, the step sizes dz1 to dz2 calculated in the immediately preceding step S9 are updated. N and dispersion coefficients β1 to β N and the nonlinear phase rotation amount φ1 to φ N The SSFM processing unit 31 of the optical receiving device 1c outputs the step sizes dz1 to dz2 determined to be optimal by the optical transmission characteristic estimation device 12c to the control unit 60 of the SSFM processing unit 31 of the optical receiving device 1c. N and the step size dz1~dz N The dispersion coefficients β1 to β N and nonlinear phase rotation amounts φ1 to φ N Based on this, it becomes possible to compensate for chromatic dispersion and nonlinear phase rotation.

[0116] In addition, other configuration examples similar to the other configuration examples (1), (2), and (3) corresponding to the optical receiving device 1 of the first embodiment may be configured based on the optical receiving device 1a of the second embodiment.

[0117] In the above-described first and second embodiments and other configuration examples, the step size update units 16, 16a, and 16c update the step sizes dz1 to dz2 in step S10. N When determining the convergence of β, the determination is made using equation (5). However, instead of equation (5), the following equation (8) using the dispersion coefficient as an index for determination may be applied to the determination process in step S10. In this case, however, the previous β is stored in the internal storage area of ​​the step size update units 16, 16a, and 16c. k-1 (z1)~β k-1 (z N ) is stored in the area, and β0(z1) to β0(z N ) must be written in advance as an initial value. The subscript under the letter β indicates the number of times the process is repeated by the step size update units 16, 16a, and 16c. When both equations (5) and (8) are satisfied, the step size update units 16, 16a, and 16c update the step sizes dz1 to dz N It may be determined that the step sizes dz1 to dz2 have converged, or when either the formula (5) or the formula (8) is satisfied, N It may be determined that the convergence has occurred.

[0118]

number

[0119] The step size update units 16, 16a, and 16c may apply, instead of equation (5), equation (9) in which the mean square error of the estimated optical power is used as an index for determination in the determination process of step S10.

[0120]

number

[0121] The step size update units 16, 16a, and 16c may apply the following equations (10) and (11) to the determination process in step S10, instead of equation (5).

[0122]

number

[0123]

number

[0124] In the above equations (10) and (11), z m is a position on the optical transmission line 2 where the true optical power can be measured, such as the start point or end point of each of the multiple optical fibers included in the optical transmission line 2. At such positions, the true input optical power, output optical power, etc. can be measured by using other monitoring means such as an OTDR. In equations (10) and (11), P true (z m ) indicates the true optical power obtained by measurement. In equations (10) and (11), m is an integer between 1 and M, where M is an integer equal to or greater than 1. The larger the value of M is, that is, the more positions at which the true optical power can be measured, the higher the accuracy of the determination in step S10 can be.

[0125] In the first and second embodiments and other configuration examples described above, the step sizes dz1 to dz2 stored in the internal storage area of ​​the control unit 60 of the SSFM processing unit 31 are N The initial value of step size dz1 to dz2 is the length of one section when the optical transmission line 2 is equally divided into N sections. N As the initial value of step size dz1~dz N A randomly selected length may be applied under the condition that the sum of these becomes the distance "L" between the start point and the end point of the optical transmission line 2.

[0126] In the first and second embodiments and other configuration examples described above, the step size update units 16, 16a, and 16c calculate new step sizes dz1 to dz2 by binary search. NHowever, by using a search method other than binary search, new step sizes dz1 to dz N may be detected.

[0127] For example, it is assumed that the magnitude of the optical power of the optical signal transmitted by the optical transmitter 3 is known, the type of optical fiber provided in the optical transmission line 2 is known, and the function α(z) and nonlinear constant indicating the loss distribution for each type of optical fiber are disclosed as specifications. In this case, the nonlinear phase rotation amounts φ1 to φ2 to be stored in the internal storage area of ​​the control unit 60 of the SSFM processing unit 31 of the optical receiving devices 1 and 1a of the first and second embodiments and the optical receiving device 1b of another configuration example, and the optical transmission characteristic estimation device 12c are N As the initial value of , the magnitude of the known optical power and the nonlinear constants γ1 to γ ​​that can be determined from the disclosed function α(z) and the disclosed nonlinear constants N The nonlinear phase rotation amounts φ1 to φ can be calculated from N Similarly, when the type of optical fiber included in the optical transmission line 2 is known and the dispersion coefficient for each type of optical fiber is disclosed as a specification, the dispersion coefficients β1 to β2 stored in the internal storage area of ​​the control unit 60 can be calculated. N The initial values ​​of the dispersion coefficients β1 to β2 can be determined from the disclosed dispersion coefficients. N may be applied.

[0128] In the first and second embodiments and other configuration examples described above, the nonlinear constants γ1 to γ N However, as mentioned above, the type of optical fiber is unknown, and the nonlinear constants γ1 to γ N cannot be determined in advance, the transmission characteristic calculation unit 15 cannot calculate the estimated optical power P(z). In this case, the transmission characteristic calculation unit 15 calculates the nonlinear phase rotation amounts φ1 to φ, which are proportional to the estimated optical power P(z), instead of the estimated optical power P(z). NThe step size update units 16 and 16a perform processing using the function φ(z) output by the transmission characteristic calculation unit 15 instead of the function P(z).

[0129] In the above-described first and second embodiments and other configuration examples, the Fourier transform units 41-1 to 41-N of the SSFM processing unit 31 are described as performing an FFT, but may be configured to perform a Fourier transform other than an FFT, for example, a DFT (Discrete Fourier Transform). Similarly, the inverse Fourier transform units 43-1 to 43-N of the SSFM processing unit 31 are described as performing an IFFT, but may be configured to perform an inverse Fourier transform other than an IFFT, for example, an IDFT (Inverse Discrete Fourier Transform).

[0130] In the above-described first and second embodiments and other configuration examples, the optical transmitting device 3 transmits a polarization multiplexed QPSK optical signal, but may transmit an optical signal optically modulated by another modulation method. In this case, the coherent receiving unit 11 performs detection by a demodulation method corresponding to the other modulation method applied to the optical transmitting device 3.

[0131] In the first and second embodiments and other configuration examples described above, the SSFM processor 31 executes SSFM, but the solution to the nonlinear Schrödinger equation may be obtained by a numerical analysis method other than SSFM. For example, an approximation model based on the perturbation method of the nonlinear Schrödinger equation may be used.

[0132] In the first and second embodiments and other configuration examples described above, the SSFM processor 31 estimates transmission characteristics by applying SSFM to the received signal, performing back propagation, and restoring the transmitted signal. Alternatively, the SSFM processor 31 may apply SSFM to the transmitted signal, performing forward propagation, and restoring the received signal.

[0133] In the configurations of the first and second embodiments described above, a determination process using an inequality sign is performed in the determination processes shown in steps S6 and S10 of the first embodiment and in the process of step Sb1 of the subroutine of step S9 of the second embodiment. However, the present invention is not limited to these embodiments, and the determination processes of "exceeding or not" and "below or not" are merely examples, and may be replaced with determination processes of "greater than or equal to or not" and "less than or equal to or not" depending on how the thresholds are defined. The thresholds used in the determination processes are also merely examples, and different thresholds may be applied in each case.

[0134] The optical transmission characteristic estimating units 12, 12a, 12b and the optical transmission characteristic estimating device 12c included in the optical receiving devices 1, 1a, 1b in the first and second embodiments and other configuration examples described above may be implemented by a computer. In this case, a program for implementing this function may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read and executed by a computer system. Note that the term "computer system" as used herein includes hardware such as an operating system (OS) and peripheral devices. Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, read-only memories (ROMs), and CD-ROMs, as well as storage devices such as hard disks built into a computer system. Furthermore, the term "computer-readable recording medium" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs via a network such as the Internet or communication lines such as telephone lines, and devices that store programs for a certain period of time, such as volatile memory within a computer system that serves as a server or client in such cases. Furthermore, the above program may be one that realizes part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in a computer system, or may be one that is realized using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0135] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Industrial Applicability]

[0136] The present invention can be applied to compensation processing in digital coherent optical transmission systems. [Explanation of symbols]

[0137] 1...optical receiving device, 2...optical transmission path, 3...optical transmitting device, 11...coherent receiving unit, 12...optical transmission characteristic estimating unit, 13...compensation processing unit, 14...coefficient updating unit, 15...transmission characteristic calculating unit, 16...step size updating unit, 17...estimation result output unit, 31...SSFM processing unit, 32...adaptive equalization unit, 33...frequency offset compensating unit, 34...carrier phase noise compensating unit, 100...optical transmission system

Claims

1. 1. An optical transmission characteristic estimation device that models an optical effect of an optical transmission line by a nonlinear Schrodinger equation, divides the optical transmission line into N sections (N is an integer of 2 or more), and finds a solution to the nonlinear Schrodinger equation by performing N-step calculations, with each step size being the section length of the N divided sections, to estimate transmission characteristics of the optical transmission line, a step size updating unit that updates each of the N step sizes based on an estimated optical power distribution and an estimated variance distribution indicated by the transmission characteristics; An optical transmission characteristic estimation device comprising:

2. The step size update unit detect new N step sizes so that a rate of change in the step sizes obtained by arranging the N step sizes in order and linearly interpolating them is proportional to a rate of change indicated by a third distribution obtained by multiplying a first distribution obtained by raising the estimated optical power distribution to a predetermined exponent p and a second distribution obtained by multiplying the estimated dispersion distribution by a square of the signal bandwidth of the optical transmission line and then raising the result to a predetermined exponent q; The optical transmission characteristics estimation device according to claim 1 .

3. The step size update unit If the dispersion coefficient indicated by the estimated dispersion distribution of the optical transmission line is the same value at any position of the optical transmission line, a power p is set to −1 to detect new N step sizes. The optical transmission characteristics estimation device according to claim 2 .

4. a split-step Fourier method processing unit that applies a split-step Fourier method, which is a solution method for the nonlinear Schrodinger equation, to the received signal after coherent detection as a method for performing the N steps of calculation; The split-step Fourier method processing unit applying the split-step Fourier transform to the received signal, with each of the N initial step size values ​​being the length of one section when the length of the optical transmission line is equally divided into N sections; The optical transmission characteristic estimation device according to any one of claims 1 to 3.

5. The step size update unit updating each of the N step sizes using a loss distribution of the optical transmission line instead of the estimated optical power distribution indicated by the transmission characteristics; The optical transmission characteristic estimation device according to claim 1 .

6. The step size update unit As the loss distribution of the optical transmission line, a loss distribution calculated in advance from the remaining distribution obtained by removing the measurement dead zone portion from the estimated optical power distribution is used, or a loss distribution predetermined for each type of optical fiber included in the optical transmission line is used. The optical transmission characteristics estimation device according to claim 5 .

7. 1. An optical transmission characteristic estimation method for estimating transmission characteristics of an optical transmission line, comprising: modeling an optical effect of an optical transmission line by a nonlinear Schrodinger equation; dividing the optical transmission line into N sections (N is an integer of 2 or more); and calculating N steps in which each of the N divided sections has a step size equal to the section length, thereby obtaining a solution to the nonlinear Schrodinger equation and estimating transmission characteristics of the optical transmission line, updating each of the N step sizes based on an estimated optical power distribution and an estimated variance distribution indicated by the transmission characteristics; Optical transmission characteristics estimation method.

8. On the computer, a step of modeling an optical effect of an optical transmission line by a nonlinear Schrodinger equation and dividing the optical transmission line into N sections (N is an integer equal to or greater than 2); a step of estimating the transmission characteristics of the optical transmission line by solving the nonlinear Schrodinger equation through calculations of N steps, each step size being the length of one of the N divided sections; updating each of the N step sizes based on an estimated optical power distribution and an estimated variance distribution indicated by the transmission characteristics; A program to execute.

Images