Optical transmission characteristic estimation device

Abstract

An optical transmission characteristic estimation device according to the present invention comprises: a matrix calculation unit that calculates a matrix representing the characteristics of a virtual transmission path on the basis of a transmission signal, a known signal, or a reception signal input to an optical transmission path; and an estimation unit that, on the basis of the matrix and a reception signal obtained by receiving the transmission signal or the known signal by a coherent detection scheme or a signal obtained on the basis of the reception signal, estimates an optical power distribution in the transmission path by using a linear least squares method with a regularization term including a real number λ.　

Description

Optical transmission characteristics estimation device

[0001] The present invention relates to an optical transmission characteristics estimation device.

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

[0003] Therefore, it is important to know the characteristics of optical fibers in 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. Knowing the characteristics of these devices is also important in 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 dedicated measuring instruments such as an OTDR (Optical Time Domain Reflectometer) or an optical spectrum analyzer. However, measurements using dedicated measuring instruments require direct measurement for each optical node and optical fiber, which poses the problem of increased equipment and operating costs.

[0005] To solve this problem, DLM (Digital Longitudinal Monitoring) has been proposed in recent years, which is a technology for detecting the characteristics of various devices in an optical transmission system by digital signal processing on the receiving side of the optical transmission system (see, for example, Patent Document 1 and Non-Patent Documents 1 and 2). DLM is based on a digital coherent optical transmission system, and monitors the characteristics of the optical transmission line, such as the optical power distribution along the longitudinal direction of the optical fiber, by performing digital signal processing on the received signal obtained by coherently detecting the optical signal transmitted by the optical transmission line.

[0006] Non-Patent Document 1 uses a method using correlation to estimate the optical power distribution, which will be referred to as the correlation method here. Non-Patent Document 2 uses a least squares method using a gradient method. Non-Patent Document 3 uses a linear least squares method using a generalized inverse matrix.

[0007] International Publication No. 2021 / 124415

[0008] T. Tanimura, et al., “Fiber-Longitudinal Anomaly Position Identification Over Multi-Span Transmission Link Out of Receiver-end Signals”, JLT, 38(9), 2020.T. Sasai, et al., “Digital longitudinal monitoring of Optical Fiber Communication Link”, JLT, 40(8), 2022.T. Sasai, et al., “Proposal of Linear Least Squares for FiberNonlinearity-Based Longitudinal Power Monitoring in Multi-Span Link”, OECC2023, W3B-1, 2023.

[0009] In the optical power distribution estimation using the correlation method described in Non-Patent Document 1, spatial resolution is limited in principle, and a true optical power level diagram cannot be obtained, limiting sensitivity and measurement accuracy. On the other hand, the correlation method described in Non-Patent Document 1 has excellent noise resistance because it does not involve multiplication of an inverse matrix.

[0010] The least-squares method for estimating optical power distributions described in Non-Patent Documents 2 and 3 can obtain true power level diagrams and achieve high spatial resolution or sensitivity. On the other hand, the least-squares method described in Non-Patent Documents 2 and 3 generates strong noise enhancement due to inverse matrix multiplication or an equivalent effect. As a result, measurement accuracy is limited in environments where noise and signal distortion generated within optical transmission systems cannot be ignored. Thus, conventional methods cannot achieve both the high noise tolerance advantage of Non-Patent Document 1 and the high spatial resolution or high sensitivity advantage of Non-Patent Documents 2 and 3, and the estimation of true optical power level diagrams. Therefore, measurement accuracy is limited.

[0011] In view of the above circumstances, an object of the present invention is to provide a technology that can improve measurement accuracy in optical power distribution estimation by achieving high noise tolerance, high spatial resolution or high sensitivity, and true power level diagram estimation at the same time.

[0012] One aspect of the present invention is an optical transmission characteristics estimation device including: a matrix calculation unit that calculates a matrix representing characteristics of a virtual transmission path based on any one of a transmission signal, a known signal, or a received signal input to an optical transmission path; and an estimation unit that estimates an optical power distribution in the transmission path using a linear least squares method with a regularization term including a real number λ, based on a received signal obtained by receiving the transmission signal using a coherent detection method or a signal obtained based on the received signal and the matrix.

[0013] The present invention achieves high noise resistance, high spatial resolution or high sensitivity, and true power level diagram estimation, thereby enabling power distribution estimation with high measurement accuracy.

[0014] FIG. 1 is a diagram (part 1) for explaining an overview of the first embodiment. FIG. 2 is a diagram (part 2) for explaining an overview of the first embodiment. FIG. 3 is a diagram illustrating an example of the configuration of an optical receiving device in the first embodiment. FIG. 4 is an example of a flowchart showing the processing flow of the optical receiving device in the first embodiment. FIG. 5 is a diagram illustrating simulation results for comparing a conventional method with the technique in the first embodiment. FIG. 6 is a diagram illustrating an overview of the second embodiment. FIG. 7 is a diagram illustrating an example of the configuration of an optical receiving device in the second embodiment. FIG. 8 is an example of a flowchart showing the processing flow of the optical receiving device in the second embodiment. FIG. 9 is a diagram illustrating an overview of the third embodiment. FIG. 10 is a diagram illustrating an example of the configuration of an optical receiving device in the third embodiment. FIG. 11 is an example of a flowchart showing the processing flow of the optical receiving device in the third embodiment.

[0015] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0016] First Embodiment First, an outline of the first embodiment will be described. In order to calculate the optical power distribution P(z) in an optical transmission path (optical fiber) in an optical transmission system including an optical transmitter, an optical receiver, and an optical transmission path connecting the optical transmitter and the optical receiver, it is sufficient to calculate γ'(z) in the nonlinear Schrödinger equation shown in the following equation (1). The nonlinear Schrödinger equation is an equation that describes the propagation of light waves in the optical transmission path. Note that γ'(z) in equation (1) is expressed by the following equation (2).

[0017]

[0018]

[0019] In equation (1), z represents the distance (km) of the optical transmission line, t represents time (s), and A represents ∫|A(z, t)| ２ dt represents the optical electric field normalized to 1, and β ２ is the group velocity dispersion coefficient (ps ２ / km), and γ represents the nonlinear constant (1 / W / km). In equation (2), P(z) represents the power (W) in the optical transmission line.

[0020] Here, as a method for estimating γ'(z) = γP(z) in an actual optical transmission path, there is a method for preparing a virtual transmission path (a digital twin of an actual optical transmission path, a simulation) by a linear regular perturbation method, as shown in FIG. 1. FIG. 1 is a diagram (part 1) for explaining the outline of the first embodiment. The output (virtual received signal) from the virtual transmission path prepared by the linear regular perturbation method is expressed as A. ｄ [L]. Received signal A of the virtual transmission line ｄ The parameter γ in the virtual transmission path that makes [L] closest to the actual received signal A[L] (the square error is minimized) ｋ This problem can be formulated as shown in the following equation (3).

[0021]

[0022] The electric field waveform A[L] at the position z=L after transmission through the optical transmission line can be expressed as the following equation (4) using the first-order regular perturbation method.

[0023]

[0024] A in formula (4) ０ (L) and A １ (L) is calculated based on the following equations (5) and (6), respectively.

[0025]

[0026]

[0027] By substituting the equation for the linear regular perturbation shown in the above equation (4) into equation (3), it can be seen that this problem can be reduced to a linear least squares problem as shown in Figure 2. The solution is expressed by equation (7), as shown in Figure 2. G in equation (7) is a matrix representing the characteristics of the virtual transmission path.

[0028]

[0029] FIG. 2 is a diagram (part 2) for explaining the outline of the first embodiment. In FIG. 2, ^γ' ｋ (The "^" is placed above γ') represents the true power level diagram ｋrepresents an estimated value of . ^γ' is expressed by the following equation (8).

[0030]

[0031] In the conventional method, the cost function is calculated by the received signal A ｄ The cost function γ' is calculated as the squared error between [L] and the actual received signal A[L]. In contrast, in the first embodiment, a regularization term is added to the cost function as shown in the following equation (9). Note that in equation (9), R represents a K × K square matrix, λ represents an arbitrary real number, and G represents a matrix representing the characteristics of the virtual transmission channel.

[0032]

[0033] A solution that minimizes the cost function in equation (9) is expressed as equation (10) below.

[0034]

[0035] For example, in the case of a unit matrix R=I, the cost function shown in equation (9) is expressed as in equation (11) below.

[0036]

[0037] As shown in equation (11), a constraint is imposed so that the absolute value (square of) of the estimated parameter ^γ' does not become too large. As the constant λ becomes larger, the regularization term λ||^γ|| ２ becomes dominant, so the absolute value of the estimated parameter becomes small.

[0038] In addition, various matrices can be considered for R depending on the desired estimation result. For example, in order to suppress oscillations in the estimated value γ', it is possible to use the sum of squares of second-order differences of γ' as a regularization term, as shown in the following equation (12). In this case, R can be expressed as a Toeplitz matrix as shown in the following equation (13).

[0039]

[0040]

[0041] A specific configuration for estimating the optical power distribution (optical transmission characteristics) of an optical transmission line using the cost function with a regularization term shown in equation (10) above will be described below.

[0042] 3 is a diagram showing an example of the configuration of an optical receiving device 10 according to the first embodiment. The optical receiving device 10 receives a transmission signal transmitted from an optical transmitting device connected via an optical transmission path. The optical receiving device 10 includes a coherent receiver 11, a chromatic dispersion compensator 12, an adaptive equalizer 13, a frequency offset compensator 14, a carrier phase noise compensator 15, and a transmission characteristics estimator 16.

[0043] The coherent receiver 11 is connected to the optical transmission line and receives and coherently detects an optical signal transmitted through the optical transmission line. The coherent receiver 11 separates the received optical signal into X-polarized and Y-polarized waves. The coherent receiver 11 detects the I- and Q-components of each of the X- and Y-polarized waves by causing interference between each of the separated X- and Y-polarized optical signals and a laser beam emitted from a local oscillator light source provided inside the coherent receiver 11. The coherent receiver 11 converts each of the I- and Q-component optical signals of each of the X- and Y-polarized waves into four analog electrical signals, and then converts the four analog signals into four digital signals using four internal analog-to-digital converters and outputs them. Hereinafter, the four digital signals output by the coherent receiver 11 will be referred to as received signals.

[0044] The chromatic dispersion compensator 12 estimates the chromatic dispersion experienced in the optical transmission line, compensates for the estimated chromatic dispersion in the received signal output from the coherent receiver 11 , and outputs the result to the adaptive equalizer 13 .

[0045] The adaptive equalizer 13 is a functional unit that compensates for distortions that occur in the waveform of the optical signal in the optical transmission path using the received signal output from the chromatic dispersion compensator 12. That is, the adaptive equalizer 13 is a functional unit that corrects code errors that occur in the optical signal due to inter-symbol interference in the optical transmission path. The adaptive equalizer 13 performs adaptive equalization processing using an FIR (Finite Impulse Response) filter in accordance with set tap coefficients.

[0046] The frequency offset compensator 14 performs a process of compensating for the frequency offset of the received signal that has been subjected to adaptive equalization processing.

[0047] The carrier phase noise compensator 15 performs a process of compensating for the phase offset of the received signal whose frequency offset has been compensated for.

[0048] The demodulated signal B[0] is obtained by the processes performed by the chromatic dispersion compensator 12, adaptive equalizer 13, frequency offset compensator 14, and carrier phase noise compensator 15.

[0049] The transmission characteristic estimating unit 16 estimates the optical power distribution (optical transmission characteristics) of the optical transmission path. The transmission characteristic estimating unit 16 includes a chromatic dispersion adding unit 161, a decoding unit 162, a transmission waveform restoring unit 163, a linear solution estimating unit 164, a perturbation term estimating unit 165, a matrix calculating unit 166, and an estimating unit 167. The transmission characteristic estimating unit 16 is one aspect of an optical transmission characteristic estimating device.

[0050] The chromatic dispersion adding unit 161 estimates the chromatic dispersion experienced in the optical transmission path and adds the estimated chromatic dispersion to the demodulated signal B[0] output from the carrier phase noise compensating unit 15. As a result, the chromatic dispersion adding unit 161 generates a signal from which polarization rotation, frequency offset, phase noise, etc. have been removed from the received signal received by the coherent receiver 11. In other words, the chromatic dispersion adding unit 161 generates a signal from which polarization rotation, frequency offset, phase noise, etc. have been removed from the received signal obtained by receiving an optical signal using a coherent detection method. The received signal generated by the chromatic dispersion adding unit 161 is a pre-chromatic dispersion compensated signal from which polarization rotation, frequency offset, phase noise, etc. have been removed. Hereinafter, the received signal generated by the chromatic dispersion adding unit 161 will be referred to as received signal A[L].

[0051] The decoding unit 162 decodes the received signal output from the carrier phase noise compensation unit 15 .

[0052] The transmission waveform restoration unit 163 restores the waveform of the transmission signal transmitted by the optical transmitting device based on the received signal decoded by the decoding unit 162. The waveform of the transmission signal restored by the transmission waveform restoration unit 163 is the waveform of the transmission signal input to a virtual transmission path represented by a digital twin (linear regular perturbation method) of the optical transmission path shown in FIG. 1. However, depending on the embodiment, the transmission signal may be known. In this case, a known signal may be used instead of the output of the transmission waveform restoration unit 163. Hereinafter, the waveform of the transmission signal restored by the transmission waveform restoration unit 163 will be referred to as transmission signal A[0].

[0053] The linear solution estimation unit 164 estimates a linear solution that has been affected only by chromatic dispersion (linear phenomenon) using the transmission signal A[0] restored by the transmission waveform restoration unit 163 or a known signal output from the transmission waveform restoration unit 163. Hereinafter, the linear solution estimated by the linear solution estimation unit 164 will be referred to as the linear solution A[0]. ０ It is written as [L].

[0054] The perturbation term estimation unit 165 calculates the received signal A[L] generated by the chromatic dispersion adding unit 161 and the linear solution A[L] estimated by the linear solution estimation unit 164. ０ The perturbation term estimation unit 165 receives the received signal A[L] and calculates the linear solution A[L]. ０ The perturbation term A excluding the component of [L] １ In this way, the perturbation term estimation unit 165 estimates the linear solution A[L] of the light wave propagation equation from the received signal A[L]. ０ By subtracting [L], the perturbation term A １ However, this block may be bypassed and the received signal A[L] may be output as is without performing this subtraction process. ０ Although there is a possibility that the estimation accuracy will be improved by subtracting [L], it is also possible to make an estimation by bypassing this block, in which case the effect of reducing the amount of calculation can be expected.

[0055] The matrix calculation unit 166 calculates a matrix G that represents the characteristics of the virtual transmission path based on the transmission signal A[0] restored by the transmission waveform restoration unit 163.

[0056] The estimation unit 167 estimates the perturbation term A １The nonlinear coefficient γ' of the optical wave propagation equation is estimated based on [L] and the matrix G calculated by the matrix calculation unit 166. The estimation unit 167 estimates the optical power distribution in the optical transmission path using the nonlinear coefficient γ' in the estimated optical wave propagation equation.

[0057] 4 is an example of a flowchart showing the processing flow of the optical receiving device 10 in the first embodiment. The coherent receiver 11 of the optical receiving device 10 receives an optical signal transmitted from an optical transmitting device (step S101). The coherent receiver 11 outputs the received signal to the chromatic dispersion compensator 12, which performs chromatic dispersion compensation on the received signal output from the coherent receiver 11 (step S102). The chromatic dispersion compensator 12 outputs the received signal after wavelength separation processing to the adaptive equalizer 13.

[0058] The adaptive equalization unit 13 performs adaptive equalization processing to compensate for distortion occurring in the waveform of the received signal after wavelength dispersion processing output from the chromatic dispersion compensator 12 (step S103). The adaptive equalization unit 13 outputs the received signal after adaptive equalization processing to the frequency offset compensator 14. The frequency offset compensator 14 performs frequency offset compensation processing to compensate for a frequency offset on the received signal after adaptive equalization processing output from the adaptive equalizer 13 (step S104). The frequency offset compensator 14 outputs the received signal after frequency offset compensation processing to the carrier phase noise compensator 15.

[0059] The carrier phase noise compensator 15 performs carrier phase compensation processing to compensate for the phase offset on the received signal after the frequency offset compensation processing output from the frequency offset compensator 14 (step S105). The carrier phase noise compensator 15 outputs the received signal after the carrier phase compensation processing to the chromatic dispersion adding unit 161 and the decoding unit 162.

[0060] The chromatic dispersion adding unit 161 adds chromatic dispersion to the received signal after carrier phase compensation processing output from the carrier phase noise compensating unit 15 (step S106). As a result, the chromatic dispersion adding unit 161 generates a received signal A[L]. The chromatic dispersion adding unit 161 outputs the generated received signal A[L] to the perturbation term estimating unit 165. The decoding unit 162 decodes the received signal after carrier phase compensation processing output from the carrier phase noise compensating unit 15 (step S107). The decoding unit 162 outputs the decoded received signal to the transmission waveform restoring unit 163.

[0061] The transmission waveform restoration unit 163 restores the waveform of the transmission signal transmitted by the optical transmitting device based on the reception signal decoded by the decoding unit 162 (step S108). The transmission waveform restoration unit 163 outputs the transmission signal A[0] indicated by the restored waveform to the linear solution estimation unit 164 and the matrix calculation unit 166. The linear solution estimation unit 164 uses the transmission signal A[0] restored by the transmission waveform restoration unit 163 to estimate the linear solution A[0] that has been affected only by chromatic dispersion (linear phenomenon). ０ [L] is estimated based on the following equation (14) (step S109). ０ [L] is the received waveform that has been subjected to only chromatic dispersion (linear phenomenon) without being subjected to nonlinearity.

[0062]

[0063] The linear solution estimation unit 164 estimates the linear solution A ０ The perturbation term estimation unit 165 outputs the received signal A[L] output from the chromatic dispersion adding unit 161 and the linear solution A[L] output from the linear solution estimation unit 164. ０ [L] and the perturbation term A based on the following equation (15): １ The perturbation term estimation unit 165 estimates the linear solution A[L] from the received signal A[L] (step S110). ０ The process of removing the component of [L] is carried out. The intention is to remove the component of E[||A-cA ０ || ２]. Note that ^c (where "^" is placed above c) represents a complex number. Note that ^c shown in equation (15) is merely an example, and other values ​​may be used as long as the optical power distribution can be ultimately estimated.

[0064]

[0065] The perturbation term estimation unit 165 estimates the perturbation term A １ [L] to the estimation unit 167. The matrix calculation unit 166 uses the transmission signal A[0] restored by the transmission waveform restoration unit 163 to calculate a matrix G representing the characteristics of the virtual transmission channel based on the following equation (16) (step S111).

[0066]

[0067] The matrix calculation unit 166 outputs the calculated matrix G to the estimation unit 167. The estimation unit 167 estimates the perturbation term A output from the perturbation term estimation unit 165. １ Based on [L] and the matrix G output from the matrix calculation unit 166, the nonlinear coefficient γ' of the optical wave propagation equation is estimated. The estimation unit 167 uses the estimated nonlinear coefficient γ' to estimate the optical power distribution P(z) of the optical transmission path based on the above equation (10) (step S112). That is, the estimation unit 167 estimates the optical power distribution in the transmission path by estimating the nonlinear coefficient γ' of the optical wave propagation equation using the regularized linear least squares method. At this time, various methods can be considered for executing equation (10). The matrix G may be calculated first and then executed as per equation (10), or equation (10) may be calculated by singular value decomposition, or the inverse matrix part Re[G^†G]^-1 and the correlation part Re[G^†A1] may be calculated separately. When the inverse matrix part and the correlation part are calculated separately, the calculation methods for the inverse matrix part and the correlation part are also arbitrary. For example, the inverse matrix part may be calculated as Re[G^†G]^-1, or may be calculated approximately from statistical information of the signal. The correlation part may be calculated as Re[G^†A1], or the constituent vector g of the matrix G expressed by the determinant (16) may be calculated as ｋ Repeat for each Re [g ｋ †A1] and then arrange them in the column direction to calculate Re[G^†A1].

[0068] (Simulation Results) A simulation was performed to compare the conventional method (least squares method, correlation method) with the method of the present invention under the following conditions: Modulation method: Probabilistically shaped 64QAM 128GBd Root raised cosine: roll-off 0.1 Transmission path model: Split-step Fourier method Fiber loss: a = 0.20 (dB / km) Fiber dispersion: β ２ = -21.7 (ps ２ / km) Fiber nonlinearity coefficient: g = 1.30 (W －１ km －１ ) Measurement grain size Δz: 0.5 (km) 50 km x 3 spans

[0069] FIG. 5 shows simulation results comparing the conventional methods (least squares method and correlation method) with the method of the first embodiment. To simulate anomalous loss in the optical transmission line, optical power was attenuated at the 75-km point. It can be seen that the proposed method corresponds to the least squares method when λ = 0 and to the correlation method when λ → ∞ (where R = I (unit matrix)). The least squares method can estimate the true power level diagram, but exhibits large oscillations in the latter half of the transmission span. In environments where noise and distortion in the optical transmission system cannot be ignored, these oscillations emphasize the noise and distortion, resulting in a noisy power distribution. On the other hand, the correlation method exhibits reduced spatial resolution and sensitivity to anomalous loss, and is unable to estimate the true power level diagram. In contrast, the proposed method: 1. suppresses noise enhancement; 2. maintains high spatial resolution or high sensitivity to anomalous loss; and 3. is able to estimate the true power level diagram. This improves the measurement accuracy of the optical power distribution.

[0070] Note that λ may be a value that minimizes the variance of the estimated value, or may be a value determined from the statistics (variance value, etc.) of the amount of noise contained in the received signal.

[0071] The optical receiving device 10 configured as described above includes a matrix calculation unit 166 that calculates a matrix representing the characteristics of a virtual transmission path based on a transmission signal or a known signal input to an optical transmission path, and a signal calculation unit 167 that calculates a received signal obtained by receiving the transmission signal using a coherent detection method or a signal obtained based on the received signal (for example, a linear solution A[L] from the received signal A[L]). ０ and an estimation unit 167 that estimates the optical power distribution in the transmission path using equation (10), which is identical to the equation for the regularized linear least squares method using the real number λ, based on the signal (from which the component [L] has been removed) and the matrix G calculated by the matrix calculation unit 166. This makes it possible to arbitrarily balance the trade-off between spatial resolution or sensitivity and noise tolerance by selecting an appropriate real number λ. Therefore, it is possible to achieve both high noise tolerance, high spatial resolution or high sensitivity, and true power level diagram estimation, thereby improving the estimation accuracy of the optical power distribution.

[0072] (Second embodiment) In the first embodiment, the received signal A of the virtual transmission line ｄ In the above example, a configuration is shown for determining the parameter γ' that minimizes the squared error between the received signal A[L] and the actual received signal A[L]. In contrast, in the second embodiment, a configuration is described for determining the parameter γ' that minimizes the squared error between the received signal A[L] and the actual received signal A[L] obtained by applying chromatic dispersion compensation to the received signal A[L] and the actual received signal A[L].

[0073] 6 is a diagram for explaining an outline of the second embodiment. In the second embodiment, a received signal A ｄ [L] and the actual received signal A[L], respectively, Ｌ －１ By applying ｄ The signal B[0] after chromatic dispersion compensation is obtained by subtracting chromatic dispersion compensation ^D[0] from the actual received signal A[L]. Ｌ －１ The signal B after chromatic dispersion compensation is obtained by applying ｄ [0] is the received signal A of the virtual transmission line ｄ For [L], chromatic dispersion compensation ^D Ｌ －１is the signal obtained by applying

[0074] In the second embodiment, the resulting chromatic dispersion compensated signal B ｄ It is sufficient to find the parameter γ' in the virtual transmission path that makes [0] closest to the signal B[0] after chromatic dispersion compensation (that minimizes the squared error). A specific configuration for realizing the above processing will be described below.

[0075] 7 is a diagram showing an example of the configuration of an optical receiving device 10a according to the second embodiment. The optical receiving device 10a receives a transmission signal transmitted from an optical transmitting device connected via an optical transmission line. The optical receiving device 10a includes a coherent receiver 11, a chromatic dispersion compensator 12, an adaptive equalizer 13, a frequency offset compensator 14, a carrier phase noise compensator 15, and a transmission characteristics estimator 16a.

[0076] The optical receiving device 10a differs in configuration from the optical receiving device 10 in that it includes a transmission characteristics estimating unit 16a instead of the transmission characteristics estimating unit 16. The other configurations of the optical receiving device 10a are similar to those of the optical receiving device 10. Therefore, the following description will focus on the differences from the optical receiving device 10.

[0077] The transmission characteristic estimator 16a estimates the optical power distribution (optical transmission characteristics) of the optical transmission path. The transmission characteristic estimator 16a includes a decoder 162, a transmission waveform restorer 163, a perturbation term estimator 165a, a matrix calculator 166, and an estimator 167a. The transmission characteristic estimator 16a is one aspect of an optical transmission characteristic estimation device. The transmission characteristic estimator 16a differs in configuration from the transmission characteristic estimator 16 in that it does not include the chromatic dispersion adder 161 and the linear solution estimator 164, and in that it includes a perturbation term estimator 165a and an estimator 167a instead of the perturbation term estimator 165 and the estimator 167. The following description will focus on the differences from the transmission characteristic estimator 16.

[0078] The perturbation term estimation unit 165a receives the demodulated signal B[0] and the transmitted signal A[0] restored by the transmitted waveform restoration unit 163. The perturbation term estimation unit 165a calculates the perturbation term B[0] by removing the component of the transmitted signal A[0] from the input demodulated signal B[0]. １In this way, the perturbation term estimation unit 165a estimates the perturbation term B[0] by subtracting the transmission signal A[0] from the demodulated signal B[0]. １ However, it is also possible to bypass this block and output demodulated signal B[0] as is without performing this subtraction processing. Although there is a possibility that estimation accuracy will be improved by subtracting transmitted signal A[0] from demodulated signal B[0], it is also possible to estimate by bypassing this block, in which case the effect of reducing the amount of calculation can be expected.

[0079] The estimation unit 167a estimates the perturbation term B １ The nonlinear coefficient γ' of the optical wave propagation equation is estimated based on [0] and the matrix G calculated by the matrix calculation unit 166. The estimation unit 167a estimates the optical power distribution in the optical transmission path using the nonlinear coefficient γ' in the estimated optical wave propagation equation.

[0080] FIG. 8 is an example of a flowchart showing the processing flow of the optical receiving device 10a according to the second embodiment. In FIG. 8, the same processes as those in FIG. 4 are denoted by the same reference numerals as those in FIG. 4, and their description will be omitted. After the processes from step S101 to step S105 are completed, the decoder 162 decodes the demodulated signal B[0] after carrier phase compensation processing output from the carrier phase noise compensator 15 (step S201). The decoder 162 outputs the decoded demodulated signal to the transmission waveform restorer 163. The transmission waveform restorer 163 restores the waveform of the transmission signal transmitted by the optical transmitting device based on the received signal decoded by the decoder 162 (step S202). The transmission waveform restorer 163 outputs the transmission signal A[0] represented by the restored waveform to the perturbation term estimator 165a and the matrix calculator 166.

[0081] The perturbation term estimation unit 165a uses the demodulated signal B[0] and the transmission signal A[0] output from the transmission waveform restoration unit 163 to calculate the perturbation term B based on the following equation (17): １ The perturbation term estimation unit 165a performs processing to remove the component of the transmission signal A[0] from the demodulated signal B[0].

[0082]

[0083] The perturbation term estimation unit 165a estimates the perturbation term B １ The matrix calculation unit 166 uses the transmission signal A[0] restored by the transmission waveform restoration unit 163 to calculate the matrix G representing the characteristics of the virtual transmission channel based on the above equation (16) (step S204). The matrix calculation unit 166 outputs the calculated matrix G to the estimation unit 167a.

[0084] The estimation unit 167a estimates the perturbation term B output from the perturbation term estimation unit 165a. １ Based on [0] and the matrix G output from the matrix calculation unit 166, the nonlinear coefficient γ' of the light wave propagation equation is estimated based on the following equation (18).

[0085]

[0086] Thereafter, the estimation unit 167a uses the estimated nonlinear coefficient γ' to estimate the optical power distribution P(z) of the optical transmission line based on the above equation (2) (step S205).

[0087] The optical receiving device 10a configured as described above can obtain the same effect as in the first embodiment by using the squared error of each signal obtained by applying chromatic dispersion compensation to the received signal of the virtual transmission path and the actual received signal.

[0088] Furthermore, the optical receiving device 10a does not include the chromatic dispersion adding unit 161 and the linear solution estimating unit 164. Therefore, the amount of calculation can be reduced compared to the first embodiment.

[0089] (Third embodiment) In the first embodiment, the received signal A of the virtual transmission line ｄ In the third embodiment, a configuration is described in which a parameter γ' is obtained that minimizes the squared error between a signal obtained by nonlinearly backpropagating an actual received signal and a transmitted pilot signal.

[0090] 9 is a diagram for explaining an outline of the third embodiment. In the third embodiment, nonlinear backpropagation is applied to an actual received signal A[L], and a signal B[0]+G ｒｘIn the third embodiment, the transmitted signal A[0] is converted into the signal B[0]+G ｒｘ It is sufficient to find the parameter γ' in the virtual transmission path that is closest to γ' (that minimizes the squared error). A specific configuration for realizing the above processing will be described below.

[0091] 10 is a diagram illustrating an example of the configuration of an optical receiving device 10b according to the third embodiment. The optical receiving device 10b receives a transmission signal transmitted from an optical transmitting device connected via an optical transmission line. The optical receiving device 10b includes a coherent receiver 11, a chromatic dispersion compensator 12, an adaptive equalizer 13, a frequency offset compensator 14, a carrier phase noise compensator 15, and a transmission characteristics estimator 16b.

[0092] The optical receiving device 10b differs in configuration from the optical receiving device 10 in that it includes a transmission characteristics estimating unit 16b instead of the transmission characteristics estimating unit 16. The other configurations of the optical receiving device 10b are similar to those of the optical receiving device 10. Therefore, the following description will focus on the differences from the optical receiving device 10.

[0093] The transmission characteristic estimator 16b estimates the optical power distribution (optical transmission characteristics) of the optical transmission path. The transmission characteristic estimator 16b includes a perturbation term estimator 165b, a matrix calculator 166b, an estimator 167b, a known symbol reader 168, and a corresponding sequence extractor 169. The transmission characteristic estimator 16b is one aspect of an optical transmission characteristic estimation device.

[0094] The transmission characteristics estimation unit 16b differs in configuration from the transmission characteristics estimation unit 16 in that it does not include the chromatic dispersion adding unit 161, the decoding unit 162, the transmission waveform restoration unit 163, and the linear solution estimator 164, that it includes a perturbation term estimator 165b, a matrix calculator 166b, and an estimator 167b instead of the perturbation term estimator 165, the matrix calculator 166, and the estimator 167, and that it newly includes a known symbol readout unit 168 and a corresponding sequence extractor 169. The following description will focus on the differences from the transmission characteristics estimation unit 16.

[0095] The matrix calculation unit 166b calculates a matrix G representing the characteristics of the virtual transmission path based on the input demodulated signal B[0]. ｒｘ,оｒｇ Calculate.

[0096] The known symbol reading unit 168 reads out a known symbol C[0] such as a training sequence or a pilot signal contained in the signal.

[0097] The corresponding series extraction unit 169 extracts the matrix G calculated by the matrix calculation unit 166b. ｒｘ,оｒｇ , a part of the matrix G corresponding to the known symbol C[0] read by the known symbol readout unit 168 ｒｘ For example, if the known symbols are present at positions 1, 2, 11, and 12 of the signal sequence, the corresponding sequence extraction unit 169 extracts the matrix G ｒｘ,оｒｇ The first, second, eleventh, and twelfth rows of the matrix G ｒｘ The matrix G ｒｘ is a matrix representing the characteristics of a virtual transmission path calculated based on the received signal.

[0098] The perturbation term estimation unit 165b receives the demodulated signal B[0] and the known symbol C[0]. The perturbation term estimation unit 165b extracts a part of the signal B[0] corresponding to the known symbol C[0] from the input demodulated signal B[0]. part The perturbation term estimation unit 165b extracts the extracted signal B part The perturbation term B obtained by removing the component of the known symbol C[0] from [0] １ In this way, the perturbation term estimation unit 165b estimates the signal B corresponding to the known symbol. part The perturbation term B is obtained by subtracting the known symbol C[0] from [0]. １ However, this subtraction process is not performed, and the signal B is bypassed. part Alternatively, [0] may be output as is. Although there is a possibility that estimation accuracy will be improved by subtracting the transmitted signal A[0] from the demodulated signal corresponding to the known symbol, estimation is also possible by bypassing this block, in which case the amount of calculation can be expected to be reduced.

[0099] The estimation unit 167b estimates the perturbation term B １ [0] and the matrix G extracted by the corresponding sequence extraction unit 169 ｒｘ The estimation unit 167b estimates the nonlinear coefficient γ' of the optical wave propagation equation based on the above equation. The estimation unit 167b estimates the optical power distribution in the optical transmission line using the nonlinear coefficient γ' in the estimated optical wave propagation equation.

[0100] 11 is an example of a flowchart showing the flow of processing in the optical receiving device 10b according to the third embodiment. In FIG. 11, the same processes as those in FIG. 4 are denoted by the same reference numerals as those in FIG. 4, and the description thereof will be omitted. When the processing from step S101 to step S105 is completed, the matrix calculation unit 166b calculates a matrix G representing the characteristics of the virtual transmission path based on the input demodulated signal B[0]. ｒｘ,оｒｇ Specifically, the matrix calculation unit 166b calculates the matrix G based on the following equation (19) (step S301). ｒｘ,оｒｇ The matrix calculation unit 166b calculates the calculated matrix G ｒｘ,оｒｇ to the corresponding sequence extraction unit 169.

[0101]

[0102] The known symbol readout unit 168 reads out known symbols such as a training sequence or a pilot signal included in the transmission signal A[0] (step S302). The known symbol readout unit 168 outputs the read out known symbols to the corresponding sequence extraction unit 169. The corresponding sequence extraction unit 169 extracts the matrix G calculated by the matrix calculation unit 166b. ｒｘ,оｒｇ From the matrix G corresponding to the known symbols read by the known symbol readout unit 168, ｒｘ (Step S303). The corresponding sequence extraction unit 169 extracts the extracted matrix G ｒｘ to the estimation unit 167b.

[0103] The perturbation term estimation unit 165b extracts a part of the signal B[0] corresponding to the known symbol C[0] from the demodulated signal B[0]. part The perturbation term estimation unit 165b extracts the extracted signal B part [0] and the known symbol C[0] are used to calculate the perturbation term B １ [0] is estimated (step S304).

[0104]

[0105] The perturbation term estimation unit 165b performs a process of removing the component of the known symbol C[0] from the demodulated signal B[0].１ The estimation unit 167b outputs the perturbation term B output from the perturbation term estimation unit 165b to the estimation unit 167b. １ [0] and the matrix G output from the corresponding sequence extraction unit 169 ｒｘ Based on this, the nonlinear coefficient γ′ of the light wave propagation equation is estimated based on the following equation (21).

[0106]

[0107] Thereafter, the estimation unit 167b uses the estimated nonlinear coefficient γ' to estimate the optical power distribution P(z) of the optical transmission line based on the above equation (2) (step S305).

[0108] The optical receiving device 10b configured as described above can obtain the same effect as in the first embodiment by using the squared error between the signal obtained by nonlinearly backpropagating the actual received signal and the known symbol C[0].

[0109] Furthermore, unlike the first embodiment, the optical receiving device 10b uses a received signal sequence to calculate a matrix G ｒｘ Therefore, the optical receiving device 10b does not need to know the entire transmission signal sequence in advance. Therefore, it is possible to estimate the power distribution even during operation by using known symbols such as training sequences and pilot symbols that are partially included in the signal during operation.

[0110] (Modification 1 common to the first to third embodiments) In this description, a quadratic form γ † Although the configuration using Rγ′ has been shown, the present invention is not limited to this and any other function may be used.

[0111] (Modification 2 common to the first to third embodiments) The order of processing by the chromatic dispersion compensator 12, adaptive equalizer 13, frequency offset compensator 14, and carrier phase noise compensator 15 included in the optical receiving devices 10, 10a, and 10b may be any order. The algorithms performed by the chromatic dispersion compensator 12, adaptive equalizer 13, frequency offset compensator 14, and carrier phase noise compensator 15 are not particularly limited. The optical receiving devices 10, 10a, and 10b may be configured to include compensation units not shown in FIGS. 3, 7, and 10.

[0112] Some or all of the functional units of the optical receiving devices 10, 10a, and 10b described above are realized as software by a processor such as a CPU (Central Processing Unit) executing a program stored in a storage device having a non-volatile storage medium (non-transitory storage medium) and a storage unit. The program may be recorded on a computer-readable non-transitory storage medium. Examples of computer-readable non-transitory storage media include portable media such as flexible disks, magneto-optical disks, ROMs (Read Only Memory), and CD-ROMs (Compact Disc Read Only Memory), and storage devices such as hard disks built into computer systems.

[0113] Some or all of the functional units of the optical receiving devices 10, 10a, and 10b described above may be realized using hardware including an electronic circuit (electronic circuit or circuitry) using, for example, an LSI (Large Scale Integrated circuit), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0114] 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.

[0115] The present invention can be applied to a technique for estimating transmission characteristics in a digital coherent optical transmission system.

[0116] REFERENCE SIGNS LIST 10, 10a, 10b...optical receiving device, 11...coherent receiver, 12...chromatic dispersion compensator, 13...adaptive equalizer, 14...frequency offset compensator, 15...carrier phase noise compensator, 16, 16a, 16b...transmission characteristic estimator, 161...chromatic dispersion adder, 162...decoder, 163...transmission waveform restorer, 164...linear solution estimator, 165, 165a, 165b...perturbation term estimator, 166, 166b...matrix calculator, 167, 167a, 167b...estimator, 168...known symbol readout unit, 169...corresponding sequence extractor

Claims

1. An optical transmission characteristic estimation device comprising: a matrix calculation unit that calculates a matrix representing the characteristics of a virtual transmission path based on either a transmission signal, a known signal, or a received signal input to an optical transmission path; and an estimation unit that estimates optical power distribution in the transmission path using a linear least squares method with a regularization term including a real number λ, based on a received signal obtained by receiving the transmission signal using a coherent detection method or a signal obtained based on the received signal and the matrix.

2. The optical transmission characteristic estimation device according to claim 1, wherein the estimation unit estimates the optical power distribution using the linear least squares method with the regularization term so as to minimize a squared error between a received signal in a virtual transmission path and a received signal obtained by receiving using a coherent detection method.

3. The optical transmission characteristic estimation device according to claim 1, wherein the estimation unit estimates the optical power distribution using the linear least squares method with the regularization term so as to minimize the squared error of each signal obtained by performing chromatic dispersion compensation on a received signal of a virtual transmission path and a received signal obtained by receiving using a coherent detection method.

4. The optical transmission characteristic estimation device according to claim 1, wherein the estimation unit estimates the optical power distribution using the linear least squares method with the regularization term so as to minimize a squared error between the known signal and a signal obtained by applying nonlinear backpropagation to a received signal obtained by receiving using a coherent detection method.

Images