JP2022103064A - Optical network device and transmission line monitoring method - Google Patents

Abstract

To provide an optical network device and a method for monitoring an optical transmission line, which reduces a time required to identify a place where a polarization dependent loss occurs.SOLUTION: In an optical network device 1 which receives a polarized multiplex optical signal, a polarization separation unit separates an electric field information signal of the polarized multiplex optical signal into mutually orthogonal first and second polarization components. A polarization control unit controls the first and second polarization components in a coordinate system which indicates mutually orthogonal first and second polarization directions, so as to generate mutually orthogonal third and fourth polarization components. A feature extraction unit calculates an evaluation value corresponding to the power of the third polarization component for each of a plurality of positions on the transmission line. A change amount calculation unit calculates the change amount of the evaluation value with respect to a control amount by the polarization control unit, for each of the plurality of positions. Based on the comparison result of the change amount of the evaluation values at mutually neighboring first and second positions, a determination unit determines whether the first position is an extraction target.SELECTED DRAWING: Figure 9

Description

The present invention relates to an optical network device and a method for monitoring an optical transmission line.

Due to the increase in the amount of communication traffic, optical fiber communication for long-distance transmission and large-capacity transmission is required. On the other hand, since communication is performed with performance close to the limit of the transmission device, a change in the state of the optical transmission line may cause deterioration of transmission characteristics. One of the causes of such deterioration of transmission characteristics is polarization-dependent loss (PDL). PDL is a phenomenon in which insertion loss differs depending on the angle of polarization, and it occurs not only during device manufacturing, but also during operation due to route changes at optical nodes, reconnection of optical fibers, bending of optical fibers, etc. Occur.

In the presence of PDL, for example, the receiving power of any of the polarizations is low and the transmission quality is poor. Therefore, it is important to detect the state of PDL during operation in order to shorten the period during which the performance of the transmission system is deteriorated.

In the measurement of PDL, for example, while inputting a single-polarized test light from a transmission node to an optical fiber transmission line, the angle of polarization of the test light is gradually changed. Then, in the receiving node, the power of the light output from the optical fiber transmission line is measured for each polarization, and the PDL is calculated based on the maximum value and the minimum value among those measured values.

Further, as a phenomenon that causes a power difference between polarizations as in PDL, polarization-dependent gain (PDG: Polarization Dependent Gain) is known. PDG is a phenomenon in which the gain differs depending on the polarization state in an optical amplifier, and may occur in a distributed Raman amplifier, for example.

In an optical transmission system, a method of identifying the position and cause of a failure has been proposed (for example, Patent Document 1). Further, a device for estimating the non-linearity in the channel has been proposed (for example, Patent Document 2).

Japanese Unexamined Patent Publication No. 2020-08628 Japanese Unexamined Patent Publication No. 2013-07425

In the above-mentioned method (that is, the method of calculating the PLD from the maximum and minimum values of the optical power at the receiving node while changing the angle of polarization), it is detected at which position on the optical transmission path the PDL is generated. It's difficult to do. In addition, in this method, since a dedicated test light for measuring PDL is input to the optical transmission line, it is necessary to stop the communication service at the time of measuring PDL.

If the optical signal is branched at a plurality of positions on the optical transmission path and the optical power is monitored, it may be possible to identify the position where the PDL is generated. However, this method requires a large amount of labor and time to identify the position where the PDL is generated. In particular, enormous labor and time are required to investigate an optical transmission line of several hundred kilometers.

An object of one aspect of the present invention is to reduce the time required to identify a location on an optical transmission path where a polarization dependent loss occurs.

The optical network device of one aspect of the present invention receives a polarized multiplex optical signal transmitted from a transmitting node. This optical network device has a polarization separation unit that separates an electric field information signal representing an electric field of the polarization multiplex optical signal into a first polarization component and a second polarization component that are orthogonal to each other, and a second that is orthogonal to each other. In the coordinate system representing the polarization direction of 1 and the second polarization direction, the third polarization component and the third polarization component orthogonal to each other are controlled by controlling the first polarization component and the second polarization component. The third polarization component or the fourth polarization component for each of a plurality of positions on the optical transmission path between the transmission node and the optical network device and the polarization control unit that generates the polarization component of 4. A feature extraction unit that calculates an evaluation value corresponding to at least one of the powers of the polarization component, and a fluctuation amount that calculates the fluctuation amount of the evaluation value with respect to the control amount by the polarization control unit for each of the plurality of positions. The calculation unit, the fluctuation amount of the evaluation value at the first position among the plurality of positions, and the fluctuation amount of the evaluation value at the second position adjacent to the first position among the plurality of positions. A determination unit for determining whether or not the first position is the extraction target based on the comparison result is provided.

According to the above-described embodiment, the time required to identify the location where the polarization-dependent loss occurs on the optical transmission path can be shortened.

It is a figure which shows an example of the method of measuring the power of an optical signal at an arbitrary position on a transmission path. It is a figure which shows an example of the function of a digital signal processing unit. It is a figure which shows an example of the change of the power and the wavelength dispersion of an optical signal. It is a flowchart which shows an example of the process of measuring the power of an optical signal at a plurality of positions on a transmission path. It is a figure which shows an example of the method of specifying the occurrence position of the polarization dependent loss. It is a figure which shows the rotation conversion in the direction which a loss occurs. It is a figure which shows an example of a power profile. It is a figure which shows an example of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows the Example of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the transmission waveform reconstruction part. It is a figure explaining the rotation of the polarization of an optical signal. It is a flowchart which shows an example of the method of specifying the occurrence position of the polarization dependent loss. It is a figure which shows an example of the method of creating a power profile. It is a figure which shows an example of the position specifying part. It is a figure which shows an example of the method of specifying the position where the polarization dependent loss occurs. It is a figure which shows the other example of the method of identifying the position where the polarization dependent loss occurs. It is a figure which shows an example of an optical transmission system composed of a plurality of spans. It is a figure which shows typically the process of an optical network apparatus. It is a figure which shows the 1st variation of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows the 2nd variation of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows the 3rd variation of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows the 4th variation of the optical network apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the change of the power of an optical signal output from a transmitting node. It is a figure explaining the amplification by the optical amplifier mounted on the optical transmission path. It is a figure which shows an example of the optical network apparatus which concerns on 2nd Embodiment of this invention. It is a figure which shows an example of a span specific part. It is a figure which shows an example of a span and a power profile. It is a figure which shows an example of the power profile in the optical transmission line shown in FIG. 27. It is a figure (the 1) which shows the example of the change of the power profile with respect to the polarization rotation operation. It is a figure (the 2) which shows the example of the change of the power profile with respect to the polarization rotation operation. It is a figure which shows the power difference for every span and the determination result. It is a flowchart which shows an example of the method of specifying the occurrence position of the polarization dependent loss in 2nd Embodiment. It is a figure explaining distribution Raman amplification. It is a figure explaining PDG generated in distribution Raman amplification. It is a figure which shows an example of the span identification part used in the optical transmission system which includes a Raman amplifier.

The present application relates to a technique for detecting polarization-dependent loss (PDL) and estimating its occurrence location. However, as described above, polarization-dependent gain (PDG) is known as a phenomenon that causes a power difference between polarizations as in PDL. Both PDL and PDG are phenomena in which a power difference occurs between polarizations in a specific polarization state. Therefore, PDG can be detected by substantially the same method as the method for detecting PDL. Therefore, although the method of detecting PDL will be described below, the method of detecting PDL can also be applied to the case of detecting PDG.

The optical network device according to the embodiment of the present invention has a function of measuring optical power at an arbitrary position on a transmission path based on a received optical signal, and a position where polarization-dependent loss is generated by using the measurement result. It has a function to identify. Therefore, before explaining the function of specifying the position where the polarization-dependent loss occurs, the function of measuring the power of the optical signal at an arbitrary position on the transmission path will be described.

FIG. 1 shows an example of a method of measuring the power of an optical signal at an arbitrary position on a transmission path. In this example, the optical signal transmitted from the transmission node 100 is transmitted to the optical network device 1 via the optical transmission line (optical fiber).

The optical network device 1 includes a coherent receiver 11, an analog-to-digital converter (ADC) 12, a digital signal processing unit 13, a simulated transmitter 14, a memory circuit 15, and a feature extraction unit 16. .. The optical network device 1 may have other functions or circuits not shown in FIG.

The coherent receiver 11 includes a 90 degree optical hybrid circuit to generate an electric field information signal (or electric field data) representing the electric field of the received optical signal. The electric field information signal includes an in-phase (I: In-phase) component and an orthogonal (Q: Quadrature) component of the received optical signal. When the optical signal is a polarization multiplex optical signal, the electric field information signal includes the I component and the Q component of the H polarization and the I component and the Q component of the V polarization. The ADC 12 converts the electric field information signal into a digital signal.

FIG. 2 shows an example of the function of the digital signal processing unit 13. As shown in FIG. 2, the digital signal processing unit 13 includes a fixed equalizer 13a, an adaptive equalizer 13b, a phase regenerator 13c, and a classifier 13d. Then, the digital signal processing unit 13 processes the electric field information of the received optical signal.

The fixed equalizer 13a compensates for known waveform degradation components (eg, transmission line wavelength dispersion). The adaptive equalizer 13b adaptively equalizes. For example, the adaptive equalizer 13b can compensate for residual dispersion. When the received optical signal is a polarization-multiplexed optical signal, the adaptive equalizer 13b includes a function of separating the received optical signal for each polarization. The phase regenerator 13c estimates the frequency offset between the transmitting node 100 and the optical network device 1. Then, the phase regenerator 13c reproduces the phase of the optical signal transmitted from the transmission node 100. That is, for each symbol, the signal point on the constellation is reproduced. The classifier 13d reproduces the transmission data based on the constellation information (phase and amplitude) output from the phase regenerator 13c. The classifier 13d may be included in the digital signal processing unit 13 or may be provided on the output side of the digital signal processing unit 13. Further, the digital signal processing unit 13 may include an error correction circuit on the output side of the classifier 13d.

The simulated transmitter 14 generates an electric field information signal by mapping the transmission data reproduced by the digital signal processing unit 13 (or the classifier 13d) on the constellation. Here, this mapping is the same as the mapping performed by the transmitting node 100. Therefore, the electric field information signal generated by the simulated transmitter 14 is substantially the same as the electric field information signal for generating the optical signal at the transmission node 100. That is, the output signal of the simulated transmitter 14 represents the electric field of the optical signal at the transmission node 100.

The memory circuit 15 stores an electric field information signal representing the electric field of the received optical signal. It should be noted that this electric field information signal represents a state before the wavelength dispersion of the transmission line is compensated.

The feature extraction unit 16 includes a first dispersion compensation unit 16a, a nonlinear compensation unit 16b, a second dispersion compensation unit 16c, and a correlation calculation unit 16d. Compensates for non-linear distortion. The first dispersion compensating unit 16a compensates the electric field information signal for a part of the wavelength dispersion of the transmission line (hereinafter referred to as the first wavelength dispersion). The non-linear compensation unit 16b compensates for the non-linear distortion of the transmission line with respect to the output signal of the first dispersion compensation unit 16a. The second dispersion compensating unit 16c compensates the output signal of the non-linear compensation unit 16b for the remaining wavelength dispersion of the wavelength dispersion of the transmission line (hereinafter referred to as the second wavelength dispersion). The correlation calculation unit 16d calculates the correlation between the output signal of the second dispersion compensation unit 16c and the output signal of the simulated transmitter 14. Here, the output signal of the simulated transmitter 14 represents the electric field of the optical signal at the transmitting node 100, as described above. That is, the correlation calculation unit 16d calculates the correlation between the electric field information signal compensated for wavelength dispersion and nonlinear distortion and the electric field information signal representing the electric field of the optical signal at the transmission node 100. It is preferable that the output signal of the second dispersion compensation unit 16c and the output signal of the simulated transmitter 14 are appropriately normalized.

This correlation value represents the power of an optical signal transmitted via a transmission line. That is, the optical network device 1 can measure the power of the optical signal transmitted via the transmission line by calculating this correlation value. Here, with reference to FIG. 3, the relationship between the correlation value and the power of the optical signal will be described.

FIG. 3 shows an example of changes in the power and wavelength dispersion of an optical signal. In this example, an optical signal is transmitted from the transmission node 100 to the optical network device 1. An optical amplifier is provided on the transmission path.

The power of the optical signal attenuates as it moves away from the transmitting node 100. Then, the optical signal is amplified by the optical amplifier. After that, the power of the optical signal is attenuated as it goes away from the optical amplifier. The cumulative wavelength dispersion added to the optical signal increases in proportion to the distance from the transmitting node 100. The CD shown in FIG. 3 represents the total wavelength dispersion of the transmission line between the transmission node 100 and the optical network device 1.

Here, the optical network device 1 measures the power of the optical signal at the position P shown in FIG. The wavelength dispersion of the transmission line between the optical network device 1 and the position P is CD1. The wavelength dispersion between position P and the transmitting node 100 is CD2. The sum of CD1 and CD2 is CD.

As described above, the feature extraction unit 16 compensates for wavelength dispersion and non-linear distortion. That is, the first dispersion compensating unit 16a compensates for the wavelength dispersion CD1 with respect to the electric field information signal representing the received optical signal. The non-linear compensation unit 16b compensates for the non-linear distortion with respect to the output signal of the first dispersion compensation unit 16a. At this time, the nonlinear compensation unit 16b compensates for a predetermined amount of nonlinear strain. Then, the second dispersion compensating unit 16c compensates for the wavelength dispersion CD2 with respect to the output signal of the nonlinear compensation unit 16b.

Here, the amount of non-linear distortion generated in the optical transmission line depends on the power of the optical signal. Specifically, the larger the power of the optical signal, the larger the amount of non-linear distortion. Then, in this example, the non-linear compensation unit 16b is designed to compensate for the non-linear distortion that occurs when the power of the optical signal is sufficiently large. As an example, although not particularly limited, the non-linear compensation unit 16b is designed to compensate for the non-linear distortion generated with respect to the output optical power of the transmission node 100.

On the other hand, the correlation value calculated by the correlation calculation unit 16d correlates with the electric field information signal in which the wavelength dispersion and nonlinear distortion are compensated in the feature extraction unit 16 and the electric field information signal representing the electric field of the optical signal in the transmission node 100. show. Therefore, when the non-linear distortion is appropriately compensated by the non-linear compensation unit 16b, the correlation value calculated by the correlation calculation unit 16d becomes large.

That is, when the power of the optical signal at the position P is large, the amount of the non-linear distortion at the position P is large, and the difference between the amount of the non-linear distortion at the position P and the amount of the non-linear distortion compensated by the non-linear compensation unit 16b is small. Become. As a result, the non-linear distortion is appropriately compensated by the non-linear compensation unit 16b, and the correlation value calculated by the correlation calculation unit 16d becomes large. On the other hand, when the power of the optical signal at the position P is small, the amount of the nonlinear strain at the position P is small, and the difference between the amount of the nonlinear strain at the position P and the amount of the nonlinear strain compensated by the nonlinear compensator 16b is large. Become. As a result, the non-linear distortion is not appropriately compensated by the non-linear compensation unit 16b, and the correlation value calculated by the correlation calculation unit 16d becomes small. Therefore, the correlation value calculated by the feature extraction unit 16 substantially represents the power of the optical signal at a predetermined position (position P in FIG. 3) on the transmission path. Note that this correlation value is an example of an evaluation value corresponding to the power of the optical signal on the transmission path.

Further, the position P shown in FIG. 3 is designated by the combination of the wavelength dispersion CD1 and the wavelength dispersion CD2. Therefore, the feature extraction unit 16 can measure the power of the optical signal at a desired position on the transmission path by changing the combination of the wavelength dispersion CD1 and the wavelength dispersion CD2 with respect to the electric field information signal representing the electric field of the received optical signal. ..

FIG. 4 is a flowchart showing an example of a process of measuring the power of an optical signal at a plurality of positions on a transmission path. This process is executed when the optical network device 1 receives the optical signal transmitted from the transmission node 100 via the transmission path.

In S1, the feature extraction unit 16 acquires the transmission electric field information signal generated by the simulated transmitter 14. This electric field information signal represents the electric field of the optical signal at the transmission node 100. In S2, the feature extraction unit 16 acquires the electric field information signal of the received optical signal. It is assumed that this electric field information signal is generated by the coherent receiver 11 and stored in the memory circuit 15.

In S3, the feature extraction unit 16 initializes the wavelength dispersion CD1 to “zero”. The value of the wavelength dispersion CD1 corresponds to the transmission distance with respect to the optical network device 1. Further, the wavelength dispersion CD2 is calculated from "CD1 + CD2 = CD". CD represents the total wavelength dispersion of the transmission line between the transmission node 100 and the optical network device 1, and its value is known. In S4, the feature extraction unit 16 determines whether the wavelength dispersion CD1 is CD or less. Then, when the wavelength dispersion CD1 is CD or less, the processing of the feature extraction unit 16 proceeds to S5.

In S5, the feature extraction unit 16 sequentially executes compensation for the wavelength dispersion CD1, non-linear compensation, and compensation for the wavelength dispersion CD2 with respect to the electric field information signal of the received optical signal. In S6, the feature extraction unit 16 calculates the correlation between the electric field information signal compensated in S6 and the transmission electric field information signal acquired in S1.

In S7, the feature extraction unit 16 adds ΔCD to the wavelength dispersion CD1. After that, the processing of the feature extraction unit 16 returns to S4. That is, in S4 to S7, the feature extraction unit 16 calculates the correlation value while increasing the wavelength dispersion CD1 by ΔCD until the wavelength dispersion CD1 becomes larger than the CD. Here, the value of the wavelength dispersion CD1 corresponds to the transmission distance with respect to the optical network device 1. Therefore, the step of increasing the wavelength dispersion CD1 by ΔCD is equivalent to the step of shifting the position on the transmission path by the distance corresponding to ΔCD. Therefore, the feature extraction unit 16 repeatedly executes the processes of S4 to S7 to calculate the correlation values at a plurality of positions on the transmission path.

When the wavelength dispersion CD1 becomes larger than the CD, in S8, the feature extraction unit 16 outputs the correlation value calculated in S4 to S7. Here, the correlation value substantially represents the power of the optical signal at a predetermined position on the transmission path corresponding to the combination of the wavelength dispersion CD1 and CD2. That is, the feature extraction unit 16 can detect the power of the optical signal at a plurality of positions on the transmission path. In the following description, information representing the power of an optical signal at a plurality of positions on a transmission path may be referred to as a "power profile".

In this way, the optical network device 1 can measure the power of the optical signal at a desired position on the transmission path. Then, the optical network device 1 utilizes this function to estimate the polarization-dependent loss at a desired position on the transmission path.

FIG. 5 shows an example of a method for specifying the position where the polarization-dependent loss occurs. In this example, the transmission node 200 includes a polarization rotation unit 201, a digital-to-analog converter (DAC) 202, and a modulation unit 203.

The polarization rotating unit 201 controls the polarization of the optical signal. However, the polarization rotating unit 201 does not directly control the polarization of the optical signal, but controls the polarization of the optical signal by correcting the electric field information signal for generating the optical signal. In this case, the polarization rotation is realized, for example, by multiplying the electric field information signal representing the main signal by the Jones matrix. Further, the amount of rotation of the polarization is changed by 10 degrees from zero degree to 180 degrees, for example.

The DAC 202 converts the output signal of the polarization rotating unit 201 into an analog signal. The modulation unit 203 generates a modulated optical signal based on the output signal of the DAC 202. Therefore, the polarization of this optical signal is controlled by the polarization rotating unit 201. The optical signal transmitted from the transmission node 200 is propagated through the transmission path. Then, the optical network device 300 receives this optical signal via the transmission line. It is assumed that the transmission node 200 notifies the optical network device 300 of the amount of the polarization rotation given by the polarization rotation unit 201.

The optical network device 300 includes a coherent receiver 301, an ADC 302, a digital signal processing unit 303, a memory circuit 304, a feature extraction unit 305, and a position specifying unit 306. The coherent receiver 301, ADC 302, digital signal processing unit 303, memory circuit 304, and feature extraction unit 305 correspond to the coherent receiver 11, ADC 12, digital signal processing unit 13, memory circuit 15, and feature extraction unit 16 shown in FIG. do. Further, the transmitted electric field information signal is generated by the same function as that of the simulated transmitter 14 shown in FIG.

The digital signal processing unit 303 reproduces the main signal based on the electric field information signal representing the electric field of the received optical signal. Further, this electric field information signal is stored in the memory circuit 304. The optical network device 300 acquires information representing the amount of polarization rotation from the transmission node 200.

As described with reference to FIGS. 1 to 4, the feature extraction unit 305 can detect the power of the optical signal at an arbitrary position on the transmission path by calculating the correlation between the transmission signal and the reception signal. can. In this embodiment, the feature extraction unit 305 detects the power of the optical signal at a plurality of positions on the transmission path. Further, the feature extraction unit 305 detects the power of the optical signal for each polarization rotation amount of the optical signal. That is, the feature extraction unit 305 detects the power of the optical signal for each polarization rotation amount at a plurality of positions on the transmission path.

The position specifying unit 306 determines whether or not a polarization-dependent loss has occurred for each position (that is, a measurement position) on the transmission path. For example, if the difference between the maximum value and the minimum value of the detected optical power for a plurality of polarization rotation amounts is larger than the threshold value, it is determined that the polarization-dependent loss has occurred. Then, the position specifying unit 306 identifies the position where the polarization-dependent loss occurs based on this determination result.

As described above, in the example shown in FIG. 5, the position where the polarization-dependent loss occurs is specified by rotating the polarization of the optical signal transmitted from the transmission node 200. However, the optical network device 300 estimates the polarization-dependent loss by acquiring the electric field information of the received optical signal when the rotation amount of the polarization of the optical signal is held at a constant value. Therefore, in the method shown in FIG. 5, the time required to identify the position where the polarization-dependent loss occurs becomes long.

<Embodiment>
In the embodiment of the present invention, the polarization multiplex optical signal is transmitted from the transmission node to the optical network device. The polarization multiplex optical signal is generated by multiplexing a set of optical signals having polarizations orthogonal to each other.

When the electric field information of the transmitted optical signal is represented by Esx and Esy and the transfer function of the transmission line is represented by A, the electric field information Ex and Ey of the received optical signal are represented by the equation (1).

In this case, in the configuration shown in FIG. 1, electric field information Ecx and Ecy are generated by compensating for dispersion and nonlinear distortion in the electric field information Ex and Ey representing the electric field of the received optical signal. Ex, Ey, Ecx, and Ecy are complex numbers having information on the electric field strength and the optical phase, respectively. Then, in the method described with reference to FIGS. 1 and 4, the correlation between the absolute value of Ecx and the absolute value of Esx and / or the correlation between the absolute value of Ecy and the absolute value of Esy is calculated. Then, the optical power at an arbitrary position on the transmission path is measured.

The polarization-dependent loss is represented by a rotational transformation in the direction in which the loss occurs, an addition of the loss, and a reverse rotational transformation to return to the original polarization state. That is, the polarization-dependent loss is expressed by Eq. (2). φ represents the amount of rotation of the rotation conversion in the direction in which the loss occurs, and Γ (0 <Γ≤1) represents the loss.

FIG. 6 shows a rotation conversion in the direction in which the loss occurs. The direction in which the loss occurs is represented by φ. In this case, the rotation conversion in the direction in which the loss occurs is expressed by the equation (3). Erx and Ery represent the polarization component after the rotation conversion. That is, Ery corresponds to the sum of the projection component of Esx and the projection component of Esy with respect to the PDL axis. Further, Erx corresponds to the sum of the projection component of Esx and the projection component of Esy with respect to the axis orthogonal to the PDL axis. Erx, Ery, Esx, and Esy are complex numbers having information on the electric field strength and the optical phase, respectively.

Here, in the rotation conversion, the direction of polarization of each electric field component (direction of the arrow in FIG. 6) changes, but the power does not change. Therefore, when measuring the optical power by the method described with reference to FIGS. 1 and 4, Erx and Ery shown in FIG. 6 (b) are used instead of Esx and Esy shown in FIG. 6 (a). be able to. That is, the polarization of the electric field information Ecx and Ecy representing the electric field compensated for the dispersion and non-linear distortion of the received light signal is rotated, and the polarization of the electric meter information Esx and Esy representing the electric field of the transmitted light signal is rotated. By calculating the correlation with the signal, the power of the optical signal at any position on the transmission path is measured.

Further, when the polarization-dependent loss occurs in the PDL axis direction shown in FIG. 6B, the optical phase information of Erx does not change, but the optical phase information of Ery is caused by the nonlinear phase noise due to the polarization-dependent loss. Then, it changes by the amount that the phase rotation amount becomes smaller. Therefore, the polarization-dependent loss can be detected by measuring the optical power from the optical phase information including the Erx / Ery phase noise by the method described with reference to FIGS. 1 and 4.

FIG. 7 shows an example of a power profile. As shown in FIG. 7A, the power of the optical signal gradually decreases almost linearly according to the transmission distance. However, when a local loss occurs on the transmission path, the power of the optical signal drops sharply. In FIG. 7 (b), a local loss occurs at the position P.

When a local polarization-dependent loss occurs, the power profile shown in FIG. 7B can be obtained if the polarization direction of the optical signal coincides with the PDL axis. On the other hand, when the direction of polarization of the optical signal is orthogonal to the PDL axis, the power profile shown in FIG. 7A is obtained. However, the direction in which the polarization-dependent loss occurs (φ in FIG. 6B) is unknown. Therefore, the optical network device of the embodiment of the present invention performs rotational conversion on the electric field information signal with a plurality of different rotation amounts θi, and creates a power profile for each rotation amount θi. Then, when the power profile shown in FIG. 7B is found, the position where the power drops sharply (that is, the position P) is detected.

FIG. 8 shows an example of an optical network device according to an embodiment of the present invention. In the optical network device 1 according to the embodiment of the present invention, the electric field information signal generated by the coherent receiver 11 is guided to the monitor signal rotating unit 24 as a monitor signal. This electric field information signal represents the electric field of the received light signal. The monitor signal rotation unit 24 executes the rotation conversion shown in FIG. 6 (that is, the rotation conversion represented by the equation (3)) with respect to the monitor signal. The rotation amount θ is designated by the rotation amount control unit 26. The reference signal represents the electric field of the optical signal transmitted from the transmission node 100 shown in FIG. The reference signal is obtained, for example, by the simulated transmitter 14 in the configuration shown in FIG. Then, the reference signal rotation unit 25 executes the rotation conversion shown in FIG. 6 (that is, the rotation conversion represented by the equation (3)) with respect to the reference signal. Here, the rotation amount of the rotation conversion with respect to the monitor signal and the rotation amount of the rotation conversion with respect to the reference signal are the same.

The feature extraction unit 27 creates a power profile for each rotation amount θ (θ1, θ2, ...). The position specifying unit 28 detects the maximum power value and the minimum power value for each measurement position on the transmission path. For example, at position k, if the power Pi calculated for θi is the largest and the power Pj calculated for θj is the smallest, it is estimated that θi is in the direction of the PDL axis. Further, the difference D between the maximum power Pi and the minimum power Pj corresponds to the magnitude of the polarization-dependent loss. In this case, the position where the difference D changes abruptly is considered to be the place where the polarization-dependent loss occurs. Therefore, the position specifying unit 28 can specify the place where the polarization-dependent loss occurs by calculating the difference D at each position.

In this way, the position specifying unit 28 determines whether or not the predetermined position on the transmission path is the position where the polarization-dependent loss occurs. Then, the position specifying unit 28 outputs information indicating the position where the polarization-dependent loss occurs. Therefore, the position specifying unit 28 is an example of a determination unit that determines whether or not a predetermined position on the transmission path is a position where a polarization-dependent loss occurs. The information output from the position specifying unit 28 is stored in, for example, a server computer. In this case, the network administrator can check the state of the transmission line by using this server computer.

FIG. 9 shows an example of an optical network device according to an embodiment of the present invention. The optical network device 1 receives an optical signal transmitted from the transmission node 100 shown in FIG. The transmission node 100 generates and transmits a polarization multiplex optical signal. The polarization multiplex optical signal is generated by combining a set of optical signals (that is, an X-polarized optical signal and a Y-polarized optical signal) whose polarizations are orthogonal to each other. The polarized wave multiplex optical signal output from the transmission node 100 is transmitted via an optical fiber transmission line. Then, the optical network device 1 receives the polarization-multiplexed optical signal via the optical fiber transmission line.

As shown in FIG. 9, the optical network device 1 refers to a coherent receiver 11, a digital signal processing unit 13, a monitor signal storage unit 21, a transmission data storage unit 22, a transmission waveform reconstruction unit 23, and a monitor signal rotation unit 24. It includes a signal rotation unit 25, a rotation amount control unit 26, a feature extraction unit 27, and a position specifying unit 28. The ADC provided between the coherent receiver 11 and the digital signal processing unit 13 is omitted. Further, the optical network device 1 may include other circuits or functions not shown in FIG.

The coherent receiver 11 generates electric field information (or electric field data) representing the electric field of the received optical signal. As described with reference to FIG. 2, the digital signal processing unit 13 includes a fixed equalizer 13a, an adaptive equalizer 13b, a phase regenerator 13c, and a classifier 13d. Therefore, the digital signal processing unit 13 reproduces the data transmitted from the transmission node 100 shown in FIG. 1 based on the electric field information representing the received optical signal. At this time, the adaptive equalizer 13b separates the electric field information of the received polarization-multiplexed optical signal into a set of polarization components (Ex, Ey) orthogonal to each other. As a result, the X-polarized optical signal and the Y-polarized optical signal are extracted. Then, the digital signal processing unit 13 reproduces data from the X-polarized optical signal and the Y-polarized optical signal, respectively.

A set of electric field information (Ex, Ey) extracted by the adaptive equalizer 13b is output as a monitor signal. Then, this monitor signal is stored in the monitor signal storage unit 21. The monitor signal storage unit 21 is realized by, for example, a memory.

The transmission data storage unit 22 stores the data reproduced by the digital signal processing unit 13. Here, as described above, the digital signal processing unit 13 reproduces the data transmitted from the transmission node 100 shown in FIG. That is, the transmission data is stored in the transmission data storage unit 22. The transmission data storage unit 22 is realized by, for example, a memory.

FIG. 10 shows an example of the transmission waveform reconstruction unit 23. As shown in FIG. 10, the transmission waveform reconstruction unit 23 includes a symbol mapper 23a, a sampling speed adjustment unit 23b, and a Nyquist filter 23c. Then, the transmission waveform reconstruction unit 23 performs substantially the same signal processing as the transmission node 100. That is, the symbol mapper 23a maps the transmission data stored in the transmission data storage unit 22 for each symbol. That is, electric field information representing transmission data is generated according to the specified modulation method. The modulation method is the same for the transmission node 100 and the symbol mapper 23a. The sampling speed adjusting unit 23b adjusts the sampling speed of the electric field information signal. The Nyquist filter 23c corrects the electric field information signal in order to suppress intersymbol interference. As a result, the electric field information of the optical signal transmitted from the transmission node 100 is reconstructed. In the following description, the signal representing the electric field of the transmission signal obtained by the sampling speed adjusting unit 23b may be referred to as a “reference signal”. Further, the transmission waveform reconstruction unit 23 corresponds to the simulated transmitter 14 shown in FIG.

FIG. 11 is a diagram illustrating the rotation of the polarization of the optical signal. Here, it is assumed that the direction of polarization is represented by using a polarization plane coordinate system composed of an H axis corresponding to horizontal polarization and a V axis corresponding to vertical polarization. The horizontally polarized waves H and the vertically polarized waves V are orthogonal to each other.

The monitor signal rotation unit 24 rotates the monitor signal in the coordinate system shown in FIG. Here, it is assumed that a set of polarization components (Ex, Ey) extracted by the adaptive equalizer 13b are in the state shown in FIG. 11A. Further, the monitor signal rotation unit 24 rotates the polarization of the monitor signal by θ. In this case, as shown in FIG. 11B, the signal Erx and the signal Ery are obtained.

The signal Ery is represented by a component obtained by projecting the polarization component Ex on the θ axis and a component obtained by projecting the polarization component Ey on the θ axis. Further, the signal Erx is represented by a component obtained by projecting the polarization component Ex on the orthogonal axis and a component obtained by projecting the polarization component Ey on the orthogonal axis.

As shown in FIG. 11C, the monitor signal rotation unit 24 executes the rotation operation. That is, the rotation matrix is multiplied by the H polarization signal Ex and the V polarization signal Ey. The rotation matrix includes a parameter θ representing the amount of rotation of the polarization. The H polarization signal Ex, the V polarization signal Ey, and the parameter θ may be once written to the register. In this case, the monitor signal rotation unit 24 reads Ex, Ey, and θ from the register for each symbol and executes the rotation operation. As a result, the signal Erx and the signal Ery shown in FIG. 11B are obtained.

The configuration and processing of the reference signal rotating unit 25 are substantially the same as those of the monitor signal rotating unit 24. That is, the reference signal rotation unit 25 executes the rotation operation on the reference signal obtained by the transmission waveform reconstruction unit 23.

The rotation amount of the monitor signal rotation unit 24 and the reference signal rotation unit 25 is designated by the rotation amount control unit 26. The rotation amount control unit 26 has, for example, a plurality of rotation amount values (0, 10, 20, ..., 180) in the range from zero to 180 degrees. Then, the rotation amount control unit 26 sequentially gives these values to the monitor signal rotation unit 24 and the reference signal rotation unit 25. Alternatively, when the rotation amount change step Δθ is constant, the rotation amount control unit 26 has parameters n (n = 0, 1, 2, ...) With respect to the monitor signal rotation unit 24 and the reference signal rotation unit 25. May be instructed. In this case, the monitor signal rotation unit 24 and the reference signal rotation unit 25 each perform a rotation operation of nΔθ.

The feature extraction unit 27 calculates the correlation value between the monitor signal and the reference signal by the method described with reference to FIGS. 3 to 5. Here, the monitor signal represents the electric field strength and the optical phase of the received optical signal. However, the monitor signal is compensated for dispersion and non-linear distortion. The reference signal represents the electric field strength and the optical phase of the optical signal transmitted from the transmission node 100. Here, the correlation value depends on the power of the optical signal, as described above. In this embodiment, the feature extraction unit 27 is designed so that the correlation value becomes large when the optical power is large. Then, the feature extraction unit 27 detects the power of the optical signal by calculating the correlation value between the monitor signal and the reference signal.

Further, the feature extraction unit 27 detects the power of the optical signal at a plurality of positions on the transmission path. The position on the transmission path is specified by a combination of the dispersion compensation amounts CD1 and CD2, for example, as described with reference to FIGS. 3 to 4. In addition, the feature extraction unit 27 detects the power of the optical signal for each rotation amount θ controlled by the rotation amount control unit 26. That is, the power of the optical signal is detected for each position and each rotation amount. Further, the feature extraction unit 27 detects the power of the optical signal for each polarization component. For example, the optical power of the polarization component in the θ-axis direction and / or the polarization component in the orthogonal axis direction shown in FIG. 11 is detected.

The position specifying unit 28 determines whether or not a polarization-dependent loss occurs on the transmission path based on the optical power detected for each position and each rotation amount. Further, when the polarization-dependent loss occurs, the position specifying unit 28 identifies the position where the polarization-dependent loss occurs.

FIG. 12 is a flowchart showing an example of a method of specifying the position where the polarization-dependent loss occurs. The processing of this flowchart is mainly executed by the monitor signal rotation unit 24, the reference signal rotation unit 25, the rotation amount control unit 26, the feature extraction unit 27, and the position specifying unit 28. The processing of this flowchart is executed for each symbol.

In S11, the monitor signal and the reference signal of the target symbol are input. The monitor signal represents the electric field of the received optical signal and is stored in the monitor signal storage unit 21. The reference signal represents the electric field of the transmission optical signal of the transmission node 100, and is generated by the transmission waveform reconstruction unit 23.

In S12, the rotation amount control unit 26 sets the rotation amount θi. The rotation amount θi is set at intervals of 10 degrees, for example, in the range of zero to 180 degrees. In this case, for example, the initial value of the rotation amount θi is zero, and the rotation amount θi is incremented by 10 degrees each time S12 to S15 are executed.

In S13, the monitor signal rotation unit 24 rotates the monitor signal by θi in the polarization plane coordinate system. The monitor signal is represented by a set of polarization components (H polarization signal and V polarization signal) that are orthogonal to each other. Here, as shown in FIG. 11A, it is assumed that the H polarization signal and the V polarization signal of the monitor signal are Ex and Ey, respectively. In this case, the output signals Erx and Ery of the monitor signal rotating unit 24 are represented by the equation (4). Note that Ex, Ey, Erx, and Ery are complex numbers expressing the electric field strength and the optical phase, respectively.

Similarly, the reference signal rotation unit 25 rotates the reference signal by θi in the polarization plane coordinate system. Here, when the H-polarized signal and the V-polarized signal of the reference signal are Esx and Esy, respectively, the output signals Esrx and Esry of the reference signal rotating unit 25 are represented by the equation (5). Note that Esx, Esy, Esrx, and Esry are complex numbers expressing the electric field strength and the optical phase, respectively.

In S14, the feature extraction unit 27 calculates the correlation value between the monitor signal and the reference signal while changing the position on the transmission path between the transmission node 100 and the optical network device 1. As described above, the position on the transmission path is specified by the combination of the dispersion compensation amounts CD1 and CD2. Here, this correlation value represents the power of the optical signal at each position. Therefore, the feature extraction unit 27 calculates the power of the optical signal at a plurality of positions on the transmission path. The feature extraction unit 27 calculates the optical power of the polarization component in the θ-axis direction and / or the polarization component in the orthogonal axis direction. That is, the feature extraction unit 27 calculates the power in the θ-axis direction and / or the power in the orthogonal axis direction of the optical signal at a plurality of positions on the transmission path. In the following description, the position where the power of the optical signal is calculated may be referred to as a "measurement position".

For example, as shown in FIG. 13A, it is assumed that the transmission node 100 and the optical network device 1 are connected by an optical fiber of 5 km. The wavelength dispersion of this optical fiber is 20 ps / nm / km. That is, the total wavelength dispersion of the transmission line between the transmission node 100 and the optical network device 1 is 100 ps / nm.

The feature extraction unit 27 calculates the correlation value at the position of 0 km (that is, the transmission end), 1 km, 2 km, 3 km, 4 km, and 5 km (that is, the reception end) with respect to the transmission node 100. Here, each position on the transmission path is represented by a combination of wavelength dispersion CD1 and CD2. For example, the position 1 km ahead of the transmission node 100 toward the optical network device 1 is represented by "CD1 = 80, CD2 = 20". Further, the position 2 km ahead of the transmission node 100 toward the optical network device 1 is represented by "CD1 = 60, CD2 = 40". As a result, for example, when the rotation amount set by the rotation amount control unit 26 is “θ0”, the feature extraction unit 27 obtains the correlation values C00 to C50 shown in FIG. 13B. Note that C00, C10, C20, C30, C40, and C50 are located at positions 0 km, 1 km, 2 km, 3 km, 4 km, and 5 km advanced from the transmission node 100 toward the optical network device 1 when the rotation amount is θ0, respectively. Represents the correlation value corresponding to (that is, the power of the optical signal). Assuming that there is no polarization-dependent loss, the power of the optical signal on the transmission path is gradually attenuated substantially linearly according to the transmission distance, as shown in FIG. 13 (a). In this way, in S14, a power profile for one rotation amount θi is generated.

In S15, the feature extraction unit 27 determines whether or not a power profile has been created for all rotation amounts (θ0 to θN). Then, if the amount of rotation for which the power profile has not been created remains, the processing of the feature extraction unit 27 returns to S12. That is, the feature extraction unit 27 creates a power profile for each rotation amount. As a result, as shown in FIG. 13B, optical power corresponding to a plurality of measurement positions on the transmission path can be obtained for each rotation amount θ0 to θ4. At this time, the feature extraction unit 27 creates a power profile for the polarization component in the θ-axis direction and / or the polarization component in the orthogonal axis direction. In the example shown in FIG. 13B, the constant N used in the flowchart is 4.

In S21, the position specifying unit 28 calculates the difference (dk) between the maximum power value and the minimum power value for the measurement position k. For example, when the process of S21 is executed for the position X2, the position specifying unit 28 extracts the maximum value and the minimum value from the correlation values C20 to C24. Then, the position specifying unit 28 calculates the difference between the extracted maximum value and the minimum value. That is, the power difference dk representing the difference between the maximum power value and the minimum power value is calculated. The difference between the maximum power value and the minimum power value calculated in S21 (or the difference between the maximum value and the minimum value extracted from a plurality of correlation values) is an example of the fluctuation amount of the evaluation value.

Here, when a polarization-dependent loss occurs on the transmitting side of the position X2, the power of the optical signal detected at the position X2 may decrease depending on the angle of polarization of the optical signal. For example, when the direction of polarization of one of a set of optical signals constituting a polarization-multiplexed optical signal coincides with or substantially coincides with the direction in which polarization-dependent loss occurs, the power of the optical signal decreases. do. Therefore, when the polarization-dependent loss is generated, if the electric field information signal representing the monitor signal and the reference signal is rotated, the amount of fluctuation in the power of the optical signal detected at the position X2 becomes large. Therefore, when the fluctuation amount of the correlation values C20 to C24 representing the power of the optical signal is large when the monitor signal and the reference signal are rotated, it is estimated that the polarization-dependent loss occurs on the transmitting side of the position X2. To. That is, when the difference between the maximum value and the minimum value among the correlation values C20 to C24 is large, it is estimated that the polarization-dependent loss occurs on the transmitting side of the position X2.

The difference between the maximum value and the minimum value of the correlation value obtained when the rotation amount of the rotation calculation is changed within a predetermined range for each measurement position is a set of signals constituting the polarization multiplex optical signal. The power difference between the two signals when one of the signals (eg, the X polarization signal) is set in the PDL axis direction and the other signal (eg, the Y polarization signal) is set in the orthogonal axis direction. Corresponds to. Therefore, the power difference calculated in S21 corresponds to the power difference between the polarizations caused by the polarization-dependent loss.

In S22, the position specifying unit 28 calculates the difference (ddk) between the power difference dk calculated for the measurement position k and the power difference dk-1 previously obtained for the measurement position k-1. do. The measurement position k-1 represents a measurement position adjacent to the transmission side of the measurement position k.

S23 is provided to execute S21 to S22 for all measurement positions. Then, when the processing of S21 to S22 is completed for all the measurement positions, the processing of the position specifying unit 28 proceeds to S24.

In S24, the position specifying unit 28 extracts the measurement position k whose difference ddk is larger than the threshold value. Here, when the difference ddk calculated with respect to the measurement position k is larger than the threshold value, the position specifying unit 28 states that a polarization-dependent loss between the measurement position k and the measurement position k-1 has occurred. judge. After that, the position specifying unit 28 provides information indicating the position where the polarization-dependent loss is presumed to occur (actually, the range where the polarization-dependent loss is presumed to occur) in S25. Output. The position information output from the position specifying unit 28 is stored in, for example, a server computer. In this case, the network administrator can check the state of the transmission line by using this server computer. When the difference ddk is smaller than the threshold value, the position specifying unit 28 determines that the polarization-dependent loss occurring between the adjacent measurement position and the target measurement position has not occurred. In this case, the position specifying unit 28 may output information indicating that no polarization-dependent loss has occurred.

The threshold value of S24 may be zero. However, when this threshold value is zero (or a value close to zero), an erroneous determination result may occur due to noise or the like. Therefore, it is preferable to set a predetermined positive value for this threshold value in consideration of noise and the like.

FIG. 14 shows an example of the position specifying unit 28. The position specifying unit 28 includes a power profile storage unit 28a, a plurality of difference calculation units 28b (# 0 to #K), and a change point search unit 28c. The power profile storage unit 28a stores the power profile generated by the feature extraction unit 27. As shown in FIG. 13 (b), the power profile represents an optical power value (or a correlation value) with respect to a combination of position and rotation amount. A plurality of difference calculation units 28b are provided corresponding to a plurality of measurement positions on the transmission path. Each difference calculation unit 28b acquires a plurality of optical power values obtained for the corresponding measurement positions, and extracts the maximum power value and the minimum power value from them. Then, each difference calculation unit 28b calculates a power difference representing the difference between the maximum power value and the minimum power value. The power difference calculated by each difference calculation unit 28b corresponds to the polarization-dependent loss at the corresponding position. Further, the processing of each difference calculation unit 28b corresponds to S21 shown in FIG.

The change point search unit 28c arranges the power differences (d1, d2, ... dN) obtained by the difference calculation unit 28b in order based on the transmission distance from the transmission node 100. Then, the change point search unit 28c searches for a position where the change in power difference between the measurement positions adjacent to each other is larger than the threshold value. For example, in FIG. 13A, the power difference d1 obtained with respect to the position X1 (that is, the PDL detected at the position X1) and the difference value d2 obtained with respect to the position X2 (that is, detected at the position X2). When the difference dd2 from PDL) is larger than the threshold value, the change point search unit 28c determines that a polarization-dependent loss has occurred between the position X1 and the position X2.

FIG. 15 shows an example of a method for identifying the position where the polarization-dependent loss occurs. In this embodiment, the polarization rotation amounts of the monitor signal and the reference signal are θ1, θ2, and θ3. Here, the correlation value obtained by the feature extraction unit 27 may be referred to as a “power value”.

FIG. 15A shows the power values C11 to C31 calculated for the positions X1 to X3 when the rotation amount is θ1. Similarly, FIG. 15B shows power values C12 to C32 calculated for positions X1 to X3 when the amount of rotation is θ2. FIG. 15C shows the power values C13 to C33 calculated for the positions X1 to X3 when the rotation amount is θ3. Then, the position specifying unit 28 determines whether or not a polarization-dependent loss has occurred by using the power value calculated for each measurement position (here, X1 to X3).

At the position X1, the power values C11, C12, and C13 are obtained for the rotation amounts θ1, θ2, and θ3, respectively. In this embodiment, the power values C11, C12, and C13 are substantially the same as each other. In this case, since the difference between the maximum power value and the minimum power value (that is, the power difference d1) is almost zero, it is determined that the polarization-dependent loss has not occurred.

At the position X2, the power values C21, C22, and C23 are obtained for the rotation amounts θ1, θ2, and θ3, respectively. In this embodiment, the power values C21 and C23 are substantially the same as each other, but the power value C22 is smaller than the other power values. Then, the difference between the maximum power value and the minimum power value obtained at the position X2 (that is, the power difference d2) becomes larger than the power difference d1 obtained at the position X1. That is, the power difference d2 obtained at the position X2 is larger than the power difference d1 obtained at the measurement position adjacent to the transmitting side of the position X2. Therefore, "position X2" is extracted in S24.

At the position X3, the power values C31, C32, and C33 are obtained for the rotation amounts θ1, θ2, and θ3, respectively. In this embodiment, the power values C31 and C33 are substantially the same as each other, but the power value C32 is smaller than the other power values. However, the difference between the maximum power value and the minimum power value obtained at the position X3 (that is, the power difference d3) is almost the same as the power difference d2 obtained at the position X2. Therefore, "position X3" is not extracted in S24. As a result, in the case shown in FIG. 15, it is determined that the polarization-dependent loss occurs between the position X1 and the position X2.

FIG. 16 shows another example of how to identify the location where the polarization dependent loss is occurring. Also in this embodiment, as shown in FIG. 13A, the transmission node 100 and the optical network device 1 are connected by an optical fiber of 5 km. Then, the measurement result shown in FIG. 16 is obtained. The value shown in FIG. 16 represents a correlation value (or optical power of an optical signal) obtained for a combination of a position and a rotation amount θ. In the following description, the value shown in FIG. 16 may be referred to as an “evaluation value”. Further, the "difference" shown in FIG. 16 represents the difference between the maximum value and the minimum value among the evaluation values calculated for each measurement position.

At position X0, the maximum and minimum evaluation values are "81" and "80", respectively, and the difference d0 is "1". Further, at the position X1, the maximum evaluation value and the minimum evaluation value are “71” and “69”, respectively, and the difference d1 is “2”. Here, in this example, it is assumed that the threshold value of S24 is “5”. Then, the result of subtracting the difference d0 from the difference d1 is smaller than the threshold value. Therefore, the position specifying unit 28 determines that no polarization-dependent loss has occurred on the transmitting side from the position X1.

At position X2, the maximum and minimum evaluation values are "60" and "50", respectively, and the difference d2 is "10". Here, the result of subtracting the difference d1 from the difference d2 is "8", which is larger than the threshold value. Therefore, in this case, the position specifying unit 28 determines that a polarization-dependent loss has occurred between the position X1 and the position X2.

At position X3, the maximum and minimum evaluation values are "50" and "40", respectively, and the difference d3 is "10". However, the result of subtracting the difference d2 from the difference d3 is "0", which is smaller than the threshold value. Therefore, the position specifying unit 28 determines that no polarization-dependent loss has occurred between the position X2 and the position X3. Hereinafter, the same determination result as that of the position X3 can be obtained for the positions X4 and X5.

In this way, the optical network device 1 can identify the position where the polarization-dependent loss occurs by analyzing the electric field information of the received optical signal. Therefore, the labor and / or time required to identify the position where the polarization-dependent loss is occurring can be reduced.

Also in the method shown in FIG. 5, the position where the polarization-dependent loss occurs can be specified by analyzing the electric field information of the received optical signal. However, in the method shown in FIG. 5, the polarization of the optical signal is actually rotated in the transmission node 200. Then, the optical network device 300 estimates the polarization-dependent loss by acquiring the electric field information of the received optical signal when the rotation amount of the polarization of the optical signal is held at a constant value. Therefore, in the method shown in FIG. 5, the time required to identify the position where the polarization-dependent loss occurs becomes long. On the other hand, in the embodiment of the present invention, the polarization-dependent loss can be detected by performing rotational conversion on the electric field information of the received optical signal without actually rotating the polarization of the optical signal. Therefore, according to the embodiment of the present invention, the time required to specify the position where the polarization-dependent loss occurs can be shortened as compared with the method shown in FIG.

FIG. 17 shows an example of an optical transmission system composed of a plurality of spans. In this embodiment, the length of the transmission line between the transmission node (Tx) 100 and the optical network device (Rx) 1 is 14 km, and optical amplifiers are mounted at positions X5 and X10, respectively. Then, a local polarization-dependent loss occurs between the position X6 and the position X7.

The optical network device 1 measures the optical power at each measurement position (X1 to X14) based on the electric field information of the received optical signal. Then, the optical network device 1 estimates the position where the polarization-dependent loss occurs based on the change in power between adjacent measurement positions. Therefore, regardless of the number of spans of the optical transmission system, the position where the polarization-dependent loss occurs can be estimated by the same method.

FIG. 18 schematically shows the processing of the optical network device 1. As described above, the transmission node 100 generates and transmits a polarized wave multiplex optical signal. The electric field of the polarization multiplex optical signal is represented by Esx and Esy. Esx and Esy are orthogonal to each other.

The polarization of the polarization multiplex optical signal rotates in the optical transmission line. Further, the polarization-dependent optical signal is affected by the polarization-dependent loss in the transmission line. Then, the coherent receiver 11 of the optical network device 1 generates electric field information Ex0 and Ey0.

The adaptive equalizer 13b corrects the output signal of the coherent receiver 11 to correct the first polarization component Ex and the second polarization component that are orthogonal to each other in the electric field information signal representing the electric field of the polarization multiplex optical signal. Separated from Ey. In this embodiment, Ex and Ey are used as monitor signals.

The classifier 13d reproduces the transmission data based on the output signal of the adaptive equalizer 13b. Then, the transmission waveform reconstruction unit 23 generates electric field information by mapping the reproduced transmission data. This electric field information is substantially the same as the electric field of the polarized wave multiplex optical signal generated in the transmission node 100. That is, the electric field information Esx and Esy are reconstructed. In this embodiment, the reconstructed Esx and Esy are used as reference signals.

The monitor signal rotation unit 24 generates electric field information Erx, Ery by rotating the monitor signal (Ex, Ey) by θ around the origin of the polarization plane coordinate system. Similarly, the reference signal rotation unit 25 generates electric field information Esrx and Esry by rotating the reference signal (Esx, Esy) by θ with respect to the origin of the polarization plane coordinate system. The rotation amount θ is designated by the rotation amount control unit 26. The feature extraction unit 27 calculates the optical power for each rotation amount θ by calculating the correlation between the monitor signal (Erx, Ery) after the rotation conversion and the reference signal (Esrx, Esry) after the rotation conversion. The feature extraction unit 27 may calculate the optical power of the X polarization signal by calculating the correlation between the polarization component Erx of the monitor signal after the rotation conversion and the polarization component Erx of the reference signal after the rotation conversion. good. Further, the feature extraction unit 27 calculates the optical power of the Y polarization signal by calculating the correlation between the polarization component Ery of the monitor signal after the rotation conversion and the polarization component Esry of the reference signal after the rotation conversion. You may. At this time, the optical power is calculated for each of the plurality of measurement positions on the transmission path. As a result, a power profile for each rotation amount θ is created. Further, for each of the X polarization signal and the Y polarization signal, a power profile for each rotation amount θ is created.

The flowchart shown in FIG. 12 is an embodiment, and the present invention is not limited to this procedure. For example, in the flowchart shown in FIG. 12, the polarization-dependent loss is detected based on the difference between the maximum power value and the minimum power value, but the polarization-dependent loss is detected based on the ratio between the maximum power value and the minimum power value. You may. Alternatively, the difference between the optical power of the X-polarized signal and the optical power of the Y-polarized signal may be monitored.

Further, the flowchart shown in FIG. 12 is executed for one symbol, but the present invention is not limited to this method. That is, considering the noise on the transmission path, it is preferable to calculate the correlation value (that is, optical power) for a plurality of symbols and specify the position where the polarization-dependent loss occurs based on the average of them.

Further, in the configuration shown in FIG. 9, the monitor signal represents the electric field of the optical signal whose dispersion is compensated by the fixed equalizer 13a (and the adaptive equalizer 13b). On the other hand, the feature extraction unit 27 designates a position on the transmission path by utilizing the dispersion added to the received optical signal. Therefore, the optical network device 1 may have a function of adding the variance compensated by the fixed equalizer 13a to the input side or the output side of the monitor signal rotating unit 24.

The digital signal processing unit 13, the monitor signal rotation unit 24, the reference signal rotation unit 25, the rotation amount control unit 26, the feature extraction unit 27, and the position identification unit 28 are realized by, for example, one or a plurality of processors. In this case, a program describing the functions of the digital signal processing unit 13, the monitor signal rotation unit 24, the reference signal rotation unit 25, the rotation amount control unit 26, the feature extraction unit 27, and the position specifying unit 28 is stored in a memory (not shown). Will be done. Then, by executing the program by the processor, the functions of the digital signal processing unit 13, the monitor signal rotation unit 24, the reference signal rotation unit 25, the rotation amount control unit 26, the feature extraction unit 27, and the position specifying unit 28 are provided. Will be done. In addition, these functions may be realized by ASIC, FPGA or the like. Further, these functions may be realized by a combination of software and hardware circuits.

<Variations>
FIG. 19 shows a first variation of the optical network device according to the embodiment of the present invention. In FIG. 19, the coherent receiver 11 and the digital signal processing unit 13 shown in FIG. 9 are omitted.

The optical network device 1B includes a plurality of calculation units 30 (# 1 to #N). Each calculation unit 30 includes a monitor signal rotation unit 24, a reference signal rotation unit 25, and a feature extraction unit 27. The monitor signal rotation unit 24, the reference signal rotation unit 25, and the feature extraction unit 27 are substantially the same in FIGS. 9 and 19. Therefore, the calculation unit 30 performs rotation conversion on the monitor signal and the reference signal. Then, the calculation unit 30 calculates the optical power at each measurement position by calculating the correlation value between the monitor signal after the rotation conversion and the reference signal.

The calculation unit 30 is provided for each rotation amount of rotation conversion. For example, the calculation unit 30 # 1 executes the conversion of the rotation amount θ1, and the calculation unit 30 # 2 executes the conversion of the rotation amount θ2. Therefore, when calculating the optical power every 10 degrees in the range of zero to 180 degrees, 19 calculation units 30 are mounted in parallel. Here, most of the load of the process of specifying the position where the polarization-dependent loss occurs is the load of the process of creating the power profile by the feature extraction unit. Therefore, by mounting N calculation units 30 in parallel, the calculation time can be reduced to about 1/N. For example, in the optical network device 1B, the processes of S12 to S14 are executed in parallel by the plurality of calculation units 30. The plurality of calculation units 30 may be realized, for example, by allocating CPU cores to each calculation unit 30. Further, since the rotation amount θ of each calculation unit 30 is predetermined, the optical network device 1B does not need to include the rotation amount control unit 26.

FIG. 20 shows a second variation of the optical network device according to the embodiment of the present invention. In the optical network device 1 shown in FIG. 9, the output signal of the adaptive equalizer 13b is used as a monitor signal. On the other hand, in the optical network device 1C shown in FIG. 20, the output signal of the fixed equalizer 13a is used as a monitor signal. That is, the electric field information signal before being processed by the adaptive equalizer 13b is used as the monitor signal.

Here, the adaptive equalizer 13b corrects the polarization rotation generated in the transmission line. That is, in the second variation, a monitor signal in which the polarization rotation is not corrected is generated. Therefore, the optical network device 1C includes a polarization rotation correction unit 41. The polarization rotation correction unit 41 corrects the polarization rotation generated in the transmission line for the monitor signal. At this time, the polarization rotation correction unit 41 may acquire the parameters used in the adaptive equalizer 13b and correct the polarization rotation based on the parameters. For example, when the adaptive equalizer 13b is realized by a digital filter, the parameter representing the correction amount of the polarization rotation corresponds to the tap coefficient of the digital filter. In this case, the polarization rotation correction unit 41 acquires the tap coefficient from the adaptive equalizer 13b and corrects the polarization rotation. Alternatively, the polarization rotation correction unit 41 calculates the correlation value with the reference signal before the polarization rotation while rotating the polarization of the monitor signal little by little, and monitors the rotation amount at which the correlation value becomes maximum. The polarization of the signal may be corrected.

Further, the polarization rotation correction unit 41 may determine the correction amount of the polarization rotation and notify the monitor signal rotation unit 24 of the determined correction amount. In this case, the monitor signal rotation unit 24 adds the correction amount determined by the polarization rotation correction unit 41 to the rotation amount θ notified from the rotation amount control unit 26, and executes the rotation calculation of the polarization.

By the way, the adaptive equalizer 13b not only corrects the polarization rotation but also compensates for the residual dispersion. At this time, the nonlinear distortion component in the electric field information signal of the received optical signal is affected. On the other hand, the feature extraction unit 27 calculates the optical power at a desired position on the transmission path by utilizing the nonlinear distortion component in the electric field information signal of the received optical signal. Therefore, in the configuration shown in FIG. 9, the optical power is calculated based on the electric field information in which the nonlinear strain component is corrected. On the other hand, in the second variation shown in FIG. 20, the optical power is calculated using the electric field information in which the nonlinear strain component is not corrected. Therefore, according to the second variation shown in FIG. 20, it is expected that the calculation accuracy of the optical power will be improved as compared with the configuration shown in FIG.

FIG. 21 shows a third variation of the optical network device according to the embodiment of the present invention. In the optical network device 1 shown in FIG. 9, the output signal of the adaptive equalizer 13b is used as a monitor signal. On the other hand, in the optical network device 1D shown in FIG. 21, the output signal of the coherent receiver 11 is used as a monitor signal. That is, the electric field information signal before being processed by the fixed equalizer 13a and the adaptive equalizer 13b is used as the monitor signal.

In addition to the configuration shown in FIG. 9, the optical network device 1D includes a polarization rotation correction unit 41 and a dispersion compensation unit 42. Since the polarization rotation correction unit 41 is substantially the same in FIGS. 20 and 21, the description thereof will be omitted.

The configuration and function of the dispersion compensation unit 42 are substantially the same as those of the fixed equalizer 13a. That is, the fixed equalizer 13a and the dispersion compensating unit 42 compensate for known waveform deterioration components (for example, wavelength dispersion of the transmission line). However, the fixed equalizer 13a needs to process continuous input signals in real time. Therefore, the accuracy of the compensation process by the fixed equalizer 13a is not necessarily high. On the other hand, the dispersion compensation unit 42 processes the signal stored in the monitor signal storage unit 21. Here, the process of monitoring the polarization-dependent loss does not necessarily have to be executed in real time. That is, the dispersion compensation unit 42 can increase the processing time per symbol as compared with the fixed equalizer 13a. Therefore, the dispersion compensating unit 42 can compensate the dispersion with higher accuracy than the fixed equalizer 13a. For example, in the dispersion compensation unit 42, the frequency resolution may be increased or the number of taps of the digital filter may be increased. Therefore, since an accurate monitor signal (that is, electric field information of the received optical signal) can be obtained, the position where the polarization-dependent loss occurs can be accurately specified.

FIG. 22 shows a fourth variation of the optical network device according to the embodiment of the present invention. In FIG. 22, only the feature extraction unit and the function connected to the feature extraction unit are drawn, and other circuits or functions are omitted.

In the fourth variation, the monitor signal rotation unit 24 and the reference signal rotation unit 25 are mounted in the feature extraction unit 27. Then, the monitor signal is compensated by the first dispersion compensation unit 16a, the nonlinear compensation unit 16b, and the second dispersion compensation unit 16c, and then processed by the monitor signal rotation unit 24. The method of performing the rotation calculation after the compensation process and the method of performing the compensation process after the rotation calculation are mathematically equivalent to each other.

<Second embodiment>
As described above, the optical network device according to the embodiment of the present invention is polarization-dependent by creating a power profile on an optical transmission path using a received optical signal and specifying a position where the optical power changes abruptly. Estimate the location where the loss is occurring. However, in the embodiment described with reference to FIGS. 6 to 22 (hereinafter, may be referred to as “first embodiment”), the occurrence of polarization-dependent loss may not be detected accurately.

FIG. 23 shows an example of a change in the power of an optical signal output from a transmitting node. The power of the optical signal decreases as the transmission distance from the transmitting node increases. In the example shown in FIG. 23A, an optical signal of 1 mW is transmitted from the transmission node, and the optical power gradually decreases as the transmission distance increases.

When a polarization-dependent loss occurs on the optical transmission path, the optical power in the direction in which the loss occurs drops sharply as shown in FIG. 23 (b). Then, in the first embodiment, the optical network device 1 calculates the power of the optical signal at predetermined intervals and identifies the position where the optical power changes abruptly, so that the position where the polarization-dependent loss occurs is generated. To estimate.

On the other hand, in the case where the polarization-dependent loss occurs at the position where the power of the optical signal is small, the amount of change ΔP of the optical power due to the polarization-dependent loss is also as shown in FIG. 23 (c). small. Here, the optical signal propagating in the optical transmission line contains noise. Therefore, when the amount of change ΔP of the optical power due to the polarization-dependent loss is small, the optical network device 1 may not be able to accurately detect the occurrence of the polarization-dependent loss.

On the other hand, in an optical transmission system that realizes long-distance transmission, one or more optical amplifiers are mounted between a transmission node and a reception node. In the example shown in FIG. 24, two optical amplifiers are mounted between the transmitting node and the receiving node. Then, the optical power has a peak at the position where the optical amplifier is mounted, and the optical power gradually decreases as the transmission distance from the optical amplifier increases. In the following description, the section from the peak of the optical power generated by each optical amplifier to the peak of the optical power generated by the adjacent optical amplifier may be referred to as "span".

Here, it is assumed that a polarization-dependent loss occurs on the optical transmission path. In the example shown in FIG. 24, the optical power of the X-polarized signal is 0.1 mW and the optical power of the Y-polarized signal is 0.05 mW at the receiving port of the optical amplifier. In this case, it is difficult to detect the polarization-dependent loss because the power difference between the polarizations is small.

However, when the polarization multiplex optical signal is transmitted between the nodes, the optical amplifier usually amplifies the X polarization signal and the Y polarization signal without separating the polarization. As a result, for example, in the output port of the optical amplifier, the optical power of the X-polarized signal is amplified to 1 mW, and the optical power of the Y-polarized signal is amplified to 0.5 mW. In this case, the power difference between the polarizations is sufficiently large, and the polarization-dependent loss can be easily detected.

Therefore, the optical network device of the second embodiment is a state of light at a point where the power of an optical signal has a peak in an optical transmission system in which one or more optical amplifiers are mounted between a transmitting node and a receiving node. Estimate the occurrence of polarization-dependent loss based on. According to this method, the accuracy of detecting the polarization-dependent loss is in span units. That is, it is detected in which of the plurality of spans the polarization-dependent loss occurs.

FIG. 25 shows an example of an optical network device according to a second embodiment of the present invention. The configuration of the optical network device 2 according to the second embodiment of the present invention is substantially the same as that of the optical network device 1 shown in FIG. That is, the optical network device 2 includes a coherent receiver 11, a digital signal processing unit 13, a monitor signal rotation unit 24, a reference signal rotation unit 25, a rotation amount control unit 26, and a feature extraction unit 27. However, the optical network device 2 includes a span specifying unit 51 instead of the position specifying unit 28 shown in FIG.

Similar to the position specifying unit 28, the span specifying unit 51 identifies the location where the polarization-dependent loss occurs based on the power profile generated by the feature extraction unit 27. Therefore, the span specifying portion 51 is an example or variation of the position specifying portion 28. However, the span specifying unit 51 specifies the location where the polarization-dependent loss occurs on a span-by-span basis. That is, the span specifying unit 51 sets a plurality of spans by dividing the optical transmission path based on the position of the optical amplifier provided on the optical transmission path, and the polarization-dependent loss occurs in any of the plurality of spans. Detect if it is occurring.

FIG. 26 shows an example of the span specifying portion 51. In this embodiment, the span specifying unit 51 includes a power profile storage unit 51a, a peak detection unit 51b, a power difference calculation unit 51c, and a determination unit 51d.

The power profile storage unit 51a stores the power profile generated by the feature extraction unit 27. The peak detection unit 51b detects the position where the power peak of the optical signal appears on the optical transmission path based on the power profile. In the following description, this position may be referred to as a "peak position". The power difference calculation unit 51c calculates the difference between the power of the X polarization signal and the power of the Y polarization signal for each polarization rotation amount at each peak position. At this time, the power difference calculation unit 51c may obtain the maximum value of the difference between the power of the X polarization signal and the power of the Y polarization signal at each peak position. When a power difference larger than a predetermined threshold value is detected, the determination unit 51d estimates that a polarization-dependent loss occurs in the span including the peak position where the power difference is detected.

FIG. 27 shows an example of a span and power profile. The position on the optical transmission path (that is, the transmission distance from the transmission node) is represented by a dispersion compensation amount as described with reference to FIG. 3 or FIG. In this embodiment, the position on the optical transmission path is represented using the dispersion compensation amount CD2 compensated by the second dispersion compensation unit 16c shown in FIG. For example, when the wavelength dispersion of the optical fiber is 20 ps / nm / km, "CD2 = 20" corresponds to "1 km".

In the example shown in FIG. 27, optical amplifiers are mounted at positions corresponding to "CD2 = 100" and "CD = 200", respectively. Then, the power of the optical signal decreases as the transmission distance from the transmitting node increases, and is amplified in the optical amplifier. Further, the power of the optical signal decreases as the transmission distance from the optical amplifier increases.

As described above, "span" represents the interval from the peak of optical power to the adjacent peak. Therefore, in this embodiment, the section between the transmitting node and the optical amplifier A1 is the span S1, the section between the optical amplifier A1 and the optical amplifier A2 is the span S2, and the optical amplifier A2 and the receiving node. The section between them is span S3.

FIG. 28 shows an example of a power profile in the optical transmission line shown in FIG. 27. In this example, the optical powers of the X-polarized signal and the Y-polarized signal at each measurement position P0 to P15 are estimated. Each measurement position is actually represented by the dispersion compensation amounts CD1 and CD2. As described with reference to FIGS. 1 to 4, the power of the optical signal correlates with the received signal (monitor signal) and the transmitted signal (reference signal) after the wavelength dispersion and the nonlinear distortion are compensated. Estimated from the represented correlation value. Further, the power profile is created for each of the plurality of polarization rotation amounts. In this example, power profiles are created for 0 degrees, 45 degrees, 90 degrees, and 135 degrees, respectively. The polarization rotation amount corresponds to the rotation amount of the rotation calculation by the monitor signal rotation unit 24 and the reference signal rotation unit 25.

For example, when the amount of rotation of polarization is zero, x00 to x0f are obtained as the estimated optical power of the X-polarized signal at positions P0 to P15, and the estimated optical power of the Y-polarized signal at positions P0 to P15. y00 to y0f are obtained. The same applies to other polarization rotation amounts.

Next, the operation of the span specifying unit 51 when the power profile shown in FIG. 28 is created for the optical transmission line shown in FIG. 27 will be described. It is assumed that the power profile is created by the feature extraction unit 27 and stored in the power profile storage unit 51a.

The peak detection unit 51b sets a plurality of spans by dividing the optical transmission line based on the span information. In the example shown in FIG. 27, a transmission node and a reception node are mounted at positions P0 and P15, respectively. Further, the optical amplifier A1 and the optical amplifier A2 are mounted at the positions P5 and P10, respectively. Therefore, the positions P0 to P5 correspond to the span S1, the positions P5 to P10 correspond to the span S2, and the positions P10 to P15 correspond to the span S3. The optical network device 2 according to the embodiment of the present invention is mounted on the receiving node.

The span information represents the position where the optical amplifiers A1 and A2 are mounted in this example. Further, the span information is given to the optical network device 2 by, for example, the network administrator.

Subsequently, the peak detection unit 51b detects the position where the peak of the optical power appears on the optical transmission path based on the power profile. At this time, the peak detection unit 51b may specify the position where the maximum value of the optical power is obtained for each span. However, in this embodiment, since the position on the optical transmission path is estimated by using the dispersion compensation amount, an error is included. Therefore, it is preferable that the range for searching the position where the maximum value of the optical power is obtained for each span is shifted to the transmitting node side with respect to the actual span. For example, in the example shown in FIG. 27, the search range corresponding to the span S1 is the position P0 to P4, the search range corresponding to the span S2 is the position P4 to P9, and the search range corresponding to the span S3 is the position P9 to. It is P14. In this case, the peak detection unit 51b identifies the position where the maximum value of the optical power can be obtained from the positions P0 to P4 with respect to the span S1, and the optical power from the positions P4 to P9 with respect to the span S2. The position where the maximum value of the light is obtained is specified, and the position where the maximum value of the optical power is obtained is specified from the positions P9 to P14 with respect to the span S3.

The peak detection unit 51b may detect the peak position based on the optical power of the X polarization signal, may detect the peak position based on the optical power of the Y polarization signal, or may be X-biased. The peak position may be detected based on the sum of the optical powers of the wave signal and the Y polarization signal. Further, the peak detection unit 51b can detect the peak position based on the optical power obtained for an arbitrary polarization rotation amount.

29 to 30 show an example of the change of the power profile with respect to the polarization rotation calculation. 29 (a), 29 (b), 30 (a), and 30 (b) are respectively for the polarized wave multiplex optical signal transmitted via the optical transmission line shown in FIG. 27. The power profile when the polarization rotation amount is 0 degree, 45 degree, 90 degree, 135 degree is shown. The solid line represents the power profile of the X-polarized signal, and the broken line represents the power profile of the Y-polarized signal. However, in FIGS. 29 (b) and 30 (b), the two power profiles match each other over the entire section. Further, in FIGS. 29 (b) and 30 (b), the two power profiles coincide with each other at positions P0 to P7.

In this embodiment, a polarization-dependent loss occurs between positions P6 and P7. Further, it is assumed that a loss occurs in the direction of X polarization. In this case, when the polarization rotation amount is "0 degree" by the monitor signal rotation unit 24 and the reference signal rotation unit 25, a loss occurs for the X polarization signal as shown in FIG. 29 (a). .. At this time, no loss occurs for the Y polarization signal. That is, a power difference is generated between the X-polarized signal and the Y-polarized signal. Further, when the optical signal is amplified by the optical amplifier at the position P10, the power difference thereof also becomes large.

When the amount of polarization rotation is "45 degrees", the loss for the X polarization signal and the loss for the Y polarization signal are the same. Therefore, as shown in FIG. 29 (b), no power difference occurs between the X-polarized signal and the Y-polarized signal.

When the amount of polarization rotation is "90 degrees", a loss occurs for the Y polarization signal and no loss occurs for the X polarization signal. That is, as shown in FIG. 30A, a power difference occurs between the X-polarized signal and the Y-polarized signal. However, when the polarization rotation amount is "0 degrees", the optical power of the X polarization signal is smaller than the optical power of the Y polarization signal, but when the polarization rotation amount is "90 degrees", The optical power of the Y polarization signal is smaller than the optical power of the X polarization signal.

The power profile when the polarization rotation amount is "135 degrees" is substantially the same as the power profile when the polarization rotation amount is "45 degrees". That is, as shown in FIG. 30 (b), no power difference occurs between the X-polarized signal and the Y-polarized signal.

For such a power profile, the peak detection unit 51b detects the peak position for each span. Specifically, the peak detection unit 51b detects peak positions in sections P0 to P4, sections P4 to P9, and sections P9 to P14, respectively, as described with reference to FIG. 27. As a result, P0, P5, and P10 are obtained as peak positions for the spans S1, S2, and S3, respectively.

As shown in FIG. 31A, the power difference calculation unit 51c calculates the power difference for each polarization rotation amount at each peak position. For example, as the power difference at the peak position P10, "x0a-y0a" is calculated for the polarization rotation amount "0 degree", and "x1a-y1a" is calculated for the polarization rotation amount "45 degrees". "X2a-y2a" is calculated for the polarization rotation amount "90 degrees", and "x3a-y3a" is calculated for the polarization rotation amount "135 degrees". The same applies to the power difference at the peak positions P0 and P5.

The determination unit 51d compares each power difference obtained by the power difference calculation unit 51c with a predetermined threshold value. Then, when a power difference larger than the threshold value is detected, the determination unit 51d determines that a polarization-dependent loss has occurred on the optical transmission path. In the example shown in FIGS. 29 to 30, at the peak position P0, the difference (that is, the power difference) between the optical power of the X polarization signal and the optical power of the Y polarization signal is for any polarization rotation amount. Is almost zero. That is, at the peak position P0, a power difference larger than the threshold value does not occur. The same applies to the peak position P5. On the other hand, at the peak position P10, when the amount of rotation of the polarization is "0 degree", a large power difference "x0a-y0a" is generated as shown in FIG. 29 (a). Even when the amount of polarization rotation is "90 degrees", a large power difference "x2a-y2a" is generated as shown in FIG. 30A. As a result, the determination result shown in FIG. 31 (b) can be obtained.

Further, the determination unit 51d identifies the location where the polarization-dependent loss occurs. Specifically, the determination unit 51d causes a polarization-dependent loss between the peak position where a power difference larger than the threshold value is detected and the peak position adjacent to the transmission node side with respect to the peak position. Is determined. In this example, a power difference larger than the threshold value is detected at the peak position P10. Further, the peak position adjacent to the transmission node side with respect to the peak position P10 is "P5". Therefore, in this case, it is determined that the polarization-dependent loss occurs in the section between the position P5 and the position P10.

FIG. 32 is a flowchart showing an example of a method of specifying the position where the polarization-dependent loss occurs in the second embodiment. It is assumed that the feature extraction unit 27 has created a power profile for each polarization rotation amount before the processing of this flowchart is started.

In S31, the peak detection unit 51b sets a plurality of spans on the optical transmission path between the transmission node and the optical network device 2 based on the span information. In S32, the power profile corresponding to the polarization rotation amount nΔθ is acquired. The variable n specifies the amount of polarization rotation by the monitor signal rotation unit 24 and the reference signal rotation unit 25. Further, it is assumed that the variable n is initialized to zero at the start of this flowchart. Δθ represents the step of the amount of rotation of polarization. In the examples shown in FIGS. 29 to 30, n is 0, 1, 2, or 3, and Δθ is 45 degrees. In S33, the peak detection unit 51b detects the peak position for each span. At this time, the peak detection unit 51b detects, for example, the measurement position where the maximum optical power is detected in the search range set for each span.

In S34, the power difference calculation unit 51c calculates the power difference at each peak position. That is, the difference between the optical power of the X-polarized signal and the optical power of the Y-polarized signal at each peak position is calculated.

In S35 to S36, the span specifying unit 51 increments the variable n. Then, when the variable n is smaller than the threshold value N, the processing of the span specifying unit 51 returns to S32. The threshold value N specifies the range of polarization rotation by the monitor signal rotation unit 24 and the reference signal rotation unit 25. In the examples shown in FIGS. 29 to 30, N is 4. Then, when the variable n reaches the threshold value N, the processing of the span specifying unit 51 proceeds to S41. That is, the span specifying unit 51 repeatedly executes S31 to S36 to calculate the power difference at each peak position for each polarization rotation amount.

In S41, the determination unit 51d acquires the power difference corresponding to the kth span. Specifically, the power difference corresponding to each polarization rotation amount detected at the peak position corresponding to the kth span is acquired. The variable k specifies the span on the optical transmission path. Further, it is assumed that the variable k is initialized to 1 at the start of this flowchart. That is, at the beginning of this flowchart, the span containing the transmitting node is selected.

In S42, the determination unit 51d compares each power difference acquired in S41 with a predetermined PDL threshold value. For example, in the case shown in FIG. 31A, the four power differences “x00-y00”, “x10-y10”, “x20-y20”, “x30-y30” and the PDL threshold value are compared with respect to the span S1. Will be done. Then, when a power difference larger than the PDL threshold value is not found in this comparison, the determination unit 51d increments the variable k in S43 to S44. Then, if the unprocessed span remains, the processing of the determination unit 51d returns to S41. That is, the determination unit 51d selects the next span. In this way, the determination unit 51d selects spans one by one from the transmitting node to the receiving node and executes the corresponding processing until a power difference larger than the PDL threshold value is found. Then, when a power difference larger than the PDL threshold value is not found in all the spans, the determination unit 51d determines in S45 that no polarization-dependent loss has occurred.

When a power difference larger than the PDL threshold value is found in the process for the kth span, the determination unit 51d determines in S46 that a polarization-dependent loss has occurred. Further, the determination unit 51d identifies the location where the polarization-dependent loss occurs. Specifically, information indicating that the polarization-dependent loss occurs in the section between the peak position corresponding to the k-1st span and the peak position corresponding to the kth span is output.

As described above, in the second embodiment, the presence or absence of the polarization-dependent loss is determined based on the power difference between the polarizations detected in the section where the optical power is large. Therefore, even in the case where noise is added to the optical signal, the presence or absence of polarization-dependent loss can be accurately determined.

In the embodiment described with reference to FIGS. 25 to 32, the polarization-dependent loss is detected by using the difference between the optical power of the X polarization signal and the optical power of the Y polarization signal. The second embodiment is not limited to this method. That is, the optical network device 2 according to the second embodiment may specify the location where the polarization-dependent loss occurs by using either the optical power of the X polarization signal or the optical power of the Y polarization signal. Alternatively, the optical network device 2 according to the second embodiment identifies the location where the polarization-dependent loss occurs based on the polarization component in the θ-axis direction or the polarization component in the orthogonal axis direction shown in FIG. 11 (b). You may.

<Raman amplification>
The above-mentioned optical amplifier is, for example, a rare earth-added fiber amplifier such as an EDFA (Erbium Doped Fiber Amplifier). However, the optical amplifier provided on the optical transmission path is not particularly limited, and may be, for example, a distributed Raman amplifier.

FIG. 33 is a diagram illustrating distribution Raman amplification. As shown in FIG. 33 (a), the Raman amplifier supplies excitation light to a transmission line optical fiber propagating an optical signal. Then, the optical signal is amplified by the energy of the excitation light. Here, the excitation light is often supplied to the optical transmission line so as to propagate in the direction opposite to the optical signal. In this case, the optical signal is amplified in a region close to the Raman amplifier, as shown in FIG. 33 (b). That is, a power profile is obtained in which the optical power gradually decreases as the distance from the transmitting node increases and the optical power increases within the range reached by the excitation light. In the following description, the region where the optical signal is amplified by the excitation light may be referred to as an “amplification section”. Further, other areas of the optical transmission line may be referred to as a "transmission section".

Even in an optical transmission line on which a Raman amplifier is mounted, when a polarization-dependent loss occurs, a power profile in which the power of the optical signal has a "step" can be obtained. For example, when the polarization-dependent loss occurs in the transmission section, the power profile shown in FIG. 33 (c) is obtained, and when the polarization-dependent loss occurs in the amplification section, the power profile shown in FIG. 33 (d) is obtained. In any case, since the power difference occurs, the location where the polarization-dependent loss occurs can be specified.

By the way, in distributed Raman amplification, PDG having different gains may occur depending on the state of polarization. For example, when PDG due to distributed Raman amplification occurs in the case where there is no polarization-dependent loss, the difference between the optical power of the X polarization signal and the optical power of the Y polarization signal gradually increases in the amplification section, as shown in FIG. 34. It will become. Therefore, if the amount of polarization rotation by the monitor signal rotation unit 24 and the reference signal rotation unit 25 is controlled, the power profile shown by the solid line and the power profile shown by the broken line in FIG. 34 can be obtained.

Here, the optical network device 1 according to the first embodiment detects a "step" in the power profile. Therefore, although the optical network device 1 can detect the polarization-dependent loss, it is difficult to detect the PDG caused by the distributed Raman amplification. On the other hand, the optical network device 2 according to the second embodiment detects the optical power of each polarization at the peak position where the peak of the optical power appears. Therefore, the optical network device 2 can detect the power difference when the power difference is caused by the PDG. That is, the optical network device 2 can detect a power difference when a power difference occurs due to a polarization-dependent loss or PDG.

On the other hand, the location where the Raman amplifier is mounted is known. In addition, the power profile obtained in the optical transmission line on which the Raman amplifier is mounted has a characteristic shape as shown in FIG. 33 (b). Therefore, if the power profile is acquired when the power difference is detected at the peak position, it can be determined whether the cause of the power difference is the polarization-dependent loss or the PDG.

FIG. 35 shows an example of a span identification unit used in an optical transmission system including a Raman amplifier. In this case, the span specifying unit 51R includes a profile selection unit 51e in addition to the configuration shown in FIG. 26.

The power profile storage unit 51a, the peak detection unit 51b, the power difference calculation unit 51c, and the determination unit 51d are substantially the same in the span identification unit 51 shown in FIG. 26 and the span identification unit 51R shown in FIG. 35. However, when the determination unit 51d mounted on the span specifying unit 51R detects a power difference larger than a predetermined threshold value, the determination unit 51d outputs information representing the amount of polarization rotation that caused the power difference. For example, in the case where a power difference larger than the threshold value is detected when the amount of polarization rotation by the monitor signal rotation unit 24 and the reference signal rotation unit 25 is "45 degrees", the determination unit 51d determines "polarization rotation amount =". "45 degrees" is output.

The profile selection unit 51e selects a corresponding power profile from the power profile storage unit 51a based on the information indicating the amount of polarization rotation output from the determination unit 51d. When "polarization rotation amount = 45 degrees" is output from the determination unit 51d, the profile selection unit 51e selects a power profile corresponding to "polarization rotation amount = 45 degrees" from the power profile storage unit 51a. Output. Then, the network administrator can determine whether the cause of the power difference is the polarization-dependent loss or the PDG based on the shape of the power profile selected by the profile selection unit 51e.

<Polarization rotation>
As described above, the optical network devices 1 and 2 according to the embodiment of the present invention perform an operation to rotate the polarization with respect to the monitor signal and the reference signal. However, the operation for controlling the polarization of the signal is not limited to the rotation operation. That is, the optical network devices 1 and 2 may perform an operation for controlling the phase difference between the two polarizations in addition to the rotation operation. Specifically, the operation of the equation (6) may be performed on the monitor signal instead of the equation (4). Note that Ex and Ey represent the polarization components of the monitor signal. θ represents the amount of polarization rotation. δ represents the phase difference between the two polarizations.

In this case, the operation of the equation (7) is performed on the reference signal instead of the equation (5). Esx and Esy represent the polarization components of the reference signal.

Then, the optical network devices 1 and 2 create a power profile while controlling the polarization rotation amount θ and the phase difference δ. At this time, the monitor signal rotation unit 24 controls the polarization rotation amount θ and the phase difference δ of the monitor signal. Therefore, the monitor signal rotating unit 24 generates a third polarization component and a fourth polarization component that are orthogonal to each other by controlling the first polarization component and the second polarization component that represent the monitor signal. This is an example of a polarization control unit. Further, the reference signal rotation unit 25 controls the polarization rotation amount θ and the phase difference δ of the reference signal. Therefore, the reference signal rotation unit 25 is an example of a second polarization control unit that controls the polarization of the reference signal.

1, 1B to 1D, 2 Optical network device 11 Coherent receiver 13 Digital signal processing unit 13a Fixed equalizer 13b Adaptive equalizer 16a First dispersion compensation unit 16b Non-linear compensation unit 16c Second dispersion compensation unit 16d Correlation calculation Unit 21 Monitor signal storage unit 22 Transmission data storage unit 23 Transmission waveform reconstruction unit 24 Monitor signal rotation unit 25 Reference signal rotation unit 26 Rotation amount control unit 27 Feature extraction unit 28 Position identification unit 28a Power profile storage unit 28b Difference calculation unit 28c Change point search unit 30 Calculation unit 41 Polarization rotation correction unit 42 Dispersion compensation unit 51, 51R Span identification unit 51a Power profile storage unit 51b Peak detection unit 51c Power difference calculation unit 51d Judgment unit 51e Profile selection unit 100 Transmission node

Claims (16)

An optical network device that receives polarized multiplex optical signals transmitted from a transmitting node.
A polarization separation unit that separates an electric field information signal representing the electric field of the polarization multiplex optical signal into a first polarization component and a second polarization component orthogonal to each other.
In a coordinate system representing a first polarization direction and a second polarization direction orthogonal to each other, a third deviation orthogonal to each other by controlling the first polarization component and the second polarization component. A polarization control unit that generates a wave component and a fourth polarization component,
Evaluation corresponding to at least one of the power of the third polarization component or the fourth polarization component for each of the plurality of positions on the optical transmission path between the transmission node and the optical network device. A feature extractor that calculates the value and
A fluctuation amount calculation unit that calculates the fluctuation amount of the evaluation value with respect to the control amount by the polarization control unit for each of the plurality of positions.
Based on the comparison result between the fluctuation amount of the evaluation value at the first position among the plurality of positions and the fluctuation amount of the evaluation value at the second position adjacent to the first position among the plurality of positions. Then, a determination unit for determining whether the first position is the extraction target, and
An optical network device equipped with. The polarization control unit performs a rotation calculation on the first polarization component and the second polarization component in the coordinate system to perform the rotation calculation on the third polarization component and the fourth polarization component. Generates wave components,
The optical network device according to claim 1, wherein the fluctuation amount calculation unit calculates a fluctuation amount of the evaluation value with respect to the rotation amount of the rotation calculation by the polarization control unit. The second position is adjacent to the transmission node side with respect to the first position.
The determination unit is characterized in that when the fluctuation amount of the evaluation value at the first position is larger than the fluctuation amount of the evaluation value at the second position, the determination unit determines that the first position is the extraction target. The optical network device according to claim 1. The fluctuation amount of the evaluation value is the maximum value and the minimum value of the evaluation value obtained when the polarization control unit rotates the first polarization component and the second polarization component in the coordinate system. The optical network device according to claim 1, wherein the difference from the value is used. The determination unit outputs information indicating the first position when the difference between the fluctuation amount of the evaluation value at the first position and the fluctuation amount of the evaluation value at the second position is larger than a predetermined threshold value. The optical network apparatus according to claim 1. The feature extraction unit
A first dispersion compensating unit that compensates for the first wavelength dispersion of the wavelength dispersion of the optical transmission line with respect to the electric field information signal.
A non-linear compensation unit that compensates for the non-linear distortion of the optical transmission line with respect to the output signal of the first dispersion compensation unit.
A second dispersion compensating unit that compensates for the remaining wavelength dispersion of the wavelength dispersion of the optical transmission line with respect to the output signal of the non-linear compensation unit.
A calculation unit for calculating an evaluation value corresponding to a combination of the first wavelength dispersion and the remaining wavelength dispersion based on the output signal of the second dispersion compensation unit is provided.
The first aspect of claim 1 is characterized in that the feature extraction unit calculates evaluation values at a plurality of positions on the optical transmission path by calculating the evaluation values while changing the amount of the first wavelength dispersion. The optical network device described. The optical network according to claim 6, wherein the evaluation value represents a correlation between a reference signal representing an electric field of the polarization-multiplexed optical signal at the transmitting node and an output signal of the second dispersion compensator. Device. Further, a second polarization control unit for controlling the polarization of the reference signal in the coordinate system is provided.
The control amount by the polarization control unit and the control amount by the second polarization control unit are the same as each other.
The seventh aspect of claim 7, wherein the evaluation value represents a correlation between a reference signal whose polarization is controlled by the second polarization control unit and an output signal of the second dispersion compensation unit. Optical network device. The polarization control unit includes a plurality of rotation processing units that rotate the first polarization component and the second polarization component in the coordinate system at different rotation amounts.
The optical network device according to claim 1, wherein the feature extraction unit includes a plurality of evaluation value calculation units that calculate corresponding evaluation values from output signals of the plurality of rotation processing units. A coherent receiver that generates an electric field information signal that represents the electric field of the polarization multiplex optical signal, and
A fixed equalizer that corrects the electric field information signal generated by the coherent receiver and compensates for the wavelength dispersion of the optical transmission line.
When the polarization separation unit generates the first polarization component and the second polarization component from the output signal of the fixed equalizer, the first deviation generated by the polarization separation unit. The optical network device according to claim 1, further comprising a reproduction unit that reproduces transmission data based on a wave component and the second polarization component. A coherent receiver that generates an electric field information signal that represents the electric field of the polarization multiplex optical signal, and
A fixed equalizer that corrects the electric field information signal generated by the coherent receiver and compensates for the wavelength dispersion of the optical transmission line.
An adaptive equalizer that corrects the output signal of the fixed equalizer to compensate for residual dispersion,
Further, a reproduction unit for reproducing transmission data based on the output signal of the adaptive equalizer is provided.
The optical network device according to claim 1, wherein the polarization separating unit generates the first polarization component and the second polarization component from the output signal of the fixed equalizer. A coherent receiver that generates an electric field information signal that represents the electric field of the polarization multiplex optical signal, and
A fixed equalizer that corrects the electric field information signal generated by the coherent receiver and compensates for the wavelength dispersion of the optical transmission line.
An adaptive equalizer that corrects the output signal of the fixed equalizer to compensate for residual dispersion,
A playback unit that reproduces transmission data based on the output signal of the adaptive equalizer, and
Further provided with a dispersion compensating unit that corrects the electric field information signal generated by the coherent receiver and compensates for the wavelength dispersion of the optical transmission line.
The optical network device according to claim 1, wherein the polarization separation unit generates the first polarization component and the second polarization component from the output signal of the dispersion compensation unit. The peak of the power of at least one of the third polarization component and the fourth polarization component based on the evaluation value calculated by the feature extraction unit for each of the plurality of positions on the optical transmission path. Further equipped with a peak detection unit that detects the position where
The determination unit is adjacent to the fluctuation amount of the evaluation value at the first peak position among the plurality of peak positions detected by the peak detection unit and the first peak position among the plurality of peak positions. It is characterized in that it is determined whether the section between the first peak position and the second peak position is the extraction target section based on the comparison result with the fluctuation amount of the evaluation value at the second peak position. The optical network device according to claim 1. When the fluctuation amount of the evaluation value at the first peak position is larger than the predetermined threshold value, the difference between the power of the third polarization component and the power of the fourth polarization component is made larger than the predetermined value. The optical network device according to claim 13, wherein a first control amount by the polarization control unit is specified, and an evaluation value corresponding to the first control amount at each of the plurality of positions is output. A transmission line monitoring method for monitoring an optical transmission path between the transmission node and the optical network device in an optical network device that receives a polarization-multiplexed optical signal transmitted from a transmission node.
The electric field information signal representing the electric field of the polarization multiplex optical signal is separated into a first polarization component and a second polarization component orthogonal to each other, and the first polarization direction and the second polarization direction are orthogonal to each other. By rotating the first polarization component and the second polarization component in the coordinate system representing the above, a third polarization component and a fourth polarization component orthogonal to each other are generated.
For each of the plurality of positions on the optical transmission path, an evaluation value corresponding to at least one of the power of the third polarization component or the fourth polarization component is calculated.
For each of the plurality of positions, the fluctuation amount of the evaluation value with respect to the rotation amount is calculated.
Based on the comparison result between the fluctuation amount of the evaluation value at the first position among the plurality of positions and the fluctuation amount of the evaluation value at the second position adjacent to the first position among the plurality of positions. A transmission line monitoring method comprising determining whether or not the first position is an output target. A transmission line monitoring method for monitoring an optical transmission path between the transmission node and the optical network device in an optical network device that receives a polarization-multiplexed optical signal transmitted from a transmission node.
The electric field information signal representing the electric field of the polarization multiplex optical signal is separated into a first polarization component and a second polarization component orthogonal to each other, and the first polarization direction and the second polarization direction orthogonal to each other are separated. In the coordinate system representing the above, a third polarization component and a fourth polarization component orthogonal to each other are generated by performing a rotation operation on the first polarization component and the second polarization component. ,
The power of the third polarization component and the power of the fourth polarization component when the rotation amount of the rotation calculation is changed within a predetermined range for each of the plurality of positions on the optical transmission path. Judging whether the difference between the above exceeds a predetermined threshold,
When the difference exceeds the threshold value at the first position among the plurality of positions, the difference at the first position and the second position adjacent to the first position among the plurality of positions. A transmission line monitoring method comprising determining whether or not the first position is an output target based on a comparison with the difference at a position.

Images