JP2019140666A - State estimation device and communication system - Google Patents

Abstract

To perform more detailed precursor detection of failures that occur in data transmission while reducing costs.SOLUTION: A state estimation device includes a preprocessing unit and an estimation unit. The preprocessing unit obtains data representing one or more of the phase, reception strength, reception quality, voltage after conversion to an electric signal, and signal processing parameters used in reception processing of a signal that is transmitted by a transmission unit of a transmission device and received by a reception unit of another transmission device via a transmission path, and processes the obtained data to feature data used for state estimation. The estimation unit estimates the state of the transmission path, the abnormal state of the transmission unit, or the abnormal state of the reception unit on the basis of the feature data.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a state estimation device and a communication system.

  FIG. 24 is a diagram illustrating an example of a conventional optical digital coherent optical communication system. The optical communication system shown in the figure includes a plurality of transponders 90. The transponder 90 on the transmitting side is referred to as a transponder 90A, and the transponder 90 on the receiving side is referred to as a transponder 90B. The transponder 90A and the transponder 90B have the same configuration. The transponder 90A and the transponder 90B communicate with each other by an optical fiber (single core fiber, multicore fiber), copper, radio, or the like. In the figure, a case where communication is performed using an optical fiber will be described as an example.

  A signal input from an external device to the transponder 90A is terminated in an OTN (Optical Transport Network) framer and mapped to an OTN frame. The OTN frame is subjected to error correction code assignment, transmission linear equivalent processing, and digital-to-analog conversion processing in a subsequent DSP (Digital Signal Processor), and then converted from an electrical signal to an optical signal by an optical transmission unit. To the opposite transponder 90B.

  The optical signal transmitted from the optical transmission unit of the transponder 90B is received by the optical reception unit of the transponder 90A via the optical fiber and converted into an electrical signal. In the DSP, analog-digital conversion processing, reception linear equivalent processing, and transmission path estimation are performed on the electrical signal, and the OTN frame signal is demodulated. In the error correction coding processing unit in the subsequent stage, the OTN frame signal obtained by correcting the signal that could not be demodulated partially correctly is converted into a signal in a format suitable for the external device by the OTN framer and output. The

  When a failure occurs in an optical fiber or the like and deterioration exceeding the equivalent processing capability of the reception linear equivalent processing unit is added to the signal, communication is interrupted. When communication is interrupted, the system recognizes that a failure has occurred, and the failure isolation process is implemented by introducing the system and personnel. When an optical fiber breaks, an OTDR (Optical Time Domain Reflectometer) is used to identify the location.

  When an optical fiber failure occurs, equipment costs and operating costs increase, for example, the equipment is equipped with a test function separately from that for communication, or a transmission device is removed and a test machine is installed. In the invention of Patent Document 1, it is possible to detect a precursor of a failure occurring on a transmission line by collecting and statistically processing a signal or physical state data associated with signal processing.

JP 2017-85355 A

  However, although the method of Patent Document 1 can detect a failure or a precursor of an optical fiber transmission line using physical state data, the failure cause (bending, pulling, compressing, breaking, vibration, etc.) and its extent (bending) (Diameter, tensile strength, vibration width, vibration period, etc.) are not known. For this reason, there is a problem that in order to quickly perform a recovery operation when a failure occurs, the cause and the extent of the failure are detected, and detailed failure sign detection is required.

  In view of the above circumstances, an object of the present invention is to provide a state estimation device and a communication system that can perform more detailed precursor detection of a failure that occurs in data transmission while reducing costs.

  One aspect of the present invention is a phase of a signal transmitted from a transmission unit of a transmission device and received by a reception unit of another transmission device via a transmission path, reception intensity, reception quality, voltage after conversion into an electrical signal, And based on the said feature data, the pre-processing part which acquires the signal reception data showing 1 or more among the signal processing parameters used in the reception process, and processes the acquired said signal reception data into the feature data used for state estimation And an estimation unit that estimates the state of the transmission path, the abnormal state of the transmission unit, or the abnormal state of the reception unit.

  One aspect of the present invention is the above-described state estimation device, wherein the state of the transmission path is one or more of presence / absence of bending, a bending diameter, and a temperature change.

  One aspect of the present invention is the above-described state estimation device, wherein the feature data includes: phase plane state data representing the phase on a phase plane; polar coordinate data representing the phase on a polar coordinate plane; and the polar coordinate data at high speed. Fourier transform data obtained by Fourier transform, histogram data indicating the frequency of appearance of the signal in the phase plane state data or the polar coordinate data, and any one or more of the reception intensity, the reception quality, the voltage, or the signal processing parameter At least one of histogram data representing the frequency of appearance of the signal in the combination and time-series data representing time-series changes in the reception intensity, the reception quality, the voltage, or the signal processing parameter.

  One aspect of the present invention is the above-described state estimation device, in which the estimation unit performs estimation using machine learning.

  One aspect of the present invention is the state estimation device described above, wherein the estimation unit uses at least a part of the feature data to determine the state of the transmission path, the abnormal state of the transmission unit, or the reception unit. A plurality of partial estimators for estimating a partial state included in any of the abnormal states, and based on estimation results by the partial estimators for estimating different partial states, The abnormal state of the transmitting unit or the abnormal state of the receiving unit is estimated.

  One aspect of the present invention is the above-described state estimation device, wherein the preprocessing unit or the estimation unit is implemented by hardware, a processor that executes a program, or a combination of hardware and a processor that executes a program It is realized by.

  One aspect of the present invention is the above-described state estimation device, wherein the transmission path includes a plurality of physical paths, and the preprocessing unit acquires and acquires the signal reception data for each physical path. Feature data used for state estimation is generated from the signal reception data, the estimation unit estimates the state of the physical path for each physical path based on the feature data, and the state estimation device includes the estimation unit An integrated estimation unit for deriving a state estimation result of the transmission path based on the state of each physical path estimated by the above.

  One aspect of the present invention is the above-described state estimation device, wherein the preprocessing unit transmits the optical signal using the same light source in two or more or all of the plurality of physical paths. The signal reception data is acquired every time.

  One aspect of the present invention is the above-described state estimation device, in which the preprocessing unit receives the signal reception data for each physical path when transmitting an optical signal using a different light source to each of the plurality of physical paths. To get.

  One aspect of the present invention is the above-described state estimation device, wherein the transmission path is a multi-core fiber, a multi-mode fiber, or a multi-fiber, or a combination of two or more.

  One aspect of the present invention is the above-described state estimation device, wherein the transmission path includes a mode multiplexing unit that multiplexes a plurality of single-mode optical signals and converts them into a multi-mode optical signal, and the mode multiplexing unit includes A multimode fiber that transmits the converted multimode optical signal; and a mode separation unit that separates the multimode optical signal transmitted through the multimode fiber into a single mode optical signal.

  One aspect of the present invention is the above-described state estimation device, in which the preprocessing unit acquires the signal reception data when transmitting each bidirectional optical signal for each physical path, and the acquired signal The feature data used for state estimation is generated from the received data, and the estimation unit estimates the state of the physical path based on the feature data for each physical path for each of the bidirectional directions. A state estimation result of the transmission path is derived based on the state of each physical path estimated in both directions by the estimation unit.

  One embodiment of the present invention is a communication system including a transmission device and any one of the above-described state estimation devices.

  According to the present invention, it is possible to perform more detailed precursor detection of a failure that occurs in data transmission while reducing costs.

It is a block diagram which shows the structure of the communication system by 1st Embodiment. It is a block diagram which shows the structure of the communication system by 2nd Embodiment. It is a block diagram which shows the structure of the communication system by 3rd Embodiment. It is a block diagram which shows the structure of the communication system by 4th Embodiment. It is a block diagram which shows the structure of the communication system by 5th Embodiment. It is a block diagram which shows the structure of the communication system by 6th Embodiment. It is a block diagram which shows the structure of the communication system by 7th Embodiment. It is a figure which shows the function of the estimation part by 8th Embodiment. It is a figure which shows the function of the estimation part by 9th Embodiment. It is a figure which shows the function of the estimation part by 10th Embodiment. It is a flowchart which shows operation | movement of the estimation part by 11th Embodiment. It is a flowchart which shows other operation | movement of the estimation part by the embodiment. It is a flowchart which shows operation | movement of the estimation part by 12th Embodiment. It is a flowchart which shows other operation | movement of the estimation part by the embodiment. It is a block diagram which shows the structure of the communication system by 13th Embodiment. It is a block diagram which shows the structure of the communication system by 14th Embodiment. It is a block diagram which shows the structure of the communication system by 15th Embodiment. It is a block diagram which shows the structure of the communication system by 16th Embodiment. It is a block diagram which shows the structure of the communication system by 17th Embodiment. It is a block diagram which shows the structure of the communication system by 18th Embodiment. It is a block diagram which shows the structure of the communication system by 19th Embodiment. It is a block diagram which shows the structure of the communication system by 20th Embodiment. It is a block diagram which shows the structure of the communication system by 21st Embodiment. It is a figure which shows the optical communication system of an optical digital coherent system of a prior art.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Hereinafter, when there are Y functional units XXX, the Y functional units XXX are referred to as functional units XXX-1 to XXX-Y, respectively.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of a communication system 101 according to the first embodiment of the present invention. The communication system 101 includes a transmission device 200 and a state estimation device 510. The number of transmission apparatuses 200 included in the communication system 101 is arbitrary. The transmission apparatus 200 performs data communication via the transmission path 300. The transmission line 300 is, for example, an optical fiber.

  The transmission apparatus 200 includes a transmission unit 201 and a reception unit 202. The transmission unit 201 transmits an optical signal via the transmission path 300. The receiving unit 202 receives data via the transmission path 300. In the figure, two transmission apparatuses 200 relay user data communication using optical signals.

  The state estimation device 510 includes a preprocessing unit 511 and an estimation unit 512. The preprocessing unit 511 receives constellation data (phase plane state data) indicating the phase plane state (constellation) of each signal received by the reception unit 202 from the reception unit 202 of the transmission apparatus 200 as signal reception data. The pre-processing unit 511 plots the appearance frequency of the phase in the constellation data, converts it to two-dimensional histogram data, and outputs it to the estimation unit 512 as feature data. The estimation unit 512 estimates the abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 511 and outputs the estimation result.

  Note that the preprocessing unit 511 may generate the two-dimensional histogram data by plotting the appearance frequency of the phase after converting the constellation data plotted on the phase plane from the orthogonal coordinates to the polar coordinates. In addition, the preprocessing unit 511 performs fast Fourier transform after converting the constellation data from orthogonal coordinates to polar coordinates, and appears on a two-dimensional plane having both axes as a result of fast Fourier transform for each of the angle and the distance from the center. The frequency may be plotted as two-dimensional histogram data.

  As a result of the estimation by the estimation unit 512, the abnormal state of the transmission line 300 such as bending, pulling, misalignment (breaking), temperature change, vibration, water leakage, and twist of the optical fiber, and the transmission unit of the transmission apparatus 200 An abnormal state of the transmission module used in 201, an abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Second Embodiment)
FIG. 2 is a block diagram showing the configuration of the communication system 102 according to the second embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 102 includes a transmission device 200 and a state estimation device 520. The state estimation device 520 includes a preprocessing unit 521 and an estimation unit 522.

  The preprocessing unit 521 converts the constellation data acquired from the reception unit 202 of the transmission apparatus 200 from orthogonal coordinates to polar coordinates. The pre-processing unit 521 processes the angle and distance data from the center into fast Fourier transform data, and outputs the processed data to the estimation unit 522 as feature data. Thus, the preprocessing unit 521 extracts features by processing the constellation data plotted on the plane into Fourier transform data so that frequency analysis can be performed. The estimation unit 522 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 521, and outputs an estimation result.

  As a result of the estimation by the estimation unit 522, the abnormal state of the transmission line 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, torsion, and the degree thereof are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Third embodiment)
FIG. 3 is a block diagram showing the configuration of the communication system 103 according to the third embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 103 includes a transmission device 200 and a state estimation device 530. The state estimation device 530 includes a preprocessing unit 531 and an estimation unit 532.

  The preprocessing unit 531 processes the constellation data acquired from the reception unit 202 of the transmission apparatus 200 into coordinate conversion data, and outputs the data to the estimation unit 532 as feature data. Specifically, the preprocessing unit 531 converts the constellation data plotted on the plane from orthogonal coordinates to polar coordinates, and extracts features by plotting the appearance frequency for each angle and distance from the center, for example. To do. The estimation unit 532 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 531, and outputs an estimation result. Note that the preprocessing unit 531 may output the constellation data as it is to the estimation unit 532 as feature data.

  As a result of estimation by the estimation unit 532, the abnormal state of the transmission path 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, torsion, and the degree thereof are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Fourth embodiment)
FIG. 4 is a block diagram showing the configuration of the communication system 104 according to the fourth embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 104 includes a transmission device 200 and a state estimation device 540. The state estimation device 540 includes a preprocessing unit 541 and an estimation unit 542.

  The preprocessing unit 541 acquires constellation data of each reception signal, reception power, and Q value data as signal reception data from the reception unit 202 of the transmission apparatus 200. The preprocessing unit 541 processes the constellation data into two-dimensional histogram data, and processes the received power and the Q value data into time series data. The preprocessing unit 541 combines the received power and time-series data of the Q value data and the two-dimensional histogram data of the constellation data, and outputs the combined data to the estimation unit 542 as feature data. The estimation unit 542 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 541 and outputs an estimation result.

  As a result of the estimation by the estimation unit 542, the abnormal state of the transmission line 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, twisting, etc., and its degree are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Fifth embodiment)
FIG. 5 is a block diagram showing the configuration of the communication system 105 according to the fifth embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 105 includes a transmission device 200 and a state estimation device 550. The state estimation device 550 includes a preprocessing unit 551 and an estimation unit 552.

  The preprocessing unit 551 acquires the constellation data of the reception signal, the reception power, and the Q value data from the reception unit 202 of the transmission apparatus 200 as signal reception data. The preprocessing unit 551 processes the constellation data into fast Fourier transform data indicating the result of fast Fourier transform, and processes the received power and the Q value data into time series data. The preprocessing unit 551 combines the received power and time-series data of the Q value data and the fast Fourier transform data of the constellation data, and outputs the combined data to the estimation unit 552. The estimation unit 552 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the data received from the preprocessing unit 551 and outputs an estimation result.

  As a result of the estimation by the estimation unit 552, the abnormal state of the transmission path 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, torsion, etc. and its degree are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Sixth embodiment)
FIG. 6 is a block diagram showing the configuration of the communication system 106 according to the sixth embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 106 includes a transmission device 200 and a state estimation device 560. The state estimation device 560 includes a preprocessing unit 561 and an estimation unit 562.

  The preprocessing unit 561 acquires the reception power or Q value data of each reception signal from the reception unit 202 of the transmission apparatus 200 as signal reception data. The preprocessing unit 561 processes reception power or Q value data that changes in time series into one-dimensional histogram data. The preprocessing unit 561 outputs the received power or Q value data converted into the one-dimensional histogram data to the estimation unit 562 as feature data. The estimation unit 562 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 561, and outputs an estimation result.

  As a result of estimation by the estimation unit 562, the abnormal state of the transmission line 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, twist, and the like, and the degree thereof are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Seventh embodiment)
FIG. 7 is a block diagram showing the configuration of the communication system 107 according to the seventh embodiment of the present invention. In this figure, the same parts as those of the communication system 101 according to the first embodiment shown in FIG. The communication system 107 includes a transmission device 200 and a state estimation device 570. The state estimation device 570 includes a preprocessing unit 571 and an estimation unit 572.

  The preprocessing unit 571 acquires reception power or Q value data as signal reception data from the reception unit 202 of the transmission apparatus 200. The preprocessing unit 571 plots the received power and the Q value data on a two-dimensional plane and processes the two-dimensional histogram data representing the frequency. The preprocessing unit 571 outputs the two-dimensional histogram data to the estimation unit 572 as feature data. The estimation unit 572 estimates an abnormal state of the transmission device 200 and the transmission path 300 based on the feature data received from the preprocessing unit 571 and outputs an estimation result.

  As a result of the estimation by the estimation unit 572, the abnormal state of the transmission line 300 such as bending, pulling, misalignment of the optical fiber, temperature change, vibration, water leakage, twisting, etc., and its degree are used for the transmission unit 201 of the transmission apparatus 200. The abnormal state of the transmission module to be transmitted, the abnormal state of the reception module used in the reception unit 202 of the transmission apparatus 200, and the like can be estimated.

(Eighth embodiment)
FIG. 8 is a diagram illustrating the state estimation function of the state estimation device 580 according to the eighth embodiment of the present invention. The state estimation device 580 includes a preprocessing unit 581 and an estimation unit 582. The preprocessing unit 581 is the preprocessing unit 511, 521, 531, 541, 551, 561, 571 of the above-described embodiment, and the estimation unit 582 is the estimation unit 512, 522, 532, 542, 552 of the above-described embodiment. 562 and 572 can be used.

  The estimation unit 582 uses machine learning such as a neural network. For example, AI (artificial intelligence) such as a neural network is learned in advance for feature data corresponding to the state to be estimated, and the learned AI (neural network) is used for the estimation unit 582. The learned AI calculates the probability (likelihood) of being in a predetermined state using the feature data input from the preprocessing unit 581. The predetermined state is used for final state determination in the estimation unit 582, and is any of the state of the transmission line 300, the abnormal state of the transmission module used in the transmission unit 201, and the abnormal state of the reception module used in the reception unit 202. This is a part of the state. For example, the learned AI determines a state such as a probability that the bending of the optical fiber is large, a probability that the bending of the optical fiber is medium, and a probability that the bending of the optical fiber is small. Also, for example, the learned AI is a probability that the optical fiber is bent more than a predetermined value, a probability that the optical fiber is twisted more than a predetermined value, a probability that the transmission module is abnormal, a probability that the reception module is abnormal, etc. Judge each state.

  As described above, the estimation unit 582 uses the featured data representing the physical state of the received signal based on the learned AI, and causes of the failure (bending, pulling, compressing, breaking, vibration, etc.) of the transmission path 300 and the degree thereof. (Bending diameter, tensile strength, vibration width / vibration period, etc.) and the probability of occurrence of abnormality in each of the transmission module and the reception module can be estimated. The estimation unit 582 determines the states of the transmission path 300 and the transmission apparatus 200 by combining the calculation results of these states, and outputs the determination results. The determination function of the estimation unit 582 can be realized by a method using AND / OR logic and maximum value extraction calculation.

(Ninth embodiment)
FIG. 9 is a diagram illustrating the state estimation function of the state estimation device 590 according to the ninth embodiment of the present invention. The state estimation device 590 includes a preprocessing unit 591 and an estimation unit 592. The preprocessing unit 591 is the preprocessing unit 511, 521, 531, 541, 551, 561, 571 of the above-described embodiment, and the estimation unit 592 is the estimation unit 512, 522, 532, 542, 552 of the above-described embodiment. 562 and 572 can be used.

  The estimation unit 592 has recognition functions # 1 to #n that recognize two states. For example, the recognition function # 1 calculates the probability that there is a large bend in the optical fiber, and the recognition function # 2 calculates the probability that there is a moderate bend in the optical fiber. Also, for example, the recognition function # 1 calculates the probability that the optical fiber has a bend greater than or equal to a predetermined value, and the recognition function # 2 calculates the probability that the optical fiber has a twist or more than a predetermined value. The function # 3 calculates the probability that there is an abnormality in the transmission module and the probability that there is no abnormality in the transmission module, and the recognition function # 4 calculates the probability that there is an abnormality in the reception module. The estimation unit 592 determines the states of the transmission path 300 and the transmission device 200 based on a plurality of two states calculated by the recognition functions # 1 to #n, and outputs the determination results. The recognition functions # 1 to #n can be realized by a method using machine learning such as a statistical estimation method, a classic rule base AI, or a neural network. The determination function can be realized by a method using AND / OR logic and a maximum value extraction operation.

(Tenth embodiment)
FIG. 10 is a diagram illustrating the state estimation function of the state estimation device 595 according to the tenth embodiment of the present invention. The state estimation device 595 includes a preprocessing unit 596 and an estimation unit 597. The preprocessing unit 596 is the preprocessing unit 511, 521, 531, 541, 551, 561, 571 of the above-described embodiment, and the estimation unit 597 is the estimation unit 512, 522, 532, 542, 552 of the above-described embodiment. 562 and 572 can be used.

  The estimation unit 597 has recognition functions # 1 to #n that recognize two or more states. For example, the recognition function # 1 calculates the probability that the optical fiber has a large bend, the probability that there is a small bend, and the probability that there is no bend, and the recognition function # 2 indicates the probability that the optical fiber has a large twist and a small twist. A certain probability and a probability that there is no twist are calculated. Further, for example, the recognition function # 3 calculates the probability that there is an abnormality in the transmission module and the probability that there is no abnormality in the transmission module, and the recognition function # 4 calculates the probability that there is an abnormality in the reception module. The estimation unit 597 determines the states of the transmission line 300 and the transmission device 200 based on a plurality of two or more states calculated by the recognition functions # 1 to #n, and outputs the determination results. The recognition functions # 1 to #n can be realized by a method using machine learning such as a statistical estimation method, a classic rule base AI, or a neural network. The determination function can be realized by a method using AND / OR logic and a maximum value extraction operation.

(Eleventh embodiment)
FIG.11 and FIG.12 is a flowchart which shows the process of the estimation part with which the state estimation apparatus which concerns on the 11th Embodiment of this invention is provided. As the state estimation device of this embodiment, the state estimation device 580 of the eighth embodiment can be used. Hereinafter, a case where the learned AI used in the estimation unit 582 of the state estimation device 580 determines n types of states # 1 to #n will be described as an example.

  FIG. 11 is a flowchart when the learned AI determines each state sequentially. First, the estimation unit 582 sets i to an initial value 1 (step S105). The estimation unit 582 acquires all or part of the feature data received from the preprocessing unit 581 as the feature data used for the determination of the state #i (Step S110). The estimation unit 582 recognizes the state #i from the learned AI using the acquired feature data as an input (step S115). The estimation unit 582 holds the recognition result of the state #i (Step S120). If the estimation unit 582 determines that the value of i has not reached n and that all state recognition has not yet been performed (step S125: NO), the estimation unit 582 adds 1 to the value of i, and the processing from step S110 is performed. Is repeated (step S130). When it is determined that the value of i has reached n and all state recognition has been executed (step S125: YES), the estimation unit 582 executes the determination function using the recognition results of state # 1 to state #n, and transmits The state of the path 300 and the transmission apparatus 200 is determined (step S135).

  FIG. 12 is a flowchart when the learned AI determines each state in parallel. First, the estimation part 582 performs step S205-1 and S210-1, step S205-2 and S210-2, ..., step S205-n and S210-n in parallel. For each of i = 1 to n, the estimation unit 582 acquires all or part of the feature data received from the preprocessing unit 581 as the feature data used for the determination of the state #i (Step S205-i). Using the learned feature data as input, the state #i is recognized from the learned AI (step S210-i). The estimation unit 582 holds the recognition results of the states # 1 to #n (Step S215). The estimation unit 582 executes the determination function using the recognition results of the states # 1 to #n to determine the states of the transmission line 300 and the transmission device 200 (step S220).

  For example, the estimation unit 582 has a probability that the optical fiber bend is Amm in the state # 1 recognition, a probability that the optical fiber bend is Bmm in the state # 2 recognition, and a probability that the optical fiber bend is Xmm in the state #n recognition. Get. Based on these state recognition results, the estimation unit 582 determines that the optical fiber bend is Ymm by the determination function.

  When the processing of FIG. 11 is performed, the state determination can be performed without increasing the circuit scale of the estimation unit 582. On the other hand, when the process of FIG. 12 is performed, the time required for the estimation unit 582 to perform state determination can be shortened.

(Twelfth embodiment)
FIG.13 and FIG.14 is a flowchart which shows the process of the estimation part with which the state estimation apparatus which concerns on the 12th Embodiment of this invention is provided. As the state estimation device of this embodiment, the state estimation devices 590 and 595 of the ninth and tenth embodiments can be used. Below, the case where the estimation part 597 of the state estimation apparatus 595 has recognition function # 1-recognition function #n is demonstrated to an example.

  FIG. 13 is a flowchart when the respective recognition functions operate sequentially. First, the estimation unit 597 sets i to an initial value 1 (step S305). The estimation unit 597 acquires all or part of the feature data received from the preprocessing unit 596 as feature data used for determination of the recognition function #i (step S310). The recognition function #i of the estimation unit 597 performs state recognition using the acquired feature data as an input (step S315). The estimation unit 597 holds the recognition result of the recognition function #i (step S320). When the estimation unit 597 determines that the value of i has not reached n and the full recognition function has not yet been executed (step S325: NO), the estimation unit 597 adds 1 to the value of i, and performs the processing from step S310. Is repeated (step S330). When the estimation unit 597 determines that the value of i has reached n and that all recognition functions have been executed (step S325: YES), the estimation unit 597 executes the determination function using the recognition results of the recognition functions # 1 to #n. Then, the states of the transmission line 300 and the transmission apparatus 200 are determined (step S335).

  FIG. 14 is a flowchart for determining each recognition function in parallel. First, the estimation part 597 performs step S405-1 and S410-1, step S405-2 and S410-2, ..., step S405-n and S410-n in parallel. For each of i = 1 to n, the recognition function #i of the estimation unit 597 acquires feature data used for determination for all or part of the feature data received from the preprocessing unit 596 (step S405-i). State recognition is performed using the feature data as input (step S410-i). The estimation unit 597 holds the recognition results of the recognition functions # 1 to #n (step S415). The estimation unit 582 executes the determination function using the recognition results of the recognition functions # 1 to #n, and determines the states of the transmission path 300 and the transmission device 200 (step S420).

  For example, the estimation unit 597 determines the probability that the optical fiber is not bent or Amm in the recognition function # 1, the probability that the optical fiber is not bent or Bmm in the recognition function # 2, and the optical fiber bend in the recognition function #n. Get a probability of none or Xmm. Based on these state recognition results, the estimation unit 597 determines that the optical fiber bend is Ymm by the determination function.

  When the processing of FIG. 13 is performed, the state determination can be performed without increasing the circuit scale of the estimation units 592 and 597. On the other hand, when the processing of FIG. 14 is performed, the time required for the estimation units 592 and 597 to perform state determination can be shortened.

  In addition, instead of or in addition to the power and Q value in the above-described embodiment, a BER (Bit Error Rate) value, an OSNR (optical signal-to-noise ratio) value, an ESNR (electrical signal-to-noise ratio). One or more of the value, the voltage amplitude value after the optical signal is converted into an electrical signal in the optical receiver (Rx) included in the receiving unit 202, the tap coefficient of the Rx equalizer, and the like may be used. The tap coefficient relates to the ability to correct transmission line distortion in the Rx equalizer, and a value determined based on the characteristics of the received signal is used.

(13th Embodiment)
In the communication system of the present embodiment, the transmission path has a plurality of physical paths. Here, a case where the transmission path is a multi-core fiber having a plurality of cores (physical paths) will be described as an example. The transmission path may be a multi-fiber in which a plurality of optical fibers (physical paths) are bundled.

  FIG. 15 is a block diagram showing a configuration of a communication system 111 according to the thirteenth embodiment. The communication system 111 includes a transmission device 210 and a state estimation device 610. The number of transmission apparatuses 210 included in the communication system 111 is arbitrary. The transmission apparatus 210 includes one or more transmission units 211 and one or more reception units 212. The transmission device 210 performs data communication via the multicore fiber 310. The multi-core fiber 310 has cores 311-1 to 311-N (N is an integer of 2 or more). In the figure, a case where N = 3 is shown as an example.

  A pair of transmission unit 211 and reception unit 212 are connected by the core 311 of the multicore fiber 310. In the figure, the transmission unit 211 and the reception unit 212 connected to the core 311-n (n is an integer of 1 or more and N or less) are referred to as a transmission unit 211-n and a reception unit 212-n, respectively. The transmission units 211-1 to 211 -N may be the transmission unit 211 provided in one transmission device 210 as shown in the figure, or may be the transmission unit 211 provided in some or all different transmission devices 210. Good. Similarly, the receiving units 212-1 to 212 -N may be the receiving unit 212 provided in one transmission device 210 as shown in the figure, or the receiving units provided in some or all different transmission devices 210. 212 may be sufficient.

  The transmission unit 211-n (n is an integer of 1 to N) of the transmission apparatus 210 on the transmission side outputs an optical signal to the core 311-n of the multicore fiber 310. Since each of the transmission units 211-1 to 211 -N has a light source, optical signals generated using different light sources are output to the cores 311-1 to 311 -N, respectively. The reception unit 212-n of the transmission device 210 on the reception side receives the optical signal transmitted from the other transmission device 210 via the core 311-n of the multicore fiber 310.

  The state estimation device 610 includes preprocessing and estimation units 611-1 to 611-K (K is an integer equal to or greater than 2) and an integrated estimation unit 612. In the figure, a case where K = N = 3 is shown as an example. The preprocessing and estimation unit 611 is a set of the preprocessing unit 511 and the estimation unit 512 of the first embodiment, a set of the preprocessing unit 521 and the estimation unit 522 of the second embodiment, and the preprocessing of the third embodiment. Unit 531 and the estimation unit 532, the pre-processing unit 541 and the estimation unit 542 of the fourth embodiment, the pre-processing unit 551 and the estimation unit 552 of the fifth embodiment, before the sixth embodiment It is a set of a processing unit 561 and an estimation unit 562, or a set of a preprocessing unit 571 and an estimation unit 572 of the seventh embodiment. The pre-processing and estimation unit 611-n (n is an integer greater than or equal to 1 and less than or equal to N), such as constellation data from the reception unit 212-n of the transmission apparatus 210, receives the reception unit 202 in the first to seventh embodiments. The same data as the signal reception data acquired from is acquired, and feature data is generated. The preprocessing and estimation unit 611-n estimates the abnormal state of the core 311-n based on the generated feature data, and outputs the estimation result to the integrated estimation unit 612. The integrated estimation unit 612 uses the estimation results output from the preprocessing and estimation units 611-1 to 611-N, and the state of the multicore fiber 310 based on the transmission / reception states of all the cores 311-1 to 311-N. The estimation result is comprehensively determined, and the state estimation result of the multi-core fiber 310 is derived. For example, the integrated estimation unit 612 uses machine learning such as a statistical estimation method, a classic rule base AI, or a neural network to derive the state estimation result.

  In the present embodiment, the state estimation sensitivity of the transmission path can be improved by combining the state estimation of each physical path constituting the transmission path.

(Fourteenth embodiment)
In the thirteenth embodiment, optical signals generated using different light sources are transmitted to a plurality of physical paths included in a transmission path, and state estimation is performed. In the present embodiment, an optical signal generated using a single light source is transmitted to two or more or all of a plurality of physical paths included in the transmission path, and state estimation is performed. Below, it demonstrates centering on the difference with 13th Embodiment.

  FIG. 16 is a block diagram showing the configuration of the communication system 112 according to the fourteenth embodiment. In this figure, the same parts as those in the communication system 111 according to the thirteenth embodiment shown in FIG. The communication system 112 shown in the figure is different from the communication system 111 shown in FIG. 15 in that a transmission device 220 is provided instead of the transmission device 210. The transmission apparatus 220 includes one or more transmission units 221 and one or more reception units 212. The cores 311-1 to 311-N (N = 3 in the figure) of the multicore fiber 310 are connected to one transmission unit 211, and the cores 311-n (n is an integer of 1 to N) are reception units 212. -N is connected. The receiving units 212-1 to 212 -N may be the receiving unit 212 provided in one transmission device 220 as shown in the figure, or may be a receiving unit 212 provided in part or all of different transmission devices 220. Good.

  The transmission unit 221 of the transmission device 220 on the transmission side outputs the same optical signal using a single light source to the cores 311-1 to 311-N of the multicore fiber 310. The reception unit 212-n of the transmission device 220 on the reception side receives the optical signal transmitted by the transmission unit 221 of the transmission device 220 on the transmission side via the core 311-n of the multicore fiber 310. By using the same optical signal, the correlation of the optical signal after transmission through each of the cores 311-1 to 311-N becomes dense, so that the state estimation sensitivity of the transmission path can be improved.

(Fifteenth embodiment)
In the present embodiment, wavelength multiplexed signals are transmitted to a plurality of physical paths included in the transmission path, and state estimation is performed.
FIG. 17 is a block diagram showing the configuration of the communication system 113 according to the present embodiment. In this figure, the same parts as those in the communication system 111 according to the thirteenth embodiment shown in FIG. The communication system 113 shown in the figure is different from the communication system 111 shown in FIG. 15 in that a transmission device 230 is provided instead of the transmission device 210.

  The transmission device 230 includes one or more transmission units 231 and one or more reception units 232. The transmission unit 231 has a multi-wavelength light source, and transmits a wavelength-multiplexed optical signal to the core 311 of the multi-core fiber 310. The receiving unit 232 receives the wavelength-multiplexed optical signal from the core 311 of the multicore fiber 310.

  A pair of transmitter 231 and receiver 232 are connected by the core 311 of the multicore fiber 310. In the figure, the transmission unit 231 and the reception unit 232 connected to the core 311-n (n is an integer of 1 to N, N = 3 in the figure) are respectively connected to the transmission unit 231-n and the reception unit 232-n. It is described. The transmission units 231-1 to 231 -N may be the transmission units 231 provided in one transmission device 230 as shown in the figure, or may be transmission units 231 provided in some or all different transmission devices 230. Good. Similarly, the receiving units 232-1 to 232-N may be the receiving unit 232 provided in one transmission device 230 as shown in the figure, or the receiving units provided in some or all different transmission devices 230. 232 may be used.

  The pre-processing and estimation unit 611-n (n is an integer greater than or equal to 1 and less than or equal to N) of the state estimation device 610 includes, for example, constellation data for each wavelength from the reception unit 232-n of the transmission device 230. Data similar to the signal reception data acquired from the reception unit 202 in the seventh embodiment is acquired to generate feature data. The preprocessing and estimation unit 611-n estimates an abnormal state of the core 311-n based on feature data for each wavelength and outputs an estimation result. The integrated estimation unit 612 uses multi-core fibers based on the transmission / reception states of all the cores 311-1 to 311-N using the estimation results for each wavelength output from the preprocessing and estimation units 611-1 to 611-N. The state estimation result of 310 is comprehensively determined, and the state estimation result of the multi-core fiber 310 is derived.

  As described above, this embodiment is different from the thirteenth embodiment in that both the transmission unit 231 and the reception unit 232 correspond to multi-wavelength light sources. Each core 311 transmits an optical signal in which a plurality of different wavelengths are multiplexed. If the wavelengths are different, the physical effect received during transmission is different even in the same core 311, so that a difference appears in the signal at the time of reception, and there is also a difference in the estimation results in the estimation unit. Since the length of the wavelength and the magnitude of the physical effect received from the fiber (core) are known, the state estimation sensitivity of the transmission path can be improved by evaluating the difference between the estimation results by the estimation unit. In the above description, the case where the transmission path is a multi-core fiber has been described as an example. However, the same effect can be expected even with a single-core fiber.

(Sixteenth embodiment)
The communication system of this embodiment uses a transmission path in which a multicore fiber and a multimode fiber are bundled. Below, it demonstrates centering on the difference with 13th Embodiment.

  FIG. 18 is a block diagram illustrating a configuration of the communication system 114 according to the present embodiment. In this figure, the same parts as those in the communication system 111 according to the thirteenth embodiment shown in FIG. The communication system 114 includes a transmission device 240 and a state estimation device 610. The transmission device 240 performs data communication via the transmission path 304. The transmission path 304 is obtained by bundling a multi-core fiber 310 and a multi-mode fiber 340 together with the same clothing. A multi-fiber may be used instead of the multi-core fiber 310. The number of transmission apparatuses 240 included in the communication system 114 is arbitrary.

  The transmission device 240 includes at least one of the transmission unit 211, the transmission unit 241, the reception unit 212, and the reception unit 242. A pair of transmission unit 211 and reception unit 212 are connected by the core 311 of the multicore fiber 310. The transmission unit 211 and the reception unit 212 connected to the core 311-n (n is an integer greater than or equal to 1 and less than or equal to N, N = 3 in the figure) are referred to as transmission unit 211-n and reception unit 212-n, respectively. In addition, a pair of transmission unit 241 and reception unit 242 are connected by a multimode fiber 340. The transmission units 211-1 to 211 -N and the transmission unit 241 may be included in one transmission device 240 as illustrated in FIG. Similarly, the receiving units 212-1 to 212-N and the receiving unit 242 may be provided in one transmission device 240 as illustrated in the figure, or a part or all of them may be provided in different transmission devices 240. Also good.

  The transmission unit 211-n (n is an integer of 1 to N) of the transmission device 240 on the transmission side outputs an optical signal to the core 311-n of the multicore fiber 310. The reception unit 212-n of the transmission device 240 on the reception side receives the optical signal transmitted by the other transmission device 210 via the core 311-n of the multicore fiber 310. The transmission unit 241 of the transmission device 240 on the transmission side outputs a multimode optical signal to the multimode fiber 340. The receiving unit 242 of the transmission device 240 on the reception side receives the multimode optical signal transmitted from the other transmission device 240 via the multimode fiber 340.

  The preprocessing and estimation unit 611-n (n is an integer between 1 and N) of the state estimation device 610 is transmitted from the reception unit 212-n of the transmission device 240, and the preprocessing and estimation unit 611- (N + 1) is transmitted from the transmission device 240. Data similar to the signal reception data acquired from the reception unit 202 in the first to seventh embodiments, such as constellation data, is acquired from the reception unit 242 of 240 to generate feature data. The preprocessing and estimation unit 611-n estimates an abnormal state of the core 311-n based on the generated feature data and outputs an estimation result. The preprocessing and estimation unit 611- (N + 1) estimates an abnormal state of the multimode fiber 340 based on the generated feature data and outputs an estimation result. The integrated estimation unit 612 comprehensively determines the estimation results output from each of the preprocessing and estimation units 611-1 to 611- (N + 1), and derives the state estimation result of the transmission path 304.

  The physical effect received from each core 311 of the multi-core fiber 310 and the physical effect received from the multi-mode fiber 340 are greatly different. Therefore, a difference appears in the signal at the time of reception, and further, a difference also occurs in the result of the preprocessing and the estimation unit included in the estimation unit 611. Since the difference in physical effect received from the multicore fiber and the multimode fiber is known, the integrated estimation unit 612 evaluates the difference between the estimation results of the preprocessing and estimation units 611-1 to 611-(N + 1). By doing so, the state estimation sensitivity of the transmission line 304 can be improved. In particular, in the case of a multimode fiber, even if it is an optical signal of the same wavelength, the transmission characteristics will be different if the mode is different (that is, the physical effect will change). . Examples thereof are shown in the following seventeenth to nineteenth embodiments.

(Seventeenth embodiment)
The communication system of the present embodiment uses a multimode fiber and a mode demultiplexing unit as a transmission line having a plurality of physical paths.

  FIG. 19 is a block diagram showing the configuration of the communication system 115 according to the present embodiment. In this figure, the same parts as those in the communication system 111 according to the thirteenth embodiment shown in FIG. In the communication system 115, a mode multiplexing unit 351, a multimode fiber 340, and a mode separation unit 352 are used as transmission lines.

  The mode multiplexing unit 351 is connected to the transmission units 211-1 to 211-N and the multimode fiber 340. In the figure, N = 3. The transmission units 211-1 to 211 -N may be the transmission unit 211 provided in one transmission device 210 as shown in the figure, or may be the transmission unit 211 provided in some or all different transmission devices 210. Good. Further, the mode multiplexing unit 351 may be provided in the transmission apparatus 210.

  The mode separation unit 352 is connected to the reception units 212-1 to 212-N and the multimode fiber 340. The receiving units 212-1 to 212-N may be the receiving unit 212 provided in one transmission device 210 as shown in the figure, or may be a receiving unit 212 provided in one or all of the different transmission devices 210. Good. Further, the mode separation unit 352 may be provided in the transmission apparatus 210.

  The transmission units 211-1 to 211-N of the transmission apparatus 210 on the transmission side generate optical signals, respectively, and the mode multiplexing unit 351 mode-multiplexes the generated optical signals and outputs them to the multimode fiber 340. The mode separation unit 352 receives the multimode optical signal transmitted by the multimode fiber 340 and separates it by mode, and the optical signal transmitted by the transmission unit 211-n is received by the reception unit 212-n of the transmission device 210 on the reception side. Output to. The preprocessing and estimation unit 611-n (n is an integer between 1 and N) of the state estimation device 610 acquires signal reception data such as constellation data from the reception unit 212-n of the transmission device 210, and obtains feature data. The estimation result of the abnormal state of the transmission path generated and estimated based on the feature data is output to the integrated estimation unit 612. The integrated estimation unit 612 comprehensively determines the estimation results for each mode output from each of the preprocessing and estimation units 611-1 to 611-N, and derives a transmission path state estimation result.

  In the present embodiment, the state estimation sensitivity is improved by connecting a plurality of transmission units 211 and a plurality of reception units 212 to the multimode fiber 340 via the mode multiplexing unit 351 and the mode separation unit 352.

(Eighteenth embodiment)
In the sixteenth embodiment, optical signals generated using different light sources are transmitted to the respective cores of the multi-core fiber and the multi-mode fiber, and the state is estimated. In the present embodiment, the optical signal generated using the same light source is transmitted to each core of the multi-core fiber and the multi-mode fiber, and the state is estimated. Below, it demonstrates centering on the difference with 16th Embodiment.

  FIG. 20 is a block diagram illustrating a configuration of the communication system 116 according to the present embodiment. In this figure, the same parts as those of the communication system 114 according to the sixteenth embodiment shown in FIG. The communication system 116 shown in the figure is different from the communication system 114 shown in FIG. 18 in that a transmission device 260 is provided instead of the transmission device 240.

  The transmission device 260 includes at least one of the transmission unit 261, the reception unit 212, and the reception unit 242. The cores 311-1 to 311-N (N = 3 in the figure) of the multicore fiber 310 and the multimode fiber 340 are connected to one transmitter 261. In addition, the core 311-n (n is an integer greater than or equal to 1 and less than or equal to N) is connected to the reception unit 212-n, and the multimode fiber 340 is connected to the reception unit 242. The receiving units 212-1 to 212-N and the receiving unit 242 may be provided in one transmission device 260 as illustrated in FIG. 1, or a part or all of them may be provided in different transmission devices 260.

  The transmission unit 261 of the transmission apparatus 260 on the transmission side outputs the same optical signal using a single light source to the cores 311-1 to 311-N of the multicore fiber 310 and the multimode fiber 340. The optical signal transmitted by the transmission unit 261 is received by the reception unit 212-n of the transmission device 260 on the reception side via the core 311-n of the multicore fiber 310, and the reception unit 242 is received by the multimode fiber 340. To do.

  As described above, in the present embodiment, there is one transmission unit 261 (single light source) for all the cores 311 and the multimode fiber 340 of the multicore fiber 310. The example of this embodiment is a combination of the fourteenth embodiment shown in FIG. 16 and the method for improving the state estimation sensitivity of the transmission line shown in the sixteenth embodiment shown in FIG.

(Nineteenth embodiment)
In the sixteenth embodiment, an optical signal having one wavelength is transmitted to each core and multimode fiber of the multicore fiber to estimate the state. In the present embodiment, a wavelength multiplexed signal is transmitted to each core of the multicore fiber and the multimode fiber, and the state is estimated. Below, it demonstrates centering on the difference with 16th Embodiment.

  FIG. 21 is a block diagram showing the configuration of the communication system 117 according to the present embodiment. In this figure, the same parts as those of the communication system 113 according to the fifteenth embodiment shown in FIG. 17 and the communication system 114 according to the sixteenth embodiment shown in FIG. The communication system 117 shown in the figure is different from the communication system 114 shown in FIG. 18 in that a transmission device 270 is provided instead of the transmission device 240.

  The transmission device 270 includes at least one of the transmission unit 231, the transmission unit 271, the reception unit 232, and the reception unit 272. The transmission unit 231 has a multi-wavelength light source, and transmits a wavelength-multiplexed optical signal to the core 311 of the multi-core fiber 310. The transmission unit 271 includes a multi-wavelength light source, and transmits a wavelength-multiplexed multimode optical signal to the multimode fiber 340. The receiving unit 232 supports wavelength multiplexing, and receives the wavelength-multiplexed optical signal from the core 311 of the multicore fiber 310. The receiving unit 272 receives a wavelength-multiplexed multimode optical signal from the multimode fiber 340.

  A pair of transmitters 231-n and receivers 232-n are connected by the core 311-n (n is an integer of 1 or more and N or less, N = 3 in the figure) of the multi-core fiber 310. In addition, a pair of transmission unit 271 and reception unit 272 are connected by multimode fiber 340. The transmission units 231-1 to 231 -N and the transmission unit 271 may be provided in one transmission device 270 as shown in the figure, or a part or all of them may be provided in different transmission devices 270. Similarly, the receiving units 232-1 to 232-N and the receiving unit 272 may be provided in one transmission device 270 as shown in the figure, or some or all of them may be provided in different transmission devices 270. Also good.

  The preprocessing and estimation unit 611-n (n is an integer between 1 and N) of the state estimation device 610 is transmitted from the reception unit 212-n of the transmission device 270, and the preprocessing and estimation unit 611- (N + 1) is transmitted from the transmission device. Signal reception data such as constellation data is acquired for each wavelength from the reception unit 272 of 270 to generate feature data. The preprocessing and estimation unit 611-n estimates an abnormal state of the core 311-n based on feature data for each wavelength and outputs an estimation result. The preprocessing and estimation unit 611- (N + 1) estimates the abnormal state of the multimode fiber 340 based on the feature data for each wavelength, and outputs an estimation result. The integrated estimation unit 612 comprehensively determines the estimation result for each wavelength output from each of the preprocessing and estimation units 611-1 to 611- (N + 1), and derives the state estimation result of the transmission path 304.

(20th embodiment)
In the present embodiment, state estimation is performed by transmitting optical signals in both directions to each of a plurality of physical paths provided in the transmission path.

  FIG. 22 is a diagram showing a communication system 118 according to the twentieth embodiment. The communication system 118 includes a transmission device 280, a state estimation device 680, and data communication entities 481 and 482. The number of transmission devices 280 included in the communication system 118 is arbitrary. The transmission device 280 performs data communication via the multicore fiber 310.

  The transmission device 280 includes one or more transmission / reception units 281. The transmission / reception unit 281 has functions of a transmission unit 211 and a reception unit 212 included in the transmission device 210 of the thirteenth embodiment. The transmission / reception unit 281 may have the functions of the transmission unit 231 and the reception unit 232 included in the transmission device 230 of the fifteenth embodiment.

  A pair of transmission / reception units 281 are connected by the core 311 of the multicore fiber 310. This is the same as that two sets of transmission units and reception units are arranged on one core 311 so as to face each other. In the figure, a transmitting / receiving unit 281 connected to one end of a core 311-n (n is an integer of 1 to N, N = 3 in the same figure) is referred to as a transmitting / receiving unit 281-na and connected to the other end. The transmitting / receiving unit 281 is referred to as a transmitting / receiving unit 281-nb. The transmission / reception units 281-1a to 281-Na may be provided in one transmission device 280 as shown in the figure, or a part or all of them may be provided in different transmission devices 280. Similarly, the transmission / reception units 281-1b to 281-Nb may be provided in one transmission device 280 as shown in the figure, or a part or all of them may be provided in different transmission devices 280.

  The state estimation device 680 includes a first state estimation unit 681 and a second state estimation unit 685. The first state estimation unit 681 and the second state estimation unit 685 are realized by, for example, another computer device or another hardware.

  The first state estimation unit 681 includes preprocessing and estimation units 682-1 to 682-N. The preprocessing and estimation unit 682 has the same function as the preprocessing and estimation unit 611 included in the state estimation device 610 of the thirteenth embodiment. The second state estimation unit 685 includes a preprocessing and estimation unit 686-1 to 686 -N and an integrated estimation unit 687. The preprocessing and estimation unit 686 has the same function as the preprocessing and estimation unit 611 included in the state estimation device 610 of the thirteenth embodiment. The integrated estimation unit 687 estimates the state of the multi-core fiber 310 using the estimation results output from each of the preprocessing and estimation units 682-1 to 682-N and the preprocessing and estimation units 686-1 to 686-N. A result is comprehensively judged and a state estimation result of the multi-core fiber 310 is derived.

  The data communication entity 481-n (n is an integer of 1 to N) transmits the estimation result output by the preprocessing and estimation unit 682-n to the data communication entity 482-n. The data communication entity 482-n outputs the estimation result received from the data communication entity 481-n to the integrated estimation unit 687.

  The transmission / reception unit 281-na (n is an integer of 1 to N) outputs an optical signal to the core 311-n of the multi-core fiber 310, and the transmission / reception unit 281-nb transmits the optical signal via the core 311-n. Receive. The preprocessing and estimation unit 686-n of the second state estimation unit 685 acquires signal reception data such as constellation data from the transmission / reception unit 281-nb to generate feature data, and estimates based on the generated feature data The estimation result of the abnormal state of the core 311 -n is output to the integrated estimation unit 687.

  In addition, the transmission / reception unit 281-nb (n is an integer of 1 to N) outputs an optical signal to the core 311-n of the multi-core fiber 310, and the transmission / reception unit 281-na transmits the optical signal to the core 311-n. Receive via. The preprocessing and estimation unit 682-n of the first state estimation unit 681 acquires data such as constellation data from the transmission / reception unit 281-na, generates feature data, and is estimated based on the generated feature data. The estimation result of the abnormal state of the core 311-n is output to the integrated estimation unit 687 of the second state estimation unit 685 via the data communication entity 481-n and the data communication entity 482-n.

  The integrated estimator 687 is a bi-directional unit for each of the cores 311-1 to 311-N output from the pre-processing and estimation units 682-1 to 682-N and the pre-processing and estimation units 686-1 to 686-N. Based on the estimation result, a state estimation result of the multi-core fiber 310 is derived. The integrated estimation unit 687 uses, for example, a statistical estimation method, a classical rule base AI, or machine learning such as a neural network to derive the state estimation result.

  As described above, in the present embodiment, two sets of transmission units and reception units are arranged to face each core 311 of the multi-core fiber 310, and pre-processing is performed for each combination of the transmission unit and the reception unit. And an estimation unit. Furthermore, data communication entities 481 and 482 for transmitting the estimation unit results are provided, and estimation results at both ends of the multi-core fiber 310 can be collected in one integrated estimation unit 687. By using the estimation results at both ends of the multi-core fiber 310, the state estimation sensitivity of the transmission path can be improved. In particular, when two transmitters facing each other in one core 311 use optical signals having the same wavelength, an improvement in state estimation sensitivity similar to the example described for the communication system 112 of the fourteenth embodiment can be expected.

(21st Embodiment)
In the twentieth embodiment, optical signals in both directions are transmitted to each of a plurality of physical paths included in the transmission path, and the integrated estimation unit comprehensively determines the result of state estimation for each direction of each physical path. The state estimation result of the transmission line was derived. In the present embodiment, an integrated estimation unit is provided at each of both ends of the transmission line, and the state of the transmission line is estimated comprehensively based on the estimation results by the integrated estimation units at both ends, thereby improving the state estimation sensitivity.

  FIG. 23 is a diagram showing a communication system 119 according to the twenty-first embodiment. In this figure, the same parts as those of the communication system 118 according to the twentieth embodiment shown in FIG. 22 are denoted by the same reference numerals, and the description thereof is omitted. The communication system 119 includes a transmission device 280, a state estimation device 690, and data communication entities 491 and 492.

  The state estimation device 690 includes a first state estimation unit 691 and a second state estimation unit 695. The first state estimation unit 691 and the second state estimation unit 695 are realized by, for example, another computer device or another hardware. The first state estimation unit 691 includes a preprocessing and estimation unit 682-1 to 682-N (N = 3 in the figure) and an integrated estimation unit 693. The integrated estimation unit 693 has the same function as the integrated estimation unit 612 included in the state estimation device 610 of the thirteenth embodiment. The second state estimation unit 695 includes a preprocessing and estimation unit 686-1 to 686 -N, an integrated estimation unit 697, and a system estimation unit 698. The integrated estimation unit 697 has the same function as the integrated estimation unit 612 included in the state estimation device 610 of the thirteenth embodiment. The system estimation unit 698 combines the estimation result by the integrated estimation unit 693 of the first state estimation unit 691 and the estimation result by the integrated estimation unit 697 of the second state estimation unit 695 to obtain the state estimation result of the entire transmission path. To derive.

  The data communication entity 491 transmits the estimation result output from the integrated estimation unit 693 of the first state estimation unit 691 to the data communication entity 492. The data communication entity 492 outputs the estimation result received from the data communication entity 491 to the system estimation unit 698 of the second state estimation unit 695.

  The transmission / reception unit 281-na (n is an integer of 1 to N) outputs an optical signal to the core 311-n of the multi-core fiber 310, and the transmission / reception unit 281-nb transmits the optical signal via the core 311-n. Receive. The preprocessing and estimation unit 686-n of the second state estimation unit 695 acquires signal reception data such as constellation data from the transmission / reception unit 281-nb to generate feature data, and estimates based on the generated feature data The estimation result of the abnormal state of the core 311 -n is output to the integrated estimation unit 697. The integrated estimation unit 697 comprehensively determines the state estimation result of the multicore fiber 310 using the estimation results output from each of the preprocessing and estimation units 686-1 to 686 -N, and estimates the state of the multicore fiber 310. Derive the result. The integrated estimation unit 697 outputs the state estimation result to the system estimation unit 698.

  In addition, the transmission / reception unit 281-nb (n is an integer of 1 to N) outputs an optical signal to the core 311-n of the multi-core fiber 310, and the transmission / reception unit 281-na transmits the optical signal to the core 311-n. Receive via. The preprocessing and estimation unit 682-n of the first state estimation unit 691 acquires data such as constellation data from the transmission / reception unit 281-na, generates feature data, and is estimated based on the generated feature data. The estimation result of the abnormal state of the core 311-n is output to the integrated estimation unit 693. The integrated estimation unit 693 comprehensively determines the state estimation result of the multicore fiber 310 using the estimation results output from each of the preprocessing and estimation units 682-1 to 682-N, and estimates the state of the multicore fiber 310. Derive the result. The integrated estimation unit 693 outputs the state estimation result to the system estimation unit 698 of the second state estimation unit 695 via the data communication entity 491 and the data communication entity 492. The system estimation unit 698 uses, for example, a statistical estimation method, a classical rule base AI, or a machine learning such as a neural network, and combines the estimation result by the integrated estimation unit 693 and the estimation result by the integrated estimation unit 697, A state estimation result of the entire multi-core fiber 310 is derived.

  In the above, the integrated estimation units 693 and 697 are provided at both ends of the multi-core fiber 310, and the estimation results by the integrated estimation units 693 and 697 are transmitted to the system estimation unit 698, but instead of or in addition to these estimation results Then, signal reception data such as constellation data acquired from each core at both ends of the multi-core fiber 310, or feature data generated by performing preprocessing on the constellation data may be transmitted to the system estimation unit 698. In addition to the estimation results obtained by the integrated estimation units 693 and 697, the system estimation unit 698 can further use the information to estimate the state of the transmission path comprehensively and improve the state estimation sensitivity.

  According to the above-described embodiment, it is possible to estimate the state of the transmission apparatus including the transmission path or the transmission unit and the reception unit by detecting the cause and the level of the failure that may occur regardless of the transmission line medium. For example, the state estimation device can transmit or receive a transmission path such as a change in temperature or one or both of an unintended bending state and a bending diameter of the transmission path without interrupting user data transmission / reception. The abnormal state of the transmission apparatus including Therefore, it is possible to quickly detect a failure sign and perform recovery work when the failure occurs.

  The state estimation devices 510, 520, 530, 540, 550, 560, 570, 580, 590, 595, 610, the first state estimation units 681, 691, and the second state estimation units 685, 695 in the embodiment described above are buses. The functions in the above-described embodiment are realized by executing a state estimation program, which includes a CPU (Central Processing Unit), a memory, an auxiliary storage device, and the like connected to each other. All of the functions of the state estimation devices 510, 520, 530, 540, 550, 560, 570, 580, 590, 595, 610, the first state estimation units 681, 691, and the second state estimation units 685, 695 or Some may be realized using hardware such as an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA). The state estimation program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a storage device such as a hard disk built in the computer system. The state estimation program may be transmitted via a telecommunication line.

  The transmission apparatus 200 may include state estimation apparatuses 510, 520, 530, 540, 550, 560, 570, 580, 590, and 595 inside, and the transmission apparatus 200 includes the state estimation apparatuses 510, 520, 530, 540, and Some functions of 550, 560, 570, 580, 590, and 595 may be provided. For example, data processing in the state estimation devices 510, 520, 530, 540, 550, 560, 570, 580, 590, and 595 using a control CPU, electronic circuit, and the like of the receiving unit 202 included in the transmission device 200 An estimation process may be executed. Further, the transmission devices 210, 220, 230, 240, 260, and 270 may include the state estimation device 610 inside, or may include some functions of the state estimation device 610. Further, the transmission device 280 may include first state estimation units 681 and 691 and second state estimation units 685 and 695 inside, and the first state estimation units 681 and 691 and the second state estimation units 685 and 695 are provided. Some functions may be provided.

  According to the embodiment described above, the state estimation device includes the preprocessing unit and the estimation unit. The preprocessing unit and the estimation unit are realized by hardware, by a processor such as a CPU executing a software program, or a combination thereof. The pre-processing unit is transmitted from the transmission unit of the transmission device, and the phase of the signal received by the reception unit of the other transmission device via the transmission path, the reception strength (for example, reception power, Q value), the reception quality (for example, BER value, OSNR value, ESNR value), voltage after conversion to an electric signal (voltage amplitude value of the electric signal of Rx), and signal processing parameters (for example, Rx equalizer) used in reception processing in the receiving unit Signal reception data representing one or more of the tap coefficients) is acquired, and the acquired signal reception data is processed, for example, statistically to be processed into feature data used for state estimation. The feature data includes phase plane state data representing the phase in the phase plane, polar coordinate data representing the phase in the polar coordinate plane, Fourier transform data obtained by fast Fourier transform of the polar coordinate data, and frequency of appearance of the signal in the phase plane state data or polar coordinate data. And histogram data representing the frequency of appearance of a signal in any one or a combination of two or more of reception intensity, reception quality, voltage and signal processing parameters, and time series of reception intensity, reception quality, voltage or signal processing parameters And at least one of time-series data indicating changes in The estimation unit estimates the state of the transmission path, the abnormal state of the transmission unit, or the abnormal state of the reception unit based on the feature data generated by the preprocessing unit. The state of the transmission line is one or more of bending presence / absence, bending diameter, temperature change, tension, fusion point deviation, vibration, water leakage, and twist.

  The estimation unit can perform estimation using machine learning such as a statistical estimation method, a classic rule base AI, or a neural network. Further, the estimation unit uses a partial estimation unit that estimates a part of the state included in any of the transmission path state, the transmission unit abnormal state, or the reception unit abnormal state using at least a part of the feature data. A transmission path state, an abnormal state of the transmission unit, or an abnormal state of the reception unit may be estimated based on estimation results obtained by a plurality of partial estimation units that have a plurality of different partial states. For example, the partial estimation unit may estimate two states such as the recognition functions # 1 to #n included in the estimation unit 592, and may be 2 such as the recognition functions # 1 to #n included in the estimation unit 597. One or more states may be estimated.

  When the transmission path is composed of a plurality of physical paths, the preprocessing unit acquires signal reception data for each physical path, generates feature data used for state estimation from the acquired signal reception data, and the estimation unit For each route, the state of the physical route is estimated based on the feature data. The integrated estimation unit of the state estimation device derives a transmission path state estimation result based on the state of each physical path estimated by the estimation unit.

  The preprocessing unit may acquire signal reception data for each physical path when transmitting an optical signal using the same light source for two or more or all of the plurality of physical paths. Alternatively, signal reception data for each physical path when an optical signal using a different light source is transmitted may be acquired.

  The transmission path composed of a plurality of physical paths is, for example, a multicore fiber, a multimode fiber, a multifiber, or a combination of two or more. Alternatively, the transmission path includes a mode multiplexing unit that multiplexes a plurality of single-mode optical signals and converts them into a multi-mode optical signal, a multi-mode fiber that transmits the multi-mode optical signals converted by the mode multiplexing unit, A mode separation unit that separates a multimode optical signal transmitted through the mode fiber into a single mode optical signal.

  The pre-processing unit may acquire signal reception data when bi-directional optical signals are transmitted for each physical path, and generate feature data used for state estimation from the acquired signal reception data. An estimation part estimates the state of a physical path | route based on characteristic data for every physical path | route about each of bidirectional | two-way. The integrated estimation unit derives a state estimation result of the transmission path based on the state of each physical path estimated for each of the bidirectional directions by the estimation unit.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

  The present invention can be applied to state detection of transmission lines and transmission devices.

101,102,103,104,105,106,107,111,112,113,114,115,116,117,118,119 ... communication system 200,210,220,230,240,260,270,280 ... Transmission devices 201, 211-1 to 211-3, 221, 231-1 to 231-3, 241, 261, 271 ... transmission units 202, 212-1 to 212-3, 232-1 to 232-3, 242, 272... Receiving units 281-1a to 281-3a, 281-1b to 281-3b... Transmitting and receiving units 300 and 304... Transmission path 310 .. multicore fibers 311-1 to 311-3. 352... Mode separators 481-1 to 481-3, 482-1 to 482-3, 491, 492. States 510, 520, 530, 540, 550, 560, 570, 580, 590, 595, 610, 680, 690 ... state estimation devices 511, 521, 531, 541, 551, 561, 571, 581, 591, 596 ... Pre-processing units 512, 522, 532, 542, 552, 562, 572, 582, 592, 597 ... estimation units 611-1 to 611-4, 682-1 to 682-3, 686-1 to 686-3. Processing and estimation units 612, 687, 693, 697 ... integrated estimation units 681, 691 ... first state estimation unit 685, 695 ... second state estimation unit 698 ... system estimation unit

Claims (13)

Used in the reception process, the phase of the signal transmitted from the transmission unit of the transmission device and received by the reception unit of the other transmission device via the transmission path, the received intensity, the reception quality, the voltage after conversion into an electrical signal, and so on. Obtaining signal reception data representing one or more of the obtained signal processing parameters, and generating feature data used for state estimation from the acquired signal reception data;
Based on the feature data, an estimation unit that estimates the state of the transmission path, the abnormal state of the transmission unit, or the abnormal state of the reception unit;
A state estimation device comprising: The state of the transmission line is one or more of presence / absence of bending, bending diameter, and temperature change,
The state estimation apparatus according to claim 1. The feature data includes phase plane state data representing the phase in a phase plane, polar coordinate data representing the phase in a polar coordinate plane, Fourier transform data obtained by fast Fourier transforming the polar coordinate data, and the phase plane state data or the polar coordinates. Histogram data representing the appearance frequency of data, histogram data representing the appearance frequency of the signal in any one or a combination of two or more of the reception intensity, the reception quality, the voltage or the signal processing parameter, the reception intensity, Including at least one of reception quality, time series data indicating a time series change of the voltage or the signal processing parameter,
The state estimation apparatus according to claim 1 or 2. The estimation unit performs estimation using machine learning.
The state estimation apparatus as described in any one of Claims 1-3. The estimation unit uses at least a part of the feature data to estimate a part of a state included in any of the state of the transmission path, the abnormal state of the transmission unit, or the abnormal state of the reception unit Based on estimation results by the plurality of partial estimation units that have a plurality of estimation units and estimate different partial states, respectively, the transmission path state, the transmission unit abnormal state, or the reception unit abnormal state Estimate
The state estimation apparatus as described in any one of Claims 1-4. The preprocessing unit or the estimation unit is realized by hardware, a processor that executes a program, or a combination of hardware and a processor that executes a program.
The state estimation apparatus according to any one of claims 1 to 5. The transmission path is composed of a plurality of physical paths,
The preprocessing unit acquires the signal reception data for each physical path, generates feature data used for state estimation from the acquired signal reception data,
The estimation unit estimates the state of the physical path based on the feature data for each physical path,
The state estimation device further includes an integrated estimation unit that derives a state estimation result of the transmission path based on the state of each physical path estimated by the estimation unit.
The state estimation apparatus according to any one of claims 1 to 5. The pre-processing unit acquires the signal reception data for each physical path when transmitting an optical signal using the same light source for two or more or all of the plurality of physical paths.
The state estimation apparatus according to claim 7. The preprocessing unit acquires the signal reception data for each physical path when transmitting an optical signal using a different light source for each of the plurality of physical paths.
The state estimation apparatus according to claim 7. The transmission line is a multi-core fiber, a multi-mode fiber, or a multi-fiber, or a combination of two or more.
The state estimation apparatus according to claim 7. The transmission path is
A mode multiplexing unit that multiplexes a plurality of single-mode optical signals and converts them into a multi-mode optical signal;
A multimode fiber for transmitting the multimode optical signal converted by the mode multiplexing unit;
A mode separation unit for separating the multimode optical signal transmitted through the multimode fiber into a single mode optical signal;
The state estimation apparatus according to claim 7. The preprocessing unit acquires the signal reception data when transmitting each bidirectional optical signal for each physical path, and generates feature data used for state estimation from the acquired signal reception data,
The estimation unit estimates the state of the physical path based on the feature data for each physical path for each bidirectional direction,
The integrated estimation unit derives the state estimation result of the transmission path based on the state of each physical path estimated for each bidirectional by the estimation unit,
The state estimation apparatus according to claim 7.   A communication system provided with a transmission apparatus and the state estimation apparatus as described in any one of Claims 1-12.

Images