Output device

The output device and method address the challenge of identifying models for update in optical networks by analyzing configuration changes and specifying affected models, thereby reducing costs and improving efficiency in model management within optical networks.

WO2024232093A1PCT designated stage expired Publication Date: 2024-11-14NEC CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2023/017790
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-11
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

In optical networks, changes in network configuration, such as adding or removing communication paths, can deteriorate the performance of models used for anomaly detection, leading to increased costs due to the need to check trends in all sensors before and after the event, making it difficult to identify models that require updating efficiently.

Method used

An output device and method that acquires configuration change information, identifies overlapping communication paths affected by the change, and specifies the corresponding models for potential updates, allowing for targeted model identification and reduction of costs by avoiding the need to check all sensors.

Benefits of technology

This approach enables efficient identification and potential updating of models affected by network configuration changes, reducing costs by focusing on specific communication paths and models, rather than requiring comprehensive sensor trend analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2023017790_14112024_PF_FP_ABST
    Figure JP2023017790_14112024_PF_FP_ABST
Patent Text Reader

Abstract

This output device 600 includes: an acquisition unit 621 that acquires configuration change information indicating a change in a network configuration in an optical network; a path identification unit 622 that, on the basis of the configuration change information, identifies a communication path having a path that overlaps with a communication path subjected to a configuration change among set communication paths; a model identification unit 623 that identifies a trained model affected by the configuration change on the basis of the communication path identified by the path identification unit 622; and an output unit 624 that outputs information about the model identified by the model identification unit 623.
Need to check novelty before this filing date? Find Prior Art

Description

Output Device

[0001] The present invention relates to an output device, an output method, and a recording medium.

[0002] In optical networks such as WDM-NW (Wavelength Division Multiplexing Network), it is known that abnormalities are detected by inputting values ​​indicating optical intensity and transmission quality into a pre-trained model.

[0003] Furthermore, when using such a model, it may be necessary to retrain the model for some reason. Patent Document 1 (JP-A-2011-103599) provides an example of a technology used to identify a model to be retrained. For example, Patent Document 1 describes a machine learning system in which a machine learning model detects anomalies in equipment based on sensor data output by one or more sensors that detect the state of the equipment being the target of anomaly detection. According to Patent Document 1, the machine learning system manages a model identifier unique to the machine learning model that identifies the machine learning model, in association with a sensor identifier unique to the sensor that identifies the sensor that outputs sensor data that serves as input data for the machine learning model. Furthermore, the machine learning system calculates an impact level representing a change in the trend of sensor data before and after the maintenance work for each maintenance event identifier unique to the maintenance work that identifies the maintenance work performed on the equipment, and manages the impact level in association with each sensor identifier. Then, if the impact level satisfies a predetermined condition, the machine learning system determines that the sensor has been affected by the maintenance work and presents a model identifier associated with the sensor identifier of the sensor that satisfies the condition.

[0004] Japanese Patent Application Laid-Open No. 2022-175959

[0005] In optical networks, when optical communications are performed, changes to the network configuration, such as increasing or decreasing the number of communication paths, are made as needed. Such changes to the network configuration increase or decrease the number of wavelengths transmitting through the communication paths. As a result, the intensity of the transmitted light fluctuates, which may cause a deterioration in the performance of the model. To address this risk, it is desirable to identify and update models whose performance deteriorates in response to changes in the network configuration. However, the technology described in Patent Document 1 requires checking the trend changes of all sensors before and after an event such as a change in the network configuration. As a result, there is a risk that the cost of identifying models to be updated will be high. As such, there has been a problem in that it is difficult to reduce the cost of identifying models that require updating.

[0006] Therefore, an object of the present invention is to provide an output device, an output method, and a recording medium that can solve the above-mentioned problems.

[0007] In order to achieve this object, an output device that is one form of the present disclosure has a configuration including: an acquisition unit that acquires configuration change information that indicates a change in network configuration in an optical network; a path identification unit that identifies, based on the configuration change information, communication paths that have already been set and have routes that overlap with communication paths whose configuration has been changed; a model identification unit that identifies a trained model that will be affected by the configuration change, based on the communication paths identified by the path identification unit; and an output unit that outputs information about the model identified by the model identification unit.

[0008] Furthermore, an output method according to another aspect of the present disclosure is configured such that an information processing device acquires configuration change information indicating a change in network configuration in an optical network, identifies, based on the configuration change information, communication paths among already set communication paths that have a route that overlaps with a communication path whose configuration has been changed, identifies, based on the identified communication paths, a trained model that is affected by the configuration change, and outputs information about the identified model.

[0009] Another embodiment of the present disclosure is a computer-readable recording medium having recorded thereon a program for causing an information processing device to perform the following processes: acquire configuration change information indicating a change in network configuration in an optical network; identify, based on the configuration change information, communication paths among already-set communication paths that have a route that overlaps with a communication path whose configuration has been changed; identify, based on the identified communication paths, a trained model that is affected by the configuration change; and output information about the identified model.

[0010] According to the above-mentioned configurations, the above-mentioned problems can be solved.

[0011] FIG. 1 is a diagram for explaining an overview of an optical communication system according to a first embodiment of the present disclosure. FIG. 2 is a diagram for explaining an example of an optical communication path. FIG. 3 is a block diagram showing an example of a configuration of an output device. FIG. 4 is a diagram showing an example of path information. FIG. 5 is a diagram showing an example of path-model correspondence information. FIG. 6 is a flowchart showing an example of operation of an output device. FIG. 7 is a block diagram showing an example of a configuration of an output device according to a second embodiment of the present disclosure. FIG. 8 is a diagram showing an example of path-sensor correspondence information. FIG. 9 is a diagram showing an example of sensor-model correspondence information. FIG. 10 is a flowchart showing an example of operation of an output device. FIG. 11 is a block diagram showing an example of a configuration of an output device according to a third embodiment of the present disclosure. FIG. 12 is a flowchart showing an example of operation of an output device. FIG. 13 is a diagram showing an example of a hardware configuration of an output device according to a fourth embodiment of the present disclosure. FIG. 14 is a block diagram showing an example of a configuration of an output device.

[0012] [First Embodiment] A first embodiment of the present disclosure will be described with reference to Figs. 1 to 6. Fig. 1 is a diagram for explaining an overview of an optical communication system 100. Fig. 2 is a diagram for explaining an example of an optical communication path. Fig. 3 is a block diagram showing an example of the configuration of an output device 300. Fig. 4 is a diagram showing an example of path information 341. Fig. 5 is a diagram showing an example of path-model correspondence information 342. Fig. 6 is a flowchart showing an example of the operation of the output device 300.

[0013] In a first embodiment of the present disclosure, an optical communication system 100 is described that outputs information about trained models that may be affected by a change in network configuration, such as adding or removing communication paths in an optical network 200. When a change in network configuration, such as adding or removing communication paths, occurs, the number of wavelengths transmitted through the communication paths increases or decreases, which may cause fluctuations in the intensity of the transmitted light and adversely affect the performance of the models. On the other hand, even if a communication path is added or removed, it is assumed that the impact of adding or removing a communication path is small for communication paths whose routes do not overlap with the added or removed communication paths. Therefore, when the optical communication system 100 acquires configuration change information indicating a change in network configuration, it identifies communication paths that may be affected by the configuration change, such as those whose routes at least partially overlap with the added or removed communication paths, based on the acquired configuration change information. Furthermore, the optical communication system 100 identifies models that may be affected based on the identified communication paths. The optical communication system 100 then outputs information about the identified models. For example, in the optical communication system 100, a model to be used for anomaly detection, etc., is pre-trained for each of one or more communication paths. The optical communication system 100 of the present disclosure can identify communication paths that may be affected, and thereby identify models that perform anomaly detection, etc. in the identified communication paths as models that may be affected.

[0014] Specifically, for example, in the optical network 200 shown in FIG. 1 , assume that communication path 1, communication path 2, and communication path 3 have already been established, and communication path 4 has been newly added. In this case, as shown in FIG. 1 , the routes taken by optical communications for communication path 1 and communication path 3 partially overlap with communication path 4. On the other hand, for communication path 2, there is no route that overlaps with the newly added communication path 4. Therefore, when the optical communication system 100 acquires configuration change information indicating that communication path 4 has been newly added, it identifies communication path 1 and communication path 3, whose routes at least partially overlap, as communication paths that may be affected, based on the acquired configuration change information. Thereafter, the optical communication system 100 outputs information about a model for performing anomaly detection, etc., on the identified communication paths.

[0015] A communication path can be defined by a combination of optical communication devices (objects) such as transponders, amplifiers, and switches, and fiber cables. Fiber cables can be identified by specifying the output and input ports of each object. Therefore, information about a communication path can be expressed as a combination of objects and ports. When objects in the information about a communication path overlap and at least some of the ports in the overlapping objects overlap, the communication paths can be determined to be overlapping. For example, FIG. 2 shows an example of an optical network 200. In the example shown in FIG. 2, communication path 1 and communication path 2 overlap in one of the objects 210 owned by node 2, and the object 210 and the output port overlap. Therefore, in the example shown in FIG. 2, the optical communication system 100 can determine that communication path 1 and communication path 2 partially overlap in their routes, and that these are a combination that may be affected.

[0016] The model is also assumed to be trained in advance to perform anomaly detection and the like using data acquired by one or more sensors possessed by the object, etc. as input. For example, the model may be trained using any method, such as training data acquired by the sensors. In this embodiment, the optical communication system 100 trains a corresponding model in advance for each of one or more communication paths. A configuration example of the present disclosure will be described below.

[0017] Fig. 1 shows an example of the configuration of an optical communication system 100. Referring to Fig. 1, the optical communication system 100 includes an optical network 200 and an output device 300. As shown in Fig. 1, the optical network 200 and the output device 300 are connected so as to be able to communicate with each other.

[0018] The optical network 200 is a network that performs optical communications such as a WDM-NW (Wavelength Division Multiplexing Network). As illustrated in FIG. 2, the optical network 200 includes a plurality of objects 210 such as transponders, amplifiers, and switches. For example, in the optical network 200, a plurality of communication paths are set that pass through a plurality of objects 210. On the communication paths, mutual communication can be performed by overlapping communications using a plurality of channels.

[0019] The output device 300 is an information processing device that identifies models that need to be updated in accordance with configuration change information that indicates a change in the network configuration in the optical network 200, and outputs information about the identified models. The output device 300 may be configured to even update the models based on the output information. FIG. 3 shows an example configuration of the output device 300. Referring to FIG. 3, the output device 300 has, as main components, for example, an operation input unit 310, a screen display unit 320, a communication I / F unit 330, a storage unit 340, and an arithmetic processing unit 350.

[0020] 3 illustrates an example in which the functions of the output device 300 are realized using one information processing device. However, the output device 300 may be realized using multiple information processing devices, for example, on the cloud. Furthermore, the output device 300 may not include some of the components illustrated above, such as not having the operation input unit 310 or the screen display unit 320, or may have a configuration other than those illustrated above. In other words, the output device 300 may employ any modified example.

[0021] The operation input unit 310 is made up of operation input devices such as a keyboard, a mouse, etc. The operation input unit 310 detects operations of the operator who operates the output device 300 and outputs the operations to the arithmetic processing unit 350.

[0022] The screen display unit 320 is composed of a screen display device such as a liquid crystal display, an organic electroluminescence (EL) display, etc. The screen display unit 320 can display various information stored in the storage unit 340 on the screen in response to instructions from the arithmetic processing unit 350.

[0023] The communication I / F unit 330 includes a data communication circuit, etc. The communication I / F unit 330 performs data communication with an external device connected via a communication line.

[0024] The storage unit 340 is a storage device such as a hard disk or memory. The storage unit 340 stores processing information and a program 343 required for various processes in the arithmetic processing unit 350. The program 343 is read into the arithmetic processing unit 350 and executed to realize various processing units. The program 343 is read in advance from an external device or recording medium via a data input / output function such as the communication I / F unit 330, and is saved in the storage unit 340. Examples of main information stored in the storage unit 340 include path information 341 and path-model correspondence information 342. The storage unit 340 may also include information other than those exemplified above, such as model information.

[0025] The path information 341 includes information about communication paths set on the optical network 200. In other words, the path information 341 includes information indicating pairs of objects 210 and ports that exist on communication paths set on the optical network 200. For example, the path information 341 is acquired in advance using a method such as acquiring it from an external device such as an NMS (Network Management System), and is stored in the storage unit 340. Furthermore, the path information 341 can be updated in response to the acquisition of configuration change information by a configuration change information acquisition unit 351 (described later), for example.

[0026] FIG. 4 shows an example of path information 341. Referring to FIG. 4, the path information associates a path ID (identifier) ​​with path configuration information. Here, the path ID indicates identification information of a communication path. Identification information may be assigned arbitrarily for each communication path. Furthermore, the path configuration information indicates a combination of an object 210 and a port present on the communication path, such as (transponder 1, output port 1), (amplifier 1, input port 1, output port 1), .... In other words, the path configuration information includes multiple pieces of information that combine an object 210 through which the communication path passes and a port used in that object 210.

[0027] The path-model correspondence information 342 indicates, for each communication path, information about a trained model used for anomaly detection, etc. For example, the path-model correspondence information 342 is acquired in advance by inputting it using the operation input unit 310, acquiring it from an external device via the communication I / F unit 330, or the like, and is stored in the storage unit 340.

[0028] Fig. 5 shows an example of the path-model correspondence information 342. Referring to Fig. 5, in the path-model correspondence information 342, a path ID and a model ID are associated with each other. Here, the path ID indicates identification information of a communication path. The path ID may be the same as the path information 341. Furthermore, the model ID indicates identification information of a model. The model identification information may be assigned arbitrarily for each model.

[0029] The arithmetic processing unit 350 has an arithmetic device such as a CPU (Central Processing Unit) and its peripheral circuits. The arithmetic processing unit 350 reads and executes a program 343 from the storage unit 340, thereby realizing various processing units by causing the above hardware and the program 343 to work together. Main processing units realized by the arithmetic processing unit 350 include, for example, a configuration change information acquisition unit 351, a path identification unit 352, a model identification unit 353, and an output unit 354.

[0030] In addition, instead of the above-mentioned CPU, the arithmetic processing unit 350 may have a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an MPU (Micro Processing Unit), an FPU (Floating point number Processing Unit), a PPU (Physics Processing Unit), a TPU (Tensor Processing Unit), a quantum processor, a microcontroller, or a combination of these.

[0031] The configuration change information acquisition unit 351 acquires configuration change information indicating changes to the network configuration in the optical network 200, such as the addition or removal of a communication path. For example, the configuration change information acquisition unit 351 can acquire configuration change information indicating that a communication path has been added or removed by referring to network configuration information managed on an NMS. The configuration change information acquisition unit 351 may acquire configuration change information using any method other than the above examples. Furthermore, the configuration change information acquisition unit 351 can update the path information 341 based on the acquired configuration change information.

[0032] For example, the configuration change information acquisition unit 351 can acquire, as configuration change information, information indicating pairs of objects 210 and ports in a communication path to be newly added or removed, such as (transponder x, output port 1), (amplifier x, input port 1, output port 1), .... The configuration change information acquisition unit 351 may acquire, along with the above-mentioned configuration change information, information indicating the time when a change occurred in the network configuration. Furthermore, the configuration change information acquisition unit 351 may acquire, as configuration change information, a path ID indicating the communication path to be removed.

[0033] The path identification unit 352 identifies, from among the already set communication paths, communication paths that may be affected by a network configuration change, such as those whose routes at least partially overlap with those of communication paths indicated by the configuration change information, based on the configuration change information acquired by the configuration change information acquisition unit 351. In other words, the path identification unit 352 identifies, from among the already set communication paths, communication paths whose routes overlap with those of communication paths whose configuration has been changed (communication paths that have been added or removed) indicated by the configuration change information.

[0034] For example, the path identification unit 352 compares the configuration change information with the path information 341 to identify communication paths that may be affected by a network configuration change. Specifically, for example, the path identification unit 352 compares the configuration change information with the path configuration information included in the path information 341. Then, if the objects 210 overlap in the compared configuration change information and the path configuration information, and if at least some of the ports in the overlapping objects 210 overlap, the path identification unit 352 determines that the communication paths indicated by the path configuration information may be affected by a network configuration change. For example, the path identification unit 352 performs the above comparison for each piece of path configuration information included in the path information 341 to identify communication paths indicated by the path information 341 that may be affected by a network configuration change. For example, as described above, the path identification unit 352 compares the configuration change information with the path information 341 to identify overlapping objects 210 and ports, thereby enabling the path identification unit 352 to identify communication paths having routes that overlap with communication paths that have been added or removed.

[0035] The model identification unit 353 identifies a model that may be affected by a change in the network configuration based on the results identified by the path identification unit 352. For example, the model identification unit 353 can identify a model that may be affected by a change in the network configuration by referring to the path-model correspondence information 342 and identifying a model that is associated with the communication path identified by the path identification unit 352.

[0036] The output unit 354 outputs information about the model identified by the model identification unit 353. For example, the output unit 354 may output a model ID or the like. The output unit 354 can display the information about the model on the screen display unit 320 or transmit it to an external device via the communication I / F unit 330.

[0037] The above is an example of the configuration of the output device 300. The output device 300 may be configured to manage a model. In this case, the output device 300 may be configured to update the model according to the output of the output unit 354. In the present disclosure, the specific processing for updating the model is not particularly limited. The output device 300 may update the model by collecting data acquired by a sensor. Next, an example of the operation of the output device 300 will be described with reference to FIG. 6 .

[0038] 6 is a flowchart showing an example of the operation of the output device 300. Referring to Fig. 6, the configuration change information acquisition unit 351 acquires configuration change information indicating a change in the network configuration in the optical network 200, such as an addition or reduction of a communication path (step S101). For example, the configuration change information acquisition unit 351 can acquire the configuration change information by referring to network configuration information managed on the NMS.

[0039] The path identification unit 352 identifies communication paths that may be affected by the network configuration change based on the configuration change information acquired by the configuration change information acquisition unit 351 (step S102). For example, the path identification unit 352 compares the configuration change information with the path information 341 to identify communication paths that may be affected by the network configuration change.

[0040] The model identification unit 353 identifies models that may be affected by the network configuration change based on the results identified by the path identification unit 352 (step S103). For example, the model identification unit 353 can identify models that may be affected by the network configuration change by referring to the path-model correspondence information 342.

[0041] The output unit 354 outputs information about the model identified by the model identifying unit 353 (step S104).

[0042] The above is an example of the operation of the output device 300. The output device 300 may update the model based on the output information.

[0043] As described above, the output device 300 includes a path identification unit 352 and a model identification unit 353. With this configuration, the model identification unit 353 can identify models that may be affected by a change in network configuration based on the results identified by the path identification unit 352. As a result, it is possible to identify models without checking trend changes in all sensors before and after an event such as a change in network configuration. This makes it possible to reduce the cost of identifying models that require updating.

[0044] The model may be trained to take data acquired by a sensor as input and output parameters such as an S / N (signal / noise) ratio or any other arbitrary quality value. The model may also be trained to perform arbitrary class classification using data acquired by one or more sensors possessed by an object, etc. as input, such as fault classification. For example, as described above, the methods exemplified in the present disclosure may be applied to models other than those that perform anomaly detection.

[0045] Second Embodiment Next, a second embodiment of the present disclosure will be described with reference to Figs. 7 to 10. Fig. 7 is a block diagram showing an example of the configuration of an output device 400. Fig. 8 is a diagram showing an example of path-sensor correspondence information 442. Fig. 9 is a diagram showing an example of sensor-model correspondence information 443. Fig. 10 is a flowchart showing an example of the operation of the output device 400.

[0046] In a second embodiment of the present disclosure, an output device 400 will be described, which is a modified example of the output device 300 described in the first embodiment. In the first embodiment, a case where a model is trained for each of one or more communication paths will be described. In the second embodiment, an output device 400 will be described that has a configuration that can accommodate a case where a model is trained for each of one or more sensors, rather than for each communication path. As will be described later, the output device 400 of the present disclosure identifies sensors that may be affected by a network configuration change based on the results of identifying the communication paths. Furthermore, the output device 400 further identifies sensors that satisfy predetermined conditions among the identified sensors. For example, the output device 400 further identifies sensors that are actually affected among the identified sensors by calculating an influence level (described later) for the identified sensors. Then, the output device 400 identifies models corresponding to the identified sensors. In this manner, the output device 400 of the present disclosure identifies models that may be affected by a network configuration change in response to identifying affected sensors based on the results of identifying the communication paths.

[0047] In the present disclosure, the degree of influence is a value corresponding to the status of data acquired by a sensor before and after a change in network configuration, and indicates, for example, the magnitude of the impact that a change in network configuration has on data acquired by a sensor. For example, the degree of influence may be a value corresponding to a trend change in data acquired by a sensor before and after a change in network configuration. As an example, the higher the value of the degree of influence, the more the data acquired by the sensor changes before and after the change in network configuration. The degree of influence can be calculated using any method, such as calculating based on statistical values ​​of data acquired by a sensor before a change in network configuration and statistical values ​​of data acquired by a sensor after a change in network configuration.

[0048] Fig. 7 shows an example of the configuration of the output device 400. Referring to Fig. 7, the output device 400 has, as main components, for example, an operation input unit 410, a screen display unit 420, a communication I / F unit 430, a storage unit 440, and an arithmetic processing unit 450. Note that, like the output device 300, the output device 400 may be realized using a plurality of information processing devices, or may employ a modified example that does not include some of the components illustrated in Fig. 7.

[0049] The configurations of the operation input unit 410, the screen display unit 420, and the communication I / F unit 430 may be similar to those of the operation input unit 310, the screen display unit 320, and the communication I / F unit 330 described in the first embodiment.

[0050] The storage unit 440 is a storage device such as a hard disk or memory. The storage unit 440 stores processing information and programs 445 required for various processes in the arithmetic processing unit 450. The programs 445 are read into the arithmetic processing unit 450 and executed to realize various processing units. The programs 445 are read in advance from an external device or recording medium via a data input / output function such as the communication I / F unit 430, and are stored in the storage unit 440. Main information stored in the storage unit 440 includes, for example, path information 441, path-sensor correspondence information 442, sensor-model correspondence information 443, and sensor data information 444.

[0051] The path information 441 may be the same as the path information 341 described in the first embodiment.

[0052] The path-sensor correspondence information 442 indicates information about sensors present on the communication path. In other words, the path-sensor correspondence information 442 includes information about sensors possessed by each object 210 through which the communication path passes. For example, the path-sensor correspondence information 442 is acquired in advance by inputting it using the operation input unit 310 or by acquiring it from an external device via the communication I / F unit 430, and is stored in the storage unit 440.

[0053] 8 shows an example of the path-sensor correspondence information 442. Referring to FIG. 8, the path-sensor correspondence information 442 associates a path ID with sensor information. Here, the path ID indicates identification information of a communication path. The path ID may be the same as the path information 441. Furthermore, the sensor information indicates identification information of a sensor present on the communication path. The sensor information may include information indicating the type of data to be acquired, information for identifying an object 210 having a sensor or a port from which data is to be acquired, information indicating the connection relationship between each object 210, and any other information.

[0054] The sensor-model correspondence information 443 indicates, for each sensor, information on a model used when performing anomaly detection, etc. For example, the sensor-model correspondence information 443 is acquired in advance by inputting it using the operation input unit 410, acquiring it from an external device via the communication I / F unit 430, or the like, and is stored in the storage unit 440.

[0055] 9 shows an example of the sensor-model correspondence information 443. Referring to FIG. 9, in the sensor-model correspondence information 443, sensor information and a model ID are associated with each other. Here, the sensor information indicates the identification information of the sensor, etc. The sensor information may be the same as the path-sensor correspondence information 442. Furthermore, the model ID indicates the identification information of the model. The model identification information may be assigned arbitrarily for each model.

[0056] The sensor data information 444 includes data acquired by the sensors. In the sensor data information 444, the data may be associated with the time at which the data was acquired. The sensor data information 444 may also include time-series data. For example, the sensor data information 444 includes data acquired by each sensor indicated by the path-sensor correspondence information 442 and the sensor-model correspondence information 443. The sensor data information 444 is updated in response to the acquisition of data from each sensor by a sensor data acquisition unit 454 (described later), etc.

[0057] The arithmetic processing unit 450 has an arithmetic device such as a CPU and its peripheral circuits. The arithmetic processing unit 450 reads and executes a program 445 from the storage unit 440, thereby causing the above hardware and the program 445 to work together to realize various processing units. Major processing units realized by the arithmetic processing unit 450 include, for example, a configuration change information acquisition unit 451, a path identification unit 452, a sensor identification unit 453, a sensor data acquisition unit 454, an influence calculation unit 455, a model identification unit 456, and an output unit 457. Note that, like the arithmetic processing unit 350 described in the first embodiment, the arithmetic processing unit 450 may have a GPU or the like instead of the above-mentioned CPU.

[0058] The configuration change information acquisition unit 451 and the path identification unit 452 may be similar to the configuration change information acquisition unit 351 and the path identification unit 352 described in the first embodiment.

[0059] The sensor identification unit 453 identifies sensors that may be affected by a change in the network configuration based on the results identified by the path identification unit 452. For example, the sensor identification unit 453 refers to the path-sensor correspondence information 442 to identify sensors associated with the communication paths identified by the path identification unit 352, thereby identifying each sensor present on the communication paths identified by the path identification unit 352 as sensors that may be affected by a change in the network configuration.

[0060] The sensor data acquisition unit 454 acquires data acquired by each sensor included in the optical network 200. For example, the sensor data acquisition unit 454 can acquire data from each sensor along with information indicating the time the sensor data was acquired. The sensor data acquisition unit 454 also stores the acquired data in the storage unit 440 as sensor data information 444.

[0061] The impact calculation unit 455 calculates an impact, which is a value according to a change in the trend of data acquired by the sensors before and after a change in the network configuration. In the case of the present disclosure, the impact calculation unit 455 calculates an impact for the sensors identified by the sensor identification unit 453 among the sensors included in the optical network 200. In other words, it can be said that the impact calculation unit 455 calculates an impact for some of the sensors included in the optical network 200.

[0062] For example, the impact calculation unit 455 refers to the sensor data information 444 to identify data corresponding to the sensor identified by the sensor identification unit 453. Furthermore, the impact calculation unit 455 distinguishes, among the identified data, data acquired before the change in network configuration from data acquired after the change, based on, for example, the time when the change in network configuration occurred. Then, the impact calculation unit 455 calculates the impact caused by the change in network configuration based on the data acquired before the change in network configuration and the data acquired after the change. For example, the impact calculation unit 455 can calculate the impact by calculating the difference or ratio between the statistical value of the data acquired before the change in network configuration and the statistical value of the data acquired after the change. Here, the statistical value may be any value, such as the average or median. The impact calculation unit 455 may also calculate the impact using a method other than the above examples.

[0063] The model identification unit 456 identifies sensors that will be affected by the network configuration change based on the results calculated by the impact calculation unit 455. Thereafter, the model identification unit 456 identifies models that may be affected by the network configuration change based on the results of identifying the sensors.

[0064] For example, the model identification unit 456 identifies sensors that will be affected by the network configuration change based on the impact calculated by the impact calculation unit 455 and a predetermined threshold. For example, the model identification unit 456 can identify sensors whose impact is equal to or greater than a predetermined threshold as affected sensors. The model identification unit 456 then identifies models corresponding to the affected sensors by referencing the sensor-model correspondence information 443. For example, as described above, the model identification unit 456 identifies models corresponding to the sensors identified based on the impact as models that may be affected by the network configuration change.

[0065] The output unit 457 outputs information about the model identified by the model identification unit 456. The configuration of the output unit 457 may be similar to that of the output unit 354 described in the first embodiment.

[0066] The above is an example of the configuration of the output device 400. Note that the output device 400 may be configured to manage models, similar to the output device 300. In this case, the output device 400 may be configured to update the model in accordance with the output of the output unit 457. As in the first embodiment, the present disclosure does not particularly limit the specific processing when updating the model. Next, an example of the operation of the output device 400 will be described with reference to FIG. 10 .

[0067] 10 is a flowchart showing an example of the operation of the output device 400. Referring to Fig. 10, the configuration change information acquisition unit 451 acquires configuration change information indicating a change in the network configuration in the optical network 200, such as an addition or reduction of a communication path (step S201). For example, the configuration change information acquisition unit 451 can acquire the configuration change information by referring to network configuration information managed on the NMS.

[0068] The path identification unit 452 identifies communication paths that may be affected by the network configuration change based on the configuration change information acquired by the configuration change information acquisition unit 451 (step S202). For example, the path identification unit 452 compares the configuration change information with the path information 441 to identify communication paths that may be affected by the network configuration change.

[0069] The sensor identification unit 453 identifies sensors that may be affected by the change in network configuration based on the results identified by the path identification unit 452 (step S203). For example, the sensor identification unit 453 can identify sensors that may be affected by the change in network configuration by referring to the path-sensor correspondence information 442 and identifying sensors associated with the communication paths identified by the path identification unit 352.

[0070] The influence calculation unit 455 calculates the influence, which is a value according to the trend change of the data acquired by the sensor before and after the change in the network configuration (step S204). For example, the influence calculation unit 455 calculates the influence of the sensor identified by the sensor identification unit 453 among the sensors included in the optical network 200.

[0071] The model identification unit 456 identifies sensors affected by the network configuration change based on the results calculated by the impact calculation unit 455 (step S205). If the impact satisfies a predetermined condition, such as being equal to or greater than a predetermined threshold (step S205, Yes), the model identification unit 456 identifies a sensor that meets the condition and identifies a model corresponding to the identified sensor (step S206). Furthermore, the output unit 457 outputs information about the model identified by the model identification unit 456 (step S207). On the other hand, if the impact does not satisfy a predetermined condition, such as being equal to or greater than a predetermined threshold, for all of the calculated impacts, the processes of steps S206 and S207 are omitted.

[0072] The above is an example of the operation of the output device 400. The output device 400 may update the model based on the output information.

[0073] As described above, the output device 400 includes a path identification unit 352, a sensor identification unit 453, an influence calculation unit 455, and a model identification unit 456. With this configuration, the sensor identification unit 453 can identify sensors that may be affected by a network configuration change based on the results identified by the path identification unit 452. Furthermore, the influence calculation unit 455 can calculate the influence of the sensors identified by the sensor identification unit 453. As a result, the model identification unit 456 can identify sensors that will be affected by a network configuration change based on the results calculated by the influence calculation unit 455, and can also identify models that will be affected by the network configuration change based on the identified results. This allows for model identification without checking the trend changes of all sensors before and after an event such as a network configuration change, even if a model is trained on a sensor-by-sensor basis. In other words, with the above configuration, the cost of identifying models that require updating can be reduced.

[0074] In this embodiment, the sensor identification unit 453 identifies sensors that may be affected by a network configuration change by referring to the path-sensor correspondence information 442 and identifying sensors associated with the communication path identified by the path identification unit 352. That is, in the above example, the path identification unit 452 identifies all sensors on the identified communication path as sensors that may be affected. However, sensors located upstream of the overlapping object 210 or fiber cable may not necessarily be significantly affected by a network configuration change. Therefore, the sensor identification unit 453 may be configured not to identify sensors located upstream of the overlapping object 210 or fiber cable that can be identified by the path identification unit 452 as sensors that may be affected. In other words, the sensor identification unit 453 may be configured to identify sensors located downstream of the overlapping object 210 or fiber cable as sensors that may be affected. The sensor identification unit 453 may be configured to determine whether to identify each sensor located upstream of the overlapping object 210 or fiber cable as a sensor that may be affected, based on the results of checking the data included in the sensor data information 444. The sensor identification unit 453 may be configured to determine whether each sensor located upstream of an overlapping object 210 or fiber cable is a sensor that may be affected, depending on whether an amplification instruction or the like is being issued in the upstream direction. Optical communication is performed from upstream to downstream. The determination of upstream or downstream may be made based on information indicating the connection relationship of the objects 210. For example, as described above, the sensor identification unit 453 may identify sensors that may be affected by a change in the network configuration based on the results identified by the path identification unit 452 and the positional relationship of the sensors (or objects having the sensors).

[0075] Furthermore, in this embodiment, an example has been given of identifying sensors affected by a network configuration change based on the calculation results of the impact level. However, the output device 400 may identify sensors affected by a network configuration change using any data drift detection technology other than the method of calculating the impact level. In other words, when determining whether the status of data acquired by a sensor satisfies a predetermined condition, the output device 400 may make the determination based on any criterion other than whether the impact level is equal to or greater than a threshold. Furthermore, the output device 400 may employ a modification similar to that of the first embodiment.

[0076] [Third Embodiment] Next, a third embodiment of the present disclosure will be described with reference to Fig. 11 and Fig. 12. Fig. 11 is a block diagram showing an example of the configuration of an output device 500. Fig. 12 is a flowchart showing an example of the operation of the output device 500.

[0077] In a third embodiment of the present disclosure, an output device 500 will be described, which is a modified example of the output device 300 described in the first embodiment and the output device 400 described in the second embodiment. In the second embodiment, the output device 400 will be described, which identifies sensors based on the results of identifying communication paths and identifies models based on the identification results. In a third embodiment, the output device 500 will be described, which tentatively identifies models that may be affected based on the results of identifying communication paths, and then calculates the influence of each sensor corresponding to the tentatively identified models, thereby identifying models that may be affected by a network configuration change. As will be described later, the output device 500 of the present disclosure tentatively identifies models that may be affected by a network configuration change based on the results of identifying communication paths. Furthermore, the output device 500 identifies sensors corresponding to the tentatively identified models and calculates the influence of the identified sensors, as will be described later, to narrow down the identified sensors to sensors that are actually affected. Then, based on the narrowed-down results, the output device 500 identifies models that may actually be affected by the network configuration change from the tentatively identified models. In this way, the output device 500 of the present disclosure tentatively identifies the affected models based on the results of identifying the communication paths, and then identifies the models that may be affected by the network configuration change by checking the corresponding impact level.

[0078] Fig. 11 shows an example of the configuration of the output device 500. Referring to Fig. 11, the output device 500 has, as main components, for example, an operation input unit 510, a screen display unit 520, a communication I / F unit 530, a storage unit 540, and an arithmetic processing unit 550. Note that, like the output device 300 and the output device 400, the output device 500 may be realized using a plurality of information processing devices, or may employ a modified example that does not include some of the components illustrated in Fig. 11.

[0079] The configurations of the operation input unit 510, the screen display unit 520, and the communication I / F unit 530 may be similar to those of the operation input unit 310, the screen display unit 320, and the communication I / F unit 330 described in the first embodiment.

[0080] The storage unit 540 is a storage device such as a hard disk or memory. The storage unit 540 stores processing information and programs 545 required for various processes in the arithmetic processing unit 550. The programs 545 are read into the arithmetic processing unit 550 and executed to realize various processing units. The programs 545 are read in advance from an external device or recording medium via a data input / output function such as the communication I / F unit 530, and are stored in the storage unit 540. Main information stored in the storage unit 540 includes, for example, path information 541, path-model correspondence information 542, sensor-model correspondence information 543, and sensor data information 544.

[0081] The path information 541 and the path-model correspondence information 542 may be similar to the path information 341 and the path-model correspondence information 342 described in the first embodiment. The sensor-model correspondence information 543 and the sensor data information 544 may be similar to the sensor-model correspondence information 443 and the sensor data information 444 described in the second embodiment.

[0082] The arithmetic processing unit 550 includes an arithmetic device such as a CPU and its peripheral circuits. The arithmetic processing unit 550 reads and executes a program 545 from the storage unit 540, thereby causing the hardware and the program 545 to work together to implement various processing units. Major processing units implemented by the arithmetic processing unit 550 include, for example, a configuration change information acquisition unit 551, a path identification unit 552, a model provisional identification unit 553, a sensor data acquisition unit 554, an influence calculation unit 555, a model identification unit 556, and an output unit 557. Note that the arithmetic processing unit 550 may include a GPU or the like instead of the CPU, similar to the arithmetic processing unit 350 described in the first embodiment.

[0083] The configuration change information acquisition unit 551 and the path identification unit 552 may be similar to the configuration change information acquisition unit 351 and the path identification unit 352 described in the first embodiment.

[0084] The model tentative identification unit 553 tentatively identifies a model that may be affected by a change in the network configuration based on the results identified by the path identification unit 552. For example, the model tentative identification unit 553 can tentatively identify a model that may be affected by a change in the network configuration by referring to the path-model correspondence information 542 and identifying a model that is associated with the communication path identified by the path identification unit 552.

[0085] The sensor data acquisition unit 554 may be similar to the sensor data acquisition unit 454 described in the second embodiment.

[0086] The influence calculation unit 555 identifies a sensor corresponding to the model tentatively identified by the model tentative identification unit 553. Then, the influence calculation unit 555 calculates the influence of the identified sensor. The influence calculation unit 555 may calculate the influence in the same manner as the influence calculation unit 455 described in the second embodiment.

[0087] For example, the influence calculation unit 555 refers to the sensor-model correspondence information 543 to identify a sensor corresponding to the model tentatively identified by the model tentative identification unit 553. Then, the influence calculation unit 555 can calculate the influence of the identified sensor.

[0088] The model identification unit 556 identifies sensors that will be affected by the network configuration change based on the results calculated by the impact calculation unit 455. Thereafter, the model identification unit 456 identifies models that may be affected by the network configuration change from among the provisionally identified models based on the results of identifying the sensors.

[0089] For example, the model identification unit 556 identifies sensors that will be affected by the network configuration change based on the impact calculated by the impact calculation unit 555 and a predetermined threshold. For example, the model identification unit 556 can identify sensors whose impact is equal to or greater than a predetermined threshold as affected sensors. The model identification unit 556 then identifies models corresponding to the affected sensors by referring to the sensor-model correspondence information 543. For example, as described above, the model identification unit 556 identifies models that correspond to sensors identified based on the impact as models that may be affected by the network configuration change, thereby identifying models that will be affected by the network configuration change from among the provisionally identified models.

[0090] The output unit 557 outputs information about the model identified by the model identification unit 556. The configuration of the output unit 557 may be similar to that of the output unit 354 described in the first embodiment.

[0091] The above is an example of the configuration of the output device 500. Note that the output device 500 may be configured to manage models, similar to the output devices 300 and 400. In this case, the output device 500 may be configured to update the model in accordance with the output of the output unit 557. As in the first embodiment, the present disclosure does not particularly limit the specific processing when updating the model. Next, an example of the operation of the output device 500 will be described with reference to FIG. 12 .

[0092] 12 is a flowchart showing an example of the operation of the output device 500. Referring to Fig. 12, the configuration change information acquisition unit 551 acquires configuration change information indicating a change in the network configuration in the optical network 200, such as an addition or reduction of a communication path (step S301). For example, the configuration change information acquisition unit 551 can acquire the configuration change information by referring to network configuration information managed on the NMS.

[0093] The path identification unit 552 identifies communication paths that may be affected by the network configuration change based on the configuration change information acquired by the configuration change information acquisition unit 551 (step S302). For example, the path identification unit 552 compares the configuration change information with the path information 541 to identify communication paths that may be affected by the network configuration change.

[0094] The model tentative identification unit 553 tentatively identifies a model that may be affected by the network configuration change based on the results identified by the path identification unit 552 (step S303). For example, the model tentative identification unit 553 can tentatively identify a model that may be affected by the network configuration change by referring to the path-model correspondence information 542.

[0095] The influence calculation unit 555 identifies a sensor corresponding to the model tentatively identified by the model tentative identification unit 553. Then, the influence calculation unit 555 calculates the influence of the identified sensor (step S304). The influence calculation unit 555 may calculate the influence in the same manner as the influence calculation unit 455 described in the second embodiment.

[0096] The model identification unit 556 identifies sensors affected by the network configuration change based on the results calculated by the impact calculation unit 555 (step S305). If the impact satisfies a predetermined condition, such as being equal to or greater than a predetermined threshold (step S305, Yes), the model identification unit 556 identifies sensors according to the conditions and identifies models corresponding to the identified sensors (step S306). This allows the model identification unit 556 to identify affected models from among the provisionally identified models. Furthermore, the output unit 557 outputs information about the models identified by the model identification unit 556 (step S307). On the other hand, if the impact does not satisfy a predetermined condition, such as being equal to or greater than a predetermined threshold, for all of the calculated impacts, the processes of steps S306 and S307 are omitted.

[0097] The above is an example of the operation of the output device 500. The output device 500 may update the model based on the output information. The output device 500 may also employ a modified example similar to the output device 300 or the output device 400.

[0098] In this way, the output device 500 can identify models that may be affected by a change in network configuration by provisionally identifying models that may be affected based on the results of identifying the communication paths, and then calculating the degree of impact for each sensor corresponding to the provisionally identified models. Even with this configuration, it is possible to reduce the cost of identifying models that require updating.

[0099] [Fourth Embodiment] Next, a fourth embodiment of the present disclosure will be described with reference to Fig. 13 and Fig. 14. Fig. 13 is a diagram showing an example of the hardware configuration of an output device 600. Fig. 14 is a block diagram showing an example of the configuration of the output device 600.

[0100] In the fourth embodiment of the present disclosure, a configuration example of an output device 600 that outputs information about a model that is affected by a change in network configuration, such as adding or removing a communication path, will be described. Fig. 13 shows an example of the hardware configuration of the output device 600. Referring to Fig. 13, the output device 600 has the following hardware configuration, for example: - CPU (Central Processing Unit) 601 (arithmetic unit) - ROM (Read Only Memory) 602 (storage device) - RAM (Random Access Memory) 603 (storage device) - Programs 604 loaded into the RAM 603 - Storage device 605 that stores the programs 604 - Drive device 606 that reads and writes from / to a storage medium 610 outside the information processing device - Communication interface 607 that connects to a communication network 611 outside the information processing device - Input / output interface 608 that inputs and outputs data - Bus 609 that connects the various components

[0101] 14 . The program group 604 is stored in advance in, for example, the storage device 605 or the ROM 602, and is loaded into, for example, the RAM 603 and executed by the CPU 601. The program group 604 may be supplied to the CPU 601 via the communication network 611, or may be stored in advance in the recording medium 610, and the drive device 606 may read out the program and supply it to the CPU 601.

[0102] 13 shows an example of the hardware configuration of the output device 600. The hardware configuration of the output device 600 is not limited to the above-described case. For example, the output device 600 may be configured with only a part of the above-described configuration, such as excluding the drive device 606. Furthermore, the CPU 601 may be the GPU exemplified in the first embodiment.

[0103] The acquisition unit 621 acquires configuration change information indicating a change in the network configuration in the optical network. For example, the acquisition unit 621 acquires configuration change information indicating that a communication path has been added or removed.

[0104] The path identification unit 622 identifies, from among the already set communication paths, a communication path having a route that overlaps with the communication path whose configuration has been changed, based on the configuration change information.

[0105] The model identification unit 623 identifies a trained model that will be affected by the configuration change based on the communication path identified by the path identification unit 622.

[0106] The output unit 624 outputs information about the model identified by the model identifying unit 623 .

[0107] As described above, the output device 600 includes a path identification unit 622, a model identification unit 623, and an output unit 624. With this configuration, the model identification unit 623 can identify a trained model that will be affected by a configuration change, based on the communication path identified by the path identification unit 622. As a result, the output unit 624 can output information about the model identified by the model identification unit 623. This makes it possible to update the model output by the output unit 624, thereby reducing the cost involved in identifying a model that requires updating.

[0108] The output device 600 described above can be realized by incorporating a predetermined program into an information processing device such as the output device 600. Specifically, a program according to another aspect of the present invention is a program for causing an information processing device such as the output device 600 to realize processing of acquiring configuration change information indicating a change in the network configuration in an optical network, identifying, based on the configuration change information, communication paths among already set communication paths that have a route that overlaps with a communication path whose configuration has been changed, based on the identified communication paths, identifying a trained model that is affected by the configuration change, and outputting information about the identified model.

[0109] Furthermore, an output method executed by an information processing device such as the output device 600 described above is a method in which the information processing device such as the output device 600 acquires configuration change information indicating a change in the network configuration in an optical network, identifies, from among the set communication paths, a communication path that has a route that overlaps with the communication path whose configuration has been changed based on the configuration change information, identifies a trained model that is affected by the configuration change based on the identified communication path, and outputs information about the identified model.

[0110] Even if the invention is a program having the above-described configuration, or a computer-readable recording medium having the program recorded thereon, or an output method, it can achieve the same functions and effects as the above-described output device 600, and therefore can achieve the above-described objective of the present disclosure.

[0111] <Supplementary Notes> Part or all of the above-described embodiments can be described as follows: An outline of the output device and the like according to the present invention will be described below. However, the present invention is not limited to the following configuration.

[0112] (Supplementary Note 1) An output device comprising: an acquisition unit that acquires configuration change information indicating a change in network configuration in an optical network; a path identification unit that identifies, based on the configuration change information, communication paths among set communication paths that have a route overlapping with a communication path whose configuration has been changed; a model identification unit that identifies a trained model that will be affected by the configuration change based on the communication paths identified by the path identification unit; and an output unit that outputs information on the model identified by the model identification unit. (Supplementary Note 2) The output device according to Supplementary Note 1, wherein the model identification unit identifies the model corresponding to the communication paths identified by the path identification unit by referring to pre-stored path / model information that indicates information on trained models for each communication path. (Supplementary Note 3) The output device according to Supplementary Note 1, comprising: a sensor identification unit that identifies a sensor present on the communication path identified by the path identification unit by referring to pre-stored information, and the model identification unit identifies a model based on the results identified by the sensor identification unit. (Supplementary Note 4) The output device according to Supplementary Note 3, wherein the model identification unit further identifies sensors from the sensors identified by the sensor identification unit, the status of data acquired by which satisfies a predetermined condition, and identifies a model based on the identification results. (Supplementary Note 5) The output device according to Supplementary Note 3 or Supplementary Note 4, wherein the output device includes a calculation unit that calculates, for the sensors identified by the sensor identification unit, an impact level that indicates an impact of a change in network configuration on data acquired by the sensors, and the model identification unit further identifies sensors from the sensors identified by the sensor identification unit based on the impact level calculated by the calculation unit, and identifies a model based on the identification results. (Supplementary Note 6) The output device according to any one of Supplementary Notes 3 to 5, wherein the sensor identification unit identifies a sensor based on the identification results of the path identification unit and a positional relationship of the sensors.(Supplementary Note 7) The output device according to any one of Supplementary Note 3 to Supplementary Note 6, wherein the path identification unit is configured to identify a communication path having an overlapping route by identifying an overlapping object in a set communication path and a communication path whose configuration has been changed, and the sensor identification unit identifies a sensor located downstream of the object identified by the path identification unit as a sensor present on the communication path identified by the path identification unit. (Supplementary Note 8) The output device according to Supplementary Note 5, wherein the calculation unit calculates the degree of impact based on data acquired by a sensor before a network configuration change and data acquired by a sensor after a network configuration change. (Supplementary Note 9) The output device according to Supplementary Note 1, further comprising a model provisional identification unit that provisionally identifies a trained model affected by a configuration change based on the communication path identified by the path identification unit, and the model identification unit identifies a sensor corresponding to the model provisionally identified by the model provisional identification unit and identifies a model based on the identification result. (Supplementary Note 10) The output device according to Supplementary Note 9, wherein the model identification unit further identifies sensors from the identified sensors whose acquired data status satisfies a predetermined condition, and identifies a model based on the identification result. (Supplementary Note 11) The output device according to Supplementary Note 9 or Supplementary Note 10, wherein the model identification unit further identifies sensors corresponding to the model tentatively identified by the model tentative identification unit, and calculates, for the identified sensors, an impact degree indicating an impact of a change in network configuration on data acquired by the sensors, and the model identification unit further identifies sensors from the sensors identified by the calculation unit based on the impact degree calculated by the calculation unit, and identifies a model from the models tentatively identified by the model tentative identification unit based on the identification result. (Supplementary Note 12) The output device according to any one of Supplementary Notes 1 to 11, wherein the acquisition unit acquires configuration change information indicating that a communication path has been added or removed.(Supplementary Note 13) An output method in which an information processing device acquires configuration change information indicating a change in network configuration in an optical network, identifies, from already set communication paths, a communication path having a route overlapping with a communication path whose configuration has been changed, based on the configuration change information, identifies a trained model that will be affected by the configuration change, based on the identified communication path, and outputs information about the identified model. (Supplementary Note 14) A computer-readable recording medium having recorded thereon a program for implementing processes in an information processing device: acquires configuration change information indicating a change in network configuration in an optical network, identifies, from already set communication paths, a communication path having a route overlapping with a communication path whose configuration has been changed, based on the identified communication path.

[0113] The programs described in the above embodiments and appendices may be stored in a storage device or a computer-readable recording medium, such as a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0114] Although the present invention has been described above with reference to the above-mentioned embodiments, the present invention is not limited to the above-mentioned embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0115] 100 Optical communication system 200 Optical network 210 Object 300 Output device 310 Operation input unit 320 Screen display unit 330 Communication I / F unit 340 Storage unit 341 Path information 342 Path-model correspondence information 343 Program 350 Arithmetic processing unit 351 Configuration change information acquisition unit 352 Path identification unit 353 Model identification unit 354 Output unit 400 Output device 410 Operation input unit 420 Screen display unit 430 Communication I / F unit 440 Storage unit 441 Path information 442 Path-sensor correspondence information 443 Sensor-model correspondence information 444 Sensor data information 445 Program 450 Arithmetic processing unit 451 Configuration change information acquisition unit 452 Path identification unit 453 Sensor identification unit 454 Sensor data acquisition unit 455 Impact calculation unit 456 Model identification unit 457 Output unit 500 Output device 510 Operation input unit 520 Screen display unit 530 Communication I / F unit 540 Storage unit 541 Path information 542 Path-model correspondence information 543 Sensor-model correspondence information 544 Sensor data information 550 Arithmetic processing unit 551 Configuration change information acquisition unit 552 Path identification unit 553 Model tentative identification unit 554 Sensor data acquisition unit 555 Impact calculation unit 556 Model identification unit 557 Output unit 600 Output device 601 CPU 602 ROM 603 RAM 604 Program group 605 Storage device 606 Drive device 607 Communication interface 608 Input / output interface 609 Bus 610 Recording medium 611 Communication network 621 Acquisition unit 622 Path identification unit 623 Model identification unit 624 Output unit

Claims

1. An output device having: an acquisition unit that acquires configuration change information indicating a change in network configuration in an optical network; a path identification unit that identifies, based on the configuration change information, communication paths among already-set communication paths that have a route that overlaps with a communication path whose configuration has been changed; a model identification unit that identifies a learned model that will be affected by the configuration change based on the communication path identified by the path identification unit; and an output unit that outputs information about the model identified by the model identification unit.

2. An output device according to claim 1, wherein the model identification unit identifies the model corresponding to the communication path identified by the path identification unit by referring to path / model information indicating pre-stored, learned model information for each communication path.

3. An output device as claimed in claim 1, further comprising a sensor identification unit which identifies sensors present on a communication path identified by said path identification unit by referring to pre-stored information, and wherein said model identification unit identifies a model based on the results identified by said sensor identification unit.

4. An output device according to claim 3, wherein the model identification unit further identifies sensors from among the sensors identified by the sensor identification unit, the conditions of the data acquired from the sensors satisfying predetermined conditions, and identifies a model based on the identification results.

5. An output device as described in claim 3, further comprising a calculation unit that calculates an impact level indicating an impact of a change in network configuration on data acquired by a sensor identified by the sensor identification unit, and the model identification unit further identifies sensors from among the sensors identified by the sensor identification unit based on the impact level calculated by the calculation unit, and identifies a model based on the identification results.

6. An output device according to claim 3, wherein the sensor identification unit identifies a sensor based on the result identified by the path identification unit and the positional relationship of the sensor.

7. An output device as described in claim 3, wherein the path identification unit is configured to identify a communication path having an overlapping route by identifying an overlapping object in a set communication path and a reconfigured communication path, and the sensor identification unit identifies a sensor located downstream of the object identified by the path identification unit as a sensor existing on the communication path identified by the path identification unit.

8. An output device according to claim 5, wherein the calculation unit calculates the degree of influence based on data acquired by a sensor before a change in network configuration and data acquired by the sensor after a change in network configuration.

9. An output device as described in claim 1, further comprising a model tentative identification unit that tentatively identifies a trained model that is affected by a configuration change based on the communication path identified by the path identification unit, and the model identification unit identifies a sensor corresponding to the model tentatively identified by the model tentative identification unit, and identifies a model based on the identification result.

10. An output device according to claim 9, wherein the model identification unit further identifies sensors from among the identified sensors, the state of data acquired from the sensors satisfying a predetermined condition, and identifies a model based on the identification results.

11. An output device as described in claim 9, further comprising a calculation unit that identifies a sensor corresponding to a model tentatively identified by the model tentative identification unit, and calculates an impact level for the identified sensor that indicates an impact of a change in network configuration on data acquired by the sensor, wherein the model identification unit further identifies a sensor from among the sensors identified by the calculation unit based on the impact level calculated by the calculation unit, and identifies a model from among the models tentatively identified by the model tentative identification unit based on the identification result.

12. An output device according to claim 1, wherein the acquisition unit acquires configuration change information indicating that a communication path has been added or removed.

13. An output method in which an information processing device acquires configuration change information indicating a change in network configuration in an optical network, identifies, based on the configuration change information, communication paths among already-set communication paths that have a route that overlaps with a communication path whose configuration has been changed, identifies a learned model that is affected by the configuration change based on the identified communication path, and outputs information about the identified model.

14. A computer-readable recording medium having recorded thereon a program for causing an information processing device to perform the following processes: acquire configuration change information indicating a change in network configuration in an optical network; identify, based on the configuration change information, communication paths among already-set communication paths that have a route that overlaps with a communication path whose configuration has been changed; identify, based on the identified communication paths, a learned model that is affected by the configuration change; and output information about the identified model.

Citation Information

Patent Citations

  • Communication device and communication method

    JP2020088402A

  • Machine learning system and method of managing machine learning model by machine learning system

    JP2022175959A

  • Information system, control apparatus, method of controlling virtual network, and program

    WO2011118585A1