Communication service management device, method, and program

Abstract

According to one aspect of the present invention, when a plurality of communication services provided by a communication network formed from a plurality of communication devices and a transmission line connecting the communication devices are managed, a communication resource corresponding to each of the plurality of communication services is managed in association with the communication service for each communication service, and past measurement data related to the communication resource in a state in which each communication service is stable and in an unstable state is acquired for each communication service. Then, for each communication service, a learning model is trained using the acquired past measurement data as training data, thereby creating a state determination model for outputting a determination result of the state of the communication service as a response variable when current measurement data related to the communication resource is input as an explanatory variable.

Description

Communication service management device, method, and program

[0001] One aspect of the present invention relates to a communication service management device, method, and program having a function of determining the status of a communication service provided by a communication network.

[0002] A communication network is composed of multiple communication devices and transmission paths connecting these communication devices. The configuration of the communication network is represented by topology information in which communication devices are represented as points (also called nodes) and transmission paths are represented as edges, and communication resources are assigned to each edge. The communication resources include multiple elements such as the number of paths, bandwidth, and service class.

[0003] Generally, communication networks have redundancy. For example, if a failure occurs in a communication device or transmission path due to an earthquake or other disaster, the communication network identifies the location of the failure and switches the communication path to bypass the failure in order to maintain the communication service currently being provided. However, when the communication path is switched, the state of the communication service may become unstable depending on the communication resources of the switched communication path. Here, an unstable communication service refers to a state in which it is possible to transmit information data, but it is difficult to achieve the same quality as under normal circumstances.

[0004] Therefore, as one of the maintenance tasks for communication services, a method has been proposed for visualizing unstable points as well as fault points for each communication service. For example, Patent Document 1 proposes a method for measuring the quality of a communication path for each communication service, determining whether the QoS of the service is satisfied, and displaying the results. Furthermore, Patent Document 2 describes a model in which information about users is stored in association with an information object related to the logical layer, and information about subscribers associated with the acquired information object related to the fault location is identified as information about the subscriber affected by the fault.

[0005] Japanese Patent Publication No. 2019-145873 Japanese Patent No. 6655524

[0006] However, in conventional methods for visualizing unstable points, system administrators are required to manually determine the threshold for determining whether a communication service is stable or unstable for each communication service. Generally, the communication resource elements and their values ​​assigned to each edge are not identical. Therefore, determining an accurate threshold for each communication service requires identifying the communication resource corresponding to the communication service from a vast number of possible combinations of communication resources. However, conventional methods manage information representing the nodes and edges of each communication service, but do not manage the communication resource elements. Therefore, determining the judgment threshold for a communication service requires the skill and experience of a system administrator, and also requires a significant amount of work time. Furthermore, the determined thresholds are prone to variance, which can make it difficult to accurately determine the status of a communication service.

[0007] This invention has been made in light of the above circumstances, and aims to provide a technology that makes it possible to accurately determine the status of a communication service when switching communication paths in a short amount of time, without relying on the skills or experience of the user.

[0008] In order to solve the above problem, one aspect of a communication service management device or management method according to the present invention manages a plurality of communication services provided by a communication network consisting of a plurality of communication devices and transmission paths connecting these communication devices, by managing, for each of the communication services, information representing communication resources corresponding to the communication service in association with the communication service, and acquiring, for each communication service, past measurement data of the communication resources when the communication service is stable and when the communication service is unstable.Then, for each communication service, the acquired past measurement data is used as learning data to cause a learning model to learn, thereby creating a state determination model that outputs, as a response variable, a determination result of the current state of the communication service when current measurement data related to the communication resources is input as an explanatory variable.

[0009] According to one aspect of the present invention, when a communication path is switched due to a disaster, for example, the status of the communication service using the switched communication path can be determined quickly and accurately by using the status determination model, without relying on the skills or experience of the administrator.

[0010] In other words, according to one aspect of the present invention, it is possible to provide a technology that enables the status of a communication service when a communication path is switched to be accurately determined in a short time without relying on the user's skills or experience.

[0011] FIG. 1 is a diagram showing an example of the configuration of a communication network system according to an embodiment of the present invention. FIG. 2 is a block diagram showing an example of the hardware configuration of a communication service management device according to an embodiment of the present invention. FIG. 3 is a block diagram showing an example of the software configuration of a communication service management device according to an embodiment of the present invention. FIG. 4 is a flowchart showing an example of the processing procedure and processing content of a model learning process in a learning phase executed by a control unit of the communication service management device shown in FIG. 3. FIG. 5 is a flowchart showing an example of the processing procedure and processing content of a state determination process in an operation phase executed by a control unit of the communication service management device shown in FIG. 3. FIG. 6 is a diagram showing an example of communication paths of a communication network providing communication services A and B. FIG. 7 is a diagram showing an example of communication resource elements used by communication service A in the communication network shown in FIG. 6 and their measured values. FIG. 8 is a diagram showing an example of communication resource elements used by communication service B in the state of the communication network shown in FIG. 6 and their measured values. FIG. 9 is a diagram for explaining an example of the learning processing of a state determination model corresponding to communication service A. FIG. 10 is a diagram for explaining an example of the learning processing of a state determination model corresponding to communication service B. FIG. 11 is a diagram showing an example of the configuration of a communication path after switching when switching of a communication path is performed in the communication network shown in FIG. 6. Fig. 12 is a diagram showing an example of elements and measurement values ​​of communication resources used by communication service A after switching of the communication path shown in Fig. 11. Fig. 13 is a diagram showing an example of elements and measurement values ​​of communication resources used by communication service B after switching of the communication path shown in Fig. 11. Fig. 14 is a diagram showing an example of a determination result of the state of communication service A output from the state determination model when elements and measurement values ​​of communication resources used by communication service A after switching of the communication path shown in Fig. 12 are input to the trained state determination model. Fig. 15 is a diagram showing an example of a determination result of the state of communication service B output from the state determination model when measurement values ​​of communication resources used by communication service B after switching of the communication path shown in Fig. 13 are input to the trained state determination model.

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0013] [One Embodiment] (Configuration Example) (1) System FIG. 1 is a diagram showing an example of a communication network system according to one embodiment of the present invention.

[0014] In a communication network system according to one embodiment, a communication service management device CS is connected to a communication network NW that connects multiple communication devices R1 to R7 distributed across multiple service areas via transmission lines L12 to L67. Note that TM is an administrator terminal used by a system administrator, and is connected to the communication service management device CS directly or via the communication network NW.

[0015] The communication devices R1 to R7 are devices that make up communication facilities such as gateways and routers housed in a central office, and provide communication services to network users. Usually, one communication device R1 to R7 is housed in one central office, but there are also cases where multiple devices are housed in one central office.

[0016] (2) Communication Service Management Device CS The communication service management device CS is configured by, for example, a server computer located in a management center. Note that the communication service management device CS may also be configured by, for example, a personal computer.

[0017] 2 and 3 are block diagrams showing an example of the hardware and software configurations of the communication service management device CS according to an embodiment of the present invention.

[0018] The communication service management device CS includes a control unit 1 using a hardware processor such as a central processing unit (CPU). A storage unit having a program storage unit 2 and a data storage unit 3, and a communication interface unit (hereinafter referred to as a communication I / F unit) 4 are connected to the control unit 1 via a bus 5.

[0019] Under the control of the control unit 1, the communication I / F unit 4 uses a communication protocol defined by a communication network NW including, for example, the Internet, to send and receive various management data between the communication devices R1 to R7 and the administrator terminal TM.

[0020] The program storage unit 2 is configured by combining, for example, a non-volatile memory such as a hard disk drive (HDD) or a solid state drive (SSD) as a storage medium that can be written to and read at any time, and a non-volatile memory such as a read only memory (ROM), and stores application programs necessary for executing various control processes according to one embodiment of the present invention, in addition to middleware such as an operating system (OS).

[0021] The data storage unit 3 is, for example, a combination of a non-volatile memory such as an HDD or SSD as a storage medium that can be written to and read from at any time, and a volatile memory such as a RAM (Random Access Memory), and its storage area is provided with a communication resource information storage unit 31, a measurement data storage unit 32, and a state determination model storage unit 33.

[0022] The communication resource information storage unit 31 stores information representing a combination of communication resource elements managed for each communication service by the communication resource management processing unit 11 of the control unit 1, which will be described later, in association with the identification information of the communication service (hereinafter referred to as the communication service ID).

[0023] The measurement data storage unit 32 stores the measurement data of each element of the communication resource acquired for each communication service by the measurement data acquisition processing unit 12 of the control unit 1, which will be described later, in association with information representing the communication service ID and the measurement time.

[0024] The state determination model storage unit 33 stores the trained state determination model created for each communication service by the model learning processing unit 14 of the control unit 1 .

[0025] The control unit 1 includes, as processing units according to one embodiment of the present invention, a communication resource management processing unit 11, a measurement data acquisition processing unit 12, a communication resource information judgment processing unit 13, a model learning processing unit 14, a state judgment processing unit 15, and a judgment result output processing unit 16.

[0026] The processing units 11 to 16 are all realized by causing a hardware processor in the control unit 1 to execute an application program stored in the program storage unit 2. Note that some or all of the processing units 11 to 16 may be realized using hardware such as an LSI (Large Scale Integration) or an ASIC (Application Specific Integrated Circuit).

[0027] The communication resource management processing unit 11 acquires communication resource information representing a combination of communication resource elements predefined for each communication service, for example, from the administrator terminal TM via the communication I / F unit 4, and stores the acquired communication resource information in the communication resource information storage unit 31 in association with the communication service ID.

[0028] The measurement data acquisition processing unit 12 acquires, for each communication service, measurement data of each element of a communication resource corresponding to the communication service from the communication devices R1 to R7 on the communication path used for that communication service via the communication I / F unit 4. Then, the acquired measurement data is stored in the measurement data storage unit 32 in association with a communication service ID. Note that an example of the configuration of the communication resource elements and their measurement data will be illustrated in an operation example.

[0029] During the learning phase, when the communication resource information determination processing unit 13 receives a request to create a state determination model, for example, from the administrator terminal TM, it refers to the communication resource information stored in the communication resource information storage unit 31 and determines the communication resource elements corresponding to the requested communication service.

[0030] The model learning processing unit 14 reads, for each communication service, a set of previously acquired measurement data for each element of the communication resource determined by the communication resource information determination processing unit 13 from the measurement data storage unit 32. Then, the read set of past measurement data is used as learning data to cause the learning model to perform learning. At this time, the set of past measurement data includes measurement data when the communication service is in a stable state and measurement data when the communication service is in an unstable state, and label information indicating stability / instability is assigned to each measurement data. The model learning processing unit 14 uses the label information indicating stability / instability as training data. An example of the model learning process will be described in the operation example.

[0031] In the operation phase, the state determination processing unit 15 acquires, for each communication service, current measurement data on each element of the communication resource corresponding to the communication service from the communication devices R1 to R7 that configure the post-switching communication path used by the communication service via the communication I / F unit 4. Then, the state determination processing unit 15 inputs the acquired current measurement data into a trained state determination model corresponding to the communication service stored in the state determination model storage unit 33, and acquires a determination result of the current state of the communication service from this state determination model. An example of this state determination processing will also be described in the operation example.

[0032] The judgment result output processing unit 16 generates information for displaying the judgment result based on the judgment result of the current state of the communication service, and transmits the generated display information from the communication I / F unit 4 to the administrator terminal TM.

[0033] (Example of Operation) Next, an example of operation of the device configured as above will be described.

[0034] (1) Advance Preparation Information representing a combination of communication resource elements corresponding to a communication service is generated in advance for each communication service, for example, at a management center.

[0035] The control unit 1 of the communication service management device CS acquires information representing the combination of communication resource elements generated for each of the above communication services from the administrator terminal TM, and stores the acquired information representing the combination of communication resource elements in the communication resource information storage unit 31 in association with the communication service ID.

[0036] Furthermore, under the control of the measurement data acquisition processing unit 12, each time a communication service is used in the communication network NW, the control unit 1 of the communication service management device CS acquires measurement data relating to each element of the communication resource corresponding to the communication service from the communication devices R1 to R7 constituting the communication path used for the communication service via the communication I / F unit 4. Then, the acquired measurement data of the communication resource is stored in the measurement data storage unit 32 in association with the communication service ID.

[0037] As mentioned above, the measurement data is provided with label information indicating the state of the communication service as determined by, for example, a system administrator, i.e., whether the state is stable and capable of transmitting information data at the desired quality, or unstable and capable of transmitting information data but in which it is difficult to achieve transmission quality equivalent to that in stable conditions.

[0038] The set of measurement data stored in the measurement data storage unit 32 is used as learning data when creating a state determination model for each communication service.

[0039] (2) Learning Phase When the learning phase is set, the control unit 1 of the communication service management device CS executes the process of creating a state determination model for each communication service as follows.

[0040] (2-1) Overall Description of Learning Process FIG. 4 is a flowchart showing an example of the processing procedure and processing contents of the model learning process executed by the control unit 1 of the communication service management device CS in the learning phase.

[0041] That is, when a learning request is transmitted from, for example, the administrator terminal TM and the control unit 1 of the communication service management device CS detects this learning request in step S10, first, in step S11, it selects one communication service corresponding to the state determination model to be learned under the control of the communication resource information determination processing unit 13. Subsequently, in step S12, the communication resource information determination processing unit 13 determines the communication resource elements corresponding to the selected communication service by referring to the communication resource information stored in the communication resource information storage unit 31.

[0042] Next, in step S13, under the control of the model learning processing unit 14, the control unit 1 of the communication service management device CS reads past measurement data related to the elements of the communication resources from the measurement data storage unit 32, and uses the read past measurement data as learning data to cause the learning model to learn.

[0043] More specifically, each piece of measurement data is provided with label information indicating whether the communication service is in a stable state or an unstable state in addition to the measurement data itself. The model learning processing unit 14 inputs the measurement data itself as explanatory variables and the label information as training data into a learning model, and causes learning to occur. Note that a support vector machine (SVM), for example, is used as the learning model.

[0044] In step S15, the model learning processing unit 14 determines whether the difference between the judgment result output from the learning model and the teacher data has converged to less than a threshold value. If convergence has not yet occurred, the process returns to step S13, reads the next measurement data corresponding to the communication service, and in step S14 causes the learning model to learn using the measurement data. Thereafter, in the same manner, the model learning processing unit 14 repeatedly executes the process of sequentially reading past measurement data and causing the learning model to learn, in steps S13 and S14, until the judgment result output from the learning model converges.

[0045] In the above explanation, we have described the case where the learning model is trained using past measurement data that has been stored in advance, but it is also possible to train the learning model using measurement data that is subsequently obtained in real time.

[0046] On the other hand, when the judgment result of the learning model converges, the model learning processing unit 14 judges that the learning of the learning model corresponding to the communication service has been completed. Then, in step S16, the model learning processing unit 14 stores the learning model for which the learning has been completed, i.e., the trained state judgment model, in the state judgment model storage unit 33 in association with the communication service ID.

[0047] When the control unit 1 of the communication service management device CS has finished creating a state learning model corresponding to one communication service as described above, it determines in step S17 whether or not the creation of state determination models corresponding to all communication services for which creation is to be performed has been completed. If, as a result of this determination, there is a state determination model that has not yet been created, the process returns to step S11 to select the next communication service, and thereafter executes the series of state determination model creation processes described above in steps S12 to S17. Thereafter, the state determination model creation process is similarly repeated for each communication service.

[0048] When it is determined in step S17 that the creation process of state determination models corresponding to all communication services has been completed, the control unit 1 of the communication service management device CS ends the creation process of state determination models and returns to a standby state.

[0049] (2-2) Specific Example of Learning Process Next, a specific example of the state determination model creation process described above will be described. Here, as shown in FIG. 6, an example will be described in which two communication services, communication services A and B, are used. Each of communication services A and B is represented by a physical layer, a transmission layer, and an IP layer. The physical layer represents the connection configuration of communication devices used for the communication service. In the example of FIG. 6, communication service A uses a communication route (communication path) connecting communication devices RA-1, RA-2, RA-3, and RA-4, and communication service B uses a communication path connecting communication devices RB-1, RB-5, RB-6, and RB-7. Note that this example illustrates a case in which communication devices RA-2, RB-2, RA-3, RB-3, RA-5, RB5, RA-6, and RB6 are each accommodated in the same station building.

[0050] 7 shows an example of a combination of communication resource elements defined for the above-mentioned communication service A, and the measurement values ​​for each element. In this example, the communication resources defined for communication service A are represented by a combination of six types of elements: "number of paths," "bandwidth," "number of sessions," "number of accommodated lines," "service class," and "alarm," and the measurement values ​​for each element except for "service class" are obtained from the IP layer and the transmission layer, respectively.

[0051] 8 shows an example of a combination of communication resource elements defined for the above-mentioned communication service B and the measurement values ​​for each element. In this example, the communication resources defined for the communication service B are also expressed by six types of elements, namely, "number of paths," "bandwidth," "number of sessions," "number of accommodated lines," "service class," and "alarm," similar to the above-mentioned communication service A, and the measurement values ​​of each element except for "service class" are obtained from the IP layer and the transmission layer, respectively.

[0052] The communication service management device CS acquires past measurement data D when the state of the communication services is stable for the communication services A and B, as shown in, for example, FIGS. 9 and 10. Ａ１ , D Ｂ１ and past measurement data D when the communication service state is unstable. Ａ２ , DＢ２ are acquired and stored in the measurement data storage unit 32.

[0053] Then, the communication service management device CS acquires, for each of the communication services A and B, past measurement data D when the state of the communication services is stable as shown in FIGS. 9 and 10. Ａ１ , D Ｂ１ and the past measurement data D when unstable Ａ２ , D Ｂ２ The learning model is trained using

[0054] The learning model is based on the input measurement data D Ａ１ , D Ｂ１ and D Ａ２ , D Ｂ２ Then, the communication resources are plotted in n-dimensional space based on the measurement data D Ａ１ , D Ｂ１ and D Ａ２ , D Ｂ２ The stable / unstable label information assigned to each of the above data J is used as training data. Ａ１ , J Ｂ１ and unstable state data J Ａ２ , J Ｂ２ A spherical threshold Q that separates Ａ , Q Ｂ Learn.

[0055] Then, the communication service management device CS stores the learning models corresponding to the communication services A and B for which the learning process has been completed in the state determination model storage unit 33 as state determination models.

[0056] (3) Operation Phase (3-1) Overall Description of State Determination Process FIG. 5 is a flowchart showing an example of the processing procedure and processing contents of the state determination process executed by the control unit 1 of the communication service management device CS in the operation phase.

[0057] It is assumed that, for example, a disaster occurs during operation of the communication network, causing a change in the communication path, and that a status determination request is transmitted from the manager terminal TM in response to this change.

[0058] When the control unit 1 of the communication service management device CS detects the above-mentioned status determination request in step S20, under the control of the status determination processing unit 15, executes the following process to determine the status of the communication service using the switched communication path.

[0059] That is, the state determination processing unit 15 first acquires a communication service ID indicating the communication service in use from the administrator terminal TM in step S21, and then selects a model corresponding to the communication service from among the state determination models stored in the state determination model storage unit 33 based on the acquired communication service ID in step S22.

[0060] Next, in step S23, the state determination processing unit 15 acquires current measurement data of each element included in the communication resources corresponding to the communication service from each of the communication devices R1 to R7 that make up the post-switching communication path used by the communication service, via the communication I / F unit 4. Then, in step S24, the state determination processing unit 15 inputs the acquired current measurement data as an explanatory variable into a state determination model corresponding to the communication service, and acquires in step S25 a determination result of the current state of the communication service that is output from the state determination model as a target variable.

[0061] The above-described process of determining the status of a communication service is performed for each of all communication services designated as the objects of determination from the administrator terminal TM.

[0062] When the determination result of the state of the communication service after the communication path switching is obtained, the control unit 1 of the communication service management device CS generates information for displaying the determination result in step S26 under the control of the determination result output processing unit 16. Then, the determination result output processing unit 16 transmits the generated display information from the communication I / F unit 4 to, for example, the administrator terminal TM.

[0063] As a result, the manager terminal TM displays the result of the determination of the state of the communication service after the communication path switching, that is, the display information indicating whether the state remains stable or becomes unstable, transmitted from the communication service management device CS. Therefore, the system manager can immediately check the state of the communication service after the path switching by using the display information, without needing skills or experience.

[0064] (3-2) Specific example of status determination processing Now, for example, suppose a failure occurs between communication devices RA-2, RB-2 and communication devices RA-3, RB3 as shown in Figure 11, and the communication path for communication service A is switched to a path that passes through communication devices RA-1, RA-2, RA-5, RA-6, and RA-7, and the communication path for communication service B is switched to a path that passes through communication devices RB-1, RB-2, RB-5, RB-6, and RB-7.

[0065] In this case, the communication service management device CS acquires current measurement data for each element of the communication resources corresponding to communication services A and B from the communication devices RA-1, RA-2, RA-5, RA-6, RA-7 and RB-1, RB-2, RB-5, RB-6, RB-7 on the communication paths after switching for each of communication services A and B.

[0066] 12 and 13 show examples of current measurement data relating to communication resources acquired from communication paths after switching between communication services A and B, respectively.

[0067] Next, the communication service management device CS inputs the current measurement data into the state determination models corresponding to the communication services A and B, respectively, as shown in Figures 14 and 15, and obtains the determination results of the current states of the communication services A and B from the state determination models. In the example shown in Figures 14 and 15, the determination result J for the communication service A indicates an "unstable" state. Ａ２ is output from the state determination model, and for communication service B, a determination result J indicating a "stable" state is Ｂ１ is output from the state determination model.

[0068] The communication service management device CS determines the result of the determination J Ａ２ , J Ｂ１ and presents the display information representing the above to the system administrator.

[0069] (Effect) As described above, in one embodiment, information representing a combination of communication resource elements defined for each communication service is acquired in advance and stored in the communication resource information storage unit 31. In this state, first, in the learning phase, for each communication service, a learning model is made to learn using past measurement data of the communication resources corresponding to that communication service, and a state determination model is created. Then, in the operation phase, for each communication service, current measurement data related to the communication resources corresponding to that communication service is acquired from each communication device constituting the communication path after path switching, and the acquired current measurement data is input to the state determination model, and a determination result of the state of the communication service is obtained from the state determination model. Then, display information representing the determination result is generated and output.

[0070] That is, for each communication service, information representing communication resources used in the communication service is associated with the communication service and managed. At the same time, for each communication service, a learning model is trained using past measurement data of the communication resources corresponding to the communication service as learning data, thereby creating a state determination model that outputs a determination result of the current state of the communication service as a response variable when current measurement data related to the communication resources is input as an explanatory variable.

[0071] Therefore, according to one embodiment, when a communication path is switched due to a disaster, for example, the status of the communication service using the switched communication path can be determined quickly and accurately by using the status determination model, without relying on the skills or experience of the administrator.

[0072] [Other Embodiments] (1) In the above embodiment, for example, information representing a combination of communication resource elements for each communication service, which is generated in a management center, is acquired in advance by the communication service management device CS and stored in the communication resource information storage unit 31. However, the present invention is not limited to this. For example, in the learning phase, when generating a state determination model corresponding to a communication service, the information may be acquired from the management center for each communication service.

[0073] (2) In the above embodiment, the state determination model is trained using only a set of past measurement data. However, each time new measurement data is acquired in the operation phase, the trained state determination model may be trained using the acquired measurement data. This can further improve the determination accuracy of the state determination model.

[0074] (3) In the above embodiment, the state determination model is used to determine whether the communication service on the switched communication path is stable or unstable. However, if the state is determined to be unstable, information indicating the degree of instability may also be generated and output. This can be achieved, for example, by determining how far the unstable state determination value is from a spherical threshold value indicating the range of stable states.

[0075] (4) In addition, the functional configuration of the communication service management device, its processing procedures and processing contents, the type of neural network used to create the state determination model, and its learning algorithm can be modified and implemented in various ways without departing from the spirit of this invention.

[0076] Although the embodiments of the present invention have been described in detail above, the above description is merely an example of the present invention in every respect. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. In other words, when implementing the present invention, specific configurations according to the embodiments may be appropriately adopted.

[0077] In short, this invention is not limited to the above-described embodiments, and in the implementation stage, the components can be modified and embodied without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in the embodiments. Furthermore, components from different embodiments may be appropriately combined.

[0078] CS...communication service management device NW...communication network TM...administrator terminal R1 to R7...communication devices (nodes) L11 to L67...transmission paths (edges) 1...control unit 2...program storage unit 3...data storage unit 4...communication I / F unit 5...bus 11...communication resource management processing unit 12...measured data acquisition processing unit 13...communication resource information determination processing unit 14...model learning processing unit 15...state determination processing unit 16...determination result output processing unit 31...communication resource information storage unit 32...measured data storage unit 33...state determination model storage unit

Claims

1. A communication service management device that manages a plurality of communication services provided by a communication network composed of a plurality of communication devices and a transmission path connecting these communication devices, For each of the communication services, a first processing unit that manages information representing communication resources corresponding to the communication service; For each of the communication services, a second processing unit that acquires past measurement data related to the communication resources corresponding to the communication service in a stable state and an unstable state of the communication service; For each of the communication services, a third processing unit that causes a learning model to perform learning using the acquired past measurement data as teacher data, and thereby outputs a determination result of the current state of the communication service as an objective variable when current measurement data related to the communication resources is input as an explanatory variable. A communication service management device comprising:

2. In the communication network, when the switching of the communication path that provides the communication service to be determined is performed, a fourth processing unit that acquires the current measurement data related to the communication resources corresponding to the communication service to be determined; A fifth processing unit that inputs the acquired current measurement data into the state determination model and acquires a determination result of the current state of the communication service to be determined from the state determination model; The communication service management device according to claim 1, further comprising a sixth processing unit that outputs information representing the determination result.

3. A communication service management method for managing, by an information processing apparatus, a plurality of communication services provided by a communication network configured by a plurality of communication devices and a transmission path connecting these communication devices, the method comprising: for each of the plurality of communication services, a process of managing information representing communication resources corresponding to the communication service; for each of the communication services, a process of acquiring past measurement data related to the communication resources corresponding to the communication service in a stable state and an unstable state of the communication service; and for each of the communication services, a process of causing a learning model to learn using the acquired past measurement data as teacher data, thereby creating a state determination model that outputs, as an objective variable, a determination result of the current state of the communication service when current measurement data related to the communication resources is input as an explanatory variable. A communication service management method comprising the above.

4. A program for causing a processor included in the communication service management apparatus to execute at least one of the processes executed by the first processing unit to the sixth processing unit included in the communication service management apparatus according to claim 1 or 2.

Images