COMMUNICATION MANAGEMENT DEVICE, COMMUNICATION MANAGEMENT METHOD, AND COMMUNICATION MANAGEMENT SYSTEM

Abstract

The object is to improve the throughput of each of a plurality of mobile communication terminals to be managed with a simpler configuration. [Solution] The communication management device 1 includes a generator 131 that generates pseudo communication volume data similar to the true communication volume data, using the communication volume data of the selected communication terminal with the largest total communication volume as the true communication volume data, and a learning unit 13 that performs adversarial learning of a generative model having a discriminator 132 that discriminates between the pseudo communication volume data generated by the generator 131 and the true communication volume data, and a notification unit 16 that notifies a plurality of communication terminals 2 of the learned generator 131' constructed by the learning unit 13 as communication management information.

Description

[Technical field]

[0001] The present invention relates to a communication management device, a communication management method, and a communication management system. [Background technology]

[0002] In recent years, the number of IoT terminals connected to mobile communication networks has been increasing, and the use of mobile IoT terminals in particular has also been increasing. For example, a communication module compatible with communication standards such as 5G implemented in a connected car connects to a data center via the Internet. Such a communication module realizes various information management and related services by connecting to the Internet via a mobile communication network. However, when an IoT terminal moves between communication areas, the communication area of ​​the destination may be congested. Therefore, for example, Patent Document 1 discloses a technology for dynamically allocating resources to a moving communication terminal according to the congestion state of the network, thereby optimizing throughput.

[0003] In the communication management technology disclosed in Patent Document 1, for IoT terminals with data to communicate, a scheduling coefficient according to the type of communication, such as voice communication or data communication, is calculated to determine the priority. Then, by comparing and judging the coefficients, wireless resources are preferentially allocated to IoT terminals with relatively large values ​​of the scheduling coefficient, thereby improving the throughput of moving IoT terminals throughout the network. In this way, in Patent Document 1, since the throughput is optimized throughout the network, it is difficult to improve the throughput of each of the specific moving communication terminals selected as the management targets. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2012 / 147479 Summary of the Invention [Problem to be solved by the invention]

[0005] In conventional techniques, it has been difficult to improve the throughput of each of a plurality of mobile communication terminals to be managed with a simpler configuration.

[0006] The present invention has been made to solve the above-mentioned problems, and has an object to improve the throughput of each of a plurality of mobile communication terminals to be managed with a simpler configuration. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, a communication management device according to the present invention is a communication management device that manages communications of a plurality of communication terminals that move between a plurality of communication areas, and includes: a collection unit configured to collect communication volume data of each of the communication terminals in the plurality of communication areas including the communication volume of each communication terminal while present in each communication area; a selection unit configured to calculate a sum of the communication volumes of each of the communication terminals in the plurality of communication areas through which each of the communication terminals has moved based on the collected communication volume data of each of the communication terminals, and to select the communication volume data of the communication terminal having the largest calculated total communication volume; a learning unit configured to perform adversarial learning of a generative model having a generator that generates pseudo communication volume data similar to the true communication volume data, and a discriminator that distinguishes between the pseudo communication volume data generated by the generator and the true communication volume data; and a notification unit configured to notify the plurality of communication terminals of the learned generator constructed by the learning unit as communication management information.

[0008] In addition, in the communication management device of the present invention, the collection unit may collect the communication volume data of each of the communication terminals in response to a location registration request signal transmitted by each of the communication terminals when the communication terminals move into each of the communication areas.

[0009] In the communication management device according to the present invention, the plurality of communication terminals may be a plurality of communication terminals to which common identification information is assigned.

[0010] In addition, in the communication management device according to the present invention, the collection unit may collect the communication volume data of each of the plurality of communication terminals to which the common identification information is assigned, from a user plane function provided in a core network.

[0011] In the communication management device according to the present invention, the pseudo traffic data may have a data distribution that has a minimum statistical distance from the data distribution of the true traffic data.

[0012] In order to solve the above-mentioned problems, a communication management method according to the present invention is a communication management method for managing communications of a plurality of communication terminals moving among a plurality of communication areas, and includes: a collection step of collecting communication volume data of each of the communication terminals in the plurality of communication areas including the communication volume of each communication terminal while present in each communication area; a selection step of calculating a sum of the communication volumes of each of the communication terminals in the plurality of communication areas through which each of the communication terminals has moved based on the collected communication volume data of each of the communication terminals, and selecting the communication volume data of the communication terminal with the largest calculated sum of communication volumes; a learning step of performing adversarial learning of a generative model having a generator that generates pseudo communication volume data similar to the true communication volume data, and a discriminator that distinguishes between the pseudo communication volume data generated by the generator and the true communication volume data; and a notification step of notifying the plurality of communication terminals of the learned generator constructed in the learning step as communication management information.

[0013] In addition, in the communication management method of the present invention, the collection step may collect the communication volume data of each of the communication terminals in response to a location registration request signal transmitted by each of the communication terminals when the communication terminals move into each of the communication areas.

[0014] In the communication management method according to the present invention, the plurality of communication terminals may be a plurality of communication terminals to which common identification information is assigned.

[0015] In the communication management method according to the present invention, the collecting step may collect the communication volume data of each of the plurality of communication terminals to which the common identification information is assigned, from a user plane function provided in a core network.

[0016] In addition, the communication management method of the present invention may further include an acquisition step in which each of the multiple communication terminals acquires the learned generator of the notified communication management information, a generation step in which the plurality of communication terminals uses the learned generator to generate pseudo communication volume data similar to the true communication volume data, and a communication step in which the plurality of communication terminals controls transmission of data in each communication area based on the generated pseudo communication volume data.

[0017] In order to solve the above-mentioned problems, the communication management system of the present invention is a communication management system comprising the above-mentioned communication management device and the plurality of communication terminals, each of the plurality of communication terminals comprising an acquisition unit configured to acquire the learned generator of the communication management information notified from the communication management device, a generation unit configured to generate the pseudo communication volume data similar to the true communication volume data using the learned generator, and a communication unit configured to control the transmission of data in each communication area based on the generated pseudo communication volume data. Effect of the Invention

[0018] According to the present invention, the communication volume data of the selected communication terminal with the largest total communication volume is set as the true communication volume data, and adversarial learning of a generative model having a generator that generates pseudo communication volume data similar to the true communication volume data and a discriminator that discriminates between the pseudo communication volume data generated by the generator and the true communication volume data is performed. Therefore, with a simpler configuration, it is possible to improve the throughput of each of a plurality of mobile communication terminals to be managed. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a block diagram showing a configuration of a communication management system including a communication management device according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing a configuration of a communication terminal according to the present embodiment. [Diagram 3] FIG. 3 is a diagram for explaining communication volume data collected by the communication management device according to the present embodiment. [Figure 4] FIG. 4 is a diagram for explaining communication volume data collected by the communication management device according to the present embodiment. [Diagram 5] FIG. 5 is a diagram for explaining the structure of a setting information table included in the communication management system according to the present embodiment. [Figure 6] FIG. 6 is a block diagram showing a configuration of a learning unit included in the communication management device according to the present embodiment. [Figure 7] FIG. 7 is a diagram for explaining the learning unit included in the communication management device according to the present embodiment. [Figure 8] FIG. 8 is a diagram for explaining the learning unit included in the communication management device according to the present embodiment. [Figure 9] FIG. 9 is a block diagram showing a hardware configuration of the communication management device according to the present embodiment. [Figure 10] FIG. 10 is a block diagram showing a hardware configuration of the communication management device according to the present embodiment. [Figure 11] FIG. 11 is a sequence diagram showing an outline of the operation of the communication management system according to the present embodiment. [Figure 12] FIG. 12 is a sequence diagram showing an outline of the operation of the communication management system according to the present embodiment. [Figure 13] FIG. 13 is a flowchart showing the learning process performed by the communication management device according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to FIGS.

[0021] [Communication management system configuration] First, there will be described an overview of a communication management system including a communication management device 1 according to an embodiment of the present invention and a communication terminal 2. Fig. 1 is a block diagram showing a configuration of a communication management system according to an embodiment of the present invention.

[0022] The communication management system according to the present embodiment is provided in, for example, a mobile communication network compatible with the 5G communication standard. The communication management system includes a communication management device 1, a communication terminal 2, a base station 3, a core network 4, and a data network 5. The communication management system manages, for each base station 3, the optimal data transmission timing of each of a plurality of communication terminals 2 moving between a plurality of communication areas 3A.

[0023] The communication terminal 2 is realized by a mobile communication terminal such as a smartphone, a tablet computer, or a laptop computer, a communication module, or an IoT terminal using a 5G mobile communication network, such as a wearable device. The communication terminal 2 can be configured as a communication module mounted on a connected car or an autonomous mobile robot. The communication terminal 2 includes an SIM 2a, and the contract profile of the SIM 2a stores the subscriber identification information of the user, and includes identifier information such as a subscriber identifier (IMSI: International Mobile Subscriber Identity) assigned to a mobile phone line contract, a telephone number (MSISDN: Mobile Subscriber International Subscriber Directory Number) of the subscriber user, and a SIM card number (ICCID: Integrated Circuit Card Identifier). The communication terminal 2 is uniquely identified by the IMSI.

[0024] The communication terminal 2 is also assigned a terminal IP address that uniquely identifies the terminal. The IP address is assigned to the communication terminal 2 through the SMF 42 after a session is established. In this embodiment, there are multiple communication terminals 2, and a group ID (common identification information) is assigned based on a predetermined attribute. The group ID can be assigned to terminals that have common attributes related to communication and services among multiple communication terminals 2 that are present in multiple communication areas 3A. For example, the same group ID is assigned to communication terminals 2 mounted on multiple vehicles such as taxis of a ride-hailing service managed and operated by an organization such as a company. The communication management process in this embodiment is performed for each of multiple communication terminals 2 with the same group ID.

[0025] A plurality of communication terminals 2 move between a plurality of communication areas 3A. The movement routes of each communication terminal 2 may differ. Furthermore, the time at which each communication terminal 2 starts moving, and the time and order in which it passes through each communication area 3A may differ. When the communication terminal 2 crosses over into the destination communication area 3A, it transmits a location registration request signal to the base station 3 of the destination communication area 3A. The location registration request signal includes the IMSI of the communication terminal 2. Note that each of the plurality of communication terminals 2 with the same group ID moves to all of the communication areas 3A from 1 to N, even if the order and time of passing through them differ.

[0026] The base station 3 is composed of a wireless base station compatible with the 5G communication standard, and relays communication between the communication terminal 2 present in the communication area 3A and the core network 4. The base station 3 is connected to the core network 4 via a backhaul link. The base station 3 receives data from the communication terminal 2 using predefined wireless resources. In this embodiment, it is assumed that one base station 3 covers one communication area 3A, and the base station 3 and the communication area 3A are identified by a base station ID.

[0027] The core network 4 is connected to the communication management device 1 via a network NW such as a LAN, a WAN, or the Internet. The core network 4 includes an AMF (Access and Mobility Management Function) 40, a UDM (Unified Data Management) / UDR (Unified Data Repository) 41, a SMF (Session Management Function) 42, and a PCF (Policy Control Function) 43, which are nodes in the C-plane. The core network 4 also includes a plurality of UPFs (User Plane Functions) 44 in the U-plane. Functional nodes in the U-plane and C-plane other than those described above that the core network 4 includes are omitted from the illustration.

[0028] The AMF 40 is a node that provides a mobility control function and performs mobility control such as location registration, paging, and handover.

[0029] The UDM / UDR 41 manages subscriber profiles, performs authentication, and manages mobility. In this embodiment, the UDM / UDR 41 creates and stores a setting information table T3 in which the IMSIs, group IDs, and base station IDs of multiple communication terminals 2, and the communication paths of the UPF 44 set in data communication of the multiple communication terminals 2 are associated with each other, as shown in Fig. 5, in response to an instruction from the communication management device 1. The instruction from the communication management device 1 includes grouping information in which multiple IMSIs and group IDs of the management targets that are specified in advance are associated with each other. The value of the IP address of the UPF 44 in the setting information table T3 is registered after the communication paths are set by the SMF 42 and PCF 43 described later.

[0030] The UDM / UDR 41 can configure the setting information table T3 by adding a group ID field to the subscriber profile. The UDM / UDR 41 according to this embodiment has a communication interface 41a for communicating with the communication management device 1. The setting information table T3 managed by the UDM / UDR 41 is also synchronized with the communication management device 1, and the same contents are stored (first storage unit 14). The UDM / UDR 41 may be configured as a single device including the UDM and UDR, or may be a device in which the UDM and UDR are arranged separately.

[0031] The SMF 42 is a session management function that performs the establishment, modification, release, etc. of a PDU (Packet Data Unit) session between the communication terminal 2 and a data network such as the Internet. The SMF 42 sets an appropriate communication path for data communication between the communication terminal 2 and the UPF 44 based on a PCC (Policy and Charging Control) policy from the PCF 43.

[0032] The PCF 43 determines QoS and policies and provides them to the SMF 42. The PCF 43 applies PCC rules based on the specifications of 3GPP (registered trademark), and in response to an instruction from the communication management device 1, creates a PCC policy for setting a communication path of the UPF 44 with which the communication terminal 2 communicates.

[0033] The UPF 44 is a user plane function that processes data between the base station 3 and a data network 5 such as the Internet. A plurality of UPFs 44 are provided in the core network 4. In this embodiment, the UPF 44 records in a memory data on the amount of communication while the communication terminal 2 is present in a communication area 3A covered by each base station 3. In this embodiment, the UPF 44 includes a communication interface 44a for communicating with the communication management device 1.

[0034] 3 is a diagram for explaining communication volume data recorded in each UPF 44 and collected by a collection unit 11 of the communication management device 1 described later. As shown in table T1 in FIG. 3, each UPF 44 records in memory a log of communication volume for each IMSI and group ID during the period when the device is present in each communication area 3A. i (i=1, 2, . . . , N) represents the base station IDs of the N base stations 3 in the range of the communication terminal 2. i represents a binary communication amount for each base station ID, taking the value "1" if there is data communication while the communication terminal 2 is in range, and "0" if there is no data communication. Note that the communication amount data recorded by the UPF 44 further includes the communication amount [G] at each base station ID before being converted into the binary communication amount.

[0035] FIG. 4 is a diagram for explaining communication volume data. Table T2 in FIG. 4 shows communication volume data in a communication terminal 2 of IMSI_1. As shown in FIG. 4, the communication volume data is data in which a base station ID is associated with a communication volume [G] at the base station ID and data obtained by converting the communication volume [G] into a binary communication volume. The communication terminal 2 of IMSI_1 performs data communication of a communication volume of 3 [G] while in a communication area 3A covered by a base station 3 having a base station ID of "t1", and the communication volume is 0 while in a communication area 3A covered by a base station 3 having a base station ID of "t2". In addition, since the communication volume is 3 [G] at the base station ID "t1", the binary communication volume takes a value of "1" (data communication present), and since no communication volume is generated at the base station ID "t2", the binary communication volume takes a value of "0" (no data communication). The communication volumes corresponding to other base station IDs are also converted into binary communication volumes in the same manner. As described above, the communication volume data (t i ,x i ) is expressed as data including a sequence of binary communication volumes. Communication volume data of other IMSIs can be calculated in a similar manner. Note that the process of converting the communication volume [G] into binary communication volumes may be performed by the collection unit 11 of the communication management device 1 instead of the UPF 44.

[0036] The data network 5 is an external network such as the Internet, a cloud network, etc. The communication terminal 2 performs data communication with the data network 5 via the base station 3 and the UPF 44 of the core network 4.

[0037] [Communication management device functional block] 1, the communication management device 1 includes a setting unit 10, a collecting unit 11, a selecting unit 12, a learning unit 13, a first storage unit 14, a second storage unit 15, and a notifying unit 16. The communication management device 1 learns a generative model by adversarial learning using the timing of data transmission per base station 3 of the communication terminal 2 with the best throughput among multiple moving communication terminals 2 as true communication volume data, and notifies the multiple communication terminals 2 to be managed of the learned generator constructed by learning as communication management information.

[0038] The setting unit 10 designates an IMSI and a group ID based on a location registration request signal from each communication terminal 2, and instructs the setting of the UPF 44 of each communication terminal 2. Specifically, the setting unit 10 transmits grouping information in which the IMSIs and group IDs of the management targets registered in advance are associated with each other to the UDM / UDR 41. Based on the grouping information from the setting unit 10, the UDM / UDR 41 creates a setting information table T3 (FIG. 5). Furthermore, the setting unit 10 instructs the setting of a user plane function for data communication of each communication terminal 2 based on the group ID associated with the IMSI included in the location registration request signal from each communication terminal 2.

[0039] The setting unit 10 requests the PCF 43 of the core network 4 to create a PCC policy by specifying the group ID, the IMSI, and the IP address of the communication management device 1. In response to the creation request, the PCF 43, the SMF 42, and the UDM / UDR 41 of the core network 4 cooperate to set an appropriate communication path of the UPF 44 for data communication of the communication terminal 2. When the communication path of the UPF 44 for data communication of the communication terminal 2 is set, the setting information table T3 updated in the UDM / UDR 41 is synchronized with, for example, the first storage unit 14, and the updated setting information table T3 is stored in the first storage unit 14.

[0040] The collection unit 11 collects communication volume data of each communication terminal 2 in a plurality of communication areas 3A, including the communication volume of each communication terminal 2 while it is present in each communication area 3A. More specifically, the collection unit 11 collects the communication volume [G] of each communication terminal 2 during the period when it is present in each communication area 3A and the binary communication volume obtained by converting the communication volume [G] into binary data, triggered by a location registration request signal transmitted by each communication terminal 2 when it moves to each communication area 3A. That is, the communication volume in each destination communication area 3A (base station ID) and the corresponding binary communication volume are collected, and finally, communication volume data in a plurality of communication areas 3A is collected, triggered by a location registration request signal transmitted by the communication terminal 2 every time it moves to the communication area 3A.

[0041] Furthermore, the collection unit 11 collects, from the UPF 44 included in the core network 4, the communication volume [G] for each communication area 3A (for each base station ID) of each of the multiple communication terminals 2 to which a group ID is assigned, and communication volume data represented by a series of binary communication volumes obtained by converting the communication volumes into binary data. The collection unit 11 collects the communication volume data of each of the multiple communication terminals 2 for each group ID via the network NW from the UPF 44 related to the setting of a communication path for data communication of the communication terminals 2 by the setting unit 10. As shown in FIG. 3, the collection unit 11 collects the base station IDs (t1, t2, ..., t N), where each communication amount [G] is represented by binary data, the communication amount data (t i ,x i ) to collect.

[0042] The selection unit 12 selects all communication areas 3A (base station IDs: t1 to t2) through which each communication terminal 2 has moved based on the collected communication volume data of each communication terminal 2. N ) is calculated, and the communication volume data (t i ,x i More specifically, the selection unit 12 selects the communication volume data (t i ,x i ), for each IMSI with a common group ID, each base station ID (base station ID: t1 to t N The communication volume data of the communication terminal 2 having the largest total communication volume [G] in each communication area 3A is selected. N The communication terminal 2 with the largest total value of the communication volume [G] in the group ID is the communication terminal 2 with the highest throughput among the multiple communication terminals 2 included in the group ID.

[0043] More specifically, the selection unit 12 can compare the total values ​​of communication volumes when all IMSIs having a common group ID are present in each communication area 3A, and select N pieces (N is a positive integer of 2 or more) of communication volume data in descending order of the largest total value. The selection of N pieces of data can be a number necessary to obtain sufficient accuracy in adversarial learning by the learning unit 13 described later, such as the top 4%, for example.

[0044] The learning unit 13 performs adversarial learning of a generation model having a generator 131 that generates pseudo communication volume data similar to the true communication volume data, with the communication volume data of the selected communication terminal 2 having the largest total communication volume being regarded as true communication volume data, and a discriminator 132 that discriminates between the pseudo communication volume data generated by the generator 131 and the true communication volume data.

[0045] 6, the learning unit 13 adversarially learns a GAN (Generative Adversarial Network) having a generator 131 and a discriminator 132. A trained generator 131′ is constructed by the learning of the learning unit 13.

[0046] 7 and 8 are diagrams showing schematic configurations of the neural network of the generator 131 and the discriminator 132 of the GAN used by the learning unit 13. As shown in FIG. 7, the generator 131 is configured of a neural network having an input layer, a hidden layer, and an output layer. The hidden layer can be configured of a fully connected layer and a deconvolution layer. The generator 131 is a model that generates pseudo communication volume data from random noise. For example, m samples of a Gaussian noise vector are randomly sampled and input to the input node of the generator 131 (z1 to z m The generator 131 performs a multiplication and accumulation operation on the input and weight parameters and performs threshold processing using an activation function to output G(z1) to G(z N ) to output.

[0047] The classifier 132 shown in Fig. 8 is configured as a neural network having an input layer, a hidden layer, and an output layer. The hidden layer can be configured as a convolutional layer, and the output layer can be configured as a fully connected layer. In the example of Fig. 8, communication volume data x1 to x N As described above, the communication volume data is expressed as a binary sequence. Therefore, the base station ID (t i 6, true communication volume data is input to the classifier 132 as training data 134.

[0048] The discriminator 132 outputs, for example, 1 or 0 after performing a product-sum operation on the input and weight parameters and threshold processing using an activation function. When the discriminator 132 correctly discriminates that the training data related to the input true communication volume data is true communication volume data, it outputs an output y=1. On the other hand, when the discriminator 132 correctly discriminates that the training data related to the input pseudo communication volume data is pseudo communication volume data, it outputs an output y=0. In this way, the discriminator 132 is a model that distinguishes the model distribution generated by the generator 131 from the data distribution of the training data, which is the true distribution. The pseudo communication volume data has a data distribution that has a minimum statistical distance from the data distribution of the true communication volume data.

[0049] FIG. 6 is a block diagram for explaining the adversarial learning of GAN by the learning unit 13. The generator 131 of the GAN adopted by the learning unit 13 is represented as a function G, and the discriminator 132 is represented as a function D. Furthermore, the true communication volume data is represented as x, the predicted value output by the discriminator 132 is represented as y, and the correct label is represented as t. The correct label t is set to 1 for the true communication volume data and 0 for the pseudo communication volume data generated by the generator 131. At this time, the discriminator 132 calculates the cross entropy E CE It can be expressed as:

[0050]

number

[0051] The first term in the brace of the above equation (1) represents t n lny n In the above, the predicted value y n is the correct label of the true traffic volume data, t n = 1. On the other hand, the second term in the braces represents (1-t n )ln(1-y n ), the predicted value y n The correct label value (1-t n ) = 0. In this way, the cross entropy E CEis maximized when the predicted value matches the correct label value.

[0052] Here, the generator 131 constituting the GAN has parameters w G ,θ G and function G(w G ,θ G ) The discriminator 132 also uses the parameter w D ,θ D and function D(w D ,θ D ) The cross entropy E CE The objective function E of the GAN including the generator 131 and the discriminator 132 based on the above can be expressed by the following equation (2).

number

[0053] The first term of the above formula (2) represents E D(x)=1 lnD(w D ,θ D ) is the expected value at which the discriminator 132 discriminates true traffic data as true traffic data. D(x)=0 ln(1-D(G(w G ,θ G ),w D ,θ D )) is an expected value at which the discriminator 132 discriminates the pseudo communication volume data generated by the generator 131 as pseudo communication volume data. In GAN learning, the generator 131 and the discriminator 132 are adversarially trained by min-max optimization of the objective function E. Therefore, the generator 131 is trained so as to generate pseudo communication volume data that deceives the discriminator 132, and the discriminator 132 is trained so as to discriminate the pseudo communication volume data generated by the generator 131 as pseudo communication volume data.

[0054] In learning by the classifier 132, when true communication volume data is given, the classifier 132 maximizes the first term of the objective function E in the above formula (2) by outputting an output close to y=1. On the other hand, when pseudo communication volume data is given, the classifier 132 learns to maximize the second term of the objective function E by outputting an output close to y=0.

[0055] In the learning of the generator 131, D(G(w G ,θ G ),w D ,θ D )(D(G(z)) in Figure 6) is close to 1. G ,θ G ) (G(z) in FIG. 6 ) to minimize the objective function E. The learning unit 13 uses a learning procedure in which the parameters of the generator 131 and the classifier 132 are alternately updated. Details of the learning procedure of the generator 131 and the classifier 132 by the learning unit 13 will be described later.

[0056] When the objective function E of the GAN is optimized, the learning unit 13 stores the trained generator 131′ in the second storage unit 15.

[0057] The first storage unit 14 stores a setting information table T3 (FIG. 5) in which the IP address, IMSI, group ID, and base station ID of the UPF 44 related to the setting of the communication path by the setting unit 10 are associated with each other. The first storage unit 14 can synchronize with the setting information table T3 created and updated by the UDM / UDR 41. Alternatively, the first storage unit 14 may be configured to periodically acquire the setting information table T3 created and updated by the UDM / UDR 41, or to refer to and reflect the updated setting information table T3.

[0058] The second storage unit 15 stores the trained generator 131′ constructed by the learning unit 13. Since the trained generator 131′ is a generator constructed for each group ID, the trained generator 131′ is associated with a group ID.

[0059] The notification unit 16 notifies the plurality of communication terminals 2 of the trained generator 131' constructed by the learning unit 13 as communication management information. More specifically, the notification unit 16 notifies the plurality of communication terminals 2 having the same group ID of the trained generator 131' constructed for each group ID via the network NW.

[0060] [Communication terminal function block] Next, functional blocks of the communication terminal 2 according to the present embodiment will be described with reference to Fig. 2. Note that the multiple communication terminals 2 are configured with the same functional blocks.

[0061] The third storage unit 20 stores the learned generator 131′ notified by the notification unit 16 of the communication management device 1 as communication management information.

[0062] The acquisition unit 21 acquires the learned generator 131′ of the communication management information notified from the communication management device 1. Specifically, the acquisition unit 21 reads out the learned generator 131′ from the third storage unit 20.

[0063] The generating unit 22 generates pseudo communication volume data similar to the real communication volume data by using the learned generator 131'. The generated pseudo communication volume data is information indicating a data communication volume that is very similar to the data communication volume of the base station 3 (communication area 3A) of the communication terminal 2 having the best throughput among the group IDs. For example, the pseudo communication volume data (t i ,x i ) is the base station ID (t1, t2, t3, t4, . . . , t N In the communication area 3A of each base station 3, (x1, x2, x3, x4, . . . , x N Consider the case where t1 is the base station ID and t2 is the communication area 3A, and t3 is the base station ID. In this case, while the base station ID is in the communication area 3A of t1, the data is not transmitted, and while the base station ID is in the communication area 3A of t2, the data is transmitted, and .. The presence or absence of data transmission is represented by a binary communication volume series.

[0064] The communication unit 23 controls data transmission in each communication area 3A based on the generated pseudo communication volume data. For example, while the base station ID is in the communication area 3A of t1, the communication unit 23 can control the data transmission to "0" which does not transmit data. On the other hand, while the base station ID is in the communication area 3A of t2, the communication unit 23 can control the data transmission to "1" which transmits data. The communication unit 23 can control the communication by, for example, a grant-free method in which data is transmitted at a determined timing without requiring additional permission from the base station 3.

[0065] [Hardware configuration of communication management device] Next, an example of a hardware configuration for realizing the communication management device 1 having the above-mentioned functions will be described with reference to FIG.

[0066] 9, the communication management device 1 can be realized by, for example, a computer including a processor 102, a main memory device 103, a communication interface 104, an auxiliary memory device 105, and an input / output I / O 106 connected via a bus 101, and a program that controls these hardware resources. The communication management device 1 can also include a display device 107 connected via the bus 101.

[0067] The processor 102 is realized by a CPU, a GPU, an FPGA, an ASIC, or the like.

[0068] The main memory device 103 prestores programs for the processor 102 to perform various controls and calculations. The processor 102 and the main memory device 103 implement the various functions of the communication management device 1, such as the setting unit 10, the collecting unit 11, the selecting unit 12, the learning unit 13, and the notifying unit 16 shown in FIG.

[0069] The communication interface 104 is an interface circuit for connecting the communication management device 1 to various external electronic devices via a network.

[0070] The auxiliary storage device 105 is composed of a readable / writable storage medium and a drive for reading and writing various information such as programs and data from and to the storage medium. The auxiliary storage device 105 can use a semiconductor memory such as a hard disk or a flash memory as the storage medium.

[0071] The auxiliary storage device 105 has a program storage area for storing the GAN learning program and communication management program executed by the communication management device 1. The auxiliary storage device 105 also has an area for storing grouping information in which a group ID and an IMSI are associated with each other. The auxiliary storage device 105 realizes the first storage unit 14 and the second storage unit 15 described in FIG. 1. Furthermore, the auxiliary storage device 105 may have, for example, a backup area for backing up the above-mentioned data and programs.

[0072] The input / output I / O 106 is an input / output device that inputs signals from an external device and outputs signals to an external device.

[0073] The display device 107 is configured by an organic EL display, a liquid crystal display, or the like.

[0074] [Hardware configuration of communication terminal] Next, an example of a hardware configuration for realizing the communication terminal 2 having the above-mentioned functions will be described with reference to FIG.

[0075] 10, the communication terminal 2 can be realized by a computer including a processor 202, a main memory device 203, a communication interface 204, an auxiliary memory device 205, and an input / output I / O 206, which are connected via a bus 201, and a program that controls these hardware resources. Furthermore, the communication terminal 2 can include a display device 207 connected via the bus 201.

[0076] The main memory device 203 stores in advance programs for the processor 202 to perform various controls and calculations. The processor 202 and the main memory device 203 realize the functions of the communication terminal 2, such as the acquisition unit 21, the generation unit 22, and the communication unit 23 shown in FIG.

[0077] The auxiliary storage device 205 has a program storage area for storing the computation program of the trained generator 131′ executed by the communication terminal 2. It also has a program storage area for storing the communication program executed by the communication terminal 2. The auxiliary storage device 205 realizes the third storage unit 20 described in FIG. 2. Furthermore, it may have, for example, a backup area for backing up the above-mentioned data, programs, and the like.

[0078] [Operation of the communication management device] Next, the operation of the communication management device 1 having the above-mentioned configuration will be described with reference to the sequence diagrams of FIG. 11 and FIG. 12 and the flowchart of FIG.

[0079] 11 is a sequence diagram showing grouping of communication terminals 2 by the communication management system and setting of communication paths for data communication. First, the communication management device 1 transmits grouping information associated with a group ID and an IMSI to the UDM / UDR 41 (step S100). The grouping information is stored in advance in the communication management device 1. Next, the UDM / UDR 41 creates a setting information table T3 (FIG. 5) based on the received grouping information (step S101). In step S101, the values ​​of "group ID" and "IMSI" in the setting information table T3 are registered, and there are no entries yet for the values ​​of the other items.

[0080] After that, the communication terminal 2 that has moved to a certain communication area 3A transmits a location registration request signal from the base station 3 to the core network 4 (step S102). The location registration request signal includes the IMSI of the communication terminal 2 and the base station ID of the base station 3 in the communication area 3A in which the communication terminal 2 is present. When the location registration request signal is received by the UDM / UDR 41 via the AMF 40, the UDM / UDR 41 associates a group ID with the base station ID and IMSI included in the received location registration request signal based on the information in the setting information table T3 (FIG. 5) created in step S101, and further transfers the location registration request signal to the communication management device 1 (step S103). Note that in step S103, the value of the base station ID in the setting information table T3 is set.

[0081] Next, the setting unit 10 of the communication management device 1 requests the PCF 43 to create a PCC policy by specifying the IMSI, base station ID, group ID, and further the IP address of the device itself, contained in the location registration request signal received in step S103 (step S104). Then, the PCF 43 creates a PCC policy based on the specified requirements (step S105). The PCF 43 creates a PCC policy including information on which UPF 44 the data communication of the communication terminal 2 will pass through. The PCC policy specifies the optimal UPF 44 as a communication path for data communication from the communication terminal 2 to the data network 5.

[0082] Next, the PCF 43 transmits the created PCC policy to the SMF 42 (step S106). After that, the SMF 42 transmits data communication path setting information related to the user plane function, such as setting of a communication path specified by the PCC policy, to the UPF 44 specified by the PCC policy (step S107). Next, the UPF 44 registers the received data communication path setting information in a memory (step S108).

[0083] Next, the UPF 44 transmits an ACK to the SMF 42 to notify it of the IP address of the UPF 44 (step S109). Furthermore, the SMF 42 transmits an ACK to the communications management device 1 to notify it of the IP address of the UPF 44 (step S110). Thereafter, the communications management device 1 notifies the UDM / UDR 41 of the IP address of the UPF 44, and transmits an ACK to the communications terminal 2 (step S111). At that time, the IP address of the UPF 44 that communicates with the communications terminal 2 is set in the setting information table T3. The setting information table T3 updated by the UDM / UDR 41 is also stored in the first storage unit 14 of the communications management device 1. Further, after that, a communications path for data communications between the communications terminal 2 and the UPF 44 is established (step S112). The communications terminal 2 performs data communications with the data network 5.

[0084] After that, the UPF 44 records a log of the communication amount while each of the multiple communication terminals 2 is present in the communication area 3A (step S113). The log of the communication amount recorded in the UPF 44 is the communication amount [G] while each communication terminal 2 is present in each communication area 3A, and the corresponding binary communication amount, as shown in table T1 in FIG. 3. In addition, the communication amount data (t i ,x i ) is the communication volume [G] for each communication area 3A (base station ID) and the corresponding binary communication volume sequence data. Therefore, every time the communication terminal 2 reaches a destination communication area 3A, the processes from step S102 to step S112 are repeated, and in step S113, a log of the communication volume during the period when the communication terminal 2 was present in each communication area 3A is recorded. After that, the communication management device 1 executes the communication management process (step S114).

[0085] Next, the communication management process (step S114) described in Fig. 11 will be described in more detail with reference to the sequence in Fig. 12. First, the collection unit 11 of the communication management device collects from the UPF 44 communication traffic data, which is data recording communication traffic over a plurality of communication areas 3A (base station IDs) during a period when an IMSI (communication terminal 2) with a common group ID was present in each communication area 3A (step S1). The collection unit 11 refers to the setting information table T3 stored in the first storage unit 14, for example, to identify the IP address "IP11" of the UPF 44 with which IMSI_1 to IMSI_N of the group ID "1" communicate, and collects communication traffic data of communication traffic or sequence data from the UPF 44 of the identified IP address "IP11" (table T1 in Fig. 3). In addition, when the UPF 44 to which the communication terminal 2 is connected changes as a result of moving to a different base station 3, the communication volume or binary communication volume at the destination base station ID is collected from the newly set UPF 44, and communication volume data expressed as a series of communication volumes and corresponding binary communication volumes in each communication area 3A (each base station ID) across multiple communication areas 3A is collected.

[0086] Next, the selection unit 12 of the communication management device 1 selects the communication volume data of the communication terminal 2 having the largest total communication volume [G] in the multiple communication areas 3A based on the collected communication volume data (step S2). In step S2, the selection unit 12 selects the communication volume data (t i ,x i ), for each IMSI with a common group ID, N ) is the largest, and the communication volume data of the communication terminal 2 having the best throughput is selected.

[0087] In step S2, the N pieces of communication volume data with the highest throughput are obtained. For example, the total communication volumes of all IMSIs can be compared, and the N pieces of data can be collected in descending order of the largest value. The communication volume data selected in step S2 is used as training data in adversarial learning, i.e., true communication volume data.

[0088] Next, the learning unit 13 of the communication management device 1 performs a learning process of a GAN including a generator 131 and a discriminator 132 (step S3). More specifically, in step S3, the learning unit 13 performs adversarial learning of a generation model having a generator 131 that generates pseudo communication volume data similar to the true communication volume data, and a discriminator 132 that discriminates between the pseudo communication volume data generated by the generator 131 and the true communication volume data, using the communication volume data of the communication terminal 2 that has the largest total communication volume for each communication area 3A (base station ID) selected in step S2 as the true communication volume data (step S3).

[0089] 13 is a flowchart showing a learning process of the GAN by the learning unit 13 included in the communication management device 1. First, the learning unit 13 acquires true communication volume data to be used as training data (step S50). In step S50, a set of N pieces of true communication volume data is acquired. The true communication volume data indicates the presence or absence of data transmission from a communication terminal 2 having a high throughput among multiple communication terminals 2 having a common group ID.

[0090] Here, as shown in the GAN learning process by the learning unit 13 in FIG. 6, the true communication volume data acquired in step S50 is used as training data 134 to be input when the classifier 132 is trained.

[0091] Next, the learning unit 13 inputs the true traffic data to the discriminator 132 as training data 134, and adjusts the parameter w of the discriminator 132 so as to discriminate the true traffic data from the true traffic data (y=1). D ,θ D (Step S51). In Step S51, the learning unit 13 can cause the classifier 132 to learn the true communication volume data by using, for example, an error backpropagation method. By Step S51, the classifier 132 capable of discriminating the true communication volume data from the true communication volume data is constructed in advance.

[0092] Next, the learning unit 13 generates Gaussian noise 130 and provides a random vector of the generated Gaussian noise 130 as an input to the generator 131 (step S52). Next, the generator 131 generates a random vector of the generated Gaussian noise 130 based on the input z and the weighting parameter w G ,θ G and threshold processing using an activation function to generate pseudo communication volume data G(z) (step S53).

[0093] Next, the classifier 132 is trained. The classifier 132 is trained by using the parameter w D ,θ D First, the learning unit 13 provides the training data 134 of the true communication volume data collected in step S50 as an input to the discriminator 132, and calculates the parameter w using the backpropagation method or the like so that the objective function E of the above formula (2) is maximized. D ,θ D (Step S54). Note that the label of the training data 134 is set to 1 (true communication volume data).

[0094] Next, the learning unit 13 provides the pseudo communication volume data generated by the generator 131 in step S53 to the discriminator 132, and calculates the parameter w using the backpropagation method or the like so that the objective function E in the above formula (2) is maximized. D ,θ D (Step S55). That is, in steps S54 and S55, in order to maximize the objective function E in the above formula (2), the first term is updated as D(w D ,θ D )=1 is output, and the second term is D(G(w G ,θ G ),w D ,θ D )=0. In addition, in the training data 134, the label is set to 0 (pseudo communication volume data).

[0095] The learning of the classifier 132 in steps S54 and S55 corresponds to the dashed arrow ("Training") shown in FIG. 6, in which a classifier error is calculated in block 135 of the objective function E based on the output 133 from the classifier 132, and then the error is back-propagated to the classifier 132.

[0096] Next, the generator 131 is trained. In the training of the generator 131, the parameters of the discriminator 132 are fixed. The training unit 13 trains the generator 131 so that pseudo communication volume data is generated when random Gaussian noise 130 is given to the generator 131. Specifically, in order to minimize the objective function E in the above formula (2), the training unit 13 calculates an error and trains the parameter w G ,θ G is updated (step S56).

[0097] The learning in step S56 corresponds to the dashed arrow indicating backpropagation of error to the generator 131 shown in Fig. 6. That is, step S56 corresponds to the dashed arrow ("training") in which pseudo communication volume data generated by the generator 131 in Fig. 6 is input to the discriminator 132, a generator error is calculated from the output 133 in the block 135 of the objective function E, and the error is further backpropagated to the generator 131.

[0098] Thereafter, until the value of the objective function E reaches a Nash equilibrium and converges (step S57: NO), the learning of the discriminator 132 and the generator 131 from step S53 to step S56 is repeated. On the other hand, if the value of the objective function E converges (step S57: YES), the processing from step S51 to step S57 is repeated until the learning of the generator 131 and the discriminator 132 is performed using the remaining N-1 pieces of true communication volume data among the N pieces of true communication volume data (step S58: NO).

[0099] After that, when the generator 131 and the discriminator 132 are trained using the remaining N-1 pieces of true communication volume data (step S58: YES), the training unit 13 stores the trained generator 131' in the second storage unit 15 (step S4). The trained generator 131' is constructed by the above process.

[0100] Next, returning to FIG. 12, the notification unit 16 notifies the multiple communication terminals 2 having the same group ID of the learned generator 131' constructed for each group ID as communication management information via the network NW (step S5). Then, the acquisition unit 21 of the communication terminal 2 acquires the learned generator 131' (step S6). Next, the generation unit 22 of the communication terminal 2 uses the learned generator 131' to generate pseudo communication volume data similar to the true communication volume data (step S7). Then, the communication unit 23 of the communication terminal 2 controls data transmission in the communication area 3A to which the terminal has moved based on the pseudo communication volume data generated in step S7 (step S8).

[0101] Similarly, the processes from step S6 to step S8 are performed for each of the other communication terminals 2 (IMSI_2 to IMSI_N) to be managed that have the same group ID (group 1). Similarly, the processes from step S1 to step S8 are performed for the communication management process for the communication terminals 2 that have different group IDs.

[0102] As described above, according to the communication management device 1 of the present embodiment, communication volume data of a communication terminal 2 with high throughput is selected as true data in adversarial learning, and a generator 131 that generates pseudo communication volume data similar to the communication volume data of the communication terminal 2 with high throughput is learned by GAN learning. Then, the learned generator 131' is notified to the multiple communication terminals 2 as communication management information. Therefore, with a simpler configuration, it is possible to improve the throughput of each of the multiple communication terminals 2 that are mobile as managed targets.

[0103] Furthermore, according to the communication management device 1 of this embodiment, a plurality of mobile communication terminals 2 to be managed are grouped in advance, and the optimal data transmission timing for each base station 3 is learned for each group. Therefore, it is possible to improve the throughput of a specific group of communication terminals 2, rather than the throughput of the entire network.

[0104] Furthermore, according to the communication management device 1 of this embodiment, the communication terminal 2 collects communication volume data at the base station 3 in which it is located, triggered by a location registration request signal transmitted when the communication terminal 2 crosses a destination communication area 3A. Therefore, with a simpler configuration, it is possible to improve the throughput of each of a specific number of communication terminals 2.

[0105] In the above embodiment, a communication management system that complies with 5G has been exemplified, but the communication management system may also be one that complies with 4G / LTE, 6G, etc.

[0106] In addition, in the embodiment described above, the case where GAN is used as the generative model is described, but the generative model is not limited to GAN. For example, VAE (Variational Autoencoder), Autoregressive Models, Energy-Based Models, Transformer-Based Generative Models, etc. can be used.

[0107] The above describes the embodiments of the communication management device, communication management method, and communication management system of the present invention, but the present invention is not limited to the described embodiments, and various modifications that a person skilled in the art can imagine can be made within the scope of the invention described in the claims. [Explanation of symbols]

[0108] 1...communication management device, 10...setting unit, 11...collection unit, 12...selection unit, 13...learning unit, 14...first memory unit, 15...second memory unit, 16...notification unit, 2...communication terminal, 2a...SIM, 3...base station, 3A...communication area, 4...core network, 40...AMF, 41...UDM / UDR, 42...SMF, 43...PCF, 44...UPF, 5...data network, 101, 201...bus, 102, 202...processor , 103, 203...main memory device, 41a, 44a, 104, 204...communication interface, 105, 205...auxiliary memory device, 106, 206...input / output I / O, 107, 207...display device, 130...noise, 131...generator, 132...discriminator, 131'...trained generator, 133...output, 134...training data, 135...block of objective function E, T1, T2, T3...table, NW...network.

Claims

1. A communication management device that manages communications of a plurality of communication terminals moving between a plurality of communication areas, a collection unit configured to collect communication volume data of each of the communication terminals in the plurality of communication areas, the communication volume data including communication volume of each of the communication terminals while the communication terminals are present in each of the communication areas; a selection unit configured to calculate a total of the communication volumes of the communication terminals in the plurality of communication areas through which the communication terminals have moved based on the collected communication volume data of the communication terminals, and to select the communication volume data of the communication terminal having the largest calculated total of the communication volumes; a learning unit configured to perform adversarial learning of a generative model having a generator that generates pseudo-communication volume data similar to the true communication volume data by using the communication volume data of the selected communication terminal having the largest total communication volume as true communication volume data, and a classifier that distinguishes between the pseudo-communication volume data generated by the generator and the true communication volume data; a notification unit configured to notify the plurality of communication terminals of the trained generator constructed by the learning unit as communication management information; A communication management device comprising:

2. 2. The communication management device according to claim 1, The collection unit collects the communication volume data of each of the communication terminals in response to a location registration request signal transmitted when each of the communication terminals moves to each of the communication areas. A communication management device comprising:

3. 2. The communication management device according to claim 1, The plurality of communication terminals are a plurality of communication terminals to which common identification information is assigned. A communication management device comprising:

4. 4. The communication management device according to claim 3, The collection unit collects the communication volume data of each of the plurality of communication terminals to which the common identification information is assigned, from a user plane function included in a core network. A communication management device comprising:

5. 2. The communication management device according to claim 1, The pseudo traffic data has a data distribution that has a minimum statistical distance from the data distribution of the true traffic data. A communication management device comprising:

6. A communication management method for managing communications of a plurality of communication terminals moving among a plurality of communication areas, comprising: a collection step of collecting communication volume data of each of the communication terminals in the plurality of communication areas, the data including the communication volume of each of the communication terminals while the communication terminals are present in each of the communication areas; a selection step of calculating a total of the communication volumes of each of the communication terminals in the plurality of communication areas through which each of the communication terminals has moved based on the collected communication volume data of each of the communication terminals, and selecting the communication volume data of the communication terminal having the largest calculated total of the communication volumes; a learning step of performing adversarial learning of a generative model having a generator that generates pseudo-communication volume data similar to the true communication volume data by using the communication volume data of the selected communication terminal having the largest total communication volume as the true communication volume data, and a classifier that distinguishes between the pseudo-communication volume data generated by the generator and the true communication volume data; a notification step of notifying the plurality of communication terminals of the trained generator constructed in the learning step as communication management information; A communication management method comprising:

7. 7. The communication management method according to claim 6, The collecting step collects the communication volume data of each of the communication terminals in response to a location registration request signal transmitted when each of the communication terminals moves to each of the communication areas. A communication management method comprising:

8. 7. The communication management method according to claim 6, The plurality of communication terminals are a plurality of communication terminals to which common identification information is assigned. A communication management method comprising:

9. 9. The method of claim 8, further comprising: The collecting step collects the communication volume data of each of the plurality of communication terminals to which the common identification information is assigned, from a user plane function included in a core network. A communication management method comprising:

10. 7. The communication management method according to claim 6, Furthermore, each of the plurality of communication terminals An acquisition step of acquiring the learned generator of the notified communication management information; A generating step of generating the pseudo communication volume data similar to the true communication volume data using the trained generator; a communication step of controlling data transmission in each communication area based on the generated pseudo communication volume data; A communication management method comprising:

11. A communication management device according to any one of claims 1 to 5, The plurality of communication terminals; A communication management system comprising: Each of the plurality of communication terminals an acquisition unit configured to acquire the learned generator of the communication management information notified from the communication management device; A generation unit configured to generate the pseudo communication volume data similar to the true communication volume data by using the trained generator; a communication unit configured to control data transmission in each communication area based on the generated pseudo communication volume data; A communication management system comprising:

Images