Communication control device and communication control method

Abstract

To infer the total data communication volume used by a communication terminal and lead to effective use of wireless resources.SOLUTION: A communication control device 1 includes a collection unit 10 configured to collect a data communication volume used by a communication terminal 2 for each preset first period, a calculation unit 12 configured to provide the data communication volume collected for each first period as an unknown input to a trained machine learning model, perform calculations on the trained machine learning model, and output a total data communication volume used by the communication terminal 2 for a second period that is longer than the first period, and a presentation unit 15 configured to present information regarding the total data communication volume output by the calculation unit 12.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] The communication capacity available to a communication device such as a smartphone is determined based on the data communication plan provided by the communication carrier. For example, data communication plans are provided that allow unlimited data communication, up to 20GB, 10GB, 2GB, etc. per month at high speed. If the data communication limit of the contracted data communication plan is exceeded, the communication carrier may impose communication restrictions on the communication device, for example, shifting the communication speed from high speed to low speed.

[0003] Among users of communication services, there are users who reach the upper limit of the data communication amount of the contracted data communication plan, while there are users who do not use up the upper limit. For example, Patent Document 1 discloses a technology in which a parent-child relationship is established between multiple communication terminals (child devices) and a mobile router (parent device), and based on this parent-child relationship, the communication speed based on the rate plan subscribed by the users of each communication terminal in communication via the mobile router is temporarily changed based on the sum of the upper limit speeds of the multiple child devices and the parent device.

[0004] However, even if the technology described in Patent Document 1 allows temporary changes in communication speed, it still allocates communication bandwidth for high-speed communication to users who do not use the data communication volume up to the upper limit. This is because the total communication bandwidth secured by a communication carrier is generally calculated by multiplying the number of subscribers, i.e., the number of communication terminals, by the communication speed during high-speed communication. Therefore, it can be said that the overall communication bandwidth is being unnecessarily expanded, and the communication bandwidth is not being effectively utilized by the communication carrier as a whole. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2014-192757 A Summary of the Invention [Problem to be solved by the invention]

[0006] In the conventional technology, the total data traffic used by a communication terminal is not estimated, and therefore, the conventional technology does not effectively utilize wireless resources based on the estimated total data traffic.

[0007] The present invention has been made to solve the above-mentioned problems, and has an object to infer the total data communication volume used by a communication terminal, leading to effective use of wireless resources. [Means for solving the problem]

[0008] In order to solve the above-mentioned problems, the communication control device of the present invention includes a collection unit configured to collect data communication volume used by a communication terminal for each predetermined first period, a calculation unit configured to provide the data communication volume collected for each first period as an unknown input to a trained machine learning model, perform calculations on the trained machine learning model, and output a total data communication volume used by the communication terminal for a second period that is longer than the first period, and a presentation unit configured to present information related to the total data communication volume output by the calculation unit.

[0009] In addition, the communication control device of the present invention further includes a learning unit configured to learn the relationship between the value for the data communication volume used by the communication terminal for each first period and the total data communication volume used by the communication terminal in the second period using a machine learning model, and a memory unit configured to store the trained machine learning model constructed by the learning unit, and the calculation unit may read out the trained machine learning model from the memory unit and perform calculations on the trained machine learning model.

[0010] In addition, the communication control device of the present invention may further include a communication control unit configured to change the communication speed set for the communication terminal via a core network of a specified communication standard based on the value of the total data communication volume output by the calculation unit.

[0011] In addition, in the communication control device according to the present invention, the machine learning model may be a neural network having an input layer, a hidden layer, and an output layer.

[0012] In order to solve the above-mentioned problems, the communication control method of the present invention includes a first step of collecting a data communication volume used by a communication terminal for each preset first period, a second step of providing the data communication volume collected for each first period as an unknown input to a trained machine learning model, performing calculations on the trained machine learning model, and outputting a total data communication volume used by the communication terminal for a second period that is longer than the first period, and a third step of presenting information related to the total data communication volume output in the second step.

[0013] In addition, the communication control method of the present invention further includes a fourth step of learning, using a machine learning model, the relationship between the value for the data communication volume used by the communications terminal for each first period and the total data communication volume used by the communications terminal in the second period, and a fifth step of storing the trained machine learning model constructed in the fourth step in a memory unit, and the second step may read out the trained machine learning model from the memory unit and perform calculations on the trained machine learning model.

[0014] In addition, the communication control method of the present invention may further include a sixth step of changing the communication speed set for the communication terminal through a core network of a predetermined communication standard, based on the value of the total data communication volume output in the second step. Effect of the Invention

[0015] According to the present invention, the data communication amount collected for each first period is given as an unknown input to a trained machine learning model, and the trained machine learning model is operated to output the total data communication amount used by the communication terminal in a second period that is longer than the first period. In this way, the total data communication amount is inferred, which leads to effective use of wireless resources. [Brief description of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing a configuration of a communication control system including a communication control device according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram for explaining an overview of the communication control device according to the present embodiment. [Diagram 3] FIG. 3 is a block diagram showing a hardware configuration of the communication control device according to the present embodiment. [Figure 4] FIG. 4 is a diagram for explaining the learning process performed by the learning unit according to the present embodiment. [Diagram 5] FIG. 5 is a flowchart showing the operation of the communication control device according to the present embodiment. [Figure 6] FIG. 6 is a flowchart showing the operation of the communication control device according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] BEST MODE FOR CARRYING OUT THE PRESENT DISCLOSURE Hereinafter, preferred embodiments of the present invention will be described in detail with reference to FIGS.

[0018] [Communication control system configuration] First, an overview of a communication control system including a communication control device 1 according to an embodiment of the present invention will be described. As shown in Fig. 1, the communication control system includes a communication control device 1, multiple communication terminals 2a, 2b, and 2c, a base station 3, and a core network 4.

[0019] The communication control system estimates, for example, the total data communication volume of each of the communication terminals 2a, 2b, and 2c for one month by calculation of a trained machine learning model based on the data communication volume used by each of the communication terminals 2a, 2b, and 2c collected in a preset period, for example, every day. Furthermore, the communication control system changes the communication speed in the data communication of each of the communication terminals 2a, 2b, and 2c based on the estimated total data communication volume for one month.

[0020] As shown in FIG. 1, the communication terminals 2a, 2b, and 2c are realized by a mobile communication terminal such as a smartphone equipped with a SIM, a PDA (Personal Digital Assistant), a tablet computer, or a laptop computer (a so-called notebook computer).

[0021] In this embodiment, each user of the communication terminals 2a, 2b, and 2c has a contract with the same communication carrier, and each of the communication terminals 2a, 2b, and 2c is uniquely identified by an IMSI (International Mobile Subscriber Identity), which is a subscriber identification number that identifies a SIM. Each user also has a contract with one of the data communication plans provided by the communication carrier. The communication carrier is a mobile communication operator that has a core network 4 such as a line network and communication facilities, a base station 3, and the like, and provides communication services such as data communication by lending lines to users.

[0022] Each of the communication terminals 2a, 2b, and 2c can perform high-speed data communication within the monthly upper limit of data communication volume according to the data communication plan subscribed to by each user. For example, the communication terminal 2a is subscribed to a data communication plan that allows data communication up to 20 GB per month. Also, for example, the communication terminals 2b and 2c are subscribed to a data communication plan that allows unlimited data communication volume per month.

[0023] The communication terminal 2a can perform data communication at high speed (e.g., communication speed of 220 Mbps) up to the monthly data communication limit of 20 GB, and when the 20 GB limit is reached, the communication speed is limited from the next day until the end of the month, and data communication is switched to low speed communication (e.g., communication speed of 128 Kbps).The communication terminals 2b and 2c can perform data communication at high speed with no upper limit on the data communication amount.

[0024] In the communication terminals 2a, 2b, and 2c, in the case where the monthly data communication volume upper limit has not been reached, data communication is performed at the communication speed for high-speed communication as a rule. In this embodiment, the communication bandwidth secured by the communication carrier for the entire communication network is calculated by multiplying the total number of subscribers, which is the number of communication terminals 2, by the communication speed during high-speed communication (for example, a communication speed of 220 Mbps). In the following, when the communication terminals 2a, 2b, and 2c are not to be distinguished from each other, they may be collectively referred to as the communication terminals 2.

[0025] The base station 3 is composed of a wireless base station compatible with LTE wireless communication and a wireless base station compatible with 5G. The base station 3 relays communication between the communication terminals 2a, 2b, and 2c located within its coverage area and the core network 4.

[0026] The core network 4 is composed of 5G SA (Stand Alone) and NSA (Non Stand Alone) core networks. The core network 4 has functions to process signals such as location registration signals and data signals transmitted by the communication terminals 2a, 2b, and 2c via the base station 3, to manage communication protocols and security functions, and to ensure communication quality and safety such as communication speed.

[0027] In the core network 4, each device constituting the NSA architecture and the SA architecture is connected via a network not shown. In the example shown in Fig. 1, the core network 4 includes a PGW-U (Packet Data Network Gateway) 40 as the NSA architecture, and other devices are omitted. In addition, as shown in Fig. 1, the core network 4 includes a UPF (User Plane Function) 41 as the SA architecture constituting the core network 4.

[0028] The PGW-U 40 and the UPF 41 are connected to the external Internet and perform processing related to the user plane function for transferring user data packets.

[0029] The PGW-U 40 also has a function of changing the communication speed for each IMSI based on the "PCC (Policy and Charging Control) Rules" defined in 3GPP TS 23.503. The PGW-U 40 counts and holds the amount of data communication used, using the IMSI of the communication terminal 2 as a key. In this embodiment, the PGW-U 40 includes an interface 40a that communicates with the communication control device 1.

[0030] The UPF 41, like the PGW-U 40, has a function of changing the communication speed for each IMSI based on the "PCC (Policy and Charging Control) Rules" defined in 3GPP TS 23.503. The UPF 41 counts and holds the amount of data communication used by using the IMSI of the communication terminal 2 as a key. In this embodiment, the UPF 41 includes an interface 41a that communicates with the communication control device 1.

[0031] [Function block of communication control device] The communication control device 1 includes a collection unit 10, a learning unit 11, a calculation unit 12, a machine learning (ML) model storage unit 13, a communication control unit 14, and a presentation unit 15. The communication control device 1 infers a total data communication volume used by the communication terminal 2, and changes a communication speed for data communication of the communication terminal 2 based on the inferred total data communication volume.

[0032] The collection unit 10 collects the data communication volume used by the communication terminal 2 for each preset first period. The length of the first period can be a period based on any collection cycle, such as one month, one day, or one hour. For example, when the communication control device 1 infers the data communication volume used by the communication terminal 2 in one day, the first period can be one hour, and the data communication volume used by the communication terminal 2 can be collected for each hour. Alternatively, when inferring the data communication volume on a monthly basis, the first period can be one day or one hour, and the data communication volume used by the communication terminal 2 can be collected for each day or each hour.

[0033] In this embodiment, as an example, a case will be described in which the collection unit 10 collects the amount of data communication used by the communication terminal 2 in a one-day cycle, with the first period being one day. The amount of data communication collected by the collection unit 10 for each first period is the cumulative value of the amount of data communication used by the communication terminal 2 up to the time of collection. Therefore, the value of the amount of data communication collected on the kth (k=1, 2, 3, . . .) day from the start of collection is the cumulative value of the amount of data communication used by the communication terminal 2 from the start of collection to the kth day.

[0034] The collection unit 10 collects, from the PGW-U 40 and the UPF 41 of the core network 4, values ​​of used data communication amounts stored in association with the IMSIs of the communication terminals 2a, 2b, and 2c.

[0035] The learning unit 11 uses a machine learning model to learn the relationship between the value of the data communication volume used by the communication terminal 2 for each first period (e.g., one day) and the total data communication volume used by the communication terminal 2 in a second period (e.g., one month) that is longer than the first period.

[0036] FIG. 2 is a diagram showing a schematic diagram of the amount of data communication used by the communication terminal 2 collected by the collection unit 10 for each first period (e.g., each day). The horizontal axis of FIG. 2 indicates the number of times X that the amount of data communication used by the communication terminal 2 has been collected. The vertical axis indicates the cumulative value Y of the amount of data communication used by the communication terminal 2. Each black dot indicates the cumulative value of the amount of data communication collected for each day. Note that FIG. 2 shows the amount of data communication used by one communication terminal 2.

[0037] The learning unit 11 causes the machine learning model to learn the relationship between the cumulative value of the amount of data communication used by the communication terminal 2 for each first period (e.g., one day) and the total amount of data communication M during a second period (e.g., one month) actually used by the communication terminal 2. The curve shown in Fig. 2 is training data indicating the cumulative value Y of the amount of data communication actually used by the communication terminal 2, and the learning unit 11 performs training using the training data expressed by the following formula (1). Y = (M / N) × X (1)

[0038] In the above formula (1), N indicates the number of times the data communication amount has been collected up to the second period in which the total data communication amount M has been used, and for example, in the case of the number of times of collection per day for one month, N=30 (times). The learning unit 11 uses the teacher data expressed by the above formula (1) to learn the learning parameters of the machine learning model so that the data communication amount used by the communication terminal 2 collected per first period (for example, per day) becomes the value of the data communication amount expressed by the above formula (1).

[0039] 4 shows a neural network structure that is adopted as an example of a machine learning model for learning by the learning unit 11. The neural network includes an input layer x, a hidden layer h, and an output layer y. In the example of FIG. 4, the input node x k and the output node y k These correspond to the number of data items of the data traffic collected by the collection unit 10.

[0040] As described in FIG. 2, the amount of data communication already used by the communication terminal 2 collected on the first day is input to the input node x1. Similarly, the amount of data communication already used by the communication terminal 2 collected on the first day is input to the input nodes x2 to x k In each of the above, the value of the used data communication amount of the communication terminal 2 collected on the second day to the kth day is input. k and N have a relationship of k≦N. In other words, when estimating the total data communication amount for one month, the daily data communication amount collected 30 times or less is used as input data.

[0041] The learning unit 11 provides the daily cumulative value of the amount of data communication used by the communication terminal 2 to the input layer of the neural network, applies an activation function to the weighted sum of the inputs, and passes the output determined by threshold processing to the output layer. In the example of Fig. 4, the output layer is provided with output nodes whose number corresponds to the number of input nodes.

[0042] The size and elements of the machine learning model, including the number of layers in the hidden layer h and the density of the connections between nodes in the neural network, are not limited as long as the design provides sufficient inference accuracy. For example, the connections between nodes may be fully connected or sparsely connected.

[0043] As shown in Fig. 4, the input nodes and output nodes of the input layer correspond, from top to bottom, to the data traffic of the communication terminal 2 collected for each day from the first day to the kth day. As described in Fig. 2, the data traffic of the communication terminal 2 collected on the first day is input to the input node x1. Similarly, the input nodes x2 to x k In each of the above, the value of the data traffic of the communication terminal 2 collected from the second day to the kth day is input.

[0044] The learning unit 11 introduces an objective function E shown in the following equation (2) to learn the parameters of the neural network so that the value of the data communication volume used by the communication terminal 2 for a first period (e.g., one day) becomes the correct label value of the data communication volume for each day based on the total data communication volume actually used in a second period (e.g., one month).

[0045]

number

[0046] In the above formula (2), y1, y2, , y k are the output values ​​of each output node, Z1, Z2, , Z k indicates the correct label. Z k is expressed by the above formula (1). The value of the objective function E is calculated by the above input values ​​x1, x2, . . . , x k , that is, the output values ​​y1, y2, . . . , y k The target outputs Z1, Z2, . . ., Z k When the objective function E coincides with the given value, the result is 0. The learning unit 11 adjusts the weight parameters of the neural network so that the objective function E is minimized, that is, becomes 0.

[0047] The learning unit 11 can optimize the objective function E by a gradient method using backpropagation, stochastic gradient descent, or the like. Furthermore, the learning unit 11 can also adjust the number k of input nodes. The learning unit 11 adjusts weight parameters of the neural network so that the objective function E is minimized in a network structure having the number k of input nodes that is optimal for finding the total data usage amount in the second period.

[0048] The learning unit 11 can be configured to learn a machine learning model common to all communication terminals 2a, 2b, and 2c. In this case, learning can be performed using teacher data in which the total data communication amount actually used by each of the communication terminals 2a, 2b, and 2c in one month is used as a correct answer label for the input of the data communication amount of each of the communication terminals 2a, 2b, and 2c for k days. In this way, the learning unit 11 can learn the neural network using learning data related to the data communication amount of many users.

[0049] Alternatively, the learning unit 11 may be configured to learn parameters of a machine learning model for each communication terminal 2. Alternatively, the learning unit 11 may be configured to learn parameters of a machine learning model for each of a plurality of communication terminals 2 that are grouped in advance. The learned parameters constructed by the learning unit 11 are stored in the ML storage unit 13, which will be described later.

[0050] Returning to Figure 1, the calculation unit 12 provides the data communication volume collected for each day, which is a preset first period, as an unknown input to the trained machine learning model, performs calculations on the trained machine learning model, and outputs the total data communication volume used by the communication terminal 2 in a second period, which is a period longer than the first period.

[0051] The calculation unit 12 loads the learned parameters from the ML storage unit 13, and calculates the input data x1 to x2 of the data traffic used by the communication terminal 2, which is collected by the collection unit 10 for the first period, i.e., for each day. k As an unknown input, the calculation unit 12 infers the total data traffic used by the communication terminal 2 during the second period of one month. Using the example of FIG. 4, the calculation unit 12 infers the total data traffic used by the communication terminal 2 during the second period of one month using input data x1 to x2 of the data traffic of the communication terminal 2 for each k days obtained by calculation of the trained neural network. k Output data y1~y k Based on this, we infer the total monthly data traffic.

[0052] For example, when estimating the total data communication volume for one month (30 days) where N=30, the calculation unit 12 calculates y1 to y k The calculation unit 12 calculates the average value of the total data communication amount for one month based on the values ​​of N and Y from the following formula (3). N can be calculated. Y N =[{(y1 / 1)×N}+{(y2 / 2)×N}+{(y3 / 3)×N}+,...{(y k / k)×N}] / k ···(3)

[0053] The calculation unit 12 calculates the total data traffic by using the average value of k pieces of output data according to the above formula (3). For example, (y k This makes it possible to reduce the variation in values ​​compared to when the total data traffic volume is calculated based only on the data traffic volume for one day, such as by using the formula (1 / k)×N.

[0054] The communication control unit 14 changes the communication speed set for the communication terminal 2 via the core network 4 of a predetermined communication standard, based on the value of the total data communication volume for one month output by the calculation unit 12. Specifically, the communication control unit 14 can transmit a control signal for changing the communication speed setting of each communication terminal 2 to the PGW-U 40 and UPF 41 of the core network 4 via the network NW.

[0055] More specifically, the communication control unit 14 transmits a control signal for changing the communication speed for each IMSI to the PGW-U 40, which has a function of changing the communication speed on an IMSI basis, and the UPF 41, based on the "PCC (Policy and Charging Control) Rules" defined in 3GPP TS 23.503.

[0056] For example, assume that the estimated total data communication volume for one month is 19 GB for a communication terminal 2a with a monthly data communication volume upper limit of 20 GB. In this case, the communication control unit 14 can transmit a control signal to the PGW-U 40 and the UPF 41 to change the communication speed of the communication terminal 2a from 220 Mbps to 209 Mbps ({220 Mbps / 20 GB}×19 GB).

[0057] As another example, if the estimated total data communication amount of the communication terminal 2a for one month is 9 GB, the communication control unit 14 can similarly transmit a control signal to the PGW-U 40 and the UPF 41 to change the communication speed of the communication terminal 2a from 220 Mbps to 99 Mbps ({220 Mbps / 20 GB}×9 GB). The communication control unit 14 can also change the communication speed for the communication terminals 2b and 2c, which have unlimited monthly data communication amounts. The communication control unit 14 can change the communication speed appropriately from the viewpoint of communication quality, etc.

[0058] The ML model storage unit 13 stores the trained parameters of the machine learning model constructed by the learning unit 11 and trained.

[0059] The presentation unit 15 presents information related to the total data communication volume of the communication terminal 2 inferred by the calculation unit 12. The presentation unit 15 can also present information related to setting information of the communication speed of the communication terminal 2 controlled by the communication control unit 14.

[0060] [Hardware configuration of communication control device] Next, an example of a hardware configuration for implementing the communication control device 1 having the above-mentioned functions will be described with reference to FIG.

[0061] As shown in FIG. 5, the communication control device 1 can be realized, for example, by a computer having 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.

[0062] 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 realize the functions of the communication control device 1, such as the collection unit 10, the learning unit 11, the calculation unit 12, and the communication control unit 14 shown in FIG.

[0063] The communication interface 104 is an interface circuit for network connection between the communication control device 1 and various external electronic devices.

[0064] 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.

[0065] The auxiliary storage device 105 has a program storage area for storing a communication control program executed by the communication control device 1. The auxiliary storage device 105 also has an area for storing a learning program for learning a machine learning model. The auxiliary storage device 105 realizes the ML model storage unit 13 described in FIG. 1.

[0066] The auxiliary storage device 105 also has an area for storing information on the upper limit of monthly data traffic corresponding to the communication plans subscribed to by the communication terminals 2a, 2b, and 2c, and may further have, for example, a backup area for backing up the above-mentioned data and programs.

[0067] 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.

[0068] [Operation of communication control device] Next, the operation of the communication control device 1 having the above-mentioned configuration will be described with reference to the flowcharts of Fig. 5 and Fig. 6. Fig. 5 is a flowchart showing a learning process by the communication control device 1. Fig. 6 is a flowchart showing a communication process for changing the communication speed of the communication terminal 2 by calculation of a trained machine learning model.

[0069] First, as shown in Fig. 5, the learning unit 11 sets the number k of input nodes of the machine learning model (step S1). Specifically, the learning unit 11 adopts a neural network as the machine learning model, and sets the settings of the input layer, hidden layer, and output layer, as well as the initial values ​​of weight parameters, thresholds, and other parameters. The learning unit 11 can adopt the number k of nodes of the input layer that can obtain sufficient accuracy in the inferred data communication amount.

[0070] Next, the learning unit 11 learns the relationship between the value of the data communication amount used by the communication terminal 2 for each first period (e.g., one day) and the total data communication amount used by the communication terminal 2 for each second period (e.g., one month) using a machine learning model (step S2). Specifically, the learning unit 11 can learn the machine learning model using teacher data in which the total data communication amount used for one month collected by the collection unit 10 is attached as a correct answer label to the value of the data communication amount used by the communication terminal 2 collected by the collection unit 10 for each day.

[0071] The learning unit 11 can employ, for example, a machine learning model having a neural network structure as shown in FIG. 4, and learn weight parameters of the neural network model using the teacher data expressed by the above formula (1).

[0072] In addition, the learning unit 11 uses the above formula (2) to calculate the k The weight parameters that minimize the objective function E are repeatedly adjusted and updated to determine the values ​​of the weight parameters so that the values ​​of the data communication volume of the communication terminal 2 from the 1st day to the kth day inputted into the parameter table become a cumulative value of the data communication volume for each day based on the value of the total data communication volume actually used by the communication terminal 2 in one month. The learning unit 11 adjusts the number k of input nodes that minimizes the objective function E, and repeatedly performs learning processing to determine the optimal number k of input nodes.

[0073] The learning unit 11 is capable of training a machine learning model that is common to all communication terminals 2, but depending on the design, it may also train a machine learning model that is common to a specified number of communication terminals 2, or train a machine learning model for each communication terminal 2.

[0074] Thereafter, the ML model storage unit 13 stores the learned weight parameters obtained by the learning process in step S2 and the number k of input nodes (step S3).

[0075] Next, a control process of the communication speed of the communication terminal 2 using the trained neural network will be described with reference to Fig. 6. First, the collection unit 10 collects the cumulative value of the amount of data communication used by the communication terminal 2 for each first period (e.g., each day) (step S10). Specifically, the collection unit 10 collects the count value of the amount of data communication for each day that is stored in association with the IMSI assigned to the communication terminal 2 via the interface 40a of the PGW-U 40 and the interface 41a of the UPF 41 of the core network 4.

[0076] Next, the calculation unit 12 loads the learned parameters from the ML storage unit 13 (step S11). Next, the calculation unit 12 provides the accumulated value of the data communication amount used by the communication terminal 2 for each first period (e.g., one day) collected by the collection unit 10 as an unknown input to the trained neural network, and performs calculation of the trained neural network (step S12). The calculation unit 12 outputs the total data communication amount used by the communication terminal 2 for a second period (e.g., one month) by the calculation in step S12 (step S13).

[0077] Next, the communication control unit 14 changes the communication speed set in the communication terminal 2 based on the total data communication volume output in step S13 (step S14). Specifically, the communication control unit 14 compares the inferred value of the total data communication volume used in the communication terminal 2 for one month output in step S13 with the upper limit value of the monthly data communication volume of the communication terminal 2, and if the value of the total data communication volume is smaller than the upper limit value, the communication speed for data communication of the communication terminal 2 can be reduced to be lower than the communication speed during high-speed communication. The communication control unit 14 transmits a control signal to the PGW-U 40 and UPF 41 of the core network 4 to change the communication speed for each communication terminal 2.

[0078] Thereafter, the presentation unit 15 presents control information on the communication speed of the communication terminal 2 by the communication control unit 14 (step S15).

[0079] In the above embodiment, the case where the communication control system complies with LTE and 5G is illustrated. However, the predetermined communication standard is not limited to LTE and 5G, and may be a communication control system that complies with 6G.

[0080] In the above embodiment, the learning unit 11 performs the learning process using a neural network as a machine learning model. However, the machine learning model may be a neural network model as described above, or a logistic regression, a random forest, a decision tree, or a deep learning model in which a neural network is multi-layered. In addition to these supervised learning models, a generative model such as a generative adversarial network or a variational autoencoder may be used as a machine learning model for performing unsupervised learning.

[0081] In the embodiment described above, the learning unit 11 that performs the learning process, the calculation unit 12 that calculates the total data communication volume based on the learned machine learning model, and the communication control unit 14 that changes the communication speed of the communication terminal 2 based on the total data communication volume are all mounted on the communication control device 1. However, in addition to the case where the learning unit 11, the calculation unit 12, and the communication control unit 14 are provided as the same hardware configuration, the learning process and the communication control process can be distributed by a plurality of servers or the like on the network NW.

[0082] As described above, according to the communication control device 1 of the present embodiment, the data communication volume of the communication terminal 2 collected for each first period is given as an unknown input to the trained machine learning model, and the trained machine learning model is operated to output the total data communication volume used by the communication terminal 2 in the second period that is longer than the first period. Therefore, based on the inferred total data communication volume, the communication speed is changed for users who do not use the data communication volume up to the upper limit, and wireless resources can be effectively utilized.

[0083] The above describes the embodiments of the communication control device and communication control method of the present invention. However, the present invention is not limited to the described embodiments, and various modifications that can be envisioned by a person skilled in the art can be made within the scope of the invention described in the claims. [Explanation of symbols]

[0084] 1...communication control device, 10...collection unit, 11...learning unit, 12...calculation unit, 13...ML model memory unit, 14...communication control unit, 15...presentation unit, 2, 2a, 2b, 2c...communication terminal, 3...base station, 4...core network, 40...PGW-U, 40a, 41a, ...interface, 41...UPF, 101...bus, 102...processor, 103...main memory device, 104...communication interface, 105...auxiliary memory device, 106...input / output I / O, NW...network.

Claims

1. A collection unit configured to collect a data communication amount used by the communication terminal for each preset first period; a calculation unit configured to provide the data communication amount collected for each first period as an unknown input to a trained machine learning model, perform a calculation on the trained machine learning model, and output a total data communication amount used by the communication terminal during a second period that is longer than the first period; a presentation unit configured to present information regarding the total data communication volume output by the calculation unit; A communication control device comprising:

2. 2. The communication control device according to claim 1, a learning unit configured to learn, using a machine learning model, a relationship between a value of the data communication amount used by the communication terminal for each first period and the total data communication amount used by the communication terminal in the second period; A storage unit configured to store the trained machine learning model constructed by the learning unit; Equipped with The calculation unit reads out the trained machine learning model from the storage unit and performs calculations on the trained machine learning model. A communication control device comprising:

3. 3. The communication control device according to claim 2, a communication control unit configured to change a communication speed set for the communication terminal via a core network of a predetermined communication standard based on the value of the total data communication volume output by the calculation unit. A communication control device comprising:

4. 4. The communication control device according to claim 3, The machine learning model is a neural network having an input layer, a hidden layer, and an output layer. A communication control device comprising:

5. A first step of collecting a data communication amount used by a communication terminal for each preset first period; a second step of providing the data communication amount collected for each of the first time periods as an unknown input to a trained machine learning model, performing a calculation on the trained machine learning model, and outputting a total data communication amount used by the communication terminal during a second time period that is longer than the first time period; a third step of presenting information regarding the total data communication amount output in the second step; A communication control method comprising:

6. 6. The communication control method according to claim 5, and a fourth step of learning, using a machine learning model, a relationship between a value of the data communication amount used by the communication terminal for each of the first time periods and the total data communication amount used by the communication terminal in the second time period. A fifth step of storing the trained machine learning model constructed in the fourth step in a storage unit; Equipped with The second step reads out the trained machine learning model from the storage unit and performs a calculation on the trained machine learning model. A communication control method comprising:

7. 7. The communication control method according to claim 6, and a sixth step of changing a communication speed set for the communication terminal via a core network of a predetermined communication standard based on the value of the total data communication amount output in the second step. A communication control method comprising:

Images