Communication control device, communication control method and communication control program

Abstract

A communication control device (100) comprises an acquisition unit (131), an identification unit (133), and a base station control unit (134). The acquisition unit (131) acquires, as learning data, first wireless communication information obtained by communication between a base station (20) included in a prescribed service area and a terminal device (30), the first wireless communication information including at least power information relating to the transmission power applied by the terminal device (30) in the uplink and interference information relating to interference received by the base station (20) from the terminal device (30). On the basis of a model trained using the learning data and second wireless communication information corresponding to the first wireless communication information, the identification unit (133) identifies, as an optimization condition for optimizing the interference received from the terminal device (30) by the base station (20) included in the prescribed service area, an optimization condition which includes a terminal condition to be used as a condition for a terminal device (30) to be controlled for the uplink transmission power thereof and a power condition to be used as a condition for the value of the transmission power. The base station control unit (134) controls the base station serving as a connection destination of the terminal device (30) to be controlled which is in a state of satisfying the terminal condition, so as to cause the terminal device (30) to be controlled to output transmission power at the power value of the power condition.

Description

COMMUNICATION CONTROL DEVICE, COMMUNICATION CONTROL METHOD, AND COMMUNICATION CONTROL PROGRAM

[0001] The present invention relates to a communication control device, a communication control method, and a communication control program.

[0002] Recently, with the widespread use of smartphones and tablet devices, wireless communication systems (wireless cellular network systems) have been established in various locations. In wireless communication systems, signal transmission in a wireless communication system may interfere with communications in other wireless communication systems, and methods for suppressing the effects of interference have been proposed.

[0003] Japanese Patent Application Laid-Open No. 2023-058263

[0004] Here, it is possible to suppress interference by controlling the radio wave strength (output level), i.e., transmission power, of the signal transmitted on the downlink (downlink communication) from the base station to the terminal device and the uplink (uplink communication) from the terminal device to the base station.

[0005] However, currently, the specifications require that each base station has an algorithm for controlling the transmission power of terminal devices under its control.

[0006] In this way, when each base station individually controls transmission power, for example, a portion of a transmission signal from a terminal device under one base station may reach another base station as an interference signal (interference wave), degrading the communication quality of the other base station. The other base station, detecting a degradation in communication quality, instructs the terminal device under its control to increase its transmission power in order to improve communication quality. The terminal device receiving the instruction increases its transmission power and transmits a signal, and a portion of this transmission signal reaches the first base station as an interference signal. This can create a vicious cycle in which the first base station also instructs the terminal device under its control to increase its transmission power, generating further interference signals.

[0007] In this way, the base station specifications are such that each base station controls the terminal devices under its control according to the reception quality of the uplink transmission signal, and therefore current wireless communication systems do not take into consideration how to control the interference that the terminal devices under its control cause to other base stations.In order to suppress the interference that the terminal devices cause to other base stations and eliminate the problem of a vicious cycle, it is thought that area-wide optimization is required, in which the transmission power is optimized among multiple base stations belonging to the area, rather than requesting control of the transmission power only for a specific base station.

[0008] However, because the above-mentioned conventional technology identifies a frequency band with low interference and causes a terminal device to use the identified frequency band, there remains the possibility that the above-mentioned vicious cycle problem will occur. In other words, the above-mentioned conventional technology does not necessarily enable optimization of transmission power for each predetermined service area for base stations belonging to the predetermined service area.

[0009] Therefore, the present invention proposes a communication control device, a communication control method, and a communication control program that can optimize the transmission power of base stations belonging to a predetermined service area in units of a predetermined service area. As will be described later, the communication control device may be implemented as, for example, a RAN Intelligent Controller (RIC).

[0010] In order to solve the above problem, one embodiment of the communication control device according to the present invention includes an acquisition unit that acquires as learning data first wireless communication information obtained by communication between a base station included in a specified service area and a terminal device, the first wireless communication information including at least power information regarding the transmission power applied by the terminal device in an uplink and interference information regarding interference received by the base station from the terminal device; an identification unit that identifies, based on a model learned using the learning data and second wireless communication information corresponding to the first wireless communication information, optimization conditions that can optimize interference received from the terminal device between the base stations included in the specified service area, the optimization conditions including a terminal condition that conditions a terminal device to be controlled that controls the transmission power of the uplink and a power condition that conditions the value of the transmission power; and a base station control unit that controls the base station to which the terminal device to be controlled is connected, so that the terminal device to be controlled, in a situation where the terminal condition is satisfied, outputs the transmission power at the power value of the power condition.

[0011] According to the present invention, it is possible to optimize the transmission power of base stations belonging to a predetermined service area over the entire predetermined service area.

[0012] FIG. 1 is a diagram for explaining the problems underlying the present invention. FIG. 2 is an explanatory diagram showing an example of the main configuration of a communication system including a communication control device according to an embodiment. FIG. 3 is a diagram conceptually showing advantages of installing a communication control device in a regional center. FIG. 4 is a diagram showing the configuration of an AI-RAN. FIG. 5 is a diagram showing an example of the configuration of a communication control device according to an embodiment. FIG. 6 is a diagram for explaining a learning method according to an embodiment. FIG. 7 is a sequence diagram showing the procedure of a communication control process according to an embodiment. FIG. 8 is a hardware configuration diagram showing an example of a computer that realizes the functions of a communication control device according to an embodiment.

[0013] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0014] One or more embodiments (including examples, modifications, and application examples) described below can be implemented independently. However, at least a portion of the embodiments described below may be implemented in appropriate combination with at least a portion of another embodiment. These embodiments may include novel features that are different from one another. Therefore, these embodiments may contribute to solving different purposes or problems and may produce different effects.

[0015] (Embodiment) [1. Introduction] As described above, when each base station has an algorithm for controlling the transmission power of terminal devices under its control, a vicious cycle may occur in which base stations (e.g., adjacent base stations) that detect degradation in communication quality due to the influence of an interference signal repeatedly issue instructions to increase the transmission power of terminal devices under their control. This point will be explained using Figure 1. Figure 1 is a diagram explaining the issues behind the present invention.

[0016] 1 shows base station gNB1 and base station gNB2 as examples of base stations for the fifth generation (5G), and these base stations are adjacent to each other. In other words, the cell formed by base station gNB1 and the cell formed by base station gNB2 are adjacent to each other. In this example, from the perspective of base station gNB1, base station gNB2 is a base station of an adjacent cell, i.e., an adjacent base station to the base station itself. Also, from the perspective of base station gNB2, base station gNB1 is also a base station of an adjacent cell, i.e., an adjacent base station to the base station itself.

[0017] 1 also shows a terminal device UE1 as an example of a terminal device connected to base station gNB1 (under the control of base station gNB1), and a terminal device UE2 as an example of a terminal device connected to base station gNB2 (under the control of base station gNB2). The terminal device UE1 is located in a cell formed by base station gNB1, and the terminal device UE2 is located in a cell formed by base station gNB2.

[0018] The terminal device UE1 transmits a signal on an uplink channel (Physical Uplink Shared Channel: PUSCH), and at this time, transmits the signal to the base station gNB1 at a transmission power (output level) according to an instruction from the base station gNB1. The terminal device UE2 also transmits a signal on an uplink channel, and at this time, transmits the signal to the base station gNB2 at a transmission power (output level) according to an instruction from the base station gNB2.

[0019] In the following embodiments, a signal transmitted on an uplink channel may be referred to as an "uplink transmission signal" (or simply as a "transmission signal" or an "uplink signal"). Also, the strength of a radio wave applied in the uplink may be referred to as "uplink transmission power" (or simply as "transmission power").

[0020] Here, according to FIG. 1(a), a portion of the uplink transmission signal corresponding to the terminal device UE1 arrives at the base station gNB2 as an interference signal, and a portion of the uplink transmission signal corresponding to the terminal device UE2 arrives at the base station gNB1 as an interference signal. In this example, the base station gNB1 detects degradation of reception quality due to the interference signal originating from the terminal device UE2, and instructs the terminal device UE1 to increase transmission power to improve reception quality. Also, the base station gNB2 detects degradation of reception quality due to the interference signal originating from the terminal device UE1, and instructs the terminal device UE2 to increase transmission power to improve reception quality. This creates a vicious cycle in which the base stations gNB1 and gNB2 repeatedly issue instructions to increase transmission power to each other. Details are explained in FIG. 1(b).

[0021] For example, suppose that the terminal device UE2 transmits an uplink signal to the base station gNB2 at a power instructed by the base station gNB2 (S1). Furthermore, suppose that the base station gNB2 detects a deterioration in the reception quality of the uplink signal when receiving the uplink signal from the terminal device UE2 (S2). In this case, the base station gNB2 instructs the terminal device UE2 to increase the uplink transmission power (S3).

[0022] The terminal device UE2 transmits an uplink signal to the base station gNB2 at the specified power (S4), but a part of the uplink signal at this time reaches the base station gNB1 as an interference signal as shown in Fig. 1 (b). Since the desired signal (desired wave) for the base station gNB1 is only the uplink signal received from the terminal device UE1 under its control, the uplink signal received from the terminal device UE2 in the adjacent cell can be said to be an interference signal from the base station gNB1.

[0023] According to the example of Fig. 1(b), the terminal device UE1 transmits an uplink signal to the base station gNB1 at the power instructed by the base station gNB1 (S5). However, the reception quality when the base station gNB1 receives the uplink signal from the terminal device UE1 may be deteriorated due to interference from the uplink signal of the terminal device UE2.

[0024] When the base station gNB1 receives an uplink signal from the terminal device UE1 and detects a deterioration in the reception quality of the uplink signal (S6), it instructs the terminal device UE1 to increase the transmission power of the uplink (S7).

[0025] The terminal device UE1 transmits an uplink signal to the base station gNB1 at the specified power (S8), but a part of the uplink signal at this time reaches the base station gNB2 as an interference signal as shown in Fig. 1 (b). Since the desired signal (desired wave) for the base station gNB2 is only the uplink signal received from the terminal device UE2 under its control, the uplink signal received from the terminal device UE1 of the adjacent cell can be said to be an interference signal from the base station gNB2.

[0026] According to the example of Figure 1 (b), the terminal device UE2 transmits an uplink signal to the base station gNB2 at the power instructed by the base station gNB2, but the reception quality when the base station gNB2 receives the uplink signal from the terminal device UE2 may deteriorate due to interference from the transmitted signal of the terminal device UE1.

[0027] Therefore, when the base station gNB2 receives an uplink signal from the terminal device UE2, it again detects deterioration in the reception quality of the uplink signal (S2), and the processes from (S3) onwards are repeated, resulting in a vicious cycle. In other words, the base station gNB1 and the base station gNB2 repeatedly issue instructions to increase the transmission power to each other, which in turn increases interference.

[0028] The inventor of the present invention considered that in order to suppress interference due to uplink transmission signals and eliminate a vicious cycle, a new measure is needed to replace the situation in which the base stations gNB1 and gNB2 individually control the transmission power. Specifically, the inventor of the present invention considered that the base station gNB1 transmits a signal to the terminal device UE1 at what value of transmission power, and the base station gNB2 transmits a signal to the terminal device UE2 at what value of transmission power, do not interfere with each other. The inventor focused on the importance of using machine learning to find an optimal solution, that is, optimization in a predetermined area unit including the base stations gNB1 and gNB2.

[0029] The background art of the present invention has been described above. Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Hereinafter, an embodiment of the present invention will be described on the assumption that it is applied to a 3GPP (registered trademark) LTE / LTE-Advanced wireless communication system and a next-generation NR (New Radio) wireless communication system of the fifth generation or later. However, the concept of the present invention can be applied to any system using a similar configuration.

[0030] A communication control device according to an embodiment described herein acquires, as training data, first wireless communication information obtained through communication between a base station included in a predetermined service area and a terminal device, the first wireless communication information including at least power information regarding transmission power applied by the terminal device in an uplink and interference information regarding interference received by the base station from the terminal device. Based on a model trained using the training data and second wireless communication information corresponding to the first wireless communication information, the communication control device identifies optimization conditions that can optimize interference received from the terminal device between base stations included in the predetermined service area, the optimization conditions including a terminal condition that conditions a target terminal device for controlling uplink transmission power and a power condition that conditions a value of the transmission power. The communication control device controls the base station to which the target terminal device is connected, so that the target terminal device, which is in a situation that satisfies the terminal condition, outputs transmission power at the power value specified by the power condition.

[0031] 2 is an explanatory diagram showing an example of the main configuration of a communication system (wireless cellular network system) Sy including a communication control device 100 according to an embodiment. The communication control device 100 executes communication control processing to optimize transmission power for each predetermined service area.

[0032] In FIG. 2, a predetermined wide area 1A is divided into a plurality of service areas 10A, which are units for optimizing transmission power, and one of these areas is shown as service area 10A(1).

[0033] For example, the wide area 1A may be a prefecture. If the wide area 1A is Tokyo, the service area 10A(1) is a local area included in Tokyo and may correspond to a city, ward, town, or village. For example, the service area 10A(1) may be Shibuya Ward. Although not shown, the wide area 1A also includes a service area 10A(2) and a service area 10A(3), each of which may correspond to a city, ward, town, or village within the wide area 1A.

[0034] In the following, the communication control process of the embodiment will be described focusing on the service area 10A(1), but the same communication control process is also applied to the other service areas 10A included in the wide area 1A.

[0035] According to the example of Figure 2, the service area 10A(1) includes a base station 20(1) and a base station 20(2), with the base station 20(1) forming a cell 20A(1) and the base station 20(2) forming a cell 20A(2). The base stations 20(1) and 20(2) may operate in the same frequency band. Figure 2 shows two base stations, 20(1) and 20(2), as an example of the base stations 20 included in the service area 10A(1), but the number of base stations within the service area 10A(1) is not limited.

[0036] Each of base station 20(1) and base station 20(2) is configured using hardware such as a computer device having a CPU, memory, etc., an external communication interface unit for core network 40, a wireless communication unit, etc., and by executing a predetermined program, it is possible to perform wireless communication with terminal device 30, send and receive information with core network devices of core network 40, and send and receive information with communication control device 100 using a predetermined communication method and wireless communication resources.

[0037] For example, the base station 20 uses radio resources (frequency resources, time resources) allocated in the cell 20A formed by the base station 20 to perform radio communication with the terminal device 30. The base station 20 may also allocate radio resources and manage schedule information indicating the allocation.

[0038] The terminal device 30 is called, for example, user equipment (UE) because it is used by a user of a communication service. Furthermore, since the terminal device 30 is mobile, it may also be called a mobile station or mobile device, or a radio device.

[0039] 2 shows terminal device 30(1) and terminal device 30(2) as examples of terminal devices 30. Terminal device 30(1) is connected to base station 20(1) because it exists within cell 20A(1) formed by base station 20(1). In other words, terminal device 30(1) is under the control of base station 20(1). Terminal device 30(2) is connected to base station 20(2) because it exists within cell 20A(2) formed by base station 20(2). In other words, terminal device 30(2) is under the control of base station 20(2).

[0040] The terminal device 30(1) can perform various communications via the base station 20(1), and the terminal device 30(2) can perform various communications via the base station 20(2). The terminal device 30(1) and the terminal device 30(2) are each configured using hardware such as a computer device having a CPU, memory, etc., and a wireless communication unit, and can perform wireless communications with the base station 20 by executing a predetermined program.

[0041] In this example, the base station 20(1) allocates radio resources to the terminal device 30(1) and transmits schedule information indicating the allocation to the terminal device 30(1). The base station 20(2) allocates radio resources to the terminal device 30(2) and transmits schedule information indicating the allocation to the terminal device 30(2).

[0042] Although FIG. 2 shows an example in which there is one terminal device 30(1) and one terminal device 30(2), the concept of each of the terminal device 30(1) and the terminal device 30(2) includes multiple devices.

[0043] The core network (for example, a 5G core network) 40 is configured with core network devices having various functions (nodes) called network functions. The core network 40 corresponds to a portion that connects the communication system Sy owned by the telecommunications carrier T to the Internet. The telecommunications carrier T here may be a company that provides interference suppression services for each service area 10A using the communication control device 100 according to the embodiment. The core network 40 may be a core network that accommodates base stations 20 included in the wide area 1A (which may also be the service area 10A).

[0044] The communication control device 100 is an information processing device that executes communication control processing according to the embodiment. For example, the communication control device 100 executes communication control processing to optimize transmission power for each service area 10A included in the wide area 1A. Specifically, the communication control device 100 uses a machine learning model to determine an optimal solution for determining, among the terminal devices 30 under the control of the base stations 20 included in the service area 10A, the conditions under which the terminal devices 30 are made to transmit uplink signals and the transmission power at which the communication quality index value can be optimized (enhanced) among all the base stations 20 included in the service area 10A, and controls the base stations 20 so that the terminal devices 30 transmit uplink signals with transmission power corresponding to the determined optimal solution. The communication control device 100 executes such communication control processing for each service area 10A.

[0045] The communication control device 100 may be provided in the core network 40, or may be installed in a remote location such as a data center. For example, the communication control device 100 may be installed in a regional center located in each wide area 1A.

[0046] 3 conceptually illustrates the advantages of installing the communication control device 100 in a regional center. For example, to support a society in which AI is rapidly evolving, it is necessary to build a next-generation social infrastructure that can handle the rapidly increasing demand for data processing and the power required for data processing. Therefore, there is an AI-RAN concept in which a large-scale server group is built in a data center for each region (for example, for each wide area 1A), and vRAN (virtual radio access network), MEC (multi-access edge computing), and AI applications are simultaneously operated and linked on the abundant computing resources.

[0047] FIG. 3 shows a scenario in which a communication control device 100 is applied to the AI-RAN concept. The communication control devices 100 present in each wide area 1A are AI-RANs having a computational platform and a learning platform, and are distributed across regions. In this way, the communication control devices 100 as AI-RANs may be cloud servers distributed as edge servers (also known as MEC servers) near the terminal devices 30, and by utilizing a closed network isolated from the Internet, high speed, large capacity, low latency, etc. can be achieved. Meanwhile, services utilizing data collected by the communication control devices 100 in each location (e.g., large-scale calculations or learning requiring large amounts of power) may be executed on the public cloud 200 via the Internet.

[0048] 3. AI-RAN Configuration Figure 4 is a diagram showing the configuration of an AI-RAN. Figure 4 shows the functional configuration of the AI-RAN possessed by the communication control device 100 corresponding to the wide area 1A. First, AI-RAN is an architecture that allows AI and RAN (base station 20) to coexist, and it can maximize the performance of the RAN using AI, while also realizing an ultra-low latency, highly secure computing infrastructure for various AI applications at the regional level.

[0049] In the example of Fig. 4, the vRAN is a 5G virtualized radio access network in which the GPU of the communication control device 100 virtualizes the RAN, i.e., the base station 20 (the base station 20 included in the wide area 1A). In other words, the communication control device 100 shown in Fig. 4 is configured as an AI-RAN by further implementing a learning platform (AI) in a virtualization platform environment in which the 5G vRAN and MEC are integrated.

[0050] Furthermore, as an example of realizing a computational infrastructure at a regional level, computational resources corresponding to the base stations 20 included in the wide area 1A are further provided.

[0051] The learning base portion of the communication control device 100 may correspond to a learning unit 132 (FIG. 5) described later.

[0052] 4. Configuration of communication control device The communication control device 100 according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a diagram showing an example configuration of the communication control device 100 according to the embodiment. As shown in Fig. 5, the communication control device 100 has a communication unit 110, a storage unit 120, and a control unit 130.

[0053] (Communication Unit 110) The communication unit 110 is realized by, for example, a network interface card (NIC), etc. For example, the communication unit 110 performs wireless communication with the base station 20 and the core network 40.

[0054] (Regarding the storage unit 120) The storage unit 120 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 may store, for example, data and programs related to the communication control process according to the embodiment.

[0055] (Regarding the control unit 130) The control unit 130 is realized by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like using RAM as a work area to execute various programs (for example, a communication control program according to the embodiment) stored in a storage device inside the communication control device 100. The control unit 130 is also realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0056] As shown in Fig. 5, the control unit 130 has an acquisition unit 131, a learning unit 132, an identification unit 133, and a base station control unit 134, and realizes or executes the functions and actions of the information processing described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in Fig. 5, and may be other configurations as long as they perform the information processing described below. Furthermore, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship shown in Fig. 5, and may be other connection relationships.

[0057] (Acquisition unit 131) The acquisition unit 131 acquires learning data. For example, the acquisition unit 131 acquires, as learning data, first wireless communication information obtained by wireless communication between the base station 20 included in the service area 10A and the terminal device 30, the first wireless communication information including at least power information on transmission power applied by the terminal device 30 in an uplink and interference information on interference the base station receives from the terminal device 30.

[0058] The power information may include any of the transmission power received by the base station 20 from the terminal device 30 connected to the base station 20, the remaining transmission power of the terminal device 30, or scheduling information determined by the base station 20 for the terminal device 30 connected to the base station 20. The interference information may include any of the reception power received by the terminal device 30 from other base stations 20 (e.g., neighboring base stations) other than the base station 20 to which the base station 20 is connected, the total transmission power which is the sum of the transmission power received by the base station 20 from terminal devices 30 in various locations, the reception quality of the uplink signal, the amount of interference received by the base station 20 from the terminal device 30, or the error rate corresponding to the uplink.

[0059] (Learning Unit 132) Using the learning data, the learning unit 132 inputs the first wireless communication information and causes a model to learn a pattern that can optimize a predetermined index value related to communication quality determined between the base stations 20 included in the service area 10 A, the index value fluctuating in response to interference. The predetermined index value here may be, for example, a value (e.g., an average value) obtained by statistically processing information on the reception quality of each of the base stations 20 included in the service area 10 A, or may be the frequency utilization efficiency.

[0060] In this embodiment, the trained model (hereinafter referred to as model M) generated by training is realized by a convolutional neural network (CNN), but other neural networks may also be used. Furthermore, model M is a machine learning model generated by training using training data (teacher data) acquired by the acquisition unit 131, and weighting coefficients have been determined.

[0061] For example, the learning unit 132 causes the model to learn patterns of combinations of the status of the terminal device 30 and the uplink transmission power as patterns that can optimize the index value, based on the relationship between the transmission power of each terminal device 30 indicated by the power information and the cause of interference estimated based on the interference information. For example, when second wireless communication information corresponding to the first wireless communication information is input, the learning unit 132 searches for an optimal solution, which is a combination of terminal conditions and power conditions according to the status indicated by the second wireless communication information, based on the pattern, and generates the model M by learning the model to output the search result.

[0062] Here, the acquisition unit 131 may further acquire, as learning data, antenna characteristics of the antennas of the base station 20. In this case, the learning unit 132 may cause the model to learn a pattern that can optimize the index value based on the relationship between the transmission power of each terminal device 30 indicated by the power information, the cause of interference estimated based on the interference information, and the antenna characteristics.

[0063] The acquisition unit 131 may further acquire, as learning data, terminal location information indicating the location of the terminal device 30 and base station location information indicating the installation location of the base station 20. In this case, the learning unit 132 may cause the model to learn a pattern capable of optimizing the index value by further using the transmission power of each terminal device 30 indicated by the power information and the positional relationship between the terminal device 30 that is the interference source estimated from the relationship with the cause of interference estimated based on the interference information and the base station 20 that is receiving interference from the terminal device 30 that is the interference source.

[0064] (Determining unit 133) Based on the model M learned using the learning data and the second wireless communication information corresponding to the first wireless communication information, the determining unit 133 determines optimization conditions that can optimize interference received from the terminal device 30 between the base station 20 included in the service area 10A and the base station 20 included in the service area 10A, including a terminal condition that conditions the terminal device 30 to be controlled for controlling uplink transmission power and a power condition that conditions the value of the transmission power. For example, the determining unit 133 determines optimization conditions based on an output result (search result) when the second wireless communication information is input to the model M. For example, the determining unit 133 determines optimization conditions that determine, among the terminal devices 30 included in the service area 10A, the conditions of the terminal devices 30 and the transmission power at which uplink signals should be transmitted in order to maximize an index value related to communication quality.

[0065] (Base station control unit 134) The base station control unit 134 controls the base station 20 to which the controlled terminal device 30 is connected, so that the controlled terminal device 30, which is in a situation that satisfies the terminal conditions, outputs transmission power at a power value specified in the power conditions. For example, the base station control unit 134 controls the base station 20 to which the controlled terminal device 30 is connected, so that the controlled terminal device 30, which is in a situation that satisfies the terminal conditions, transmits an uplink signal at a power value specified in the power conditions.

[0066] 5. Specific Example of Learning Method Next, a specific example of a learning method will be described with reference to Fig. 6. Fig. 6 is a diagram illustrating a learning method according to an embodiment. Fig. 6 illustrates a scene in which learning is performed using, as learning data, information obtained through communication between a base station 20(1) included in a service area 10A(1) and a terminal device 30(1) under the control of the base station 20(1), and information obtained through communication between a base station 20(2) included in the service area 10A(1) and a terminal device 30(2) under the control of the base station 20(2).

[0067] First, the first wireless communication information obtained by uplink communication between the base station 20(1) and the terminal device 30(1) will be described. The first wireless communication information includes power information X1 and interference information Y1.

[0068] The power information X1 may include the transmission power received by the base station 20(1) from the terminal device 30(1), the transmission power margin of the terminal device 30(1), and the scheduling information transmitted by the base station 20(1) to the terminal device 30(1). The transmission power received by the base station 20(1) from the terminal device 30(1) can be rephrased as the transmission power applied by the terminal device 30(1) when transmitting an uplink signal. From the perspective of the base station 20(1), the uplink signal received from the terminal device 30(1) is a desired signal, but the uplink signal received from the terminal device 30(2) is an interference signal.

[0069] The interference information Y1 may include the received power received by the terminal device 30(1) from the adjacent base station 20(2), the total transmission power which is the sum of the transmission power received by the base station 20(1), the reception quality with which the base station 20(1) receives the uplink signal of the terminal device 30(1), the error rate corresponding to the uplink of the terminal device 30(1), and the amount of interference received by the base station 20(1) from terminal devices 30 throughout the area.

[0070] The received power received by the terminal device 30(1) from the adjacent base station 20(2) is included in a Measurement Report that the terminal device 30(1) transmits to the base station 20(1). The Measurement Report includes information associating a base station ID with an RSRP (Reference Signal Received Power) as information on how much power the terminal device 30(1) received from the base stations 20 other than the connected base station 20(1). In the example of Fig. 6, the Measurement Report includes information associating the base station ID "20(2)" with a value indicating RSRP.

[0071] For example, when the terminal device 30(1) is located near the center of the cell formed by the base station 20(1), the received power from the adjacent base station 20(2) is weak, whereas when the terminal device 30(1) is located near the edge of the cell, the received power from the adjacent base station 20(2) is considered to be strong. For example, when the received power from the terminal device 30(1) is strong, it means that the uplink signal from the terminal device 30(1) is reaching the adjacent base station 20(2) as strong interference. Therefore, the model can estimate that the terminal device 30(1) is the source of interference interfering with the base station 20(2), and as a result, it can learn that the transmission power of the terminal device 30(1) should be reduced. On the other hand, the model can also learn that increasing the transmission power of the terminal device 30(1) located at the edge tends to optimize the index value in units of the service area 10A(1).

[0072] The total transmission power, which is the sum of the transmission power received by the base station 20(1), is a value obtained by aggregating the transmission power received by the base station 20(1) not only from the subordinate terminal device 30(1) but also from the terminal devices 30 in the entire area. In the example of Fig. 6, the total transmission power is the sum of the transmission power of the terminal device 30(1) and the transmission power of the terminal device 30(2). For example, the communication control device 100 can obtain the signal to interference plus noise ratio (SINR) and the reference signal received quality (RSRQ) as the reception quality of the base station 20(1) from the total transmission power included in the interference information Y1.

[0073] Next, the first wireless communication information obtained by uplink communication between the base station 20(2) and the terminal device 30(2) will be described. The first wireless communication information includes power information X2 and interference information Y2.

[0074] The power information X2 may include the transmission power received by the base station 20(2) from the terminal device 30(2), the transmission power margin of the terminal device 30(2), and the scheduling information transmitted by the base station 20(2) to the terminal device 30(2). The transmission power received by the base station 20(2) from the terminal device 30(2) can be rephrased as the transmission power applied by the terminal device 30(2) when transmitting an uplink signal. From the perspective of the base station 20(2), the uplink signal received from the terminal device 30(2) is a desired signal, but the uplink signal received from the terminal device 30(1) is an interference signal.

[0075] The interference information Y2 may include the received power received by the terminal device 30(2) from the adjacent base station 20(1), the total transmission power which is the sum of the transmission power received by the base station 20(2), the reception quality with which the base station 20(2) receives the uplink signal of the terminal device 30(2), the error rate corresponding to the uplink of the terminal device 30(2), and the amount of interference received by the base station 20(2) from terminal devices 30 throughout the area.

[0076] The received power received by the terminal device 30(2) from the adjacent base station 20(1) is included in the Measurement Report that the terminal device 30(2) transmits to the base station 20(2). The Measurement Report includes information associating a base station ID with an RSRP as information on how much power the terminal device 30(2) has received from the base stations 20 other than the connected base station 20(2). In the example of Fig. 6, the Measurement Report includes information associating the base station ID "20(1)" with a value indicating RSRP.

[0077] For example, when the terminal device 30(2) is located near the center of the cell formed by the base station 20(2), the received power from the adjacent base station 20(1) is weak, and when the terminal device 30(2) is located near the edge of the cell, the received power from the adjacent base station 20(1) is considered to be strong. For example, when the received power from the terminal device 30(2) is strong, it means that the uplink signal from the terminal device 30(2) is reaching the adjacent base station 20(1) as strong interference. Therefore, the model can estimate that the terminal device 30(2) is the source of interference causing interference to the base station 20(1), and as a result, it can learn that the transmission power of the terminal device 30(2) should be reduced. On the other hand, the model can also learn that increasing the transmission power of the terminal device 30(2) located at the edge tends to optimize the index value in units of the service area 10A(1).

[0078] The total transmission power, which is the sum of the transmission power received by the base station 20(2), is a value obtained by aggregating the transmission power received by the base station 20(2) not only from the terminal device 30(2) under its control but also from the terminal devices 30 in the entire area. In the example of Fig. 6, the total transmission power is the sum of the transmission power of the terminal device 30(2) and the transmission power of the terminal device 30(1). For example, the communication control device 100 can obtain the SINR or RSRQ as the reception quality of the base station 20(2) from the total transmission power included in the interference information Y2.

[0079] 6, the first wireless communication information including the power information X1 and the interference information Y1 may be acquired from the base station 20(1), and the first wireless communication information including the power information X2 and the interference information Y2 may be acquired from the base station 20(2).

[0080] The learning data may include not only the first wireless communication information but also terminal location information indicating the locations of the terminal devices 30 included in the service area 10A(1). In the example of Fig. 6, the location information of the terminal device 30(1) and the location information of the terminal device 30(2) are used as the learning data. The terminal location information may be GPS location information calculated by GPS positioning and may be acquired via the web browsers of the terminal devices 30(1) and 30(2).

[0081] The training data may also include base station design information. The base station design information may include base station location information indicating the installation locations of the base stations 20 included in the service area 10A(1) and the antenna characteristics of the antennas of the base stations 20. In the example of FIG. 6, the location information of the base station 20(1) and the location information of the base station 20(2) are used as the training data. The antenna patterns, antenna tilt angles, antenna gains, and antenna directivities of the antennas of the base stations 20(1) and 20(2) are also used as the training data. The base station design information may be acquired from a base station design information DB 70. The location of the base station design information DB 70 is not limited. For example, the communication control device 100 may have the base station design information DB 70, or each base station 20 may have its own base station design information DB 70.

[0082] 6 , the first wireless communication information and the terminal location information may be input to a screening device 60 and preprocessed in the screening device 60. The screening device 60 may screen the input first wireless communication information and terminal location information and process them into appropriate training data. In this case, the acquisition unit 131 of the communication control device 100 acquires the learning data preprocessed by the screening device 60 from the screening device 60.

[0083] The learning unit 132 then inputs the preprocessed learning data into a convolutional neural network (CNN) to generate a model M. For example, the learning unit 132 causes the model to learn patterns of combinations of the status of the terminal device 30 and uplink transmission power as patterns that can optimize a predetermined index value related to communication quality, based on the relationships between elements included in the power information (X1, X2), elements included in the interference information (Y1, Y2), terminal position information, base station position information, and antenna position information. For example, when second wireless communication information corresponding to the first wireless communication information is input, the learning unit 132 causes the model to search for an optimal solution, which is a combination of terminal conditions and power conditions according to the status indicated by the second wireless communication information, based on the patterns, and output the search result.

[0084] In this case, the model analyzes patterns of change in how the reception quality of the uplink signals of each terminal device 30 received by the base station 20 changes depending on how the transmission power of the terminal device 30 changes in a certain situation (for example, a positional relationship) relative to a specific base station 20. Then, based on the analysis results, the model searches for a pattern that optimizes the index value.

[0085] For example, the model analyzes various patterns such as, <When the terminal device 30 is in the status "A1" and the uplink transmission power is "A2," if the transmission power of terminal "A4" of the terminal devices 30 in the "A3" state is controlled to "A5," the index value will be maximized in units of the service area 10A(1)>, or, <When the terminal device 30 is in the status "B1" and the uplink transmission power is "B2," if the transmission power of terminal "B3" of the terminal devices 30 is controlled to "B4," the index value will be maximized in units of the service area 10A(1)>. In other words, the model learns combination patterns of terminal conditions that condition the terminal device 30 to be controlled, whose transmission power is to be controlled, and power conditions that condition the value of the transmission power. As a result, based on the situation indicated by the input second wireless communication information and the learned pattern, the model can predict that, in the current situation indicated by the second wireless communication information, for example, if the transmission power of terminal "C4" in the "C3" state is controlled to "C5," the index value will be optimized to the "N1 value." This prediction result may be output to the identification unit 133. The output manner of the prediction result is not limited. In this example, the identification unit 133 may identify "terminal "C4" in the "C3" state" as the terminal condition that conditions the terminal device to be controlled, and "C5" as the power condition that conditions the value of the transmission power. Furthermore, the base station control unit 134 controls the base station 20 connected to the terminal device 30 identified by "C4" so that the terminal device 30 identified by "C4" transmits an uplink signal with its transmission power set to "C5."

[0086] As another example, the learning unit 132 may have a model learn patterns of combinations of the status of the terminal device 30 and the uplink transmission power as patterns that can optimize a predetermined index value related to communication quality, based on the relationship between the transmission power of each terminal device 30 indicated by the power information (X1, X2) and the cause of interference (e.g., the terminal device 30 that is the source of interference) estimated based on the interference information (Y1, Y2).

[0087] In this case, the model analyzes various patterns, such as: "If the interference source among the terminal devices 30 is terminal "D1," controlling the transmission power of terminal "D2" to "D3" maximizes the index value for each service area 10A(1)," or "If the interference source among the terminal devices 30 is terminal "E1," controlling the transmission power of another terminal "E2" in the vicinity of terminal "E1" to "E3" maximizes the index value for each service area 10A(1)." As a result, based on the situation indicated by the input second wireless communication information and the learned pattern, the model can predict that, in the current situation indicated by the second wireless communication information, for example, controlling the transmission power of another terminal "F2" in the vicinity of terminal "F1" to "F3" will optimize the index value to the "N2 value." The model may output this prediction result to the identification unit 133. The output manner of the prediction result is not limited. In this example, the specifying unit 133 may specify <terminal "F2"> as the terminal condition that conditions the terminal device to be controlled, and <"F3"> as the power condition that conditions the value of the transmission power. Furthermore, the base station control unit 134 controls the base station 20 to which the terminal device 30 identified by "F2" is connected, so that the terminal device 30 identified by "F2" transmits an uplink signal with the transmission power set to "F3".

[0088] 7 is a sequence diagram showing the procedure of the communication control process according to the embodiment. Fig. 7 shows a scene of inference processing using the model M learned by the method described in Fig. 6. Fig. 7 also shows a scene in which optimization conditions are specified that can optimize interference received from the terminal device 30 between the base stations 20 included in the service area 10A(1), that is, for each service area 10A(1).

[0089] Although not shown, each base station 20 determines whether or not it is time to perform inference. To determine the timing, time synchronization may be established between the base stations 20. If it is not time to perform inference, each base station 20 waits until it is time to perform inference.

[0090] On the other hand, when it is time to perform inference, each base station 20 inputs second wireless communication information corresponding to the first wireless communication information used when learning the model M to the screening device 60 (step S701). The second wireless communication information input here may be the latest information at the time when it is time to perform inference.

[0091] When the screening device 60 acquires the second wireless communication information (step S702), the screening device 60 executes preprocessing to screen the acquired second wireless communication information and process it into a state suitable for learning (step S703).The screening device 60 then transmits the preprocessed second wireless communication information to the communication control device 100 (step S704).

[0092] The acquisition unit 131 of the communication control device 100 acquires the preprocessed second wireless communication information (step S705), and also acquires base station design information from the base station design information DB 70 (step S706).

[0093] The identifying unit 133 inputs the processed second wireless communication information and the base station design information into the model M, causing the model M to execute an inference process using the input information (step S707). Then, the identifying unit 133 identifies optimization conditions (terminal conditions, power conditions) based on the output result of the model M (step S708).

[0094] The base station control unit 134 also controls the base station 20 to operate in accordance with the optimization conditions (step S709). Specifically, the base station control unit 134 controls the base station 20 to be connected to the terminal device 30 to be controlled, so that the terminal device 30 satisfies the terminal conditions and outputs transmission power at the power value specified by the power conditions.

[0095] The base station 20 that has received the instruction from the base station control unit 134 performs transmission power control according to the optimization conditions for the terminal devices 30 that are the control target among the terminal devices 30 under its control (step S710). Specifically, the base station 20 that has received the instruction from the base station control unit 134 instructs the terminal devices 30 to transmit uplink signals at the power value of the power conditions.

[0096] 7. Other Embodiments The above-described communication control device 100 may be implemented in various different forms other than the above-described embodiment. Therefore, other embodiments of the communication control device 100 will be described below.

[0097] In the above embodiment, an example has been shown in which the communication control device 100 adaptively changes the transmission power control policy, which has previously been set individually for each base station 20, for each service area 10A using model M. However, the communication control device 100 may, for example, find an optimal solution for each period in which the traffic behavior of users (terminal devices 30) varies significantly, and adaptively change the policy when that period occurs. In other words, the communication control device 100 may perform semi-static control in accordance with the traffic behavior for each service area 10A.

[0098] Specifically, the identifying unit 133 may perform a process of identifying the optimization condition based on the model M corresponding to the current period out of the models M for each predetermined period learned using the first wireless communication information for each predetermined period as learning data, and the second wireless communication information acquired when the current period begins. In this case, the learning unit 132 generates the model M for each predetermined period using the first wireless communication information for each predetermined period as learning data.

[0099] For example, there may be a large difference in the flow of people between weekdays and holidays even during the same time period (e.g., the time period from 8:00 to 9:00), resulting in a large difference in traffic behavior. Therefore, by using the model M that has been trained using the first wireless communication information corresponding to the time period as training data, the identification unit 133 becomes able to identify optimization conditions appropriate for this time period.

[0100] 8. Hardware Configuration The communication control device 100 according to the embodiment may be realized, for example, by a computer 1000 configured as shown in Fig. 8. Fig. 8 is a hardware configuration diagram showing an example of a computer that realizes the functions of the communication control device 100 according to the embodiment. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM 1300, a HDD 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

[0101] The CPU 1100 operates and controls each unit based on programs stored in the ROM 1300 or the HDD 1400. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 starts up, programs that depend on the hardware of the computer 1000, and the like.

[0102] The HDD 1400 stores programs executed by the CPU 1100, data used by such programs, etc. The communication interface 1500 receives data from other devices via a predetermined communication network and sends it to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the predetermined communication network.

[0103] The CPU 1100 controls an output device such as a display and an input device such as a keyboard via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. The CPU 1100 also outputs generated data to the output device via the input / output interface 1600.

[0104] Media interface 1700 reads a program or data stored in recording medium 1800 and provides it to CPU 1100 via RAM 1200. CPU 1100 loads the program or data from recording medium 1800 onto RAM 1200 via media interface 1700 and executes the loaded program. Recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase Change Rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0105] For example, when the computer 1000 functions as the communication control device 100 according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control unit 130. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from another device via a predetermined communication network.

[0106] [9. Other] Furthermore, among the processes described in each of the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0107] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0108] Furthermore, the above-described embodiments can be combined as appropriate within the scope of not causing any contradiction in the processing content.

[0109] Although some of the embodiments of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be implemented in other forms that include the aspects described in the "present invention" section and that have been modified and improved in various ways based on the knowledge of those skilled in the art.

[0110] Sy communication system 20 base station 30 terminal device 40 core network 100 communication control device 110 communication unit 120 storage unit 130 control unit 131 acquisition unit 132 learning unit 133 identification unit 134 base station control unit

Claims

1. A communication control device comprising: an acquisition unit that acquires, as learning data, first wireless communication information obtained by communication between a base station included in a specified service area and a terminal device, the first wireless communication information including at least power information regarding transmission power applied by the terminal device on an uplink and interference information regarding interference received by the base station from the terminal device; an identification unit that identifies, based on a model learned using the learning data and second wireless communication information corresponding to the first wireless communication information, optimization conditions that can optimize interference received from the terminal device between the base stations included in the specified service area, the optimization conditions including a terminal condition that conditions a terminal device to be controlled that controls the uplink transmission power and a power condition that conditions the value of the transmission power; and a base station control unit that controls the base station to which the terminal device to be controlled is connected, so that the terminal device to be controlled, in a situation that satisfies the terminal condition, outputs the transmission power at the power value of the power condition.

2. The communication control device according to claim 1, further comprising: a learning unit that uses the learning data to input the first wireless communication information and causes the model to learn a pattern that can optimize a predetermined index value related to communication quality that is determined between base stations included in the specified service area and that varies depending on the interference, wherein the power information includes any of the transmission power received by the base station from the terminal device connected to the base station, the remaining transmission power of the terminal device, or scheduling information that the base station has determined for the terminal device connected to the base station; and the interference information includes any of the reception power received by the terminal device from base stations other than the base station to which the terminal device is connected, a total transmission power that is the sum of the transmission powers received by the base station, the reception quality of the uplink transmission signal, the amount of interference that the base station has received from the terminal device, or an error rate corresponding to the uplink; and wherein the communication control device further comprises: a learning unit that uses the learning data to input the first wireless communication information and cause the model to learn a pattern that can optimize a predetermined index value related to communication quality that is determined between base stations included in the specified service area and that varies depending on the interference; and wherein the identification unit identifies the optimization condition based on an output result when the second wireless communication information is input to the model.

3. The communication control device described in claim 2, wherein the learning unit causes the model to learn patterns of combinations of the status of the terminal device and the uplink transmission power as patterns that can optimize the index value based on the relationship between the transmission power of each of the terminal devices indicated by the power information and the cause of the interference estimated based on the interference information.

4. The communication control device described in claim 3, wherein the acquisition unit further acquires the antenna characteristics of the antennas possessed by the base station as the learning data, and the learning unit causes the model to learn a pattern that can optimize the index value based on the transmission power of each of the terminal devices indicated by the power information, the cause of the interference estimated based on the interference information, and the relationship between the antenna characteristics.

5. A communication control device according to claim 4, wherein the antenna characteristics include any one of the antenna pattern of the antenna of the base station, the tilt angle of the antenna, the antenna gain of the antenna, or the directivity of the antenna.

6. The communication control device described in claim 3, wherein the acquisition unit further acquires terminal location information indicating the location of the terminal device and base station location information indicating the installation location of the base station as the learning data, and the learning unit further uses the location relationship between an interference source terminal device among the terminal devices estimated from the relationship and a base station among the base stations that is interfered with by the interference source terminal device, thereby having the model learn a pattern that can optimize the index value.

7. The communication control device described in claim 3, wherein the learning unit, when inputting the second wireless communication information, searches for an optimal solution that is a combination of the terminal conditions and the power conditions according to the situation indicated by the second wireless communication information based on the pattern, and learns the model to output the search result.

8. The communication control device described in claim 1, wherein the identification unit identifies the optimization conditions based on a model corresponding to the current period among the models for each predetermined period learned using the first wireless communication information for each predetermined period as the learning data, and the second wireless communication information acquired when the current period begins.

9. An information processing method executed by a communication control device, comprising: an acquisition step of acquiring, as learning data, first wireless communication information obtained by communication between a base station included in a predetermined service area and a terminal device, the first wireless communication information including at least power information regarding transmission power applied by the terminal device in an uplink and interference information regarding interference received by the base station from the terminal device; an identification step of identifying, based on a model trained using the learning data and second wireless communication information corresponding to the first wireless communication information, optimization conditions that can optimize interference received from the terminal device between the base stations included in the predetermined service area and the base station, the optimization conditions including a terminal condition that conditions a terminal device to be controlled that controls the uplink transmission power and a power condition that conditions the value of the transmission power; and a base station control step of controlling the base station to which the terminal device to be controlled is connected, so that the terminal device to be controlled, which is in a situation that satisfies the terminal condition, outputs the transmission power at the power value of the power condition.

10. A communication control program that causes a computer to execute the following steps: an acquisition procedure for acquiring, as training data, first wireless communication information obtained by communication between a base station included in a specified service area and a terminal device, the first wireless communication information including at least power information regarding the transmission power applied by the terminal device in an uplink and interference information regarding interference received by the base station from the terminal device; an identification procedure for identifying, based on a model trained using the training data and second wireless communication information corresponding to the first wireless communication information, optimization conditions that can optimize interference received from the terminal device between the base stations included in the specified service area and the base station, the optimization conditions including a terminal condition that conditions a terminal device to be controlled that controls the uplink transmission power and a power condition that conditions the value of the transmission power; and a base station control procedure for controlling the base station to which the terminal device to be controlled is connected, so that the terminal device to be controlled that is in a situation that satisfies the terminal condition outputs the transmission power at the power value of the power condition.

Images