Control device and control method

Abstract

A controller (320) of a server (30) refers to MAC schedulers (11a-11c) provided to each of a plurality of base stations (A-C), and controls radio resources to be allocated to a terminal on the basis of a model generated by being trained with information relating to radio resources to be allocated to terminals connected to any base station among the plurality of base stations, and a result of the reference.

Description

Control Device and Control Method

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

[0002] Conventionally, as a method for allocating radio resources defined by time and frequency to a terminal, a method is known in which a MAC scheduler is implemented for each base station, and the allocation of radio resources to the terminal using the MAC scheduler is performed for each base station (for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2024-028977

[0004] However, if each base station performs radio resource allocation by referring only to its own MAC scheduler, there is a possibility that the same radio resources may be allocated to adjacent base stations, and there is a risk of deterioration of communication quality due to interference.

[0005] One aspect of the present invention aims to prevent deterioration of communication quality.

[0006] To solve the above problems, a control device according to one aspect of the present invention is a control device capable of communicating with a plurality of base stations, comprising a controller, the controller referring to the MAC schedulers provided in each of the plurality of base stations, and based on a model generated by learning information on radio resources to be allocated to a terminal connected to any one of the plurality of base stations and the reference result, controls the radio resources to be allocated to the terminal.

[0007] To solve the above problems, a control method according to one aspect of the present invention is a control method executed by a control device capable of communicating with a plurality of base stations, comprising a reference step of referring to the MAC schedulers provided in each of the plurality of base stations, and a control step of controlling the radio resources to be allocated to the terminal based on a model generated by learning information on radio resources to be allocated to a terminal connected to any one of the plurality of base stations and the reference result.

[0008] Each aspect of the present invention may be implemented by a computer, in which case a control program that enables the implementation of the control device by a computer by operating the computer as each part (software element) of the control device, and a computer-readable recording medium on which the program is recorded, also fall within the scope of the present invention.

[0009] According to one aspect of the present invention, it is possible to prevent deterioration of communication quality.

[0010] This figure illustrates an example configuration of a wireless communication system according to an embodiment of the present invention. This figure shows an example of a functional block of a server according to an embodiment of the present invention. This figure illustrates an example of wireless resource allocation control. This flowchart shows an example of processing performed by a server according to an embodiment of the present invention. This block diagram illustrates the configuration of a computer that can be used as a server, base station, etc.

[0011] Embodiments of the present invention will be described in detail below with reference to the drawings. In the drawings, identical or substantially identical components are denoted by the same reference numerals and will not be repeated in the description.

[0012] In this embodiment, the notations "first," "second," "third," etc., are used to distinguish one component from another, and are not intended to limit the number, order, or priority of such components. For example, if there are descriptions of "first element" and "second element," it does not mean that only two elements, "first element" and "second element," will be adopted, nor does it mean that "first element" must precede "second element."

[0013] (Example of the configuration of the wireless communication system 100) Figure 1 is a diagram illustrating an example of the configuration of the wireless communication system 100 according to this embodiment.

[0014] As shown in Figure 1, the wireless communication system 100 includes base stations A to C and a server 30 that is communicatively connected to each of the base stations A to C. The wireless communication system 100 is intended to be applied to a 5G network, but is not limited thereto. For example, the wireless communication system 100 may be applied to future networks following a 5G network, to a 4G network, or to legacy networks prior to a 4G network.

[0015] Base stations A through C communicate with terminals located within the cell, which is the wireless communication area. These terminals are typically portable smartphones. Such terminals will be referred to as "Terminal UE (User Equipment)" below. Terminal UE may also be a tablet, smartwatch, mobile phone, etc. Furthermore, Terminal UE may be an electronic device mounted on a vehicle.

[0016] Base stations A to C can accommodate one or more (for example, three) cells. Taking base station A as an example, the wireless communication system 100 may include base station A, which forms a macrocell with relatively wide coverage, and base stations A1 to A3, which are located within such a macrocell and form small cells that are narrower than the macrocell. The same applies to base stations B and C. When base stations A to C accommodate multiple cells, the entire coverage area of ​​base stations A to C can be divided into multiple smaller areas.

[0017] In Figure 1, three base stations A to C are shown for clarity, but the number of base stations connected to the server 30 is not limited to three; it may be two, or four or more.

[0018] The communication between server 30 and base stations A to C is not particularly limited, but a virtualized virtual network using NFV (Network Function Virtualization) may be used, or a physical core network may be used. Alternatively, data may be processed locally using edge computing.

[0019] This section describes the DU10a (Distributed Unit) and CU20a (Central Unit), which are functional modules of base station A. Note that the DU10b and CU20b, which are functional modules of base station B, and the DU10c and CU20c, which are functional modules of base station C, are similar to DU10a and CU20a, and therefore their descriptions are omitted.

[0020] DU10a, also known as a distributed station or distributed node, primarily performs MAC layer (Medium Access Control) processing. DU10a may be configured by being installed and virtualized on a general-purpose computer serving as a regional data center. DU10a is connected to CU20a, performs communication processing with higher layers, and is also connected to RU (Radio Unit) located at the site. DU10a includes a MAC scheduler 11a.

[0021] The MAC scheduler 11a is a function that operates at the MAC layer and is responsible for managing wireless resources, for example. Regarding the management of wireless resources, the MAC scheduler 11a manages the allocation of wireless resources between terminal UEs based on fairness, efficiency, QoS (Quality of Service), etc., and allocates appropriate wireless resources to each terminal UE.

[0022] CU20a, also known as a centralized station or centralized node, controls DU10a and the RRC (Radio Resource Control) communication protocol between terminal UE and base station A. Note that, like DU10a, CU20a may be configured by being installed and virtualized on a general-purpose computer.

[0023] Server 30 is a server capable of controlling each of the base stations A to C, and includes a controller 320. Such a controller 320 may be configured as a RIC (RAN Intelligent Controller). A RIC is a controller that manages and controls nodes such as RU (not shown), DU10a to 10c, and CU20a to 20c that constitute the 5G RAN.

[0024] (Functional Blocks of Server 30) Figure 2 is a diagram showing an example of the functional blocks of the server 30 according to this embodiment. As shown in Figure 2, the server 30 includes a communication unit 310, a controller 320, a memory 330, and a storage unit 340.

[0025] The communication unit 310 is configured as a communication interface for communicating with base stations A to C.

[0026] The controller 320 is composed of, for example, one or more processors. The processors are not particularly limited, but can be implemented as, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an MPU (Micro Processor Unit), or an FPGA (Field-Programmable Gate Array). The controller 320 reads a program from the storage unit 340, expands it into the memory 330, and executes processing according to the program. Such a program is one that causes a computer to execute at least a part of the functions described below. The program may also perform its functions in combination with other programs already stored in the storage unit 340, or in combination with other programs implemented in other devices. The program may also be distributed to the server 30 by wireless communication. In this case, the controller 320 will expand the distributed program into the memory 330 and execute processing. In other words, the program does not necessarily have to be stored in the storage unit 340.

[0027] The memory 330 is a computer-readable recording medium and consists of at least one of the following: RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM® (Electrically Erasable Programmable ROM), etc. Such memory 330 may also be called a register, cache, main memory, etc.

[0028] The storage unit 340 is a computer-readable recording medium, and is composed of, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc. Alternatively, the storage unit 340 may be a portable recording medium such as a flexible disk, optical disk, compact disk, or Blu-ray® disk. The storage unit 340 is also sometimes referred to as an auxiliary storage device. The storage unit 340 stores information that is read, written, and referenced by the controller 320. In the example shown in Figure 2, the storage unit 340 stores the training dataset 341 and the inference model 342.

[0029] The training dataset 341 and the inference model 342 will be described. The training dataset 341 is a dataset containing training data (labeled data used in machine learning). Machine learning is a technique that learns the features contained in input data and generates a "model" that predicts the result corresponding to newly input data.

[0030] The inference model 342 is an AI (Artificial Intelligence) model generated by machine learning using the training dataset 341. For example, the inference model 342 takes statistical information as input data and predicts the communication quality of a terminal UE connected to a predetermined base station (one of base stations A to C) based on that statistical information. The prediction results from the inference model 342 are used to determine what radio resources to allocate to the terminal UE. Further details on this will be described later.

[0031] The training data included in the training dataset 341 is a representation of the input data that is fed into the inference model 342, with the values ​​that the inference model 342 should output being considered as ground truth data. Such input data can also be referred to as explanatory variables. Similarly, the ground truth data can be referred to as the target variable.

[0032] The inference model 342 predicts the communication quality of a terminal UE connected to a predetermined base station (any of base stations A to C) from statistical information. Therefore, the training data included in the training dataset 341 is data that associates the communication quality of the terminal UE when statistical information is applied to "input data" that shows statistical information, with "ground truth data".

[0033] "Statistical information" includes, for example, the radio resources allocated to the terminal UE, the base station that allocated those radio resources, the number of layers at that base station, the amount of data waiting to be transmitted uplink / downlink at that base station, whether MIMO (Multiple Input Multiple Output) is available at that base station, the distance between base stations, and information indicating obstacles between base stations. Here, "distance between base stations" refers to the distance between base station A and base station B, and the distance between base station A and base station C, assuming that base station A is the base station that allocated the radio resources. It should be assumed that base stations A to C are adjacent to each other. Each of base stations A to C may also transmit such statistical information to server 30.

[0034] "Terminal UE communication quality" refers, for example, to the signal strength during communication. Signal strength is also called "Received Signal Strength Indicator" (RSSI). The terminal UE may measure this signal strength and transmit it to the connected base station. The base stations (each of base stations A to C) may transmit the "data indicating the terminal UE's signal strength" obtained from the terminal UE to the server 30. Note that the terminal UE communication quality is not limited to signal strength, but may also include communication speed, latency, packet loss rate, etc.

[0035] The training data described above is generated by associating the input data with the correct answer data and stored in the storage unit 340 as the training dataset 341.

[0036] (Example of Controller 320 Functions) The controller 320 functions as a training data acquisition unit 321, a learning unit 322, a reference unit 323, a prediction unit 324, and an allocation unit 325 by reading a program from the storage unit 340 and executing the program using the memory 330 as a working area. The units described as "~units" here may be replaced with "~circuits," "~devices," or "~equipment," or they may be replaced with "~steps," "~procedures," or "~processes." In other words, the units described as "~units" may be implemented by programs stored in the storage unit 340 as described above, or they may be implemented by hardware such as elements, devices, boards, and wiring only, or by a combination of software and hardware. These functions will be described below.

[0037] The training data acquisition unit 321, when generating the inference model 342, refers to the storage unit 340 to acquire the training dataset 341. The training data acquisition unit 321 outputs the acquired training dataset 341 to the learning unit 322.

[0038] The learning unit 322 generates an inference model 342 for predicting the communication quality of a terminal UE from statistical information by machine learning using the training dataset 341 acquired from the training data acquisition unit 321. As described above, the inference model 342 is a computational model that uses "statistical information" as an explanatory variable and "the communication quality of a terminal UE connected to a predetermined base station (any of base stations A to C)" as the objective variable. The machine learning algorithm is not particularly limited. For example, the learning unit 322 may generate the inference model 342 using a neural network, or it may generate the inference model 342 using regression analysis, random forest, etc. The inference model 342 generated by such machine learning is stored in the storage unit 340.

[0039] The reference unit 323 refers to the MAC schedulers 11a to 11c of base stations A to C. The reference unit 323 outputs to the prediction unit 324 information obtained by referring to the MAC schedulers 11a to 11c, such as which base station is allocating what kind of radio resources to which terminal UE.

[0040] The prediction unit 324 predicts the communication quality of the terminal UE using information obtained by referring to the MAC schedulers 11a to 11c and an inference model 342 generated by learning the relationship between statistical information and the communication quality of the terminal UE. The prediction unit 324 outputs the prediction result to the allocation unit 325.

[0041] The allocation unit 325 determines the radio resources to be allocated to the terminal UE based on the prediction results from the prediction unit 324. The allocation unit 325 instructs the target base station to allocate the determined radio resources to the terminal UE.

[0042] (An example of wireless resource allocation control) Next, an example of wireless resource allocation control will be explained with reference to Figure 3. Figure 3 shows resource blocks (RB) at base stations A to C. The horizontal axis in Figure 3 represents frequency, and the vertical axis represents time. Figure 3 illustrates an example of allocating wireless resources in the order of base station A, base station B, and base station C, but the order in which wireless resources are allocated is not limited to this.

[0043] The controller 320 refers to the MAC schedulers 11a to 11c of each of the base stations A to C. This allows the controller 320 to understand which base station is currently allocating which radio resources to which UE. Initially, all base stations A to C are assumed to be in a state where they have not allocated any radio resources.

[0044] In this initial state, suppose that terminal UE (referred to as the first terminal UE) connects to base station A. The controller 320 can determine that resource blocks are available at any of the base stations by referring to the MAC schedulers 11a to 11c. The controller 320 determines which radio resources to allocate to the first terminal UE based on fairness, efficiency, QoS, etc. The controller 320 notifies base station A of the information regarding the determined radio resources. Base station A allocates the radio resources to the first terminal UE according to the controller 320's decision. Here, the radio resources allocated to the first terminal UE are shown in resource block X (1 x 5).

[0045] Next, assume that another terminal UE (referred to as the second terminal UE) connects to base station A. By referring to the MAC schedulers 11a to 11c, the controller 320 can determine that resource block X has been allocated at base station A and that resource blocks are available at base stations B to C. Based on fairness, efficiency, QoS, etc., the controller 320 determines the radio resources to be allocated to the second terminal UE. The controller 320 notifies base station A of the information regarding the determined radio resources. Base station A allocates radio resources to the second terminal UE according to the determination of the controller 320. Here, the radio resources allocated to the second terminal UE are shown as resource block Y (2×3).

[0046] Next, assume that another terminal UE (referred to as the third terminal UE) connects to base station A. By referring to the MAC schedulers 11a to 11c, the controller 320 can determine that resource blocks X and Y have been allocated at base station A and that resource blocks are available at base stations B to C. Based on fairness, efficiency, QoS, etc., the controller 320 determines the radio resources to be allocated to the third terminal UE. The controller 320 notifies base station A of the information regarding the determined radio resources. Base station A allocates radio resources to the third terminal UE according to the determination of the controller 320. Here, the radio resources allocated to the third terminal UE are shown as resource block Z (1×5).

[0047] Next, assume that another terminal UE (referred to as the fourth terminal UE) connects to base station B. By referring to the MAC schedulers 11a to 11c, the controller 320 can determine that resource blocks X, Y, and Z have already been allocated in base station A and that resource blocks are available in base stations B to C. The controller 320 can decide to allocate wireless resources that do not overlap with the wireless resources already allocated in base station A to the fourth terminal UE while considering fairness, efficiency, QoS, etc. The controller 320 notifies base station B of the information regarding the determined wireless resources. Base station B allocates wireless resources to the fourth terminal UE according to the decision of the controller 320. Here, the wireless resources allocated to the fourth terminal UE are shown as resource block U (2×2).

[0048] In this way, by referring to the MAC schedulers (here, MAC schedulers 11a to 11c) provided in each base station (here, base stations A to C), the controller 320 can avoid allocating the same wireless resources as those of adjacent base stations, so that it is possible to prevent deterioration of communication quality due to interference. Note that the controller 320 may decide to allocate wireless resources that do not overlap with the wireless resources already allocated in base station A to the fourth terminal UE without considering fairness, efficiency, QoS, etc.

[0049] Next, suppose another terminal UE (referred to as the fifth terminal UE) connects to base station B. By referring to MAC schedulers 11a to 11c, the controller 320 can determine that resource blocks X, Y, and Z are already allocated at base station A, resource block U is already allocated at base station B, and resource blocks are available at base station C. Considering fairness, efficiency, QoS, etc., the controller 320 can decide to allocate radio resources to the fifth terminal UE that do not overlap with radio resources already allocated at base station A. The controller 320 notifies base station B of the information regarding the determined radio resources. Base station B allocates radio resources to the fifth terminal UE according to the controller 320's decision. Here, the radio resources allocated to the fifth terminal UE are shown in resource block W (1 x 5). Note that the controller 320 may also decide to allocate radio resources to the fifth terminal UE that do not overlap with radio resources already allocated at base station A, without considering fairness, efficiency, QoS, etc.

[0050] Next, suppose another terminal UE (referred to as the sixth terminal UE) connects to base station C. By referring to the MAC schedulers 11a to 11c, the controller 320 can determine that resource blocks X, Y, and Z are already allocated at base station A, resource blocks U and W are already allocated at base station B, and that there are no resource blocks available at base station C. In other words, by referring to the MAC schedulers 11a to 11c, the controller 320 can determine that it is difficult to allocate wireless resources at base station C that do not overlap with either base station A or base station B. In this case, the controller 320 decides to allocate the same wireless resources at base station C as either base station A or base station B.

[0051] Here, the wireless resources indicated by resource blocks X, Y, Z, U, and W are candidates for "identical wireless resources." If any of the wireless resources indicated by resource blocks X, Y, and Z are assigned, base stations A and C will be assigned the same wireless resource. Also, if any of the wireless resources indicated by resource blocks U and W are assigned, base stations B and C will be assigned the same wireless resource. If the same wireless resources are assigned, as mentioned above, there is a risk of deterioration in communication quality due to interference between base stations.

[0052] Therefore, in this embodiment, the controller 320 uses the inference model 342 to predict the communication quality of the sixth terminal UE when the same wireless resource is assigned. Specifically, the controller 320 uses the inference model 342 to predict the communication quality of the sixth terminal UE when any of the wireless resources represented by resource blocks X, Y, Z, U, and W are assigned to the base station C. Based on the prediction results, the controller 320 can determine which wireless resource has the highest communication quality to assign to the sixth terminal UE. The "wireless resource with the highest communication quality" here may be defined as a wireless resource whose communication quality does not deteriorate, or a wireless resource whose impact on the deterioration of communication quality is the smallest.

[0053] Examples of input data to the inference model 342 include: (1) information on the radio resources currently allocated to base station A, (2) information on the radio resources currently allocated to base station B, and (3) information on the radio resources that base station C will allocate in the future (any of the radio resources indicated by resource blocks X, Y, Z, U, and W). By inputting such information into the inference model 342, the communication quality of the sixth terminal UE can be obtained when any of the radio resources indicated by resource blocks X, Y, Z, U, and W are allocated to base station C. The input data may also include, as described above, the number of layers for each base station, the amount of data waiting to be transmitted uplink / downlink for each base station, whether MIMO (Multiple Input Multiple Output) can be used at each base station, the distance between base stations, and information indicating obstacles between base stations.

[0054] Thus, the controller 320 uses the inference model 342 to determine which radio resource to allocate to the base station C from among the radio resources represented by resource blocks X, Y, Z, U, and W. Here, it is assumed that the radio resource represented by resource block Z is the radio resource with the highest communication quality. The controller 320 notifies the base station C of the information regarding the determined radio resource. The base station C allocates the radio resource represented by resource block Z to the sixth terminal UE according to the controller 320's decision (see Figure 3).

[0055] As described above, according to this embodiment, the controller 320 optimizes wireless resource allocation by referring to the MAC schedulers 11a to 11c provided by each of the base stations A to C, thereby avoiding allocating the same wireless resources as adjacent base stations. Furthermore, even if it does end up allocating the same wireless resources as adjacent base stations, it is possible to prevent deterioration of communication quality.

[0056] Although Figure 3 uses three base stations (base stations A to C) for explanation, the same processing is possible even if there are two base stations or four or more.

[0057] (Processing Flow) Next, with reference to Figure 4, the processing flow executed by the server 30 will be explained. Figure 4 is a flowchart showing an example of the processing executed by the server 30.

[0058] In step S101, the controller 320 of the server 30 refers to the MAC schedulers 11a to 11c of each of the base stations A to C. If there is an available resource block at any of the base stations (A to C) to which the radio resources will be allocated (YES in step S102), the controller 320 instructs the target base station to allocate the radio resources to the available resource block (step S103). The controller 320 may consider fairness, efficiency, QoS, etc., when allocating radio resources. Here, "an available resource block" means that the base station can be allocated radio resources that do not overlap with radio resources already allocated at other adjacent base stations.

[0059] If no resource blocks are available, specifically, as explained in Figure 3, if it is difficult for base station C to allocate a wireless resource that does not overlap with either base station A or base station B (NO in step S102), the controller 320 uses the inference model 342 to predict the communication quality when the same wireless resource is allocated (step S104). If the prediction result indicates that the communication quality will not deteriorate, or the impact of the deterioration in communication quality will be minimal, even if the same wireless resource is allocated, the controller 320 instructs the target base station to allocate the same wireless resource (step S105).

[0060] As described above, the control method according to this embodiment includes a reference step (S101) which refers to the MAC schedulers 11a to 11c provided by each of the base stations A to C, and a control step (steps S104 to S105) which controls the radio resources to be assigned to a terminal UE based on an inference model 342 generated by learning information about radio resources to be assigned to a terminal UE connected to any of the base stations A to C, and the reference result.

[0061] Note that the processing flow shown in the flowchart in Figure 4 is just one example, and steps may be deleted, new steps added, or the processing order rearranged as long as it does not deviate from the main point.

[0062] (Effects) As described above, the following effects can be obtained according to this embodiment.

[0063] The server 30 (control device) is connected to multiple base stations (base stations A to C) in a communicative manner. The server 30 includes a controller 320 (see Figure 1). The controller 320 refers to the MAC schedulers 11a to 11c provided by each of the base stations A to C. The controller 320 controls the radio resources to be allocated to terminal UEs based on an inference model 342 generated by learning information about radio resources to be allocated to terminal UEs connected to any of the base stations A to C, and the reference results.

[0064] With the above configuration, the controller 320 can optimize the allocation of wireless resources by referring to the MAC schedulers 11a to 11c provided by each of the base stations A to C. This enables efficient utilization of wireless resources and provides a high-quality communication environment.

[0065] Base stations A to C are adjacent to each other. Information regarding radio resources may include at least information indicating the radio resources allocated to terminal UE, information indicating the base station that allocated the radio resources, and information indicating the communication quality of terminal UE when the radio resources were allocated. Controller 320 may predict the communication quality of terminal UE when predetermined radio resources are allocated to terminal UE using inference model 342. Controller 320 may decide which radio resources to allocate to terminal UE based on the prediction results. The "predetermined radio resources" referred to here are, for example, any of the radio resources indicated by resource blocks X, Y, Z, U, and W that will be allocated to base station C in the future (see Figure 3).

[0066] Determining wireless resources using the prediction results from inference model 342 contributes to optimizing wireless resource allocation.

[0067] Furthermore, if the controller 320 determines, based on the reference results, that it is difficult to allocate a wireless resource to the target base station that does not overlap with one or more other base stations, it may use the inference model 342 to predict the communication quality of the terminal UE connected to the target base station when it is allocated the same wireless resource as one or more other base stations. The controller 320 may then decide which wireless resource to allocate to the terminal UE based on the prediction results. An example of the "target base station" is base station C mentioned above. An example of "one or more base stations other than the target base station" is base stations A and B mentioned above.

[0068] According to the above configuration, the controller 320 can optimize wireless resource allocation by referring to the MAC schedulers 11a to 11c provided by each of the base stations A to C, thereby avoiding allocating the same wireless resources as adjacent base stations. Furthermore, even if it becomes necessary to allocate the same wireless resources as adjacent base stations, it is possible to prevent deterioration of communication quality.

[0069] The data input to the inference model 342 is not limited to the data described above, and may also include information such as the number of layers at each base station, the amount of data waiting to be transmitted uplink / downlink at each base station, whether MIMO (Multiple Input Multiple Output) can be used at each base station, the distance between base stations, and information indicating obstacles between base stations. Depending on conditions such as the distance between base stations and the presence of obstacles, interference may be less likely to occur even if the same radio resources are allocated to adjacent base stations. Therefore, by generating a model that includes this data, a model capable of accurately predicting communication quality can be realized.

[0070] [Other Embodiments] The above describes an example in which the inference model 342 is used when it is difficult to allocate wireless resources to base station C that do not overlap with those of base station A and base station B. However, it is not limited to this. Specifically, the controller 320 can also use the inference model 342 when it is possible to allocate wireless resources to base station C that do not overlap with those of base station A and base station B. That is, if the controller 320 obtains a prediction result that communication quality will not deteriorate, or that better communication quality will be obtained than when non-overlapping wireless resources are allocated, even if the same wireless resources are allocated to either base station A or base station B, or both, the controller 320 may decide to allocate the same wireless resources.

[0071] Thus, if the controller 320 determines, based on the reference results, that it is possible to allocate a wireless resource to the target base station that does not overlap with one or more other base stations, it may use the inference model 342 to predict the communication quality of the terminal UE connected to the target base station when it is allocated the same wireless resource as one or more other base stations. Based on the prediction results, the controller 320 may decide which wireless resource to allocate to the terminal UE. This optimizes the allocation of wireless resources and allows for efficient utilization of wireless resources.

[0072] (Examples of hardware configuration and software implementation) The control block of the server 30 (especially the various parts included in the controller 320) may be implemented by logic circuits (hardware) formed on an integrated circuit (IC chip), or by software using a CPU (Central Processing Unit). In the latter case, the server 30 may be configured using a computer (electronic computer).

[0073] Figure 5 is a block diagram illustrating the configuration of a computer 910 that can be used as a server 30. The computer 910 comprises an arithmetic unit 912, a main memory 913, an auxiliary storage device 914, and an input / output interface 915, all connected to each other via a bus 911. The arithmetic unit 912, the main memory 913, and the auxiliary storage device 914 may be, for example, a CPU, RAM, a solid-state drive, or a hard disk drive. An input device 920 for a user to input various information into the computer 910, and an output device 930 for the computer 910 to output various information to the user are connected to the input / output interface 915. The input device 920 and the output device 930 may be built into the computer 910 or connected to the computer 910 (externally). For example, the input device 920 may be a button, keyboard, mouse, touch sensor, etc., and the output device 930 may be a lamp, display, printer, speaker, etc. Alternatively, a device having both input device 920 and output device 930 functions, such as a touch panel that integrates a touch sensor and a display, may be used. The communication interface 916 is an interface for the computer 910 to communicate with an external device.

[0074] The auxiliary storage device 914 stores an information processing program that causes the computer 910 to operate as a server 30. The arithmetic unit 912 then loads the information processing program stored in the auxiliary storage device 914 onto the main memory 913 and executes the instructions contained in the information processing program, thereby causing the computer 910 to function as a component of the server 30. The recording medium used by the auxiliary storage device 914 to record information such as the information processing program can be any "non-temporary tangible medium" that is readable by a computer, such as tape, disk, card, semiconductor memory, or programmable logic circuit.

[0075] Furthermore, the present invention may employ a configuration in which the computer 910 functions using a program recorded on an external recording medium, or a program supplied to the computer 910 via any transmission medium (such as a communication network or broadcast waves). The present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the above program is embodied by electronic transmission.

[0076] In this disclosure, terms such as “information” and “data” refer to objects that can be processed by a computer and may be used interchangeably.

[0077] In this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. The phrase "based on" means both "based solely on" and "based on at least."

[0078] The present invention is not limited to the embodiments described above, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0079] 100 Wireless communication system 30 Server 320 Controller 342 Inference models A, B, C Base station 11a, 11b, 11c MAC scheduler

Claims

1. A control device capable of communicating with multiple base stations, comprising a controller, the controller refers to a MAC scheduler provided by each of the multiple base stations, and controls the radio resources to be allocated to a terminal connected to any of the multiple base stations based on a model generated by learning information about radio resources to be allocated to a terminal connected to any of the multiple base stations, and the reference result.

2. The control device according to claim 1, wherein the plurality of base stations are adjacent to each other, the information relating to the radio resources includes at least information indicating the radio resources allocated to the terminal, information indicating the base station that allocated the radio resources, and information indicating the communication quality of the terminal when the radio resources were allocated, and the controller predicts the communication quality of the terminal when predetermined radio resources are allocated to the terminal using the model, and determines the radio resources to be allocated to the terminal based on the prediction result.

3. The control device according to claim 2, wherein, if the controller determines, based on the reference results, that it is difficult to allocate a target base station to a target base station that does not overlap with one or more base stations other than the target base station, the controller predicts, using the model, the communication quality of the terminal connected to the target base station when it is allocated the same radio resources as those already allocated to one or more base stations, and determines, based on the prediction results, the radio resources to be allocated to the terminal.

4. The control device according to claim 2, wherein, if the controller determines, based on the reference result, that it is possible to allocate a wireless resource to the target base station among the plurality of base stations that does not overlap with one or more base stations other than the target base station, the controller predicts, using the model, the communication quality of the terminal connected to the target base station when the same wireless resource as the wireless resource already allocated to one or more base stations is allocated to the terminal, and determines the wireless resource to be allocated to the terminal based on the prediction result.

5. A control method performed by a control device capable of communicating with multiple base stations, comprising: a reference step of referencing a MAC scheduler provided by each of the multiple base stations; and a control step of controlling the radio resources to be allocated to a terminal connected to any of the multiple base stations based on a model generated by learning information about radio resources to be allocated to a terminal connected to any of the multiple base stations, and the reference result.

Images