Control device, control method, control program, and wireless communication system
The control device uses KPI information and a trained model to manage handovers by avoiding congested cells, reducing processing load and ensuring smooth transitions to non-congested cells, thereby improving communication quality and accuracy.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- SOFTBANK CORPORATION
- Filing Date
- 2025-03-26
- Publication Date
- 2026-06-02
AI Technical Summary
Existing handover techniques in wireless communication systems rely on limited parameters for congestion detection, leading to inaccurate congestion assessment and repeated handovers to congested cells, exacerbating the congestion and delaying handover timing.
A control device uses KPI information and a trained model to determine congestion, notifying surrounding base stations to avoid handovers to congested cells, and adjusting handover priorities or excluding them from aggregation and dual connectivity options.
This approach reduces processing load on congested cells, prevents communication quality deterioration, and ensures smooth handovers to non-congested cells with improved accuracy and flexibility.
Smart Images

Figure 0007869364000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a control device, a control method, a control program, and a wireless communication system.
Background Art
[0002] Regarding handover of a terminal, a technique for performing handover based on the moving speed of the terminal is known.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Means for Solving the Problems
[0004] A control device according to an aspect of the present disclosure includes an acquisition unit that acquires KPI (Key Performance Indicator) information of a cell, and a learned model that uses the KPI information as an input and outputs information indicating whether the cell is congested, and determines whether the cell is congested from the KPI information acquired by the acquisition unit; and a notification unit that notifies a base station accommodating non-congested cells around the congested cell that the cell is congested when the determination unit determines that the cell is congested.
[0005] A control method according to one embodiment of the present disclosure is a control method used in a control device, comprising: an acquisition step of acquiring KPI (Key Performance Indicator) information of a cell; a determination step of determining whether or not the cell is congested from the KPI information acquired in the acquisition step, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested; and, if the determination step determines that the cell is congested, a notification step of notifying a base station that accommodates non-congested cells surrounding the congested cell that the cell is congested.
[0006] A control program according to one embodiment of the present disclosure is a control program for causing a computer to function as a control device, wherein the computer functions as: an acquisition unit that acquires KPI (Key Performance Indicator) information of a cell; a determination unit that takes the KPI information as input and uses a trained model that outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit; and a notification unit that, if the determination unit determines that the cell is congested, notifies a base station that accommodates non-congested cells surrounding the congested cell that the cell is congested.
[0007] A wireless communication system according to one embodiment of the present disclosure is a wireless communication system including a control device and a base station, wherein the control device includes an acquisition unit that acquires KPI (Key Performance Indicator) information of a cell, a determination unit that uses a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit, and, if the determination unit determines that the cell is congested, a notification unit that notifies the base station which accommodates non-congested cells surrounding the congested cell that the cell is congested, and the base station, based on the notification from the control device, (3A) Exclude the congested cell from the candidates for handover destination, or lower the priority of the congested cell as a candidate for handover destination. (3B) Exclude the congested cell from the candidates for carrier aggregation. (3C) The congestion cell is excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (3D) Remove the congested cell from the list of candidates for NR-DC (New Radio - Dual Connectivity) by performing at least one of the above steps (3A) to (3D). [Brief explanation of the drawing]
[0008] [Figure 1] This is a diagram illustrating the outline of the wireless communication system according to Embodiment 1. [Figure 2] This is a block diagram showing an example of the hardware configuration of a control server according to Embodiment 1. [Figure 3] This is a block diagram showing an example of the functional configuration of the control server according to Embodiment 1. [Figure 4] This is a sequence chart showing an example of the processing performed by the control server, KPI server, and each base station according to Embodiment 1. [Figure 5] This flowchart shows an example of the processing performed by the control server in the modified example 1. [Figure 6] This diagram illustrates an example of base station processing according to Embodiment 2. [Figure 7] This is a block diagram showing an example of the functional configuration of a base station according to Embodiment 2. [Modes for carrying out the invention]
[0009] [Embodiment 1] Hereinafter, one embodiment of this disclosure will be described in detail with reference to the drawings. For ease of understanding, the background and issues of this disclosure will be explained first, followed by a detailed explanation of the disclosure.
[0010] A function called admission control is known to restrict handovers to congested cells (hereinafter sometimes simply referred to as "congested cells"). This function determines whether a congested cell is congested based on the number of users connected to it (CU: Connected Users) and the CPU (Central Processing Unit) usage rate of its BBU (Baseband Unit), and rejects handover requests from base stations that accommodate surrounding cells.
[0011] The above function allows a congested cell to reject handover requests from surrounding base stations, meaning the congested cell will perform the rejection process. This process may exacerbate congestion. Furthermore, if a handover request to a congested cell is rejected, the user terminal will attempt to hand over to another cell, delaying the handover timing.
[0012] Another issue is that the congestion detection using the above function relies solely on specific parameters (number of CUs, CPU usage), and its accuracy is not sufficient. Furthermore, because each base station independently determines congestion using this function, if surrounding base stations are also experiencing congestion, handovers will be repeatedly passed around. For this reason, the conditions under which this function is activated must be set to activate only in cases of extreme congestion.
[0013] Therefore, the control device according to this disclosure comprises: an acquisition unit that acquires KPI information of a cell; a determination unit that uses a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit; and a notification unit that, if the determination unit determines that the cell is congested, notifies a base station accommodating non-congested cells around the congested cell to prohibit handover to the congested cell, notifies a base station to lower the priority of the congested cell as a candidate for handover destination, notifies a base station to exclude the congested cell from the candidates for EN-DC, notifies a base station to exclude the congested cell from the candidates for NR-DC, or notifies a base station accommodating the congested cell to prohibit carrier aggregation, or notifies a base station to change the handover conditions so that handover is less likely to occur.
[0014] The above configuration suppresses the handover of user terminals to congested cells. Furthermore, since the control unit is configured to prohibit handovers, the congested cell does not need to use a function to reject handover requests. This reduces the processing load on the congested cell. In addition, user terminals can smoothly hand over from the currently connected cell to a non-congested cell, so there is no delay in the handover timing. Moreover, since KPI information containing a lot of information is used for congestion determination, it does not depend on specific parameters, contributing to improved accuracy. Furthermore, because the AI model comprehensively considers KPI information of not only the target cell but also surrounding cells to determine congestion, it can identify congested cells more advancedly and flexibly. In addition, since carrier aggregation, EN-DC, and NR-DC are suppressed, further deterioration of communication quality can be prevented. Details are explained below.
[0015] <Examples of application of this disclosure> An example of the application of this disclosure will be explained with reference to Figure 1.
[0016] As shown in FIG. 1, the wireless communication system 1 includes a control server 10, a KPI server 20, base stations 30 to 50, and a user terminal UE (User Equipment).
[0017] The wireless communication system 1 may be configured as a system that integrates an application for a radio access network (RAN) and an AI (Artificial Intelligence) application on the same platform. Such a system is also called AI-RAN. AI-RAN is a technology that uses the computing platform of the base station to provide highly timely services to users and devices around the base station with low latency. In AI-RAN, AI applications and applications for machine learning are deployed on the edge side (terminal side) of the network through the RAN. With this configuration, AI-RAN promotes the creation of new industries and solutions that take advantage of low latency and confidentiality. Note that the network may be a mobile communication system such as 3G, 4G, 5G, 6G, LTE (Long Term Evolution), Wi-Fi (registered trademark), or an in-house LAN.
[0018] <Base Station> Base stations 30 to 50 each accommodate cells 31 to 51, which are wireless communication areas. Base stations 30 to 50 perform wireless communication with user terminals UE located inside cells 31 to 51.
[0019] In the example shown in FIG. 1, for ease of viewing, three base stations 30 to 50 and one user terminal UE are shown, but it is not limited to this. The number of base stations included in the wireless communication system 1 is usually large. Also, the number of user terminals UE connected to each of the base stations 30 to 50 is usually large.
[0020] In the example shown in Figure 1, there is some overlap between cell 31, cell 41, and cell 51, which are housed in base station 30, and base station 50, respectively. The user terminal UE is connected to cell 31. Note that cells 31 through 51 do not necessarily have to overlap.
[0021] Each base station 30-50 transmits KPI information (Key Performance Indicator) for the cells (cells 31-51) that it houses to the KPI server 20. The KPI information includes a variety of information related to communication quality, such as "Peak data rate," "User plane latency," and "Area traffic capacity." In this embodiment, the KPI information is used to determine whether or not cells 31-51 are congested.
[0022] <KPIサーバ> The KPI server 20 stores KPI information for each cell 31-51 acquired from each base station 30-50 in a database. The KPI server 20 transmits the KPI information for each cell 31-51 to the control server 10 upon request from the control server 10.
[0023] <User terminal> The user terminal UE is a communication device that performs wireless communication. In Figure 1, the user terminal UE is shown as a smartphone that the user can carry, but it is not limited to a smartphone. For example, the user terminal UE may be a tablet device, a smartwatch, a mobile phone, etc. Furthermore, the user terminal UE may be an electronic device mounted in a vehicle, etc.
[0024] In the example shown in Figure 1, the user terminal UE is located within cell 31 and therefore communicates wirelessly with base station 30. Now, let's assume that the user terminal UE moves as the user moves, as indicated by the arrows in Figure 1. Since the user terminal UE is moving away from cell 31 to which it is currently connected, it needs to hand over to cell 41 or cell 51 in order to continue wireless communication. Now, let's assume that cell 51 is congested. When the user terminal UE hands over from cell 31 to the congested cell 51, it can lead to a decrease in communication quality for the user terminal UE, and the already strained resources of the congested cell 51 will be further consumed, potentially leading to a further decrease in communication quality. For this reason, there is a need to suppress the user terminal UE from handing over to the congested cell 51.
[0025] One method for suppressing handovers is for the congested cell (in this case, congested cell 51) to reject the handover request, as described above (admission control). However, as already mentioned, this method has its drawbacks.
[0026] <Control Server> Therefore, in this embodiment, the control server 10 is configured to control the handover. Specifically, the control server 10 determines whether each cell 31 to 51 is congested or not based on the KPI information of each base station 30 to 50 obtained from the KPI server 20. Here, we assume that cell 51 is determined to be congested. Based on the determination result, the control server 10 notifies the base stations (in this case, base stations 30 and 40) that accommodate non-congested cells (in this case, cells 31 and 41) within a predetermined distance (for example, within a few kilometers) from the congested cell 51 that they are prohibited from handing over to the congested cell 51. However, the method for selecting cells to be notified of the handover prohibition is not limited to a method based on a predetermined distance. As another example, past handover data to the congested cell 51 may be obtained from the history of the KPI information of each cell, and cells that have a history of handing over to the congested cell 51 may be made targets for the handover prohibition notification.
[0027] This configuration suppresses the handover of user terminal UEs to the congested cell 51. Furthermore, since the control server 10 is configured to prohibit handovers, the congested cell 51 does not need to use a function to reject handover requests. This reduces the processing load on the congested cell 51. In addition, the user terminal UE can smoothly hand over from the currently connected cell 31 to the non-congested cell 41, so there is no delay in the handover timing. Moreover, by preventing the user terminal UE from handing over to the congested cell 51, a deterioration in the communication quality of the user terminal UE can be prevented. Furthermore, since KPI information containing a lot of information is used for congestion determination, it does not depend on specific parameters, contributing to improved accuracy.
[0028] Such a control server 10 may be configured as a RIC (Radio Access Network Intelligent Controller). An RIC is a device that controls and optimizes RAN functions. RICs are classified into "Non-Real Time RICs" and "Near-Real Time RICs," and the control server 10 may be configured as either a "Non-Real Time RIC" or a "Near-Real Time RIC."
[0029] <Example of control server hardware configuration> Referring to Figure 2, an example of the hardware configuration of the control server 10 will be described. Figure 2 is a block diagram showing an example of the hardware configuration of the control server 10.
[0030] As shown in Figure 2, the control server 10 comprises a processor 11, memory 12, storage 13, and a communication interface 14. Each of these components is connected to the others via a bus 15 so that they can communicate with one another.
[0031] <Processor> The processor 11 reads the program from the storage 13, loads it into the memory 12, and executes processing according to the program. Such a program may be one that causes the computer to execute at least some of the functions described below. The program may also function in combination with other programs already stored in the storage 13, or in combination with other programs implemented in other devices. Furthermore, the program may be distributed to the control server 10 via wireless communication. In this case, the processor 11 loads the distributed program into the memory 12 and executes processing. In other words, the program does not necessarily have to be stored in the storage 13.
[0032] The processor 11 is 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). Although Figure 2 shows one processor, it is not limited to this, and multiple processors may be provided.
[0033] <memory> Memory 12 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 (Registered Trademark) (Electrically Erasable Programmable ROM), etc. Such memory 12 may also be called registers, cache, main memory, etc.
[0034] <Storage> Storage 13 is a computer-readable recording medium that stores various data and programs. Such storage 13 can be composed of, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc. Storage 13 may also be a portable recording medium such as a flexible disk, optical disk, compact disk, or Blu-ray® disc. Storage 13 is sometimes also called an auxiliary storage device. As shown in Figure 2, the AI model 131 is stored in storage 13. Details of the AI model 131 will be described later.
[0035] <Communication Interface> Communication I / F14 is an interface for communicating with other devices. Communication I / F14 is implemented, for example, as hardware such as a network adapter, various communication software, or a combination thereof, and is configured to enable wireless or wired communication over a communication network.
[0036] The hardware configuration of the control server 10 is not limited to the configuration described above. For example, the control server 10 may be equipped with an input / output interface (I / F) as hardware. Such an I / F may include, for example, information input devices such as a keyboard or pointing device (e.g., mouse, touch panel), audio input devices such as a microphone, and image input devices such as a camera. The I / F may also include image output devices such as a display and audio output devices such as a speaker.
[0037] <Examples of processor functions> Next, an example of the functions of the processor 11 of the control server 10 will be described with reference to Figure 3. Figure 3 is a block diagram showing an example of the functions of the processor 11. The processor 11 functions as an acquisition unit 111, a determination unit 112, and a notification unit 113 by reading a program from the storage 13 and executing the program using the memory 12 as a working area. What is described as "~unit" here may also be rephrased as "~step," "~procedure," or "~process."
[0038] <Acquisition part> The acquisition unit 111 acquires KPI information for each cell 31 to 51 from the KPI server 20. The acquisition unit 111 outputs the acquired KPI information to the judgment unit 112.
[0039] <Judgment part> The determination unit 112 uses the AI model 131 to determine whether cells 31 to 51 are congested or not, based on the KPI information obtained from the acquisition unit 111. The AI model 131 is a trained model generated by machine learning using the training dataset. The training data included in the training dataset is a representation of the input data that is input to the AI model 131, with the values that the AI model 131 should output being considered as ground truth data. Such input data is also called explanatory variables, and the ground truth data is also called the target variable.
[0040] Examples of "input data" and "correct answer data" will be explained.
[0041] The "input data" is the "KPI information for each cell." Here, "each cell" may include many cells other than those described in Figure 1 (cells 31-51). The KPI information includes, as mentioned above, "Peak data rate," "User plane latency," and "Area traffic capacity."
[0042] The "correct data" is "information indicating whether or not each cell is congested." Here, "each cell" may include many cells other than cells 31 to 51 as explained in Figure 1, similar to the input data. The "information indicating whether or not each cell is congested" may also be information indicating that a predetermined parameter for determining congestion (e.g., throughput or latency) exceeds a threshold. In this case, the target cell is indicated to be congested. However, it is not limited to this. The "information indicating whether or not each cell is congested" may also be information indicating that the target cell is relatively congested compared to other cells. "Relatively congested" means that the target cell shows signs of relative congestion, such as an increasing trend in response time or an increasing trend in packet loss compared to other cells. Therefore, even if the absolute threshold is not exceeded, if the cell is congested in a relative sense, information indicating that fact may be included in the "information indicating whether or not each cell is congested."
[0043] The training dataset is generated by associating such "input data" with "ground truth data." The AI model 131 is then generated by machine learning using the training dataset. The machine learning algorithm is not particularly limited. For example, the AI model 131 may be generated using a neural network, or it may be generated using support vector regression, random forest, or the like.
[0044] The determination unit 112, by using the AI model 131 generated in this way, can determine not only the absolute congestion state based on a threshold, but also the relative congestion state. Taking Figure 1 as an example, the determination unit 112 can determine not only cells that are absolutely congested, but also cells that are relatively congested, from the KPI information of cells 31, 41, and 51. Depending on the communication conditions, there may be cells that are not absolutely congested, but are relatively congested. In this case, it is preferable to suppress handover to "relatively congested cells". According to this embodiment, by using the AI model 131, it is possible to determine "relatively congested cells" in addition to "absolutely congested cells" by comprehensively considering the situation of the surrounding cells, and therefore handover to "relatively congested cells" can also be suppressed.
[0045] The determination unit 112 outputs the determination result from the AI model 131 to the notification unit 113.
[0046] <Notification section> Based on the determination result by the AI model 131, the notification unit 113 notifies the base stations that house the cells surrounding the congested cell, which has been determined to be congested, that they are prohibited from handing over to the congested cell. For example, in Figure 1, the notification unit 113 notifies the base station 30 that houses the cells 31 surrounding the congested cell 51, and the base station 40 that houses cell 41, that they are prohibited from handing over to the congested cell 51. This prevents the user terminal UE from handing over to the congested cell 51.
[0047] Furthermore, according to this embodiment, since KPI information for each cell is used, it may be determined that all of the cells (for example, all of cell 31, cell 41, and cell 51) are congested. In this case, it is difficult to divert the traffic, so the notification unit 113 does not need to notify any base station that handover is prohibited.
[0048] <Processing flow> Next, with reference to Figure 4, the processing flow executed by the control server 10, KPI server 20, and base stations 30-50 will be explained. Figure 4 is a sequence chart showing an example of the processing executed by the control server 10, KPI server 20, and base stations 30-50.
[0049] <Step S1> In step S1, the base station 30 transmits KPI information for the cell 31 that it houses to the KPI server 20.
[0050] <Step S2> In step S2, the base station 40 transmits KPI information for the cell 41 that it houses to the KPI server 20.
[0051] <Step S3> In step S3, the base station 50 transmits KPI information for the cell 51 that it houses to the KPI server 20.
[0052] <Step S4> In step S4, the KPI server 20 stores the KPI information for each cell 31-51 obtained from each base station 30-50 in a database.
[0053] <Step S5> In step S5, the KPI server 20 sends KPI information for each cell 31 to 51 to the control server 10 in response to a request from the control server 10.
[0054] <Step S6> In step S6, the acquisition unit 111 of the control server 10 acquires KPI information for each cell 31 to 51 from the KPI server 20.
[0055] <Step S7> In step S7, the determination unit 112 of the control server 10 uses the AI model 131 to determine whether each cell 31 to 51 is congested or not, based on the KPI information obtained in the processing of step S6. Here, "congestion" may refer to absolute congestion or relative congestion, as described above.
[0056] <Step S8> In step S8, the notification unit 113 of the control server 10 notifies the base stations 30 that house the surrounding cells 31 and the base station 40 that house cell 41 of the congested cell 51, which has been determined to be congested, based on the determination result by the AI model 131, that handover to the congested cell 51 is prohibited. This prevents user terminal UEs from handing over to the congested cell 51.
[0057] Note that the processing flow in the sequence chart shown 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.
[0058] [Variation 1] Next, we will describe Modification 1 with reference to Figure 5. Figure 5 is a flowchart showing an example of the processing performed by the control server 10 in Modification 1. Modification 1 concerns the processing after a handover prohibition is notified. The processing in steps S6 to S8 shown in Figure 5 is the same as the processing in steps S6 to S8 explained in Figure 4, so we will omit the explanation. We will begin the explanation from step S9 in Figure 5.
[0059] <Step S9> In step S9, the control server 10 notifies the base station 50, which houses the congested cell 51 determined to be congested, that it will shorten the KPI information acquisition cycle. This speeds up the acquisition of KPI information. However, if the acquisition of KPI information is accelerated for all cells, it may lead to an increase in network load. Therefore, in this embodiment, the configuration is such that the acquisition of KPI information is accelerated only for the congested cells determined to be congested. This acceleration allows the control server 10 to quickly grasp when the congestion has been resolved.
[0060] <Step S10> In step S10, the control server 10 uses the AI model 131 to determine whether the congestion state of the congestion cell 51 has been resolved, based on the KPI information of the congestion cell 51 obtained in the processing of step S9. If it is determined that the congestion state of the congestion cell 51 has been resolved (YES in step S10), the process proceeds to step S11.
[0061] <Step S11> In step S11, the control server 10 notifies base stations 30 and 40 that the congestion in cell 51 has been resolved and that the handover prohibition to cell 51 has been lifted, that is, that a handover to cell 51 is permitted. As a result, the user terminal UE can hand over to cell 51.
[0062] On the other hand, if it is determined that the congestion state of the congested cell 51 has not been resolved (NO in step S10), the process is repeatedly executed until the congestion state of the congested cell 51 is resolved.
[0063] [Embodiment 2] Next, Embodiment 2 will be described. Embodiment 2 describes an example of processing on the base station side. Figure 6 is a diagram illustrating an example of processing on the base station side according to Embodiment 2. Embodiment 2 is based on the premise that, as explained in Figure 4, the control server 10 notifies base stations 30 and 40 that handover to the congested cell 51 is prohibited.
[0064] As shown in Figure 6, the user terminal UE searches for a handover destination as it moves. Let's assume that the first candidate handover destination is a congested cell 51 and the second candidate handover destination is a non-congested cell 41. The user terminal UE sends the search results as a measurement report message to the BBU 32 of the base station 30. The BBU 32 may be implemented as dedicated hardware or it may be virtualized in software. If the BBU 32 is implemented as dedicated hardware, it has a real processor, real memory, etc. If the BBU 32 is virtualized in software, it has a virtual processor and virtual memory. This is also true for the BBU 42 of the base station 40 and the BBU 52 of the base station 50.
[0065] <bbu> Referring to Figure 7, an example of the functions of the BBU 32 of the base station 30 will be described. Figure 7 is a diagram illustrating an example of the functions performed by the actual or virtual processor of the BBU 32. As shown in Figure 7, the BBU 32 functions as a determination unit 321, a selection unit 322, and a transmission unit 323. The terms "~unit" described here may also be rephrased as "~step," "~procedure," or "~process."
[0066] <Decision Section> The determination unit 321 determines the congested cell 51 and the uncongested cell 41 as candidates for handover destinations based on the measurement report received from the user terminal UE connected to the base station 30. The determination unit 321 outputs the determined information to the selection unit 322.
[0067] <Selection section> Based on the notification from the control server 10, the selection unit 322 excludes the congested cell 51 from the candidates for the handover destination and selects the non-congested cell 41 as the handover destination from the congested cell 51 and non-congested cell 41 that have been determined by the decision unit 321 as candidates for the handover destination. The selection unit 322 outputs the selected information to the transmission unit 323.
[0068] <Transmitting section> The transmitting unit 323 transmits a handover request to the BBU 42 of the base station 40 that houses the uncongested cell 41 selected by the selection unit 322 (see Figure 6).
[0069] As described above, according to the configuration of Embodiment 2, the user terminal UE can smoothly hand over to the non-congested cell 41 while suppressing the handover of the user terminal UE to the congested cell 51.
[0070] [Other Embodiments] As explained above, if cell 51 is determined to be congested, the control server 10 notifies base stations 30 and 40 that handover to the congested cell 51 is prohibited, but this is not limited to this. If cell 51 is determined to be congested, the control server 10 may perform the following actions.
[0071] <Carrier Aggregation> If the control server 10 determines that cell 51 is congested, it may notify the base station 30 to which the user terminal UE is connected that the congested cell 51 should be excluded from the carrier aggregation candidates. Carrier aggregation is a technology that improves communication speed and bandwidth by simultaneously utilizing multiple frequency bands.
[0072] Here, we assume that cells 31, 41, and 51 each utilize different frequency bands. Furthermore, we assume that carrier aggregation is available with cell 31 as the anchor cell, and with combinations of cell 31 and cell 41, or cell 31 and cell 51. In such a case, if the control server 10 determines that cell 51 is congested, it may notify the base station 30 to exclude the congested cell 51 from the candidates for carrier aggregation. As a result, the base station 30 will not perform carrier aggregation with cell 51 and will instead perform carrier aggregation with another non-congested cell (in this case, cell 41).
[0073] Using carrier aggregation with the combination of cell 31 and congested cell 51 would further consume already strained resources, potentially exacerbating congestion. With the above configuration, congested cell 51 is excluded from the candidates for carrier aggregation, thus preventing further deterioration of the congestion state.
[0074] <en-dc> As another example, if the control server 10 determines that cell 51 is congested, it may notify the base station 30 to which the user terminal UE is connected that the congested cell 51 should be excluded from the EN-DC (E-UTRA New Radio - Dual Connectivity) candidates. EN-DC is a connection method that provides high-speed communication and stability to users by simultaneously utilizing LTE and NR. Taking Figure 1 as an example, since the user terminal UE is located within cell 31, cell 31 is the primary cell providing LTE, and the congested cell 51 is the secondary cell providing NR. As mentioned above, using EN-DC in the congested cell 51 would further consume already strained resources, potentially exacerbating the congestion. With the above configuration, the congested cell 51 is excluded from the EN-DC candidates, thus preventing further deterioration of the congestion state. Furthermore, applying EN-DC to the congested cell 51 has little effect on improving throughput. Therefore, the base station 30, upon receiving the exclusion notification, should select a cell that is not congested and capable of EN-DC as the EN-DC candidate.
[0075] <nr-dc> As another example, if the control server 10 determines that cell 51 is congested, it may notify the base station 30 to which the user terminal UE is connected that the congested cell 51 should be excluded from the NR-DC (New Radio - Dual Connectivity) candidates. NR-DC is a connection method that provides high-speed communication and stability to users by simultaneously utilizing two different NRs. Taking Figure 1 as an example, since the user terminal UE is located within cell 31, cell 31 becomes the primary cell providing NR, and the congested cell 51 becomes the secondary cell providing NR. As mentioned above, using NR-DC in the congested cell 51 would further consume already strained resources, potentially exacerbating the congestion. With the above configuration, the congested cell 51 is excluded from the NR-DC candidates, thus preventing further deterioration of the congestion state. Furthermore, applying NR-DC to the congested cell 51 would have little effect on improving throughput. Therefore, the base station 30, upon receiving the exclusion notification, should select a cell that is not congested and capable of NR-DC as the NR-DC candidate. Furthermore, the NR frequency band provided by cell 31 and the NR frequency band provided by congestion cell 51 are to be different.
[0076] <Priority> As another example, if the control server 10 determines that cell 51 is congested, it may notify the base station 30 to which the user terminal UE is connected that it will lower the priority of the congested cell as a candidate for handover.
[0077] In the handover process, the cell with the best RF quality (Radio Frequency) is generally selected as the target cell. Therefore, if the best cell is excluded from the candidates for the handover destination, for example, a cell with poor SINR (Signal to Interference & Noise Ratio) may be selected as the target cell, and the affected terminal will be handed over to that cell. In this case, there is a risk of a decrease in communication quality.
[0078] Therefore, depending on the congestion situation, it may be effective to lower the priority of a congested cell as a handover target rather than excluding it from the list of handover target candidates. As an example of lowering priority, in the RRC (Radio Resource Control) Connection Reconfiguration that notifies the user terminal UE of the handover measurement instruction, the order of frequencies notified by measObject may be changed. For example, as shown in Figure 1, the priority of the frequency of congested cell 51 may be lowered, or the priority of a different frequency that can guarantee the same RF quality as the source cell (cell 31 in Figure 1) may be increased.
[0079] <Change in handover conditions> As another example, if the control server 10 determines that cell 51 is congested, it may notify the base station 30 to which the user terminal UE is connected that it will change the handover conditions to make handover less likely. Such a condition change may be adopted when the frequencies of the source cell (cell 31 in Figure 1) and the congested cell 51 to which the handover is intended are equivalent. To illustrate an example of the condition change process, the control server 10 may notify the base station 30 to change the handover conditions to a setting that makes handover less likely by applying an offset to the handover conditions using CIO (Cell Individual Offset). If the congested cell 51 to which the handover is intended is a cell with good RF quality, such a condition change can delay the handover timing. As a result, it is expected that the user terminal UE will hand over to the congested cell 51 when the congestion in the congested cell 51 has been resolved. The handover conditions are not particularly limited, but for example, "signal strength" may be adopted. In this case, "offsetting the handover conditions to make handovers less likely" means setting a higher signal strength.
[0080] As explained above, prohibition instructions such as handover and carrier aggregation are executed when a cell is determined to be congested, but this is not limited to that. For example, AI model 131 may, by learning from daily KPI information, issue prohibition instructions such as handover and carrier aggregation before congestion occurs.
[0081] As a concrete example, suppose AI model 131 has learned that the cells covering a certain station become congested during specific times in the morning and / or evening. In this case, AI model 131 may predict when the congestion will begin and issue instructions to prohibit or lower the priority of handovers, or to exclude them from carrier aggregation, EN-DC, and NR-DC before the congestion occurs. AI model 131 may also issue instructions to revert the settings back to their original state when the congestion is resolved.
[0082] <Effects and Effects> As explained above, the following effects and benefits can be obtained with respect to this disclosure.
[0083] The control device (control server 10) includes an acquisition unit 111 that acquires cell KPI information, a determination unit 112 that uses a trained model (AI model 131) that takes KPI information as input and outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit 111, and a notification unit 113 that, if the determination unit 112 determines that the cell is congested, notifies the base station that accommodates the non-congested cells surrounding the congested cell that the cell is congested.
[0084] Furthermore, the acquisition unit 111 may acquire KPI information for multiple cells, including cells surrounding the target cell. The determination unit 112 may use the AI model 131 to determine, from the KPI information of each cell, which cells are relatively congested among the cells, as congested cells.
[0085] Furthermore, the notification unit 113 also notifies the base station, (1A) Notify that the congested cell will be excluded from the list of candidates for handover, or that the priority of the congested cell as a candidate for handover will be lowered. (1B) Notify that the congested cell will be excluded from the candidates for carrier aggregation. (1C) Notify that the congested cell will be excluded from the candidates for EN-DC. (1D) Notify that the congested cell will be excluded from the list of candidates for NR-DC. You may perform at least one of the above (1A) to (1D).
[0086] With the above configuration, the handover of user terminal UEs to congested cell 51 is suppressed. Furthermore, since the control server 10 is configured to prohibit handovers, the congested cell 51 does not need to use a function to reject handover requests. This reduces the processing load on the congested cell 51. In addition, the user terminal UE can smoothly hand over from the currently connected cell 31 to the non-congested cell 41, so there is no delay in the handover timing. Moreover, by preventing the user terminal UE from handing over to the congested cell 51, a deterioration in the communication quality of the user terminal UE can be prevented. Furthermore, since KPI information containing a lot of information is used for congestion determination, it does not depend on specific parameters, contributing to improved accuracy. In addition, because the AI model 131 determines congestion by comprehensively considering not only the target cell 51 but also the KPI information of the surrounding cells 31 and 41, it can identify congested cells more advancedly and flexibly.
[0087] Furthermore, with the above configuration, carrier aggregation, EN-DC, and NR-DC are suppressed, thus preventing further deterioration of congestion.
[0088] Furthermore, when the congestion state of a congested cell is resolved, the notification unit 113 will notify the base station that houses the uncongested cell. (2A) Notify that a cell whose congestion has been resolved will be added to the list of candidates for handover, or that the changed priority will be reverted. (2B) Notify that cells whose congestion has been resolved will be added as candidates for carrier aggregation. (2C) Notify that cells whose congestion has been resolved will be added as candidates for EN-DC. (2D) Notify that cells whose congestion has been resolved will be added as candidates for NR-DC. You may perform at least one of the above (2A) to (2D).
[0089] With the above configuration, the user terminal UE can hand over to cell 51 once the congestion has been resolved. In addition, carrier aggregation, EN-DC, and NR-DC become available, providing users with high-speed communication and stability.
[0090] Furthermore, the notification unit 113 may notify the base station accommodating the congested cell to shorten the KPI information acquisition cycle.
[0091] With the above configuration, the acquisition of KPI information is accelerated, allowing the control server 10 to quickly determine when congestion has been resolved.
[0092] As explained above, when the control server 10 determines that a cell is congested, it issues a prohibition order to the target base station for actions such as handover or carrier aggregation. However, this is not limited to this. For example, the control server 10 may simply notify the target base station that a cell is congested.
[0093] In this case, the base station, upon receiving notification from the control server 10, (3A) Exclude congested cells from the list of potential handover destinations, or lower the priority of congested cells as potential handover destinations. (3B) Exclude congested cells from the list of candidates for carrier aggregation. (3C) Exclude congested cells from the candidates for EN-DC. (3D) Exclude congested cells from the NR-DC candidates. You may perform at least one of the above (3A) to (3D).
[0094] Even with this configuration, the same effects as described above can be obtained.
[0095] <Examples of implementation using software> The functions of the control server 10 (hereinafter simply referred to as "device") are control programs that cause the device to function as a computer, and these can be realized by programs that cause the computer to function as each control block of the device (particularly the acquisition unit 111, the determination unit 112, and the notification unit 113).
[0096] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., memory) as hardware for executing the program. By executing the program using this control device and storage device, the functions described in each of the embodiments are realized.
[0097] The above program may be recorded on one or more computer-readable recording media, not temporary ones. These recording media may or may not be provided by the above device. In the latter case, the program may be supplied to the above device via any wired or wireless transmission medium.
[0098] Furthermore, some or all of the functions of each of the above control blocks can also be realized by logic circuits. For example, an integrated circuit in which logic circuits functioning as each of the above control blocks are formed is also included in the scope of this disclosure. In addition, it is also possible to realize the functions of each of the above control blocks by, for example, a quantum computer.
[0099] Furthermore, each of the processes described in the above embodiments may be performed by an AI. In this case, the AI may operate on the control device described above, or it may operate on another device (for example, an edge computer or a cloud server).
[0100] <Summary> A control device according to Embodiment 1 of the present disclosure includes: an acquisition unit that acquires Key Performance Indicator (KPI) information of a cell; a determination unit that uses a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit; and a notification unit that, if the determination unit determines that the cell is congested, notifies a base station that accommodates non-congested cells around the congested cell that the cell is congested.
[0101] The control device according to Embodiment 2 of this disclosure, in Embodiment 1 described above, further, the notification unit to the base station, (1A) Notify that the congested cell will be excluded from the list of candidates for handover, or that the priority of the congested cell as a candidate for handover will be lowered. (1B) Notify that the congested cell will be excluded from the candidates for carrier aggregation. (1C) Notify that the congestion cell will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (1D) Notify that the congestion cell will be excluded from the list of candidates for NR-DC (New Radio - Dual Connectivity). You may perform at least one of the above (1A) to (1D).
[0102] The control device according to Embodiment 3 of this disclosure, in Embodiment 2 described above, wherein the notification unit, when the congestion state of the congestion cell is resolved, notifies the base station, (2A) Notify that the cell whose congestion has been resolved will be added to the list of candidates for the handover destination, or that the changed priority will be reverted to its original state. (2B) Notify that the cell whose congestion has been resolved will be added as a candidate for the carrier aggregation. (2C) Notify that the cell whose congestion has been resolved will be added to the candidates for the EN-DC. (2D) Notify that the cell whose congestion has been resolved will be added as a candidate for the NR-DC. You may perform at least one of the above (2A) to (2D).
[0103] In the control device according to aspect 4 of the present disclosure, in any of aspects 1 to 3 above, the acquisition unit may acquire KPI information of the cell and KPI information of the cells surrounding the cell, and the determination unit may use the trained model to determine from the KPI information of each cell which cells are relatively congested among the cells as congested cells.
[0104] In any of the above embodiments 1 to 4, the control device according to aspect 5 of the present disclosure may, in the notification unit, notify the base station accommodating the congestion cell to shorten the acquisition cycle of the KPI information.
[0105] A control method according to aspect 6 of the present disclosure is a control method used in a control device, comprising: an acquisition step of acquiring KPI (Key Performance Indicator) information of a cell; a determination step of determining whether or not the cell is congested from the KPI information acquired in the acquisition step, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested; and a notification step of notifying a base station that accommodates non-congested cells surrounding the congested cell that the cell is congested, if the determination step determines that the cell is congested.
[0106] A control program according to aspect 7 of the present disclosure is a control program for causing a computer to function as a control device, wherein the computer functions as: an acquisition unit that acquires KPI (Key Performance Indicator) information of a cell; a determination unit that takes the KPI information as input and uses a trained model that outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit; and a notification unit that, if the determination unit determines that the cell is congested, notifies a base station that accommodates non-congested cells surrounding the congested cell that the cell is congested.
[0107] A wireless communication system according to aspect 8 of the present disclosure is a wireless communication system including a control device and a base station, wherein the control device includes an acquisition unit that acquires KPI (Key Performance Indicator) information of a cell, a determination unit that uses a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested to determine whether or not the cell is congested from the KPI information acquired by the acquisition unit, and, if the determination unit determines that the cell is congested, a notification unit that notifies the base station which accommodates non-congested cells surrounding the congested cell that the cell is congested, and the base station, based on the notification from the control device, (3A) Exclude the congested cell from the candidates for handover destination, or lower the priority of the congested cell as a candidate for handover destination. (3B) Exclude the congested cell from the candidates for carrier aggregation. (3C) The congestion cell is excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (3D) Remove the congested cell from the list of candidates for NR-DC (New Radio - Dual Connectivity) by performing at least one of the above steps (3A) to (3D).
[0108] A wireless communication system according to aspect 9 of the present disclosure, in aspect 8 described above, may include: a determination unit that determines the congested cell and the uncongested cell as candidates for the handover destination based on a measurement report received from a terminal connected to the base station; a selection unit that, based on a notification from the control device, excludes the congested cell from the candidates for the handover destination and selects the uncongested cell as the handover destination from the congested cell and the uncongested cell determined by the determination unit as candidates for the handover destination; and a transmission unit that transmits a handover request to a base station accommodating the uncongested cell selected by the selection unit.
[0109] <Additional Notes> This disclosure 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 this disclosure.
[0110] This disclosure describes how an AI model comprehensively considers the conditions of surrounding cells to determine congestion and, based on the determination result, suppresses handover to congested cells. Such a configuration will serve as an innovative technological foundation in the telecommunications business and can contribute to achieving Sustainable Development Goal 9, "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation." [Explanation of symbols]
[0111] 1. Wireless communication system 10 Control Server 20 KPI Server 30, 40, 50 base stations Cells 31, 41, and 51 32, 42, 52 BBU 111 Acquisition Department 112 Judgment section 113 Notification Department 131 AI Models 321 Decision Section 322 Selection Section 323 Transmitter < / bbu>
Claims
1. An acquisition unit that acquires KPI (Key Performance Indicator) information for cells, A determination unit determines whether or not the cell is congested based on the KPI information acquired by the acquisition unit, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested. If the determination unit determines that the cell is congested, the notification unit notifies the base station accommodating the uncongested cells surrounding the congested cell that the cell is congested. The determination unit uses the trained model to predict the time when congestion will occur in the cell. The notification unit notifies the base station before the time when congestion occurs in the cell as predicted by the determination unit, (A) Notify that the cell where congestion is predicted to occur will be excluded from the list of candidates for handover, or that the priority of the cell where congestion is predicted to occur will be lowered as a candidate for handover. (B) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for carrier aggregation. (C) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (D) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for NR-DC (New Radio - Dual Connectivity). Perform at least one of the above (A) to (D). Control device.
2. The notification unit further notifies the base station, (1A) Notify that the congested cell will be excluded from the list of candidates for handover destination, or that the priority of the congested cell as a candidate for handover destination will be lowered. (1B) Notify that the congested cell will be excluded from the candidates for carrier aggregation. (1C) Notify that the congestion cell will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (1D) Notify that the congestion cell will be excluded from the candidates for NR-DC (New Radio-Dual Connectivity). Perform at least one of the above (1A) to (1D). The control device according to claim 1.
3. When the congestion state of the congested cell is resolved, the notification unit will notify the base station, (2A) Notify that the cell whose congestion has been resolved will be added to the candidates for the handover destination, or that the changed priority will be reverted. (2B) Notify that the cell whose congestion has been resolved will be added to the list of candidates for carrier aggregation. (2C) Notify that the cell whose congestion has been resolved will be added to the candidates for the EN-DC, (2D) Notify that the cell whose congestion has been resolved will be added to the candidates for the NR-DC, Perform at least one of the above (2A) to (2D). The control device according to claim 2.
4. The acquisition unit acquires the KPI information of the cell and the KPI information of the cells surrounding the cell. The determination unit uses the trained model to determine, from the KPI information of each cell, which cells are relatively congested among the cells, as congested cells. The control device according to claim 1.
5. The notification unit notifies the base station accommodating the congestion cell to shorten the acquisition cycle of the KPI information. The control device according to any one of claims 1 to 4.
6. A control method used in a control device, The acquisition step for obtaining cell KPI (Key Performance Indicator) information, A determination step is performed using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested, to determine whether or not the cell is congested from the KPI information acquired in the acquisition step. If the determination step determines that the cell is congested, the notification step includes notifying the base station accommodating the uncongested cells surrounding the congested cell that the cell is congested. In the determination step, the time at which congestion occurs in the cell is predicted using the trained model. In the notification step, before the time when congestion occurs in the cell predicted in the determination step, the base station is notified: (A) Notify that the cell where congestion is predicted to occur will be excluded from the list of candidates for handover, or that the priority of the cell where congestion is predicted to occur will be lowered as a candidate for handover. (B) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for carrier aggregation. (C) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (D) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for NR-DC (New Radio - Dual Connectivity). Perform at least one of the above (A) to (D). Control method.
7. A control program for making a computer function as a control device, The aforementioned computer, An acquisition unit that acquires KPI (Key Performance Indicator) information for cells, A determination unit determines whether or not the cell is congested based on the KPI information acquired by the acquisition unit, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested. If the determination unit determines that the cell is congested, the notification unit notifies the base station accommodating the uncongested cells surrounding the congested cell that the cell is congested, To make it function as, The determination unit uses the trained model to predict the time when congestion will occur in the cell. The notification unit notifies the base station before the time when congestion occurs in the cell as predicted by the determination unit, (A) Notify that the cell where congestion is predicted to occur will be excluded from the list of candidates for handover, or that the priority of the cell where congestion is predicted to occur will be lowered as a candidate for handover. (B) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for carrier aggregation. (C) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (D) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for NR-DC (New Radio - Dual Connectivity). Perform at least one of the above (A) to (D). Control program.
8. A wireless communication system including a control device and a base station, The control device is An acquisition unit that acquires KPI (Key Performance Indicator) information for cells, A determination unit determines whether or not the cell is congested based on the KPI information acquired by the acquisition unit, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested. If the determination unit determines that the cell is congested, the notification unit notifies the base station that accommodates the uncongested cells surrounding the congested cell that the cell is congested, The determination unit uses the trained model to predict the time when congestion will occur in the cell. The notification unit notifies the base station before the time when congestion occurs in the cell as predicted by the determination unit, (A) Notify that the cell where congestion is predicted to occur will be excluded from the list of candidates for handover, or that the priority of the cell where congestion is predicted to occur will be lowered as a candidate for handover. (B) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for carrier aggregation. (C) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (D) Notify that the cell in which congestion is predicted to occur will be excluded from the candidates for NR-DC (New Radio - Dual Connectivity). Perform at least one of the above (A) to (D), The aforementioned base station is Based on the notification from the control device, (3A) Exclude the cell where congestion is predicted to occur from the candidates for handover destination, or lower the priority of the cell where congestion is predicted to occur as a candidate for handover destination. (3B) The cells in which congestion is predicted to occur are excluded from the candidates for carrier aggregation. (3C) The cell in which congestion is predicted to occur is excluded from the candidates for EN-DC (E-UTRA New Radio - Dual Connectivity). (3D) The cell where congestion is predicted to occur is excluded from the candidates for NR-DC (New Radio - Dual Connectivity). Perform at least one of the above (3A) to (3D). Wireless communication system.
9. The aforementioned base station is A determination unit that determines the congested cell and the uncongested cell as candidates for the handover destination based on a measurement report received from a terminal connected to the base station, Based on the notification from the control device, the selection unit excludes the congested cell from the candidates for the handover destination and selects the non-congested cell as the handover destination from the congested cell and non-congested cell determined by the determination unit as candidates for the handover destination. The system includes a transmitting unit that transmits a handover request to a base station accommodating the non-congested cell selected by the selection unit, The wireless communication system according to claim 8.
10. An acquisition unit that acquires KPI (Key Performance Indicator) information for cells, A determination unit determines whether or not the cell is congested based on the KPI information acquired by the acquisition unit, using a trained model that takes the KPI information as input and outputs information indicating whether or not the cell is congested. If the determination unit determines that the cell is congested, the notification unit notifies the base station accommodating the uncongested cells surrounding the congested cell that the cell is congested. The notification unit notifies the base station accommodating the congestion cell to shorten the acquisition cycle of the KPI information. Control device.