RAN node

Abstract

A Radio Access Network (RAN) node and a method are provided that contribute to the RAN node collecting information useful for operating a cell. The method includes a first NG-RAN node 11B in an NG-RAN (Next Generation-RAN) that supports AI / ML (Artificial Intelligence / Machine Learning) as a RAN function, sending (S1) a first Xn message that is a request for collecting information to a second NG-RAN node 12B in the NG-RAN, and receiving (S2) a response thereto. The first Xn message includes a Cell To Report List for collecting the information, the Cell To Report List includes Cell To Report Items for collecting the information, and the Cell To Report Item includes a Cell ID.

Description

[Technical Field]

[0001] The present disclosure relates to a RAN node and method. [Background technology]

[0002] 3GPP (3rd Generation Partnership Project; registered trademark) specifies communications between RAN (Radio access network) nodes that manage adjacent cells, such as HO (Handover). For example, Non-Patent Document 1 specifies signaling procedures in the radio network layer of the control plane between NG-RAN (Next Generation-Radio access network) nodes. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] 3GPP TS 38.423 V16.7.0 (2021-10), “3rd Generation Partnership Project; Technical Specification Group Radio Access Network; NG-RAN; Xn application protocol (XnAP) (Release 16)”. [Non-patent document 2] 3GPP TR 37.816 V16.0.0 (2019-07), “3rd Generation Partnership Project; Technical Specification Group Radio Access Network; Study on RAN-centric data collection and utilization for LTE and NR (Release 16)”. [Non-patent document 3] TR 37.817 v1.2.0, 3GPP TSG-RAN WG3 Meeting #115 electronic, R3-221610, Online, 21th Feb - 3rd Mar, 2022 [Non-patent document 4] TP to 37.817 on AI / ML based load balancing, 3GPP TSG-RAN WG3 Meeting #114bis-e, R3-221446, 17-26 Jan 2022, E-meeting [Non-patent document 5] Aurelien Geron, “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems 2nd Edition”. [Non-patent document 6] Charu C. Aggarwal, “Neural Networks and Deep Learning: A Textbook”. [Non-Patent Document 7] Maxim Lapan, “Deep Reinforcement Learning Hands-On: Apply modern RL methods to practical problems of chatbots, robotics, discrete optimization, web automation, and more, 2nd Edition”. Summary of the Invention [Problem to be solved by the invention]

[0004] One objective of the present disclosure is to provide a RAN node and a method that contribute to the RAN node collecting information useful for operating a cell. It should be noted that this objective is only one of multiple objectives that multiple embodiments disclosed in this specification aim to achieve. Other objectives or problems and novel features will become apparent from the description of this specification or the accompanying drawings. [Means for solving the problem]

[0005] According to one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to transmit, to other RAN nodes, a message including an information element indicating that the RAN node is connected to an Artificial Intelligence (AI) model capable of managing the other RAN nodes.

[0006] In accordance with one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to receive a message from a first RAN node, the message including an information element indicating that another first RAN node is connected to an Artificial Intelligence (AI) model capable of managing the RAN node.

[0007] According to one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to transmit, to a first RAN node, a message including an information element indicating that the RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to another first RAN node.

[0008] According to one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to receive a message from another RAN node, the message including an information element indicating that the other RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to the RAN node.

[0009] In accordance with one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to transmit a message including an information element indicating a request to perform an action to another first RAN node and another second RAN node configured to be managed for a common action.

[0010] According to one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to receive a message from a RAN node configured to be managed for a common action and the message includes an information element indicating a request to cause the other first RAN node to perform the action.

[0011] According to one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to transmit, to a second RAN node, a message including information elements related to execution of an action requested to be performed by the RAN node and the other first RAN node configured to be managed for a common action.

[0012] In accordance with one aspect, a Radio Access Network (RAN) node includes a memory, a processor coupled to the memory, and a transceiver, wherein the processor is configured to cause the transceiver to receive messages from a first RAN node and a second RAN node, the messages including information elements related to execution of an action requested of another first RAN node and another second RAN node configured to be managed for a common action.

[0013] In one aspect, a method is performed by a Radio Access Network (RAN) node and includes transmitting a message to another RAN node, the message including an information element indicating that the RAN node is connected to an Artificial Intelligence (AI) model capable of managing the other RAN node.

[0014] In one aspect, a method is performed by a Radio Access Network (RAN) node and includes receiving a message from another RAN node, the message including an information element indicating that the other RAN node is connected to an Artificial Intelligence (AI) model capable of managing the RAN node.

[0015] In one aspect, a method is performed by a Radio Access Network (RAN) node and includes causing another RAN node to transmit a message including an information element indicating that the RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to the other RAN node.

[0016] In one aspect, a method is performed by a Radio Access Network (RAN) node and includes receiving a message from another RAN node, the message including an information element indicating that the other RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to the RAN node.

[0017] In one aspect, a method is performed by a Radio Access Network (RAN) node and includes transmitting a message including an information element indicating a request to perform an action to another first RAN node and another second RAN node configured to be managed for a common action.

[0018] In one aspect, a method is performed by a Radio Access Network (RAN) node, and includes receiving a message from a second RAN node that includes an information element indicating a RAN node configured to be managed for a common action and a request for the other first RAN node to perform the action.

[0019] In one aspect, a method is performed by a Radio Access Network (RAN) node, and includes transmitting a message to a second RAN node, the message including information elements related to the execution of an action requested of a RAN node and another first RAN node configured to be managed for a common action.

[0020] In one aspect, a method is performed by a Radio Access Network (RAN) node, and includes receiving messages from a first RAN node and a second RAN node, the messages including information elements related to the execution of an action requested of another first RAN node and another second RAN node configured to be managed for a common action. [Effects of the Invention]

[0021] According to the present disclosure, it is possible to provide a RAN node and a method that helps the RAN node collect information useful for operating a cell. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 1 is a diagram illustrating a configuration example of a communication system according to a first embodiment. [Figure 2]FIG. 2 is a block diagram showing a configuration example of a RAN node according to the first embodiment. [Figure 3] FIG. 3 is a sequence diagram illustrating an example of operation of the communication system according to the first embodiment. [Figure 4A] FIG. 10 is a sequence diagram illustrating an example of operation of the communication system according to the second embodiment. [Figure 4B] FIG. 10 is a sequence diagram illustrating an example of operation of the communication system according to the second embodiment. [Figure 5] FIG. 10 is a diagram illustrating a configuration example of a communication system according to a third embodiment. [Figure 6] FIG. 11 is a sequence diagram illustrating an example of operation of the communication system according to the third embodiment. [Figure 7] FIG. 11 is a sequence diagram illustrating an example of operation of the communication system according to the fourth embodiment. [Figure 8] FIG. 10 is a diagram illustrating a configuration example of a communication system according to a fifth embodiment. [Figure 9A] FIG. 13 is a sequence diagram illustrating an example of operation of the communication system according to the fifth embodiment. [Figure 9B] FIG. 13 is a sequence diagram illustrating an example of operation of the communication system according to the fifth embodiment. [Figure 10] FIG. 10 is a sequence diagram showing an example of executing the transmission and reception shown in steps S1001 and S1002 using the Xn setup procedure. [Figure 11A] An example of the configuration of IEs in the Xn SETUP REQUEST message in FIG. 10 is shown. [Figure 11B] An example of the configuration of IEs in the Xn SETUP REQUEST message in FIG. 10 is shown. [Figure 12A] An example of the configuration of IEs in the Xn SETUP RESPONSE message in FIG. 10 is shown. [Figure 12B] An example of the configuration of IEs in the Xn SETUP RESPONSE message in FIG. 10 is shown. [Figure 12C] An example of the configuration of IEs in the Xn SETUP RESPONSE message in FIG. 10 is shown. [Figure 13]FIG. 10 is a sequence diagram showing an example of executing the transmission and reception shown in steps S1001 and S1002 using another procedure. [Figure 14] An example of the configuration of IEs in the Xn RAN AI / ML SETUP REQUEST message in FIG. 13 is shown. [Figure 15] An example of the configuration of IEs in the Xn RAN AI / ML SETUP RESPONSE message in FIG. 13 is shown. [Figure 16A] 10 illustrates an example of a Resource Status Reporting Initiation procedure used to configure cell load reporting. [Figure 16B] FIG. 10 illustrates an example of a Resource Status Reporting procedure used to obtain information about cell load. [Figure 17A] FIG. 10 is a diagram showing an example of a RESOURCE STATUS REQUEST message sent in step S4001. [Figure 17B] FIG. 10 is a diagram showing an example of a RESOURCE STATUS REQUEST message sent in step S4001. [Figure 17C] FIG. 10 is a diagram showing an example of a RESOURCE STATUS REQUEST message sent in step S4001. [Figure 18] FIG. 10 is a diagram showing an example of a RESOURCE STATUS UPDATE message transmitted in step S4003. [Figure 19A] FIG. 10 is a diagram illustrating an example of the configuration of a Radio Resource Status IE. [Figure 19B] FIG. 10 is a diagram illustrating an example of the configuration of a Radio Resource Status IE. [Figure 19C] FIG. 10 is a diagram illustrating an example of the configuration of a Radio Resource Status IE. [Figure 19D] FIG. 10 is a diagram illustrating an example of the configuration of a Radio Resource Status IE. [Figure 20]FIG. 10 is a diagram illustrating an example of the configuration of a Composite Available Capacity Group IE. [Figure 21] FIG. 10 is a diagram illustrating an example of the configuration of a Composite Available Capacity IE. [Figure 22] A diagram showing an example of the configuration of Cell Capacity Class Value IE. [Figure 23] FIG. 10 is a diagram illustrating an example of the configuration of a Capacity Value IE. [Figure 24] FIG. 10 is a diagram illustrating an example of the configuration of Slice Available Capacity IE. [Figure 25] A figure showing an example of the configuration of Number of Active UEs IE. [Figure 26] A diagram showing an example of the configuration of an RRC Connections IE. [Figure 27A] FIG. 10 is a diagram illustrating an example of the configuration of Number of RRC Connections IE. [Figure 27B] FIG. 10 is a diagram illustrating an example of the configuration of Available RRC Connection Capacity Value IE. [Figure 28] FIG. 10 is a sequence diagram showing an example of executing the transmission and reception shown in steps S1013 and S1015. [Figure 29] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION REQUEST message. [Figure 30] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message. [Figure 31] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION REQUEST message. [Figure 32A] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message. [Figure 32B] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message. [Figure 33] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION REQUEST message. [Figure 34] A figure showing an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message. [Figure 35] FIG. 2 is a block diagram illustrating a configuration example of a RAN node according to each embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the following description and drawings have been omitted or simplified as appropriate for clarity of explanation. Furthermore, in the following drawings, the same elements are given the same reference numerals, and duplicate explanations are omitted as necessary. Furthermore, in this disclosure, unless otherwise specified, "at least one of A or B (A / B)" may mean any one of A or B, or both A and B. Similarly, when "at least one" is used with respect to three or more elements, it may mean any one of these elements, or any multiple elements (including all elements).

[0024] Embodiment 1 (Communication System Configuration) FIG. 1 is a diagram illustrating a configuration example of a communication system according to a first embodiment. The communication system 10A is, for example, a fifth-generation mobile communication system (5G system). The 5G system is New Radio Access (NR), which is a fifth-generation radio access technology. The communication system 10A is not limited to a fifth-generation mobile communication system, and may be a different mobile communication system such as a Long Term Evolution (LTE) system, an LTE-Advanced system, or a sixth-generation mobile communication system. The communication system 10A may also be another wireless communication system including at least a Radio Access Network (RAN) node and user equipment (UE). The communication system 10A may also be a communication system in which an LTE evolved NodeB (ng-eNB), which is a base station in LTE (Long Term Evolution), is connected to a 5G core network (5GC) via an NG interface.

[0025] The communication system 10A includes a RAN node 11 and a RAN node 12. The RAN nodes 11 and 12 are shown as RAN nodes 1 and 2 in Fig. 1. Although only two RAN nodes are shown in Fig. 1, the communication system 10A may include three or more RAN nodes.

[0026] The RAN nodes 11 and 12 may be, for example, gNBs. The gNB is a node that terminates NR user plane and control plane protocols for UEs and connects to 5GC via an NG interface. As another example, the RAN nodes 11 and 12 may be ng-eNBs. The ng-eNB is a node that terminates E-UTRA (Evolved Universal Terrestrial Radio Access) user plane and control plane protocols for UEs and connects to 5GC via an NG interface. The RAN nodes 11 and 12 may be CUs (Central Units) in a C-RAN (cloud RAN) configuration or gNB-CUs. The gNB-CU is a logical node that hosts the gNB's RRC (Radio Resource Control) protocol, SDAP (Service Data Adaptation Protocol) protocol, and PDCP (Packet Data Convergence Protocol) protocol. Alternatively, the gNB-CU is a logical node that hosts the en-gNB's RRC protocol and PDCP protocol, which controls the operation of one or more gNB-Distributed Units (gNB-DUs). The gNB-CU terminates the F1 interface connecting to the gNB-DU. Alternatively, the RAN nodes 11 and 12 may be CP (Control Plane) Units or gNB-CU-CP (gNB-CU-Control Plane). The gNB-CU-CP is a logical node that hosts the RRC protocol and the control plane portion of the PDCP protocol of the gNB-CU for the en-gNB or gNB. The gNB-CU-CP terminates the E1 interface connecting to the gNB-CU-UP (gNB-CU-User Plane) and the F1-C interface connecting to the gNB-DU. The gNB-CU-UP is a logical node that hosts the user plane portion of the PDCP protocol of the gNB-CU for the en-gNB. The gNB-CU-UP terminates the E1 interface connecting to the gNB-CU-CP and the F1-U interface connecting to the gNB-DU.

[0027] The RAN nodes 11 and 12 may be eNBs or eNB-CUs. The RAN nodes 11 and 12 may be EUTRAN (Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network) nodes or NG-RAN (Next Generation Radio Access Network) nodes. The EUTRAN nodes may be eNBs or en-gNBs. The NG-RAN nodes may be gNBs or ng-eNBs. The en-gNBs provide NR user plane and control plane protocol terminations for UEs and operate as secondary nodes in EN-DC (NR Dual Connectivity).

[0028] The RAN nodes 11 and 12 establish an inter-node interface and communicate with each other through the inter-node interface, which may be an Xn interface (a network interface between NG-RAN nodes), an X2 interface, or any other inter-node interface.

[0029] 1, RAN nodes 11 and 12 each serve at least one cell. Each RAN node operates its own cell and connects to and communicates with UEs located in that cell. The manner in which the cell is operated varies depending on the actions performed by each RAN node.

[0030] Furthermore, the RAN node 11 is coupled to an AI (Artificial Intelligence) model, which is a management entity that manages the RAN node. This AI model may be, for example, a Machine Learning (ML) model, or a model using other types of programs or algorithms. As will be described in detail below, the RAN node 11 functions as a gateway through which the AI ​​model manages the RAN node 12. Furthermore, the AI ​​model may or may not manage RAN nodes other than the RAN node 12 (e.g., the RAN node 11 and other RAN nodes) together with the RAN node 11. The AI ​​model manages, for example, the actions of the RAN node 12, and examples of such management include, but are not limited to, RAN node offloading, UE homeownership, and control related to energy saving in communications.

[0031] Fig. 2 is a block diagram showing an example of the configuration of a RAN node. In Fig. 2, RAN nodes 11 and 12 are collectively referred to as RAN node 100. The RAN node 100 includes a communication unit 101 and a control unit 102. The communication unit 101 and the control unit 102 may be software or modules that perform processing by a processor executing a program stored in a memory. Alternatively, the communication unit 101 and the control unit 102 may be hardware such as a circuit or a chip.

[0032] The communication unit 101 connects to and communicates with other RAN nodes and core network nodes included in the access network. The communication unit 101 also connects to and communicates with UEs. More specifically, the communication unit 101 receives various types of information from other RAN nodes, core network nodes, and UEs. The communication unit 101 also transmits various types of information to other RAN nodes, core network nodes, and UEs.

[0033] The control unit 102 reads out and executes various information and programs stored in the memory, thereby executing various processes of the RAN node 100. The control unit 102 performs processing in accordance with any or all of the setting information such as various information elements (IEs), various fields, and various conditions included in the message received by the communication unit 101. The control unit 102 is configured to be able to execute processes of multiple layers. The multiple layers may include a physical layer, a media access control (MAC) layer, a radio link control (RLC) layer, a PDCP layer, an RRC layer, and a non-access stratum (NAS) layer.

[0034] The configuration example of the communication system shown above is common to embodiment 1-2. Furthermore, RAN node 11 is a general term for RAN nodes 11A-11B in embodiment 1-2, and RAN node 12 is a general term for RAN nodes 12A-12B in embodiment 1-2. In embodiment 1-2, different operations performed by different RAN nodes will be described.

[0035] 3 is a sequence diagram showing an example of the operation of the communication system according to the first embodiment. Hereinafter, an example of the operation of the communication system 10A will be described with reference to FIG.

[0036] In step S1, the RAN node 11A transmits a first message to the RAN node 12A, the first message including an IE indicating that the RAN node 11A is connected to an AI model capable of managing other RAN nodes. The RAN node 12A receives the first message.

[0037] In the first embodiment, the RAN node 11A transmits the first message to the RAN node 12A, whereby the RAN node 12A can know that the RAN node 11A is connected to an AI model that can manage the RAN node 12A. By knowing this information, the RAN node 12A can enter a state where it is managed by the AI ​​model. For example, as described in paragraph 0039 below, the RAN node 12A may notify the AI ​​model that it accepts being managed by the AI ​​model. By being managed by the AI ​​model, the RAN node 12A can obtain information for operating a cell from the AI ​​model. Therefore, the RAN node according to the first embodiment potentially contributes to collecting information useful for the RAN node to operate a cell.

[0038] Embodiment 2 (2A) Fig. 4A is a sequence diagram showing an example of operation of the communication system according to the second embodiment. Hereinafter, an example of operation of the communication system 10A according to the second embodiment will be described with reference to Fig. 4A.

[0039] In step S2, the RAN node 12B transmits a second message to the RAN node 11B, the second message including an IE indicating that the RAN node 12B will accept being managed by the AI ​​model connected to the RAN node 11B. The RAN node 11B receives the second message.

[0040] In this example, the RAN node 12B transmits the second message to the RAN node 11B, which allows the RAN node 11B to know that the RAN node 12B has accepted management by the AI ​​model. This allows the AI ​​model to manage the RAN node 12B via the RAN node 11B. Therefore, the RAN node according to the second embodiment contributes to collecting information useful for the RAN node to operate a cell.

[0041] (2B) In (2A), the operation in which the RAN node 12B transmits the second message to the RAN node 11B has been described. However, this operation example may be combined with the operation described in paragraph 0036.

[0042] 4B is a sequence diagram showing another example of operation of the communication system according to the second embodiment. In FIG. 4B, in step S1, the RAN node 11B transmits, to the RAN node 12B, a first message including an IE indicating that the RAN node 11B is connected to an AI model capable of managing other RAN nodes. The RAN node 12B receives this first message. In response to receiving this first message, in step S2, the RAN node 12B transmits, to the RAN node 11B, a second message including an IE indicating that it accepts that the RAN node 12B will be managed by the AI ​​model connected to the RAN node 11B. The second message is a response to the first message. The RAN node 11B receives this second message.

[0043] In this example, the RAN node 11B transmits a first message to the RAN node 12B, which allows the RAN node 12B to know that the RAN node 11B is connected to an AI model that can manage the RAN node 11B. Then, the RAN node 12B transmits a second message to the RAN node 11B, which allows the RAN node 11B to know that the RAN node 12B has accepted management by the AI ​​model. This allows the AI ​​model to manage the RAN node 12B via the RAN node 11B. Therefore, the RAN node according to the second embodiment contributes to collecting information that is useful for the RAN node to operate a cell.

[0044] Embodiment 3 5 is a diagram illustrating a configuration example of a communication system according to the third embodiment. The communication system 10B is configured by adding a RAN node 13 to the communication system 10A illustrated in FIG.

[0045] The RAN nodes 11-13 establish inter-node interfaces and communicate with each other via the inter-node interfaces. The inter-node interfaces may be Xn interfaces (network interfaces between NG-RAN nodes), X2 interfaces, or other inter-node interfaces. A detailed description of the RAN nodes 11 and 12 is omitted here because it is the same as that described in paragraphs 0026-0033. A detailed description of the RAN node 13 is also omitted here because it is the same as that of the RAN node 12. However, in the third embodiment, the management entity that manages the RAN nodes 12 and 13 may be an AI model coupled to the RAN node 11 or another device. The other device may be, for example, the RAN node 11 or a device connected to the RAN node 11.

[0046] The configuration example of the communication system shown above is common to embodiments 3-4. Furthermore, RAN node 11 is a general term for RAN nodes 11C-11D in embodiments 3-4, RAN node 12 is a general term for RAN nodes 12C-12D in embodiments 3-4, and RAN node 13 is a general term for RAN nodes 13C-13D in embodiments 3-4. In embodiments 3-4, different operations performed by different RAN nodes will be described.

[0047] 6 is a sequence diagram showing an example of operation of the communication system according to the third embodiment. Hereinafter, an example of operation of the communication system 10B according to the third embodiment will be described with reference to FIG.

[0048] In step S3, the RAN node 11C transmits a third message including an information element indicating a request to execute a common action to the RAN nodes 12C and 13C that are configured to be managed by the RAN nodes 12C and 13C. The RAN nodes 12C and 13C receive the third message. Examples of the "common action" that is the object of management in the RAN nodes 12C and 13C are as shown in paragraph 0030, and examples thereof include RAN node offloading, UE homeownership, and control related to energy saving in communications, but the objects of management are not limited to these. Note that when the management entity is an AI model, a management relationship may be established between the AI ​​model and the RAN nodes 12C and 13C, for example, by the process shown in paragraph 0042.

[0049] In the third embodiment, the RAN node 11C transmits a message to the RAN nodes 12C and 13C to be managed, causing the RAN nodes 12C and 13C to execute an action to be managed. This enables the RAN nodes 12C and 13C to execute the action. Therefore, the RAN node according to the third embodiment contributes to collecting information that is useful for the RAN node to operate a cell.

[0050] Embodiment 4 7 is a sequence diagram showing an example of the operation of the communication system according to the fourth embodiment. Hereinafter, an example of the operation of the communication system 10B according to the fourth embodiment will be described with reference to FIG.

[0051] In step S4, the RAN node 12D transmits a fourth message to the RAN node 11D, including an IE related to the execution of an action requested to be performed by the RAN nodes 12D and 13D that are configured to be managed for the common action between the RAN node 12D and the RAN node 13D. The RAN node 13D also transmits this fourth message to the RAN node 11D. The RAN node 11D receives the fourth message from the RAN nodes 12D and 13D. Examples of the "common action" that is the management target of the RAN node 12D and the RAN node 13D are as described in paragraph 0030, and examples thereof include RAN node offloading, UE home control, and control related to energy saving in communications, but the management targets are not limited to these. In addition, the IE related to the execution of the action may be, for example, an IE indicating at least one of the fact that the action has been performed, the type of the action, the execution result of the action, identification information of a cell related to the action, and the amount of the action performed. In addition, when the management entity is an AI model, a management relationship may be established between the AI ​​model and RAN nodes 12D and 13D, for example, by the processing shown in paragraph 0042.

[0052] In the fourth embodiment, the RAN nodes 12D and 13D transmit messages including execution results of the requested actions to the RAN node 11D. This enables the RAN nodes 12D and 13D to be managed by a management entity based on the execution results. Therefore, the RAN node according to the fourth embodiment contributes to collecting information useful for the RAN node to operate a cell.

[0053] Fifth embodiment Next, a description will be given of embodiment 5. In embodiment 5, a specific example of the communication system shown in embodiments 1 to 4 will be provided.

[0054] Fig. 8 is a diagram illustrating a configuration example of a communication system according to a fifth embodiment. The communication system 20 is a 5G system and includes a RAN node 21, a RAN node 22, a RAN node 23, and a RAN AI / ML model 24. The RAN nodes 21-23 are NG-RAN nodes, and are gNBs or gNB-CUs described in paragraph 0026. The RAN nodes 21-23 are illustrated as RAN nodes 1-3 in Fig. 8.

[0055] In FIG. 8, RAN node 22 provides cells 25A-25F. Specifically, RAN node 22 operates cells 25A-25F and connects to and communicates with UEs in each cell. In this example, RAN node 22 is connected to UE 30A in cell 25A and UE 30B in cell 25F. Also in FIG. 8, RAN node 23 provides cells 25G-25L. Specifically, RAN node 23 operates cells 25G-25L and connects to and communicates with UEs in each cell. Although not shown in FIG. 8, RAN node 21 also provides and operates cells, similar to RAN nodes 22 and 23. RAN nodes 21-23 establish inter-node Xn interfaces and communicate with each other via the inter-node interfaces. An example configuration of RAN nodes 22-23 is similar to the example configuration of the RAN node shown in FIG. 2, and therefore a description thereof will be omitted.

[0056] In Fig. 8, cell 25E provided by RAN node 22 is adjacent to cells 25G and 25H provided by RAN node 22. Similarly, cell 25F provided by RAN node 22 is adjacent to cells 25G and 25H. For this reason, hereinafter, RAN node 23 may also be referred to as an adjacent RAN node of RAN node 22. Because of the above-described relationship between adjacent cells, traffic offloading or handover processing can be performed between cell 25E or 25F and cell 25G or 25H, as shown by the arrows in Fig. 8.

[0057] In this situation, the RAN node 22 and the RAN node 23 recognize their neighboring relationships based on at least one of the settings made in each device and measurement reports sent from UEs located in the cells served by the nodes. That is, the RAN node 22 recognizes that the cells 25E and 25F are neighboring cells with the cells 25G and 25H, and the RAN node 23 recognizes that the cells 25G and 25H are neighboring cells with the cells 25E and 25F.

[0058] For example, the RAN node 22, the RAN node 23, and the UE 30B can execute a handover procedure such that the UE 30B is handed over from a cell 25F provided by the RAN node 22 to a cell 25H provided by the RAN node 23. After the handover procedure is completed, the UE 30B communicates with the RAN node 23 providing the cell 25H. In this case, the cell 25F from which the UE 30B has moved may be referred to as a source cell, and the cell 25H to which the UE 30B has moved may be referred to as a target cell. The RAN node 22 may be referred to as a source RAN node, and the RAN node 23 may be referred to as a target RAN node. Furthermore, the cell 25H is a serving cell after the handover, and the cell 25F may be referred to as a last serving cell.

[0059] 8, the RAN AI / ML model 24 is coupled to the RAN node 21 and can communicate with each device via the RAN node 21. The RAN AI / ML model 24 may be located within the RAN node 21, or may be located in a device external to the RAN node 21, and connected to the RAN node 21 by that device being connected to the RAN node 21. This RAN AI / ML model 24 manages a certain area including multiple RAN nodes 22 and 23. Note that the managed objects may also include RAN nodes other than the RAN nodes 22 and 23 (for example, the RAN node 21).

[0060] The RAN node 21 coupled with the RAN AI / ML model 24 may be an AI-enabled RAN node. The AI-enabled RAN node may be referred to as an AI-capable RAN node or an AI-equipped RAN node. Alternatively, the AI-enabled RAN node may be referred to as an AI-enhanced RAN node. In this disclosure, the terms "AI-enabled RAN node," "AI-equipped RAN node," and "AI-equipped RAN node" refer to a RAN node that uses an AI / ML model to control communications based on information received from other devices (other network elements).

[0061] The RAN node 21 may operate as a RAN node equipped with AI functions by, for example, communicating with a RAN intelligence device (not shown) and using an AI / ML model held by the RAN intelligence device. Alternatively, the RAN node 21 may have the functionality of a RAN intelligence device and operate as a RAN node equipped with AI functions by using an AI / ML model held by the RAN intelligence device. Alternatively, the RAN AI / ML model 24 may obtain an AI / ML model from the RAN intelligence device, and the RAN node 21 may use the AI / ML model to operate as a RAN node equipped with AI functions.

[0062] The RAN intelligence device is, for example, a control device that makes the RAN intelligent and controls communications in the RAN. The RAN intelligence device may be, for example, a RAN Intelligent Controller (RIC) defined in O-RAN (Open-RAN). The RAN intelligence device performs, for example, policy management, analysis of various RAN information, AI-based function management, load balancing for each UE, radio resource management, QoS (Quality of Service) management, and mobility management such as handover control.

[0063] The RAN AI / ML model 24 executes LB (Load Balancing) or UE HO, which will be described later, between the RAN node 22 and the RAN node 23. This enables local centralized LB control.

[0064] The RAN AI / ML model 24 has an AI function that controls communications based on information received from other devices (other network elements) including the RAN nodes 21-23 and UEs such as the UE 30, using the RAN node 21 as a gateway. In the fifth embodiment, the RAN AI / ML model 24 has a machine learning (ML) function as an example of the AI ​​function. In this example, the AI / ML function executes processing for load offloading or HO performance optimization, but the processing to be executed is not limited to this, and for example, control related to energy saving in communications may also be executed.

[0065] 2 operates as the RAN node 21, and when a RAN intelligence device is provided outside the RAN node 21, the communication unit 101 of the RAN node 21 connects to and communicates with the RAN intelligence device. In this case, in the RAN node 21, the communication unit 101 may communicate with the RAN intelligence device, and the control unit 102 may enable the AI / ML model held by the RAN intelligence device to be used. Alternatively, the communication unit 101 may communicate with the RAN intelligence device and acquire the AI / ML model held by the RAN intelligence device.

[0066] When the RAN node 100 is the RAN node 21, the control unit 102 may use the AI / ML model to perform RAN communication control based on information received by the communication unit 101. Specifically, the control unit 102 may input the information received by the communication unit 101 to the AI / ML model and cause it to output various information related to RAN communication control and / or various information related to UE communication control. The control unit 102 may control the RAN and the UE by transmitting such various information to the RAN node and the UE. The control unit 102 may machine-learn the AI / ML model based on the information received by the communication unit 101. Note that the terms "learning," "training," and "training" in the present disclosure refer to automatically adjusting parameters of an AI / ML model and constructing the model.

[0067] The fifth embodiment provides a deployment scenario in which the AI ​​function in the RAN node serves only one gNB or gNB-CU, thereby providing a fully distributed and autonomous solution. However, the AI ​​function in the RAN node may also serve multiple gNBs or gNB-CUs.

[0068] 8 is a Network Data Analytic Function (NWDAF) in this example. The CN node 40 has a function of collecting and analyzing various data acquired in the network in 5GC. The OAM (Operations, Administration and Management) device 50 has an operation and management function for the communication system 20. The RAN node 21 is connected to the CN node 40 and the OAM device 50.

[0069] (Overview of operation) 9A and 9B are sequence diagrams showing an example of the operation of the communication system according to the fifth embodiment. Below, an outline of the processing executed in the communication system 20 will be described with reference to FIGS. 9A and 9B. In this embodiment, the RAN nodes 21-23, the CN node 40, and the OAM device 50 establish inter-node interfaces with one another. Furthermore, the order of the steps shown below is not limited unless otherwise specified. Furthermore, the presence or absence of each step or the presence or absence of detailed processing of each step can be changed as appropriate.

[0070] First, in step S1001, the RAN node 21 transmits a message (hereinafter also referred to as a MANAGEMENT RELATIONSHIP REQUEST) including a proposal to become a subject of management by the RAN AI / ML model 24 to the RAN nodes 22, 23 in order to establish a management relationship between the RAN AI / ML model 24 and the managed RAN nodes 22, 23. The RAN nodes 22, 23 receive the message. Note that the message transmitted from the RAN node 21 to the RAN nodes 22, 23 may be a message including a proposal to become a subject of management for at least one of the actions of offloading and UE HO in order to establish a management relationship between the RAN AI / ML model 24 and the managed RAN nodes 22, 23.

[0071] Next, in step S1002, the RAN nodes 22 and 23 transmit, as a response to the MANAGEMENT RELATIONSHIP REQUEST, a message indicating that they accept the proposal to become a management target (hereinafter also referred to as MANAGEMENT RELATIONSHIP RESPONSE) to the RAN node 21. The RAN node 21 receives the message.

[0072] There are two options for how the RAN nodes 21-23 perform the message transmission and reception shown in steps S1001 and S1002:

[0073] (Xn setup) The Xn setup procedure indicates a procedure related to the setup of an Xn interface between RAN nodes (a configuration procedure related to communication settings), and conforms to section 8.4.1 of Non-Patent Document 1.

[0074] 10 is a sequence diagram showing an example of performing the transmission and reception shown in steps S1001 and S1002 using the Xn setup procedure. First, in step S2001, NG-RAN node R1 (corresponding to RAN node 21) sends an Xn SETUP REQUEST message to NG-RAN node R2 (corresponding to RAN nodes 22 and 23) including a proposal to be managed by the RAN AI / ML model 24. The Xn SETUP REQUEST message may be a message including a proposal to be managed for at least one of the actions of offloading and UE HO. Next, in step S2002, NG-RAN node R2 sends an Xn SETUP RESPONSE message to NG-RAN node R1 indicating that it accepts the proposal to be managed.

[0075] 11A-11B show example configurations of IEs in the Xn SETUP REQUEST message in FIG. 10. The Xn SETUP REQUEST message shown in FIGS. 11A-11B includes the IEs of the Xn SETUP REQUEST message described in section 9.1.3.1 of Non-Patent Document 1. Furthermore, in this embodiment, a RAN AI / ML setup proposal IE (RAN AI / ML SETUP PROPOSAL) may be added as an option to the Xn SETUP REQUEST message. The RAN AI / ML setup proposal IE is the underlined IE in FIG. 11A. The RAN AI / ML setup proposal IE indicates that the RAN node 21 that transmitted this message is connected to a RAN AI / ML model that can manage other RAN nodes.

[0076] 12A-12C show example configurations of IEs in the Xn SETUP RESPONSE message in FIG. 10. The Xn SETUP RESPONSE messages shown in FIGS. 12A-12C include the IEs of the Xn SETUP RESPONSE message described in section 9.1.3.2 of Non-Patent Document 1. Furthermore, in this embodiment, a RAN AI / ML setup response IE (RAN AI / ML SETUP REPLY) may be added to the Xn SETUP RESPONSE message as an option. The RAN AI / ML setup response IE is the underlined IE in FIG. 12A. The RAN AI / ML setup response IE is an IE that can take two values, TRUE or FASLE. TRUE indicates that the RAN node R2 accepts management based on the RAN AI / ML model, while FASLE indicates that the RAN node R2 rejects management based on the RAN AI / ML model. However, the formats of the RAN AI / ML SETUP PROPOSAL IE and the RAN AI / ML SETUP REPLY IE are not limited to the examples shown above.

[0077] (another procedure) Another procedure described below is based on the Xn setup procedure in section 8.4.1 of Non-Patent Document 1, and is hereinafter referred to as the "Xn RAN AI / ML setup" procedure. This procedure performs the transmission and reception shown in steps S1001 and S1002, thereby achieving a reduction in the signaling required for transmission and reception. Another advantage is that a management relationship can be established even after the Xn interface has been set up using the Xn SETUP RESPONSE message and the Xn SETUP REQUEST message.

[0078] 13 is a sequence diagram showing an example of performing the transmission and reception shown in steps S1001 and S1002 using another procedure. First, in step S3001, NG-RAN node R1 (corresponding to RAN node 21) transmits an Xn RAN AI / ML SETUP REQUEST message to NG-RAN node R2 (corresponding to RAN nodes 22 and 23) including a proposal to be managed by the RAN AI / ML model 24. The Xn RAN AI / ML SETUP REQUEST message may be a message including a proposal to NG-RAN node R2 (corresponding to RAN nodes 22 and 23) to be managed for at least one of the actions of offloading and UE HO. Next, in step S3002, NG-RAN node R2 transmits an Xn RAN AI / ML SETUP RESPONSE message to NG-RAN node R1 indicating acceptance of the proposal to be managed.

[0079] FIG. 14 shows an example of the configuration of IEs in the Xn RAN AI / ML SETUP REQUEST message in FIG. 13. The Xn RAN AI / ML SETUP REQUEST message shown in FIG. 14 includes at least the following IEs: Message Type, Global NG-RAN Node ID, and Xn RAN AI / ML Setup Proposal. The Global NG-RAN Node ID indicates the ID of the sending RAN node. The Xn RAN AI / ML Setup Proposal IE has the same content as the RAN AI / ML Setup Proposal IE in FIG. 11A. The Message Type IE indicates the message type described in section 9.2.3.1 of Non-Patent Document 1, and the Global NG-RAN Node ID IE is used to globally identify NG-RAN nodes as described in section 9.2.2.3 of Non-Patent Document 1. In the example of FIG. 14, the Presence of each IE is indicated as "M" (Mandatory).

[0080] FIG. 15 shows an example of the configuration of IEs in the Xn RAN AI / ML SETUP RESPONSE message in FIG. 13. The Xn RAN AI / ML SETUP RESPONSE message shown in FIG. 15 includes IEs: Message Type, global NG-RAN node ID, and Xn RAN AI / ML setup response IE. The global NG-RAN node ID indicates the ID of the sending RAN node. The Xn RAN AI / ML setup response IE has the same content as the RAN AI / ML setup response IE in FIG. 12A. The Message Type IE indicates the message type as described in section 9.2.3.1 of Non-Patent Document 1, and the Global NG-RAN Node ID IE is used to globally identify NG-RAN nodes as described in section 9.2.2.3 of Non-Patent Document 1. In the example of FIG. 15, the Presence of each IE is indicated as "M."

[0081] Returning to FIG. 9A, the description continues. In step S1003, a UE located in each cell of the RAN nodes 22 and 23 periodically performs measurements. The measurements are related to parameters related to load offloading and HO decisions. The parameters may include, for example, quality information measured for the cell of the RAN node 22 or 23 to which the UE belongs. The quality information may be a value of a reference signal received from a source cell (source RAN node) or a target cell (target RAN node). The quality information may also be a value of a reference signal received from a source cell (source RAN node) or a target cell (target RAN node) in response to the occurrence of an unintended event. The reference signal value may include at least one of RSRP (Reference Signal Received Power), RSRQ (Reference Signal Received Quality), or SINR (Signal to Interference plus Noise Ratio). Furthermore, a specific example of "quality information measured for the cell to which this UE belongs" may include both quality information (RSRP, etc.) of the cell where the UE is located (serving area) and quality information (RSRP, etc.) of neighboring cells. For example, in FIG. 8, UE 30B can measure both quality information of cell 25F and quality information of neighboring cell 25H or 25G in step S1003.

[0082] Then, the UE transmits the measurement results of the parameters it has performed, for example as a measurement report, to the RAN node 22 or 23 to which the UE belongs.

[0083] In step S1004, the RAN nodes 22 and 23 under management transmit the measurement report information obtained in step S1003 to the RAN AI / ML model 24 via the RAN node 21. This information may be transmitted using the procedures described in sections 8.4.10 and 8.4.11 of Non-Patent Document 1, which will be described in detail below.

[0084] Figure 16A is a diagram showing the Resource Status Reporting Initiation procedure used to transmit a cell load report to another RAN node, as described in section 8.4.10 of Non-Patent Document 1. In Figure 16A, a RESOURCE STATUS REQUEST message is transmitted from NG-RAN node R1 to NG-RAN node R2 in step S4001, and in response, NG-RAN node R2 transmits a RESOURCE STATUS RESPONSE message to NG-RAN node R1 in step S4002. The RESOURCE STATUS REQUEST message in step S4001 may include a parameter (e.g., an indicator) that instructs the UE to report the load information measured in step S1003.

[0085] Figure 16B shows the Resource Status Reporting procedure used to obtain information about cell load, as described in section 8.4.11 of Non-Patent Document 1. After transmitting the RESOURCE STATUS RESPONSE message, as shown in Figure 16B, the NG-RAN node R2 transmits a RESOURCE STATUS UPDATE message to the NG-RAN node R1. This RESOURCE STATUS UPDATE message may be used as an example of the message transmitted in step S1004. In this way, the RAN node 21 obtains load information from the RAN nodes 22 and 23.

[0086] Figures 17A-17C are diagrams showing the RESOURCE STATUS REQUEST message sent in step S4001. This RESOURCE STATUS REQUEST message is described in section 9.1.3.18 of Non-Patent Document 1. The following IE structure shown in Figure 17B defines the characteristics of which cells are reported. Note that in the structure shown in the specification and drawings, ">" indicates the data hierarchy. Cell To Report List >Cell To Report Item >>Cell ID ... There are two options for which cells to report on:

[0087] (Option 1) In this option 1, only cells of the NG-RAN node R2 that are adjacent to cells of other RAN nodes that are managed together with the NG-RAN node R2 are reported. The RAN AI / ML model 24 manages only the LB decision within the cell that is the subject of the report, while the LB decision in other cells is managed autonomously by each RAN node.

[0088] 8, the RAN node 22 reports to cells 25E and 25F, and the RAN node 23 reports to cells 25G and 25H. The RAN AI / ML model 24 manages the LB decisions in cells 25E to 25H, while the RAN node 22 manages the LB decisions in cells 25A to 25D, and the RAN node 23 manages the LB decisions in cells 25I to 25L.

[0089] (Option 2) In this option 2, it may be all cells of the NG-RAN node R2. The RAN AI / ML model 24 manages the LB decision for all cells it reports to.

[0090] 8, all cells 25A to 25F are report targets in the RAN node 22, and all cells 25G to 25L are report targets in the RAN node 23. The RAN AI / ML model 24 manages LB determination within the cells 25A to 25L.

[0091] Comparing Option 1 and Option 2, Option 1 decentralizes the management of cells in each RAN node, while Option 2 manages all cells collectively by the RAN AI / ML model 24, resulting in more efficient management in Option 2. However, while Option 2 transmits signals from RAN nodes 22 and 23 to RAN node 21 for all cells, Option 1 transmits signals from RAN nodes 22 and 23 to RAN node 21 for only a limited number of cells, resulting in a reduction in the amount of data communication required in Option 1.

[0092] Note that the cells to be reported are not limited to those shown in the above options 1 and 2. For example, some cells of the NG-RAN node R2, including cells of the NG-RAN node R2 that are adjacent to cells of other RAN nodes that are managed together with the NG-RAN node R2, and pre-configured cells of the NG-RAN node R2, may be reported.

[0093] Fig. 18 is a diagram showing the RESOURCE STATUS UPDATE message sent in step S4003. This RESOURCE STATUS UPDATE message is described in section 9.1.3.21 of Non-Patent Document 1. Some of the IEs shown in Fig. 18 will be explained below.

[0094] The Radio Resource Status IE shown in Figure 18 is the Radio Resource Status IE defined for R-16 in section 9.2.2.50 of Non-Patent Document 1. However, it may be an IE to which the Radio Resource Status IE in R-17 as of October 2021 has been further added.

[0095] 19A-19D are diagrams showing examples of the structure of such a Radio Resource Status IE. This Radio Resource Status IE is used to report the following information: At least one of the following: DL (Downlink) / UL (Uplink) usage for each beam in each cell of each RAN node, GBR (Guaranteed Bit Rate), non-GBR, or total PRB (Physical Resource Block) usage At least one of the following: DL / UL usage per slice, GBR, non-GBR, or total PRB usage per slice in each cell of each RAN node

[0096] The Composite Available Capacity Group IE shown in Figure 18 is the Composite Available Capacity Group IE defined for R-16 in section 9.2.2.51 of Non-Patent Document 1. However, it may also be an IE to which the Composite Available Capacity Group IE in R-17 as of October 2021 has been further added.

[0097] Fig. 20 is a diagram showing an example of the configuration of such a Composite Available Capacity Group IE. This group indicates the Composite Available Capacity in DL / UL / SUL (Supplementary Uplink). An example of the configuration of each Composite Available Capacity IE is shown in Fig. 21. The Composite Available Capacity IE includes a Capacity Value IE and may optionally include a Cell Capacity Class Value IE.

[0098] Fig. 22 is a diagram showing an example of the configuration of a Cell Capacity Class Value IE. Fig. 23 is a diagram showing an example of the configuration of a Capacity Value IE. The Capacity Value IE is used to report the following information: DL, UL and SUL capacities for the required cells and beams > Capacity per cell, and >Capacity per beam of each cell

[0099] Returning to Fig. 18, the description will continue. The Slice Available Capacity IE shown in Fig. 18 is defined in section 9.2.2.55 of Non-Patent Document 1, and an example of its configuration is shown in Fig. 24. This is used to report the capacity of each slice DL / UL for each PLMN (Public Land Mobile Network) in each cell for the requested cell and slice.

[0100] Fig. 25 is a diagram showing an example of the configuration of the Number of Active UEs IE shown in Fig. 18. The Mean Number of Active UEs IE shown in Fig. 25 is the Number of Active UEs IE defined in section 9.2.2.62 of Non-Patent Document 1, and is used to report the average number of active UEs per cell for the requested cells.

[0101] Figure 26 is a diagram showing an example of the configuration of the RRC Connections IE shown in Figure 18. This RRC Connections IE is the RRC Connections IE defined in sections 9.2.2.56, 9.2.2.57, and 9.2.2.58 of Non-Patent Document 1, and includes a Number of RRC Connections IE and an Available RRC Connection Capacity Value IE. Figure 27A is a diagram showing an example of the configuration of the Number of RRC Connections IE, and Figure 27B is a diagram showing an example of the configuration of the Available RRC Connection Capacity Value IE. The RRC Connections IE is used to report the number of used RRC connections and the number of available RRC connections for each requested cell.

[0102] Of the information transmitted from the RAN nodes to be managed as described above, the information that can be input to the RAN AI / ML model 24, which is an LB algorithm, and used is, for example, as follows: At least one of DL / UL usage, GBR, non-GBR, or total PRB usage per beam in each cell of each RAN node for the requested cell or beam At least one of DL / UL usage per slice, GBR, non-GBR, or total PRB usage per slice in each cell of each RAN node for the requested cell or slice. DL, UL, and / or SUL capacity (including capacity per cell and capacity per beam per cell for the requested cells and / or beams) DL / UL usage per cell per PLMN per slice for the requested cell and / or slice Number of active UEs per cell for the requested cells For the requested cells, the number of RRC connections used and / or available RRC connections per cell

[0103] Returning to FIG. 9A , the description continues. In step S1005, the CN node 40 (NWDAF) transmits network information to the RAN AI / ML model 24 via the RAN node 21. The network information can be transmitted, for example, via an existing NWDAF subscription service. The network information transmitted from the CN node 40 may include, for example, at least one of the following pieces of information: Network function load, slice load, and service experience >>Network performance >>Statistics or predictions of gNB status, resource usage, communication and mobility performance, number of UEs in an area of ​​interest >Average rate of successful handovers >UE mobility >>Time series of location statistics or predictions for a specific UE or group of UEs The RAN node 21 may obtain such network information from devices on the 5GC, not limited to the CN node 40.

[0104] In step S1006, the OAM device 50 transmits network information to the RAN AI / ML model 24 via the RAN node 21. The network information transmitted from the OAM device 50 may include area information such as the cell in which the UE is located, traffic information, and statistical information. The statistical information may include statistical information on handover and statistical information on call processing such as call connection and call disconnection. Note that the order in which steps S1004, S1005, and S1006 are performed is not limited to that described above.

[0105] In step S1007, the RAN AI / ML model 24 performs initial and periodic training of the RAN AI / ML model based on the information received in steps S1004 to S1006. After training, the RAN AI / ML model 24 can make LB decisions for selected cells of the RAN nodes 22 and 23 it manages. For example, either option 1 or option 2 described in paragraphs 0087 to 0091 can be applied as the selected cells. The LB decision is a decision regarding LB HO between the selected cells of the RAN nodes it manages. In this way, the RAN AI / ML model 24 can be configured.

[0106] Any known or new model can be used as the RAN AI / ML model 24. Examples of various known models and their implementation methods are described in Non-Patent Documents 5 to 7.

[0107] Next, the processing shown in Fig. 9B will be described. However, since steps S1008, S1009, S1010, and S1011 are the same as steps S1003, S1004, S1005, and S1006, respectively, their description will be omitted. This allows the RAN AI / ML model 24 to collect information for determining LB HO.

[0108] The information received in steps S1009 to S1011 is input to the trained RAN AI / ML model 24. In step S1012, the RAN AI / ML model 24 generates an LB decision for the selected cell of the managed RAN nodes 22, 23 in response to the input. As the selected cell, as described above, for example, either option 1 or option 2 shown in paragraphs 0087 to 0091 can be applied. The LB decision is a decision regarding LB HO between the selected cells of the managed RAN nodes.

[0109] In step S1013, the RAN node 21 transmits an LB HO request to the RAN nodes 22 and 23 based on the LB decision generated by the RAN AI / ML model 24. In response to receiving this, the RAN nodes 22 and 23 execute LB HO in step S1014. In step S1015, the RAN nodes 22 and 23 transmit an LB HO response to the RAN AI / ML model 24 via the RAN node 21, notifying that the LB HO action has been executed and including the execution result of the action.

[0110] The procedure for transmitting and receiving messages in steps S1013 and S1015 is defined in a general manner that can be applied not only to the LB HO action according to this embodiment but also to any action such as energy saving control. In the present disclosure, the message sent in step S1013 is defined as "Xn RAN AI / ML ACTION REQUEST", and the message sent in step S1015 is defined as "Xn RAN AI / ML ACTION RESPONSE".

[0111] 28 is a sequence diagram showing an example of transmitting and receiving an Xn RAN AI / ML ACTION REQUEST message and an Xn RAN AI / ML ACTION RESPONSE message in steps S1013 and S1015. In Fig. 28, an Xn RAN AI / ML ACTION REQUEST message is transmitted from NG-RAN node R1 (RAN node 21) to NG-RAN node R2 (RAN nodes 22 and 23) in step S5001. NG-RAN node R2 performs an action in response to the message, and then transmits an Xn RAN AI / ML ACTION RESPONSE message to NG-RAN node R1 in step S5002.

[0112] In the fifth embodiment, two types of actions are assumed as the actions targeted by the Xn RAN AI / ML ACTION REQUEST and Xn RAN AI / ML ACTION RESPONSE messages. "Load Offload" (loadOffload): Offloads part of the current load "UEHO": Handover some UEs However, other actions may also be set as the actions.

[0113] Below, the following three methods are explained as examples of message implementation methods. (1) Optional IE (2)CHOICE structure (3) Conditionally Existing IE

[0114] Fig. 29 shows a configuration example of an Xn RAN AI / ML ACTION REQUEST message in (1) selective IEs. The underlined IEs in Fig. 29 are IEs newly proposed in this embodiment.

[0115] The Action type IE in FIG. 29 is an indicator indicating the type of action being requested. As described in paragraph 0112, there are two possible types of requested actions: "loadOffload" and "UEHO." When either of these two actions is specified as the action type, the actor cell CGI (Cell Global Identity) IE and the neighbor cell CGI IE are used as identification information. The actor cell is a cell to which the load in that cell is to be offloaded or to which a UE serving in that cell is to be handed over. When the action type is "loadOffload," the percentage of the current load of the actor cell to be offloaded to the specified neighbor cell is indicated as an IE (indicator) of "Percentage of Load to Offload." The current load of the actor cell may be, for example, the last reported or predicted load. On the other hand, if the action type is "UE HO", the number of UEs to be handed over from the actor cell to the indicated neighbor cell is indicated as an IE (indicator) of "Number of UEs to HO". The number of active UEs can be used to generate the number of UEs for the HO value. However, the indicator of the amount of load to be offloaded is not limited to the Percentage of Load to Offload IE, and other types of indicators may also be used.

[0116] In FIG. 29, the presence of each IE related to neighboring cell CGI, Percentage of Load to Offload, and Number of UEs to HO is described as "O" (Optional), but the presence of any or all of these IEs may be "M" (Mandatory).

[0117] Figure 30 shows an example of the structure of an Xn RAN AI / ML ACTION RESPONSE message in (1) optional IEs. After the NG-RAN node R2 performs an action in response to the Xn RAN AI / ML ACTION RESPONSE, it sends this message indicating that the action has been performed. To report the result of the action, this example of the Xn RAN AI / ML ACTION RESPONSE structure may include an Action Result IE indicating the result of the action. For example, the Action Result IE may report "ActionPerformed" indicating that the action was successful, or "ActionFailed" indicating that the action failed. However, the Action Result IE may use other values ​​indicating the reason for the action failure.

[0118] In addition, to report the results of the action in more detail, the "Percentage of Load Offloaded" IE and the "Number of UEs HOed" IE can be used. When the action type is "loadOffload", the percentage of the actor cell's load before the action that was offloaded to the specified neighboring cell is indicated as the "Percentage of Load Offloaded" IE. On the other hand, when the action type is "UEHO", the number of UEs that were handed over from the actor cell to the specified neighboring cell is indicated as the "Number of UEs HOed" IE.

[0119] In FIG. 30, the presence of each IE related to Action Result, Neighboring Cell CGI, Percentage of Load Offloaded, and Number of UEs HOed is described as "O", but the presence of any part or all of these IEs may be "M".

[0120] Furthermore, as another optional IE, a Cause IE (indicator) indicating the reason for the result of the action may be added to the Xn RAN AI / ML ACTION RESPONSE. For example, assume that an Xn RAN AI / ML ACTION REQUEST requesting execution of LB HO from cell 1 to cell 2 is transmitted from RAN node R1 to RAN node R2. When RAN node R2 attempts to execute the action, the signal level of cell 2 for UEs residing in cell 1 may be lower than a predetermined threshold, and therefore there may be no UE in cell 1 that can perform HO to cell 2, and HO may not be executed. The Cause IE can be used to indicate such a reason. In this way, a Cause IE may be added to the Xn RAN AI / ML ACTION RESPONSE when the action fails and the value of the Action Result IE is ActionFailed. However, the Cause IE may also be added to the Xn RAN AI / ML ACTION RESPONSE not when the action fails but when the action is successful and the value of the Action Result IE is ActionPerformed. In this case, the Cause IE may indicate the reason why the action was successful, or may be added to the Xn RAN AI / ML ACTION RESPONSE to indicate the reason for the action result whether the action was successful or failed (i.e., whether the value of the Action Result IE is ActionPerformed or ActionFailed).

[0121] As another example, if the action is successful, an Action Result IE is provided for the Xn RAN AI / ML ACTION RESPONSE, and no Cause IE is provided, whereas if the action fails, a Cause IE is provided for the Xn RAN AI / ML ACTION RESPONSE, and no Action Result IE may be provided. If the action is successful, the value of the Action Result IE is ActionPerformed. On the other hand, if the action fails, setting the Cause IE essentially means ActionFailed, so it is not necessarily necessary to provide an Action Result IE.

[0122] Figure 31 shows an example of the configuration of an Xn RAN AI / ML ACTION REQUEST message in the (2) CHOICE structure. Figures 32A-32B show an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message in the (2) CHOICE structure. The underlined IEs in Figures 31 and 32A-32B are newly proposed IEs in this embodiment.

[0123] Compared to the Xn RAN AI / ML ACTION REQUEST message shown in Figure 29, Figure 31 differs in that a CHOICE Action Type IE is further added, and either a loadOffload IE or a UEHO IE is defined below it depending on the type of action. A Percentage of Load to Offload IE may be included below the loadOffload IE, and a Number of UEs to HO IE may be included below the UEHO IE. The explanation of these IEs is omitted because it is the same as that of the Xn RAN AI / ML ACTION REQUEST message in (1) Selective IEs.

[0124] In FIG. 31, the presence of each IE related to neighboring cell CGI, Percentage of Load to Offload, and Number of UEs to HO is described as "O", but the presence of any part or all of these IEs may be "M".

[0125] Compared to the Xn RAN AI / ML ACTION RESPONSE message shown in FIG. 30, FIGS. 32A-32B differ in that a CHOICE Action Type IE is further added, and either a loadOffload IE or a UEHO IE is defined below it depending on the type of action. The Percentage of Load Offloaded IE may be included below the loadOffload IE, and the Number of UEs HOed IE may be included below the UEHO IE. The explanation of these IEs is omitted here, as they are similar to those of the Xn RAN AI / ML ACTION RESPONSE message in (1) Selective IEs. Furthermore, a Cause IE may also be added to the Xn RAN AI / ML ACTION RESPONSE message in (2) the CHOICE structure. As described in paragraph 0120, the Cause IE may be added to the Xn RAN AI / ML ACTION RESPONSE message when the action is at least either successful or failed. Alternatively, as described in paragraph 0121, if the action is successful, an Action Result IE may be provided for the Xn RAN AI / ML ACTION RESPONSE and no Cause IE may be provided, whereas if the action is unsuccessful, a Cause IE may be provided for the Xn RAN AI / ML ACTION RESPONSE and no Action Result IE may be provided.

[0126] In FIG. 32A, the presence of each IE related to Action result, neighboring cell CGI, Percentage of Load Offloaded, and Number of UEs HOed is described as "O", but the presence of any part or all of these IEs may be "M".

[0127] Figure 33 shows an example of the configuration of an Xn RAN AI / ML ACTION REQUEST message in (3) IEs that exist under certain conditions. Also, Figure 34 shows an example of the configuration of an Xn RAN AI / ML ACTION RESPONSE message in (3) IEs that exist under certain conditions. The underlined IEs in Figures 33 and 34 are IEs that are newly proposed in this embodiment.

[0128] Compared to the Xn RAN AI / ML ACTION REQUEST message shown in FIG. 29, FIG. 33 differs in that the presence of the Percentage of Load to Offload IE and the Number of UEs to HO IE is "C" (Conditional). In particular, the Percentage of Load to Offload IE is present when the action type is "loadOffload", and the Number of UEs to HO IE is present when the action type is "UEHO". The explanation of the other IEs is omitted because they are the same as those of the Xn RAN AI / ML ACTION REQUEST message in (1) and (2).

[0129] In FIG. 33, the presence of the neighboring cell CGI IE is described as "O", but it may also be "M".

[0130] Compared to the Xn RAN AI / ML ACTION RESPONSE message shown in FIG. 30, FIG. 34 differs in that the presence of the Percentage of Load Offloaded IE and the Number of UEs HOed IE is "C" (Conditional). In particular, the Percentage of Load Offloaded IE is present when the action type is "loadOffload", and the Number of UEs HOed IE is present when the action type is "UEHO". The explanation of the other IEs is omitted here, as they are the same as those of the Xn RAN AI / ML ACTION RESPONSE message in (1) and (2). Furthermore, a Cause IE may also be added to the Xn RAN AI / ML ACTION RESPONSE message in (3) IEs that are conditionally present. As described in paragraph 0120, the Cause IE may be added to the Xn RAN AI / ML ACTION RESPONSE message when the action is at least either successful or failed. Alternatively, as described in paragraph 0121, if the action is successful, an Action Result IE may be provided for the Xn RAN AI / ML ACTION RESPONSE and no Cause IE may be provided, whereas if the action is unsuccessful, a Cause IE may be provided for the Xn RAN AI / ML ACTION RESPONSE and no Action Result IE may be provided.

[0131] In FIG. 34, the presence of the IE of the neighboring cell CGI is described as "O", but it may also be "M".

[0132] Returning to Figure 9B, the description continues. In step S1014, the RAN nodes 22, 23 perform LB HO from the actor cell indicated in the Xn RAN AI / ML ACTION REQUEST message sent from the RAN AI / ML model 24 (via the RAN node 21) to the RAN nodes 22, 23 to the neighboring cells. In this solution, the RAN AI / ML model 24 does not need to select specific UEs for LB HO. Instead, an approximate number of UEs to be HO-targeted can be indicated as a percentage of the total cell load or as a number of UEs. However, the UEs to be HO-targeted can also be identified in other ways.

[0133] The RAN nodes 22 and 23 can autonomously select UEs for LB HO, taking into account signal level measurements of the particular UE relative to the source cell and potential neighboring cells, traffic characteristics, Quality of Service (QoS) requirements, and slices of the particular UE, etc.

[0134] After the LB HO is performed, in step S1015, the RAN nodes 22 and 23 indicate to the RAN AI / ML model 24 via the RAN node 21 that an action has been taken and the result of the action using an Xn RAN AI / ML ACTION REQUEST message.

[0135] (Effects of the Invention) As described above, in the fifth embodiment, the RAN node 21 transmits, to the RAN nodes 22 and 23, an Xn SETUP REQUEST message including a RAN AI / ML setup proposal IE (or an Xn RAN AI / ML SETUP REQUEST message including an Xn RAN AI / ML setup proposal IE) indicating that the RAN node 21 is connected to the RAN AI / ML model 24 that can manage the RAN nodes 22 and 23. Therefore, the RAN nodes 22 and 23 can know that the RAN node 21 is connected to the RAN AI / ML model 24 that can manage the RAN nodes 22 and 23. By knowing this information, the RAN nodes 22 and 23 can enter a state where they are managed by the RAN AI / ML model 24. For example, the RAN nodes 22 and 23 may notify the AI ​​model that they accept being managed by the RAN AI / ML model 24. By being managed by the AI ​​model, the RAN nodes 22 and 23 can obtain information for operating a cell from the AI ​​model. Therefore, this potentially contributes to the RAN nodes 22 and 23 collecting information useful for operating a cell.

[0136] Furthermore, in the fifth embodiment, the RAN node 22 transmits, to the RAN node 21, an Xn SETUP RESPONSE message including a RAN AI / ML setup response IE (or an Xn RAN AI / ML SETUP RESPONSE message including an Xn RAN AI / ML setup response IE), which indicates that the RAN node 22 will accept being managed by the RAN AI / ML model 24 connected to the RAN node 21. The RAN node 23 can also perform a similar process. This allows the RAN AI / ML model 24 to manage the RAN nodes 22 and 23 via the RAN node 21. This therefore contributes to the RAN nodes 22 and 23 collecting information useful for operating a cell.

[0137] The RAN AI / ML model 24 may also manage common actions for the RAN nodes 22, 23. This allows the RAN nodes 22, 23 to gather information useful for operating the cell with a view to performing the common actions.

[0138] Furthermore, the RAN node 22 may be a neighboring RAN node of the RAN node 23, and the common action may be offloading and / or UE HO, thereby allowing the RAN nodes 22 and 23 to collect information useful for operating the cell in terms of performing offloading or UE HO.

[0139] Furthermore, the Xn RAN AI / ML SETUP REQUEST message can be sent to the RAN nodes 22 and 23 even after a setup message (Xn SETUP REQUEST message) regarding communication setup between the RAN node 21 and the RAN nodes 22 and 23 has been sent. This allows the RAN AI / ML model 24 to manage the RAN nodes 22 and 23 even after the Xn interface has been set up.

[0140] Furthermore, the Xn RAN AI / ML SETUP RESPONSE message can be sent to the RAN node 21 even after a setup message (Xn SETUP RESPONSE message) regarding communication setup between the RAN node 21 and the RAN nodes 22 and 23 has been sent. This allows the RAN AI / ML model 24 to manage the RAN nodes 22 and 23 even after the Xn interface has been set up.

[0141] Furthermore, in the fifth embodiment, the RAN node 21 transmits, to the RAN nodes 22 and 23 that are configured to be managed for a common action, an Xn RAN AI / ML ACTION REQUEST message including an Action type IE indicating a request to execute the action. This enables the RAN AI / ML model 24 to manage the RAN nodes 22 and 23 via the RAN node 21. This contributes to the RAN nodes 22 and 23 collecting information that is useful for operating the cell.

[0142] The Action type IE may be an indicator of the type of action, which allows the RAN nodes 22, 23 to identify the action to be taken when multiple actions are available for the RAN nodes 22, 23. This helps the RAN nodes 22, 23 to gather information useful for operating the cell.

[0143] Furthermore, the RAN node 22 may be a neighboring RAN node of the RAN node 23, and the common action may be offloading and / or UE HO, thereby allowing the RAN nodes 22 and 23 to collect information useful for operating the cell in terms of performing offloading or UE HO.

[0144] The Xn RAN AI / ML ACTION REQUEST message may also include identification information of at least one of two neighboring cells of the RAN node 22, 23 that is the target of the action, for example in the form of an actor cell CGI IE and / or a neighboring cell CGI IE, allowing the RAN node 22, 23 to gather information that is useful when performing the action.

[0145] Furthermore, the action is offload, and the Xn RAN AI / ML ACTION REQUEST message may further include a Percentage of Load to Offload IE indicating the amount of load to be offloaded between neighboring cells, thereby allowing the RAN nodes 22, 23 to gather useful information when performing the action.

[0146] Furthermore, the action is HO of the UE, and the Xn RAN AI / ML ACTION REQUEST message may further include a Number of UEs to HO IE indicating the number of UEs to be handed over between neighboring cells, thereby allowing the RAN nodes 22, 23 to gather useful information when performing the action.

[0147] Furthermore, in the fifth embodiment, the RAN node 22 transmits, to the RAN node 21, an Xn RAN AI / ML ACTION RESPONSE message including an IE related to the execution of an action requested to be executed by the RAN nodes 22 and 23 that are configured to be managed for a common action. The RAN node 23 can also perform similar processing. This enables the RAN AI / ML model 24 to obtain feedback related to management. This contributes to the RAN nodes 22 and 23 collecting information useful for operating a cell based on the management of the RAN AI / ML model 24.

[0148] The Xn RAN AI / ML ACTION RESPONSE message may further include a Cause IE indicating the reason for the result of the action taken, which allows the RAN AI / ML model 24 to obtain detailed feedback for management purposes, thereby helping the RAN nodes 22, 23 to collect information useful for operating the cell based on the management of the RAN AI / ML model 24.

[0149] The Xn RAN AI / ML ACTION RESPONSE message may further include an Action type IE indicating the type of action, which allows the RAN AI / ML model 24 to know what action the RAN nodes 22 and 23 have performed. Therefore, under the management of the RAN AI / ML model 24, the RAN nodes 22 and 23 can collect information useful for operating the cell.

[0150] Also, the RAN node 22 may be a neighboring RAN node of the RAN node 23, and the action may be at least one of offloading and UE HO, thereby allowing the RAN nodes 22 and 23 to collect information useful for operating the cell in terms of performing offloading or UE HO under the management of the RAN AI / ML model 24.

[0151] The Xn RAN AI / ML ACTION RESPONSE message may further include identification information of at least one of the two neighboring cells of the RAN nodes 22, 23 that are the target of the action, for example in the form of an actor cell CGI IE or a neighboring cell CGI IE. This allows the RAN AI / ML model 24 to know details of the action performed by the RAN nodes 22, 23. Therefore, under the management of the RAN AI / ML model 24, the RAN nodes 22, 23 can collect information that is useful when operating the cell.

[0152] Furthermore, the action is offload, and the Xn RAN AI / ML ACTION RESPONSE message may further include a Percentage of Load Offloaded IE indicating the amount of load offloaded between neighboring cells. This allows the RAN AI / ML model 24 to know details of the action performed by the RAN nodes 22 and 23. Therefore, under the management of the RAN AI / ML model 24, the RAN nodes 22 and 23 can collect information that is useful when operating the cell.

[0153] Furthermore, the action is HO of the UE, and the Xn RAN AI / ML ACTION RESPONSE message may further include Number of UEs HOed, which indicates the number of UEs that are the target of handover between neighboring cells. This allows the RAN AI / ML model 24 to know details of the action performed by the RAN nodes 22 and 23. Therefore, under the management of the RAN AI / ML model 24, the RAN nodes 22 and 23 can collect information that is useful when operating the cell.

[0154] It should be noted that the present disclosure is not limited to the above-described embodiments and can be modified as appropriate without departing from the spirit of the present disclosure. For example, the technology described in the present disclosure is not limited to dedicated communication devices and can be applied to any device having a communication function.

[0155] (Example of hardware configuration) Next, a hardware configuration example of the RAN node (RAN nodes 11 to 13, 21 to 23) described in the above-mentioned embodiments will be described below. FIG. 35 is a block diagram showing a configuration example of a RAN node according to each embodiment. Referring to FIG. 35, the RAN node 100 includes an RF (Radio Frequency) transceiver 1001, a network interface 1003, a processor 1004, and a memory 1005. The RF transceiver 1001 performs analog RF signal processing for communication with a UE. The RF transceiver 1001 may include multiple transceivers. The RF transceiver 1001 is coupled to an antenna 1002 and the processor 1004. The RF transceiver 1001 receives modulation symbol data (or Orthogonal Frequency Division Multiplexing (OFDM) symbol data) from the processor 1004, generates a transmit RF signal, and provides the transmit RF signal to the antenna 1002. The RF transceiver 1001 also generates a baseband received signal based on the received RF signal received by the antenna 1002 and supplies it to the processor 1004 .

[0156] The network interface 1003 is used to communicate with network nodes (e.g., other core network nodes) and may include, for example, a network interface card (NIC) compliant with the IEEE (Institute of Electrical and Electronics Engineers) 802.3 series.

[0157] The processor 1004 performs data plane processing and control plane processing, including digital baseband signal processing for wireless communication. For example, in the case of LTE and 5G, the digital baseband signal processing by the processor 1004 may include signal processing of a MAC layer and a physical layer.

[0158] The processor 1004 may include multiple processors. For example, the processor 1004 may include a modem processor (e.g., a digital signal processor (DSP)) that performs digital baseband signal processing and a protocol stack processor (e.g., a central processing unit (CPU) or a microprocessor unit (MPU)) that performs control plane processing.

[0159] The memory 1005 is configured by a combination of volatile memory and nonvolatile memory. The memory 1005 may include multiple physically independent memory devices. The volatile memory may be, for example, static random access memory (SRAM), dynamic random access memory (DRAM), or a combination thereof. The nonvolatile memory may be mask read only memory (MROM), electrically erasable programmable read only memory (EEPROM), flash memory, or a hard disk drive, or any combination thereof. The memory 1005 may include storage located remotely from the processor 1004. In this case, the processor 1004 may access the memory 1005 via the network interface 1003 or an I / O interface (not shown).

[0160] The memory 1005 may store software modules (computer programs) including instructions and data for performing the processes of the RAN node 100 described in the above embodiments. In some implementations, the processor 1004 may be configured to read and execute the software modules from the memory 1005 to perform the processes of the RAN node 100 described in the above embodiments.

[0161] As described above, one or more processors included in each device in the above-described embodiments execute one or more programs including instructions for causing a computer to execute the algorithms described using the drawings. This processing enables the signal processing method described in each embodiment to be realized.

[0162] The program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, non-transitory computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disk (DVD), Blu-ray® disk or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagated signals.

[0163] In this specification, a user equipment (UE) (or mobile station, mobile terminal, mobile device, or wireless device, etc.) is an entity connected to a network via a radio interface.

[0164] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Appendix 1) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, The processor is configured to cause the transceiver to transmit, to the other RAN nodes, a message including an information element indicating that the RAN node is connected to an Artificial Intelligence (AI) model capable of managing the other RAN nodes. RAN node. (Appendix 2) the other RAN nodes include a first RAN node and a second RAN node; The AI ​​model manages common actions for the first RAN node and the second RAN node. RAN node as described in Appendix 1. (Appendix 3) a first cell of the first RAN node is a neighboring cell of a second cell of the second RAN node; The common action is at least one of offloading or HO (Handover) of UE (User Equipment), RAN node as described in Appendix 2. (Appendix 4) the message is transmitted to the other RAN node after a configuration message regarding communication configuration between the RAN node and the other RAN node has been transmitted; RAN node as defined in Supplementary Note 1 or 2. (Appendix 5) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to receive a message from the first RAN node, the message including an information element indicating that another first RAN node is connected to an Artificial Intelligence (AI) model capable of managing the RAN node. RAN node. (Appendix 6) The AI ​​model manages common actions for the RAN node and other second RAN nodes. RAN node as described in Appendix 5. (Appendix 7) a first cell of the RAN node is a neighboring cell of a second cell of the second RAN node; The common action is at least one of offloading or HO (Handover) of UE (User Equipment), RAN node as described in Appendix 6. (Appendix 8) the message is received from the first RAN node after a configuration message regarding communication configuration between the RAN node and the first RAN node is received; RAN node as defined in Supplementary Note 5 or 6. (Appendix 9) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to transmit, to the first RAN node, a message including an information element indicating that the RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to another first RAN node. RAN node. (Appendix 10) The AI ​​model manages common actions for the RAN node and other second RAN nodes. RAN node as described in Appendix 9. (Appendix 11) a first cell of the RAN node is a neighboring cell of a second cell of the second RAN node; The common action is at least one of offloading or HO (Handover) of UE (User Equipment), RAN node as defined in Supplementary Note 10. (Appendix 12) the message is transmitted to the first RAN node after a configuration message regarding communication configuration between the RAN node and the first RAN node is transmitted; 11. A RAN node as defined in Supplementary Note 9 or 10. (Appendix 13) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to receive a message from the other RAN node, the message including an information element indicating that the other RAN node accepts to be managed by an Artificial Intelligence (AI) model connected to the RAN node. RAN node. (Appendix 14) the other RAN nodes include a first RAN node and a second RAN node; The AI ​​model manages common actions for the first RAN node and the second RAN node. RAN node as described in Supplementary Note 13. (Appendix 15) a first cell of the first RAN node is a neighboring cell of a second cell of the second RAN node; The common action is at least one of offloading or HO (Handover) of UE (User Equipment), RAN node as described in Supplementary Note 14. (Appendix 16) the message is received from the other RAN node after a configuration message regarding communication configuration between the RAN node and the other RAN node is received; 15. A RAN node as defined in Supplementary Note 13 or 14. (Appendix 17) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to transmit a message to another first RAN node and another second RAN node configured to be managed for a common action, the message including an information element indicating a request to perform the action. RAN node. (Appendix 18) The message further includes an indicator of the type of the action. RAN node as described in Supplementary Note 17. (Appendix 19) the first cell of the first RAN node is a neighboring cell of the second cell of the second RAN node, and the action is at least one of offloading or HO (Handover) of a UE (User Equipment); 19. A RAN node as defined in Supplementary Note 17 or 18. (Appendix 20) The message further includes identification information of at least one of the first cell or the second cell that is the target of the action. RAN node as described in Supplementary Note 19. (Appendix 21) the action is off-roading, The message further includes an indicator indicating an amount of load to be offloaded between the first cell and the second cell. RAN node as described in Supplementary Note 19. (Appendix 22) The action is a UE home action; The message further includes an indicator indicating the number of UEs to be handed over between the first cell and the second cell. RAN node as described in Supplementary Note 19. (Appendix 23) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to receive a message from another second RAN node, the message including an information element indicating a request to cause the RAN node and another first RAN node configured to be managed for a common action to perform the action. RAN node. (Appendix 24) The message further includes an indicator of the type of the action. RAN node as described in Supplementary Note 23. (Appendix 25) a first cell of the RAN node is a neighboring cell of a second cell of the first RAN node; The action is at least one of offloading or HO (Handover) of UE (User Equipment), 25. A RAN node as defined in claim 23 or 24. (Appendix 26) The message further includes identification information of at least one of the first cell or the second cell that is the target of the action. RAN node as described in Supplementary Note 25. (Appendix 27) the action is off-roading, The message further includes an indicator indicating an amount of load to be offloaded between the first cell and the second cell. RAN node as described in Supplementary Note 25. (Appendix 28) The action is a UE home action; The message further includes an indicator indicating the number of UEs to be handed over between the first cell and the second cell. RAN node as described in Supplementary Note 25. (Appendix 29) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to transmit, to the other second RAN node, a message including an information element related to execution of the action requested to be executed by the RAN node and the other first RAN node configured to be managed for a common action. RAN node. (Appendix 30) The message further includes an indicator indicating the reason for the result of the action. RAN node as described in Supplementary Note 29. (Appendix 31) The message further includes an indicator of the type of the action. 31. The RAN node of claim 29 or 30. (Appendix 32) the first cell of the RAN node is a neighboring cell of the second cell of the first RAN node, and the action is at least one of offloading or HO (Handover) of a UE (User Equipment); RAN node as described in Supplementary Note 31. (Appendix 33) The message further includes identification information of at least one of the first cell or the second cell that is the target of the action. RAN node as described in Supplementary Note 32. (Appendix 34) the action is off-roading, The message further includes an indicator indicating an amount of load to be offloaded between the first cell and the second cell. RAN node as described in Supplementary Note 32. (Appendix 35) The action is a UE home action; The message further includes an indicator indicating the number of UEs to be handed over between the first cell and the second cell. RAN node as described in Supplementary Note 32. (Appendix 36) A Radio Access Network (RAN) node, comprising: The RAN node Memory and a processor coupled to the memory; a transceiver, the processor is configured to cause the transceiver to receive, from the first and second RAN nodes, messages including information elements related to execution of an action requested of another first RAN node and another second RAN node configured to be managed for a common action. RAN node. (Appendix 37) The message further includes an indicator indicating the reason for the result of the action. RAN node as described in Supplementary Note 36. (Appendix 38) The message further includes an indicator of the type of the action. 38. A RAN node as defined in claim 36 or 37. (Appendix 39) the first cell of the first RAN node is a neighboring cell of the second cell of the second RAN node, and the action is at least one of offloading or HO (Handover) of a UE (User Equipment); RAN node as described in Supplementary Note 38. (Appendix 40) The message further includes identification information of at least one of the first cell or the second cell that is the target of the action. RAN node as described in Supplementary Note 39. (Appendix 41) the action is off-roading, The message further includes an indicator indicating an amount of load that has been offloaded between the first cell and the second cell. RAN node as described in Supplementary Note 39. (Appendix 42) The action is a UE home action; The message further includes an indicator indicating the number of UEs that are subject to handover between the first cell and the second cell. RAN node as described in Supplementary Note 39. (Appendix 43) 1. A method performed by a Radio Access Network (RAN) node, comprising: sending a message to the other RAN node including an information element indicating that the RAN node is connected to an AI (Artificial Intelligence) model capable of managing the other RAN node; The method includes: (Appendix 44) 1. A method performed by a Radio Access Network (RAN) node, comprising: receiving a message from the other RAN node, the message including an information element indicating that the other RAN node is connected to an AI (Artificial Intelligence) model capable of managing the RAN node; The method includes: (Appendix 45) 1. A method performed by a Radio Access Network (RAN) node, comprising: and causing the other RAN node to transmit a message including an information element indicating that the RAN node accepts to be managed by an AI (Artificial Intelligence) model connected to the other RAN node. The method includes: (Appendix 46) 1. A method performed by a Radio Access Network (RAN) node, comprising: receiving a message from the other RAN node including an information element indicating that the other RAN node accepts to be managed by an AI (Artificial Intelligence) model connected to the RAN node; The method includes: (Appendix 47) 1. A method performed by a Radio Access Network (RAN) node, comprising: transmitting a message including an information element indicating a request to perform a common action to another first RAN node and another second RAN node configured to be managed for the common action; The method includes: (Appendix 48) 1. A method performed by a Radio Access Network (RAN) node, comprising: receiving a message from another second RAN node, the message including an information element indicating a request to cause the RAN node and another first RAN node configured to be managed for a common action to perform the action; The method includes: (Appendix 49) 1. A method performed by a Radio Access Network (RAN) node, comprising: sending a message to another second RAN node, the message including an information element related to the execution of the action requested to be executed by the RAN node and another first RAN node configured to be managed for a common action; The method includes: (Appendix 50) 1. A method performed by a Radio Access Network (RAN) node, comprising: receiving messages from the first and second RAN nodes, the messages including information elements related to the execution of a common action requested of the first and second RAN nodes configured to be managed by the first and second RAN nodes; The method includes:

[0165] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the disclosure.

[0166] This application claims priority based on Japanese Patent Application No. 2022-035280, filed on March 8, 2022, the disclosure of which is incorporated herein in its entirety. [Explanation of symbols]

[0167] 10, 20 Communication Systems 11, 12, 13 RAN nodes 21, 22, 23 RAN nodes 24 RAN AI / ML models 25 cells 30UE 40 CN nodes 50 OAM devices

Claims

1. A first NG-RAN node in a Next Generation - Radio Access Network (NG-RAN) that supports AI / ML (Artificial Intelligence / Machine Learning) as a RAN function, means for collecting information to support said AI / ML; The means for collecting the information includes: first means for transmitting a first Xn message, which is a request for collection of the information, to a second NG-RAN node in the NG-RAN; and second means for receiving a second Xn message, which is a response to the request for collection of the information, from the second NG-RAN node; the first Xn message includes a Cell To Report List for the collection of information; the Cell To Report List includes a Cell To Report Item for the collection of information; The Cell To Report Item includes a Cell ID. The first NG-RAN node.

2. A first NG-RAN node in a Next Generation - Radio Access Network (NG-RAN) that supports AI / ML (Artificial Intelligence / Machine Learning) as a RAN function, means for transmitting a first Xn message to a second NG-RAN node in the NG-RAN, the first Xn message being a request for an action related to the AI / ML; means for receiving a second Xn message from the second NG-RAN node in response to a request for an action related to the AI / ML; the first Xn message includes a first information element indicating a type of action associated with the AI / ML; the second Xn message includes a second information element indicating the result of an action related to the AI / ML; The first NG-RAN node.

3. The second Xn message includes a third information element indicating a reason for the result of the action related to the AI / ML. The first NG-RAN node of claim 2.

Images