Network systems and terminals

Abstract

This solution effectively provides an autonomous, next-generation network based on user intent, while leveraging the capabilities of existing networks. [Solution] The network system is composed of a plurality of first network nodes, which include one or more user terminals, one or more base stations to which one or more user terminals are connected, and network entities that constitute a core network to which one or more base stations are connected. The system also includes a first node group having a network configuration in which the plurality of first network nodes are connected, and a second node group consisting of a plurality of AI agents which are a plurality of second network nodes corresponding to the plurality of first network nodes. The plurality of AI agents monitor or manage the target first network node and communicate with each other.

Description

【Technical Field】 【0001】 The present disclosure relates to a network system and a terminal. 【Background Art】 【0002】 Conventionally, there is a fog computing architecture in an environment based on the Internet of Things (for example, Patent Document 1). 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent Translation of PCT International Publication No. 2021-512525 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 An object of the present disclosure is to provide a network system and a terminal that can suitably provide a self-driving next-generation network based on user intent while utilizing the functions of an existing network. 【Means for Solving the Problems】 【0005】 One aspect of the present disclosure is configured by a plurality of first network nodes including one or more user terminals (UEs), one or more base stations to which the one or more UEs are connected, and network entities constituting a core network to which the one or more base stations are connected, a first node group having a network configuration to which the plurality of first network nodes are connected, and a second node group configured by a plurality of AI agents that are a plurality of second network nodes corresponding to the plurality of first network nodes, wherein the plurality of AI agents monitor, operate, or manage a target first network node and communicate with each other among the plurality of AI agents is a network system. 【0006】 Another aspect of this disclosure is a terminal that communicates with one or more base stations, The terminal is characterized in that it is composed of a plurality of first network nodes, including one or more base stations and network entities that constitute a core network to which the one or more base stations are connected, and has communication means for communicating with a network system which includes a first node group having a network configuration and a second node group consisting of a plurality of AI agents that are a plurality of second network nodes corresponding to the plurality of first network nodes, the second node group including an AI agent corresponding to the terminal, and the communication means for transmitting information processed by the AI ​​agent. 【0007】 Other aspects of this disclosure may include a first network node, a second network node (AI agent) constituting a network system, information processing devices operating as such, a program that causes the information processing devices to operate as each network node, and a computer-readable medium storing the program. [Effects of the Invention] 【0008】 According to this disclosure, it is possible to suitably provide an autonomous next-generation network based on user intent while utilizing the functions of existing networks. [Brief explanation of the drawing] 【0009】 [Figure 1] Figure 1 shows an example of the configuration of a network system according to an embodiment. [Figure 2] Figure 2 shows an example of a 5G system configuration. [Figure 3] Figure 3 is a sequence diagram illustrating the procedure for establishing communication between neighboring agents during registration. [Figure 4] Figure 4 is a sequence diagram showing an example of the registration procedure for an AI agent. [Figure 5]Figure 5 is a sequence diagram showing an example of the AI ​​agent discovery procedure. [Figure 6] Figures 6A and 6B show examples of the configuration of an information processing device. [Modes for carrying out the invention] 【0010】 Embodiments of this disclosure will be described in detail with reference to the drawings. Figure 1 is a diagram showing an example configuration of a network system according to an embodiment. Figure 2 is a diagram showing an example configuration of a 5G system. In Figure 1, the network system has a two-layer structure consisting of a first layer and a second layer that is above the first layer. The first layer is an example of a "first node group," and the second layer is an example of a "second node group." However, the first layer / second layer can also be called a core node group / support node group, a main node group / assist node group, or a primary group / secondary group. 【0011】 <1st layer> The first layer is, for example, a fifth-generation mobile communication system (called 5GS or 5G network). 5GS is an example of a "mobile communication network (mobile communication system)" and a "cellular network." As shown in Figure 2, 5GS comprises one or more RANs (Radio Access Networks) 3 to which one or more user terminals (UEs (User Equipment) 2) are connected, and a core network 4 (5GC) to which one or more RANs 3 are connected. Here, a fifth-generation communication system is shown as an example, but it is not limited to this, and could be, for example, a fifth-generation next-generation communication system. 【0012】 UE2 is the user's (subscriber's) terminal. UE2 may be a mobile device such as a smartphone, or it may be a vehicle (a stationary in-vehicle terminal). RAN3 is the access network to core network 4 (5GC) and is composed of base stations (called gNBs). Core network 4 is composed of a collection of multiple network entities (called network functions (NFs)). 【0013】 The core network 4 (5GC) includes multiple types of NFs 11 as follows. However, the types of NFs 11 are not limited to the examples in FIG. 2. UPF (User Plane Function) 11a AMF (Access and Mobility Management Function) 11b SMF (Session Management Function) 11c PCF (Policy Control Function) 11d NEF (Network Exposure Function) 11e NRF (Network Repository Function) 11g NSSF (Network Slice Selection Function) 11h AUSF (Authentication Server Function) 11i UDM (Unified Data Management) 11j NWDAF (Network Data Analytics Function) 11k 【0014】 UPF 11a routes and forwards user packets (packets in the user plane transmitted and received by UE2), performs packet inspection, and QoS processing. UPF 11a is connected to the DN (Data Network) 5. DN 5 is a data network (such as the Internet) outside the 5GC. 【0015】 AMF 11b is a device for accommodating UE2 in the core network 4. AMF 11b accommodates the RAN 3 (base station) and performs subscriber authentication control, location (mobility) management of UE2, etc. UDM 11j is a database (storage device) that provides subscriber information or acquires, registers, deletes, and changes the state of UE2. 【0016】 SMF11c manages PDU (Protocol Data Unit) sessions and controls UPF11a for QoS (Quality of Service) control and policy control. A PDU session is a virtual communication channel for data exchange between UE2 and DN5. 【0017】 PCF11d performs QoS control, policy control, and billing control under the control of SMF11c. QoS control involves controlling the quality of communication, such as prioritizing packet forwarding. Policy control involves communication control such as QoS, packet forwarding eligibility, and billing based on network or subscriber information. NEF11e acts as an intermediary between external nodes (external devices) such as AF (Application Function) 12 and nodes (NF) within the control plane. In other words, NEF11e functions as a gateway (GW) between the core network 4 and the external network. AF12 is an application server (external server) located outside the core network 4 (for example, connected to DN5). 【0018】 NRF11g stores and manages information about the NFs that make up the core network 4. In response to an inquiry regarding an NF that the user wishes to use, NRF11g can return multiple candidate NFs to the inquirer. 【0019】 NSSF11h has the function of selecting the network slice to be used by the subscriber from among the network slices generated by network slicing. A network slice is a virtual network with specifications tailored to its intended use. 【0020】 AUSF11i is a subscriber authentication server that performs subscriber authentication under the control of AMF11b. NWDAF11k has the function of collecting and analyzing data from each NF11, OAM (Operations, Administration, and Maintenance) terminal, AF12, etc. NWDAF11k is an NF that provides analytical information regarding 5GS. 【0021】 Each NF consists of one or more information processing devices. These devices are installed in a special building called a data center. A data center is also called a central office. One or more data centers are located within the communication area of ​​the core network 4, and the data centers are connected by communication lines 7. Each data center is equipped with an OAM terminal for the operation, management, and maintenance of the core network 4. 【0022】 Figure 2 shows examples of NF11a~11k that make up 5GC, but core network 4 is 6G system This may include new NFs, such as new NFs in the system. Each of UE2, base station 3, each NF11, and AF12 is included in the "multiple first network nodes" that constitute the first layer. Figure 1 also illustrates, as an example, a cloud server used to provide various services to the user (UE2) (for example, providing information to UE2) and an edge server located closer to UE2 than the cloud server. The edge server is located (connected) within the core network 4, or in the backhaul connecting the base station and UPF11a, or in a data center close to UPF11a. The cloud server is a form of AF12, but the purpose of using AF12 is not limited to use as a cloud server. The cloud server and edge server are also included in the "multiple first network nodes" that constitute the first layer. 【0023】 <Second layer> The second layer is a network built above the first layer, and has a network configuration in which communication takes place between multiple second network nodes implemented on multiple first network nodes. The second network nodes are AI (Artificial Intelligence) This is called a "gent." Here, the second level is set above the first level, but this is not limited to this; the second level may be at the same level as the first level or below it. 【0024】 AI agents are, for example, intelligent (large-scale language models, or visual-language) agents. Features (such as word-behavior models) that allow the AI ​​to communicate without human intervention, either by itself or other AI agents. It is a network node (network entity) that can work together to perform specific tasks (such as cognition, planning, decision-making, and tool utilization, including understanding intents and the perceived environment). 【0025】 In the second layer, for example, in cases where an NF11 within the core network 4 performs a predetermined task or process in response to a request from AF12 and returns the processing result to AF12, or in cases where a specific NF11 operates to provide some kind of service (such as information) to UE2, queries for the location of the desired NF11 may concentrate on NRF11g, leading to an increase in the load on NRF11g. In this case, for example, the load on NRF11g can be balanced by distributing the functions of NRF11g to the second layer. 【0026】 An AI agent is a virtual network node implemented in the first network node. In other words, an AI agent can be deployed in correspondence with end devices (UE2, AF12, intelligent robots, intelligent cars, etc.) included in the first network node in the first layer, and with the first network node (base station, NF11, edge server, network intelligent assistant, etc.). 【0027】 The AI ​​agent is essentially software (application program) implemented in an information processing device (computer) that operates as the first network node. AI agents communicate with each other at the application layer, and communication processing related to layers below the application layer uses the first layer (for example, existing control plane and user plane protocols of 5GS). In other words, basic communication functions (for example, connection to core network 4, authentication, data transmission and reception, and mobility management of UE2) use the first layer. Therefore, the second layer can be described as a virtual overlay network built on top of the first layer. By utilizing 5GS as the communication infrastructure, the intelligent functions of the layers and AI agents enable the provision of an AI-native next-generation communication network (6G network), including automation of network management, optimization of network resource allocation, and provision of services tailored to user intent. 【0028】 The AI ​​agent may be implemented in a one-to-one correspondence with the first network node (UE2, base station, or NF11, etc.). In the example shown in Figure 1, one AI agent (UE agent) is implemented on one UE2, and one AI agent (base station agent) is implemented on one base station. In addition, one corresponding AI agent (edge ​​agent, NF#1 agent, NF#2 agent, GW agent, AF agent) is implemented on each of the edge servers, NF11 (NF#1, NF#2), NEF11e, and cloud server (AF12). For example, UE11 transmits the acquired user conversation content to the core network 4, and the UE agent processes the transmitted conversation content and provides a service to UE2. Here, for example, UE2 provides the user with a service such as store recommendations based on the conversation content. Here, when the UE transmits data such as conversation content, control may be implemented to indicate whether or not to set a flag requesting that the AI ​​agent process it. For example, data transmitted by the UE without setting a flag will be processed by the UE agent. No processing will be performed. 【0029】 However, a 1:n correspondence is possible, meaning one AI agent may be implemented for two or more first network nodes. For example, one AI agent may be implemented to manage two or more UE2s, or one AI agent may be implemented to manage two or more base stations. A configuration with one AI agent for multiple NF11s is also possible. However, it is not mandatory for all NF11s constituting the core network 4 to have an AI agent implemented, and there may be NF11s without an AI agent implemented. 【0030】 The AI ​​agent can operate autonomously on the first network node (UE2, base station, NF11, etc.) in which it is implemented, and can monitor, operate, and manage at least one of the resources and state of the first network node in which it is implemented (and which it manages). 【0031】 Resources to be monitored, operated, or managed include, for example, computing resources (CPU (Central Processing Unit) and GPU (Graphics Processing Unit) usage of a node), memos. This includes storage usage (free space) and communication resources (wireless, wired bandwidth, etc.). Furthermore, the resources being monitored or managed may include various types of data (system logs, collected data, data features, labels, data volume, and data quality (e.g., the contribution of the data to the training of the relevant AI model)). 【0032】 During communication between AI agents, communication between adjacent AI agents is established when the communication procedure in the first layer (5GS) is executed. For example, during the connection procedure between UE2 and the base station, communication between the UE agent and the base station agent is established. Also, during the connection between the base station and NF11 (AMF11b), communication between the base station agent and the NF agent (AMF agent) corresponding to AMF11b may be established. In the first layer, with respect to the first network node on which the AI ​​agent is implemented, information about the implemented AI agent is registered in NRF11g or UDM11j (if the first network node is UE2), and the first network node can obtain the information as needed, thereby enabling communication between AI agents. By exchanging information between AI agents, for example, a UE agent can request processing from a BS agent or request the acquisition of desired resources. 【0033】 AI agent information includes information for communication between AI agents and information indicating the attributes of the AI ​​agent. For example, AI agent identification information (agent ID), communication address information (IP address, MAC address), type (e.g., UE, base station, NF type, etc.), learning method (e.g., reinforcement learning, associative learning (FL), transfer learning), AoI (Area of ​​Interest: list of TAs or cells), This includes time windows, vendor / interoperability information, etc. 【0034】 The communication between AI agents can be configured as needed. For example, a dedicated protocol for communication between AI agents may be used. Alternatively, the communication may be conducted in a natural language-based manner (for example, if LLM is used, natural language plus prompts may be used). 【0035】 Figure 3 is a sequence diagram illustrating the procedure for establishing communication between neighboring agents during registration. In step S1, a registration request message is sent from UE2 to the base station. The registration request message contains information about the AI ​​agent, including the agent ID and IP address of the UE agent acting as the AI ​​agent. This includes addresses and other information. 【0036】 In step S2, AI agent-to-agent communication (neighboring agent-to-agent communication) is established between UE2 and the base station. Establishing this agent-to-agent communication requires authentication and authorization processing on UE2. The processing in step S2 may be performed after the processing in step S6, which will be described later. 【0037】 In step S3, a Registration Request message is sent from the base station to the AMF11b. The Registration Request message includes information about the AI ​​agent, specifically the UE agent and BS agent (Agent ID, IP address, etc.). 【0038】 In step S4, AI agent-to-agent communication (inter-neighbor agent communication) is established between UE2 and AMF11b. This agent-to-agent communication is established depending on the authority and necessity. The processing in step S4 may be performed after the processing in step S6, which will be described later. 【0039】 In step S5, the registration procedure is performed between AMF11b and a designated NF11 such as PCF11d or SMF11c. In the 5GS procedure, for example, the location of the UE2 is registered with UDM11j, and information on SMF11c, UPFa, and PCF11d is registered with NRF11g. Based on the registration information in NRF11g, AMF11b selects either SMF11c or PCF11d, and the selected SMF11c selects UPFa, establishing a PDU session with the UE2. At this time, if cooperation is required between AMF11b and PCF11d, or between AMF11b and SMF11c, depending on authority and necessity, information on the UE agent and BS agent is shared between them, and inter-agent communication is established. 【0040】 In step S6, AMF11b sends a message to UE2 indicating that the registration has been accepted (Registration Accept). 【0041】 Figure 4 is a sequence diagram showing an example of the AI ​​agent registration procedure. In step S11, NF11 (such as AMF11b) sends a message (Nnrf_NFManagement_NFRegister) to NRF11g at an appropriate time, requesting registration of the AI ​​agent's information. The AI ​​agent's information, including the agent ID and IP address, is registered by NRF11g upon receiving the message. In step S12, NRF11g replies to NF11 with a message (Register result) indicating that registration is complete. 【0042】 Figure 5 is a sequence diagram showing an example of the AI ​​agent discovery procedure. In step S21, the NF service customer sends a message (Nnrf_NFDiscovery_Request) to the NRF11g requesting the discovery of the desired NF11. This message contains information to identify the NF11 corresponding to the desired AI agent. The NF service customer is a type of NF11. The NF service customer may be configured by extending an existing NF11 so that an existing NF11 operates as an NF service customer. Alternatively, a new NF11 may be provided that operates in response to requests from an existing NF11. 【0043】 In step S22, NRF11g sends a discovery reply message (Nnrf_NFDiscovery_Reply) to the NF service customer in response to the discovery request. The discovery reply contains information about the AI ​​agent for the discovered NF11. In this way, the NF service An NF11 (such as an AMF11b) acting as a Viscus customer can, prior to the processing of step S5 shown in Figure 3, obtain information on the desired AI agent (NF agent) from the NRF11g as needed and establish inter-agent communication. 【0044】 <Configuration of information processing equipment and terminal> Figure 6A shows an example configuration of an information processing device that can operate as NF11a~11k, AF12, a cloud server, or an edge server. In Figure 6A, the information processing device 20 can be configured using a dedicated or general-purpose information processing device (computer) such as a personal computer (PC), workstation (WS), or server machine. However, the information processing device 20 may also be a collection of one or more computers (cloud). 【0045】 The information processing device 20 includes a processor 21 acting as a processing unit or control unit (controller), a storage device 22, a communication interface 23 (communication IF 23), an input device 24, and a display 25, all interconnected via a bus 26. 【0046】 The storage device 22 includes main memory and auxiliary storage. The main memory is used as at least one of the following: a program and data storage area, a program deployment area, a program work area, and a communication data buffer area. The main memory consists of RAM (Random Access Memory), or a combination of RAM and ROM (Read Only Memory). The auxiliary storage is used as a data and program storage area. Non-volatile storage media are used for the auxiliary storage. Non-volatile storage media include, for example, hard disks, solid state drives (SSDs), flash memory, or EEPROM (Electrically Erasable Programmable Read-Only Memory). The storage device 22 may also include a drive device for a disk recording medium. 【0047】 Communication IF23 is a circuit that performs communication processing. For example, communication IF23 is a network interface card (NIC). Alternatively, communication IF23 may be a wireless communication circuit that performs wireless communication (5G, wireless LAN (Wi-Fi), BLE, etc.). F23 may be a combination of a wired communication processing circuit and a wireless communication circuit. 【0048】 The input device 24 includes keys, buttons, pointing devices, and touch panels, and is used for inputting information. The display 25 is, for example, a liquid crystal display and displays information and data. 【0049】 The processor 21 performs various processes by executing various programs stored in the storage device 22. By the processor 21 executing the programs stored in the storage device 22, the information processing device 20 can operate as NF11a~11k, AF12, cloud server, or edge server, respectively. 【0050】 Figure 6B shows an example configuration of a terminal 40 (information processing device) capable of operating as UE2. The terminal 40 includes a processor 41 (an example of "providing means" or "control means"), a storage device 42, a communication interface 43 (a communication IF 43, an example of "communication means"), an input device 44, and a display 45, all interconnected via a bus 46. The processor 41, storage device 42, communication IF 43, input device 44, and display 45 can be the same as those used for the processor 21, storage device 22, communication IF 23, input device 24, and display 25. Therefore, their descriptions are omitted. The terminal 40 may also be a mobile device or a mobile device such as a vehicle. The terminal 40 may have a position detection device such as a GPS receiver, or various sensors such as radar, lidar, or a camera (image sensor). 【0051】 The information processing device 20 stores a program for it to operate as an AI agent, as well as data used when the program is executed, in the storage devices 22 and 42. For example, the storage device 22 of the information processing device 20 operating as an NRF11g or UDM11j stores AI agent information. 【0052】 Processors 21 and 41 are, for example, CPUs. A CPU is also called a Microprocessor Unit (MPU). Processors 21 and 41 can be a single processor configuration. A multi-core configuration is also possible. Alternatively, a single physical CPU connected via a single socket may have a multi-core configuration. Processors 21 and 41 may include various circuit configurations of arithmetic units such as Digital Signal Processors (DSPs) or GPUs. Furthermore, processors 21 and 41 may have configurations that interact with at least one of the following: integrated circuits (ICs), other digital circuits, and analog circuits. Integrated circuits include LSIs, Application Specific Integrated Circuits (ASICs), and programmable logic devices. This includes PLDs, for example, Field-Programmable Gate Arrays (FPGAs). This includes processors 21 and 41, which also include, for example, microcontrollers (MCUs), SoCs (System-on-a-chip), system LSIs, or chipsets. 【0053】 In the example shown in Figure 1, in a system where UE2 receives services (data or information) from a cloud server or edge server using 5GS, an AI agent is implemented to monitor or manage a first network node located on the path between UE2 and the cloud server or edge server. By monitoring or managing the first network node, the AI ​​agent can reduce the load on a specific NF11 (e.g., NRF11g) (load balancing). In other words, in 5GS, information regarding NF11 in the core network 4 is centrally managed by NRF11g, whereas the first network node is managed in a distributed manner at the second layer, thereby reducing the load on NRF11g. For example, when a user inputs information into UE2, the UE agent implemented in UE2 operates autonomously and communicates with other AI agents. At this time, queries to NRF11g that would normally occur in 5GS operation are replaced by distributed processing through AI agent communication, thereby reducing the load on NRF11. However, centralized management by NRF11g may be retained. For example, information about each AI agent is registered in NRF11g, and the Discovery of adjacent AI agents queries NRF11g. I will contact NRF11g regarding this matter. 【0054】 Furthermore, dynamic information such as load information for each NF is managed locally by each AI agent, and optimal network management (e.g., selecting an NF based on load conditions) is achieved through communication between AI agents, eliminating the need to register with NRF11g and reducing the load on NRF11g. Through communication between AI agents, for example, data held by UE2 can be transmitted to an AF agent implemented on the cloud server via the UE agent, BS agent, NF#2 agent, and GW agent. Conversely, through the same path, the UE agent implemented on UE2 can receive data from the cloud server to the UE agent (UE2). Also, under predetermined conditions, the UE agent can communicate with an edge agent implemented on an edge server through interaction with the NF#1 agent and send and receive data. In this way, through communication at the second layer, UE2 can receive data or information (service provision) from the cloud server or edge server in response to service requests. According to the network system of this embodiment, by using 5GS (first layer), it is possible to realize communication between AI agents (second layer) while keeping the scale of development under control. According to this embodiment, it is possible to suitably provide an autonomous next-generation network based on user intent while utilizing the functions of existing networks. 【0055】 The configurations shown in the embodiments can be combined as appropriate, as long as the purpose of the disclosure is achieved. Furthermore, the processes and means described in this disclosure can be freely combined and implemented, as long as no technical inconsistencies arise. Also, processes described as being performed by one device may be divided and executed by multiple devices. Conversely, processes described as being performed by different devices may be executed by a single device. In a computer system, the hardware configuration (server configuration) used to implement each function can be flexibly changed. 【0056】 The present disclosure can also be realized by supplying a computer program implementing the functions described in the embodiments above to a computer, and having one or more processors in the computer read and execute the program. Such a computer program may be provided to the computer by a non-temporary computer-readable storage medium that can be connected to the computer's system bus, or it may be provided to the computer via a network. The non-temporary computer-readable storage medium includes any type of disk, such as magnetic disks (floppy disks, hard disk drives (HDDs), etc.), optical disks (CD-ROMs, DVDs, Blu-ray discs, etc.), read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic cards, flash memory, or optical cards, and any other type of medium suitable for storing electronic instructions. 【0057】 <Note> (Note 1) A group of first nodes comprising a network configuration in which the plurality of first network nodes are connected, the group of first network nodes comprising a plurality of first network nodes including one or more user terminals (UEs), one or more base stations to which the one or more UEs are connected, and network entities that constitute a core network to which the one or more base stations are connected, It includes a group of second nodes, which consists of multiple AI agents, that are multiple second network nodes corresponding to the multiple first network nodes, The aforementioned multiple AI agents monitor or manage the target first network node and communicate with each other. Network system. (Note 2) The plurality of first network nodes further include external nodes connected to the core network. The network system described in Appendix 1. (Note 3) The second group of nodes includes an AI agent corresponding to the external node. The network system described in Appendix 2. (Note 4) The plurality of first network nodes further include one or more cloud servers, or one or more cloud servers and one or more edge servers. The network system described in Appendix 1. (Note 5) The second node group includes an AI agent corresponding to one or more cloud servers or an AI agent corresponding to one or more edge servers. The network system described in Appendix 4. (Note 6) A specific network node among the plurality of first network nodes stores information about the AI ​​agent, including information for communication of the AI ​​agent and information indicating the attributes of the AI ​​agent. The aforementioned multiple AI agents use information about each AI agent when communicating with each other. The network system described in Appendix 1. (Note 7) The second node group comprises an AI agent corresponding to one or more UEs, an AI agent corresponding to one or more base stations, and a network entity corresponding to The network system described in Appendix 1, including the AI ​​agent. (Note 8) The first node group is the network system described in Note 1, which is configured based on a fifth-generation mobile communication system. (Note 9) The aforementioned network entities are UPF (User Plane Function) and AM F(Access and Mobility Management Function), SMF(Session Management Function), PCF(Policy Control Function), NEF(Network Exposure Function), NRF(Network Repository Function), NSSF(Network Slice Selection Function) ), AUSF (Authentication Server Function), UDM (Unified Data Management ), and one or more of the following: NWDAF (Network Data Analytics Function) The network system described in Appendix 1. (Note 10) A terminal that communicates with one or more base stations, A terminal comprising a plurality of first network nodes, including one or more base stations and network entities constituting a core network to which the one or more base stations are connected; having communication means for communicating with a network system comprising a first node group having a network configuration and a second node group consisting of a plurality of AI agents which are a plurality of second network nodes corresponding to the plurality of first network nodes; the second node group includes an AI agent corresponding to the terminal; and the communication means transmits information processed by the AI ​​agent. (Note 11) The terminal according to Note 10, further comprising means for providing a service to a user based on the AI ​​agent processing the information. (Note 12) The terminal according to Note 10, characterized in that the AI ​​agent corresponding to the terminal can detect other AI agents. (Note 13) The terminal according to Note 10, further comprising control means for controlling whether or not to transmit data to be processed by the AI ​​agent corresponding to the terminal. [Explanation of Symbols] 【0058】 2. User Terminal (UE) 3···RAN 4. Core network 5···DN 11···NF 12···AF 20. Information Processing Devices 21,41... processors 22,42...Storage device 23,43...Communication Interface

Claims

[Claim 1] A group of first nodes having a network configuration in which the plurality of first network nodes are connected, comprising at least one of the following: one or more user terminals, one or more base stations to which the one or more user terminals are connected, and a group of first network nodes having a network configuration in which the plurality of first network nodes are connected, It includes a group of second nodes, which are a plurality of AI agents that are a plurality of second network nodes corresponding to the plurality of first network nodes, The aforementioned multiple AI agents monitor or manage the target first network node and communicate with each other. Network system. [Claim 2] The plurality of first network nodes further include external nodes connected to the core network. The network system according to claim 1. [Claim 3] The second group of nodes includes an AI agent corresponding to the external node. The network system according to claim 2. [Claim 4] The plurality of first network nodes further include one or more cloud servers, or one or more cloud servers and one or more edge servers. The network system according to claim 1. [Claim 5] The second group of nodes includes an AI agent corresponding to one or more cloud servers or an AI agent corresponding to one or more edge servers. The network system according to claim 4. [Claim 6] A specific network node among the plurality of first network nodes stores information about the AI ​​agent, including information for communication of the AI ​​agent and information indicating the attributes of the AI ​​agent. The aforementioned group of AI agents use information about each AI agent when communicating with each other. The network system according to claim 1. [Claim 7] The network system according to claim 1, wherein the second node group includes an AI agent corresponding to one or more user terminals, an AI agent corresponding to one or more base stations, and an AI agent corresponding to the network entity. [Claim 8] The network system according to claim 1, wherein the first group of nodes is configured based on a fifth-generation mobile communication system. [Claim 9] The aforementioned network entity includes UPF (User Plane Function), AMF (Access and Mobility Management Function), SMF (Session Management Function), and PC. F (Policy Control Function), NEF (Network Exposure Function), NRF (Network Repository Function), NSSF (Network Slice Selection Function), AUS F (Authentication Server Function), UDM (Unified Data Management), NWD Includes one or more of the AF (Network Data Analytics Function) The network system according to claim 1. [Claim 10] A terminal that communicates with one or more base stations, The system has communication means for interacting with a network system which includes a plurality of first network nodes comprising one or more base stations and network entities that constitute a core network to which the one or more base stations are connected, a first node group in which the plurality of first network nodes have a network configuration and a plurality of AI agents which are a plurality of second network nodes corresponding to the plurality of first network nodes, and a second node group which communicates among the plurality of AI agents, The terminal is characterized in that the second node group includes an AI agent corresponding to the terminal, and the communication means transmits information processed by the AI ​​agent. [Claim 11] The terminal according to claim 10, further comprising means for providing a service to a user based on the AI ​​agent processing the information. [Claim 12] The terminal according to claim 10, characterized in that the AI ​​agent corresponding to the terminal can detect other AI agents. [Claim 13] The terminal according to claim 10, further comprising control means for controlling whether or not to transmit data to be processed by the AI ​​agent corresponding to the terminal.

Images