A system and methodology that enables inter-agent discovery by leveraging network capabilities offered to agents for next-generation mobile networks.

Abstract

The invention relates to a system and method that enables the discovery and matching of suitable AI agents by jointly evaluating the capabilities, protocols, and identification information of AI agents accessed via or deployed on network components of next-generation mobile networks, along with the status of communication, sensing, and computational network capabilities offered by the mobile network; monitors network capabilities through key performance indicators and service level values; makes predictions based on historical and real-time network data; and updates agent discovery when a mismatch occurs in the ongoing matching process due to network capability issues.

Description

1 TARIFF Utilizing Network Capabilities Provided to Agents for Next-Generation Mobile Networks A system and method that enables inter-agent reconnaissance. TECHNICAL FIELD The invention utilizes the network capabilities of artificial intelligence agents in next-generation mobile networks. It relates to the inter-agent discovery system and method that enables its discovery. The invention is particularly relevant to 6G mobile networks, also known as sixth generation networks. the capabilities of artificial intelligence agents that are accessible or included in the components of these networks, protocol and identification information, as well as communication and detection provided by the mobile network. and discovery, matching, networking according to the state of computational capabilities It relates to the area of ​​estimating competencies and updating ongoing matches. 15 PREVIOUS TECHNIQUE Today, artificial intelligence-based software components are used in mobile communication systems. The user device must be run in different network locations such as edge, core network and cloud, and these 20 The components increasingly work together by accessing data, computing, and communication resources. It is becoming widespread. In such structures, one artificial intelligence agent collaborates with another agent. In order to establish a business or obtain services from it, the identification of a suitable agent and the agent's capabilities, This requires an assessment of supported communication features and network conditions. Mobile Network 25 features such as connection speed, latency, computing capacity, and sensing capabilities in networks Since their capabilities can directly affect the mission performance of agents, agent reconnaissance the mechanisms rely not only on application-level information but also on the technical information provided by the network. Being sensitive to competencies is also technically important. Starting with third-generation communication systems, their work continues with sixth-generation mobile communication. Service 30, within the 3rd Generation Partnership Project (3GPP) that continues with its systems. and System Aspects Work Group 1 (SA1), 3GPP Technical Report (TR) 22.870 It examines artificial intelligence use cases along with their potential requirements. The document describes how the artificial intelligence agent interacts with its environment, acquiring contextual information and reasoning. autonomously or independently by acting, learning on its own, making decisions, and performing tasks. 35 as an automated and intelligent unit that achieves a specific goal on behalf of another unit in a non-existent way It is defined in 3GPP use cases at the user device, at the edge, and at the core network. And it appears that artificial intelligence agents residing in the cloud perform various functions. 2 Artificial intelligence agents, especially in conjunction with 6G standard development for mobile communication systems. creating use cases that include, determining network requirements, and Research and development activities aimed at designing applications have increased. together with 3GPP, International Telecommunication Union (ITU) and Internet Engineering Task 5 Formally accepted in the standard documents of standards-developing organizations such as the International Institute for Standardization (IETF) There is currently no inter-agent communication protocol or agent discovery mechanism in place. Among the structures that have gained recognition in the market, the Agent2Agent (A2A) protocol is offered by different manufacturers and the frameworks created allow agents to discover each other, share their skills and While aiming for mutual assignment, Model Context Protocol (MCP) is a large language 10 an open-source system that provides access to data sources, tools, and workflows for its models It is designed as a communication protocol and is not directly aimed at communication between agents. In the A2A approach, an agent card contains agent capabilities, supported protocols, and identification information. Methods and endpoints to which task requests will be sent can be defined. This agent 15 Access to the card via a fixed and known web path, query to the catalog-based central unit. This can be achieved through either directives or broadcast-based dynamic discovery. However, the aforementioned in a general-purpose exploration approach accessible via mobile communication systems or mobile access Connection speed, which can affect the task performance of agents involved in the system components, Agent 20 capabilities offered by the network, such as latency and computational ability. It needs to be considered as a criterion for selection. The aforementioned shortcomings relate to agent discovery, agent capabilities alone, and general record information. If conducted via mobile network, communication, detection and The current and projected state of computing network capabilities directly influences the discovery decision. This prevents it from being reflected, and is particularly relevant in next-generation mobile networks. This highlights the need to incorporate network conditions affecting performance into the agent discovery mechanism. It extracts. As a result of research conducted in the literature, the publication numbered US20240039982A1, 'Methods and 30 devices for providing a discovery service for a network including collaborative agents patent titled "Methods and devices for providing reconnaissance services for a network including agents" The document was found. This document outlines the conditions of a collaboration request for the nodes. Identifies potential nodes for collaboration by comparing contextual information and discovery data. It relates to a discovery service that transmits information to the requesting node. The document states that there are 35 nodes currently. computing and communication resources, sensing capabilities, resource utilization status, and location. It is explained that contextual information such as this can be used in discovery. However, mobile network components... 3 by interacting with each other, the suitability of key performance indicators and service level values. a network interaction module that evaluates the services offered to agents by the mobile network Monitoring the status of communication, sensing and computing network capabilities, and this data It is used in agent matching and networking along with agent registration and payload data. The periodic changes in their competencies are predicted using the digital twin prediction module. 5 related to an agent discovery system that uses matching update features in conjunction with each other. No evidence has been found. Literature research has revealed that the 'Cross-domain agent' with publication number CN122027542A Discovery routing method and system for agent network (Interdomain agent discovery 10 for agent network) A patent document titled "guidance method and system" was found. The document includes trust lists, chaining, or bulk routing across multiple exploration areas. Cross-domain agent discovery is performed using guidance boundaries and path tracking. This is related to the document, such as average response latency and cross-domain discovery success rate. It is also explained that the guidance method can be selected according to performance indicators. However, 15 Communication, sensing, and computation capabilities offered to artificial intelligence agents by mobile networks. Key performance indicators from mobile network components regarding the agent-based status of services. Monitoring the network's capabilities through indicators and service level values. The use of agent registration and payload data as matching criteria, and 20 features of ongoing agent matching updates with digital twin prediction data No evidence has been found of it being used in conjunction with an agent detection system. Ultimately, the problems mentioned above, which cannot be solved with current technology, are the subject of this technical analysis. This has made it necessary to make an innovation in the field. A BRIEF DESCRIPTION OF THE INVENTION The invention aims to eliminate the aforementioned shortcomings and bring new advantages to the relevant technical field. using network capabilities offered to agents on next-generation mobile networks to bring about change. It relates to a system and method that enables inter-agent reconnaissance. 30 The main purpose of the invention is to create devices accessible via mobile communication systems or mobile communication systems. The artificial intelligence agents in its components are evaluated not only based on their agent capabilities, but also on the agents themselves. exploration taking into account the communication, computing and sensing network capabilities offered. and to ensure matching. 35 4 Another purpose of the invention is to improve the communication speed, latency, and detection specified in the agent discovery request. the functionality and computing capacity requirements along with the network's current service status The aim is to ensure its evaluation. Another objective of the invention is to analyze periodic changes in network capabilities, both past and current network 5. Estimating the duration of collaboration or service procurement through simulation methods using data. The goal is to ensure that the outcome of the linked agent discovery is determined according to current network conditions. Another objective of the invention is to address network capabilities-related issues in ongoing agent matching. When a discrepancy occurs, the agent discovery should be updated, and the relevant 10 of the mission or service procurement should be completed. The context, along with the new agent, is then transferred and updated match and workload information is subsequently obtained. The aim is to make it available for exploration. Accessed via or within the components of next-generation mobile networks AI agents deployed with agent capabilities are provided to the agents by the mobile network with 15 By evaluating the network capabilities offered, they can discover each other and collaborate or It is a system that enables matching for the purpose of service procurement, and its feature is;  Artificial intelligence through software running on at least one electronic device an inter-agent reconnaissance unit that operates the reconnaissance mechanism among its agents,  20 AI agents that will interact through the inter-agent discovery unit their capabilities, supported protocols, identification methods, and tasks agents who record the endpoints to which requests are sent and share this information in authorized inquiries. registry server,  with information from matched AI agents to establish collaboration or receive services Record agent payload and mission information and connect with inter-agent reconnaissance unit components 25 Shared agent match payload database,  By interacting with relevant mobile network components, key performance indicators and assessing the appropriateness of service level values ​​and providing agents with next-generation solutions. communication, sensing and computing network capabilities offered by the mobile network Network interaction module monitoring status, 30  Agent capabilities to be discovered as specified in the incoming collaboration or service request, Supported protocols and authentication methods, communication speed, and latency. network capability in terms of duration, sensing function and computational capacity requirements, agent information located on the agent registry server, agent The matches are based on data in the load-duty database and network interaction module 35. along with access status to network capabilities obtained through Agent scouting unit that matches agents by evaluating them. It includes. In order to best understand the structure of the invention and its advantages, including additional components It should be evaluated together with the figures explained below. BRIEF DESCRIPTION OF THE FIGURES Figure 1 is a representative illustration of the system structure related to the invention. Figure 2 is a representative flowchart illustrating the steps of the method described in the invention. The drawings do not necessarily need to be scaled and are not essential for understanding the invention. Details may have been omitted. Furthermore, they are at least significantly identical. Elements, or elements with similar functions, are represented by the same number. REFERENCE NUMBERS 15 100. User terminal 101. Terminal agent 200. Access network remote end area 210. First base station 20 211. First-line field agent 220. Second base station. 221. Second end-field agent 300. Core network area 310. First core network unit 25 311. First core network agent 320. Second core network unit. 321. Second core network agent 330. Third core network unit. 331. Third core network agent 30 340. Fourth core network unit. 341. Fourth core network agent 400. Cloud environment 401. First cloud agent 402. Second cloud agent 35 403. Third cloud agent 404. Fourth cloud agent 6 500. Inter-agent reconnaissance unit 501. Agent log server 502. Agent Reconnaissance Module 503. Network interaction module 504. Digital twin prediction module 5 505. Agent Matches Payload Database 1001. Accessible via mobile communication systems or included in mobile access system components. terminal agent, first endpoint agent, second endpoint agent, first core network agent, second core network agent, third core network agent, fourth core network agent, first 10 Capabilities provided by cloud agent, second cloud agent, third cloud agent and fourth cloud agent, supported protocol, identification method and endpoint to which task requests will be sent the information is transferred to and recorded on the agent log server 1002. To be discovered by an artificial intelligence agent that wants to collaborate or obtain services. The capabilities that agents will possess, the protocols that will be supported, and the identification methods will be 15. Communication speed, latency, and sensing function for tasks to be carried out jointly or services to be provided. and the agent which includes the network capabilities that must be provided within the scope of computing capacity. transfer of the reconnaissance request to the agent reconnaissance unit 1003. Agent information found on the agent log server by the agent reconnaissance unit, agent Matches are made using data from the load-duty database and the network interaction module. Access status of the acquired agents to communication, sensing and computing network capabilities By evaluating them together, agent matching is carried out in accordance with the agent discovery request. 1004. The information received from the grid by the grid interaction module by the digital twin prediction module. Traffic load can be determined using historical and real-time network data and various simulation methods. User demands, coverage quality, and resource utilization differ in the short and long term. 25 Predicting based on their intervals, detecting anomalies and faults in the network in advance, and Performing assumption analyses for different scenarios 1005. Network capabilities while agent matching continues based on agent discovery request. a mismatch originating from the network interaction module via the digital twin prediction module If detected, the agent reconnaissance unit will be informed and the agent reconnaissance report will be updated. 30 The newly discovered agent should be informed of the collaboration task or service procurement, along with the relevant context. and new match, payload and mission information are added to the agent match payload / mission database. recording DETAILED DESCRIPTION OF THE INVENTION 35 7 This detailed explanation describes the offerings provided to agents for the next-generation mobile networks that are the subject of the invention. A system and method that enables inter-agent reconnaissance using network capabilities. only with examples that will not have any limiting effect on better understanding of the subject It is explained. User terminal (100), mobile phone, smartphone, tablet, computer or internet of things It has various forms, primarily the module, and allows mobile network subscribers to listen to the network's radio. It is a device that enables the user to connect to the access network and receive network services. Terminal agent (101), user resources, information, tools and other artificial intelligence that can be accessed with this device located in its terminal (100). Intelligence agents interact with their environment through collaboration, acquire contextual information, and use reason. 10 autonomous or non-autonomous, performing tasks, learning on its own, making decisions, and carrying out tasks. It is an artificial intelligence agent that achieves a specific goal on behalf of another entity in a remote access network. The end area (200) is where the user terminal (100) can connect to the mobile network via radio access methods. central network and control signaling such as identification and service request 15 that provides wireless network connectivity enabling it to perform tasks and access network services It is infrastructure and contains application layer software when positioned as a remote endpoint. It hosts the servers. The first base station (210) is in the access network remote end area. (200) communicates with the user terminal (100) via radio air interface as the basic unit. and enables the user terminal (100) to reach the central network. The first endpoint agent (211), located in the remote end area of ​​the access network (200) and interacting with its environment, 20 contextual information acquisition, reasoning, independent learning, decision-making, and task completion. It is an artificial intelligence agent that performs its functions. The second base station (220) is the access network remote end. radio air interface with user terminal (100) as a second base station in the area (200) It communicates via and enables the user terminal (100) to reach the central network. Second The end-field agent (221) can be found in numerous end-fields (200) in the access network remote end-field. A second example of an agent is interacting with its environment, acquiring contextual information, reasoning. The core performs executive functions, independent learning, decision-making, and task execution. Network area (300), identification for subscribers using user terminal (100), It offers session management and mobility management functions, from the access network remote end area. (200) By transmitting the user data packets it receives to the relevant places, it provides data service with realities and quality. 30 It performs the billing functions. The first core network unit (310), core network It is a unit that performs some of the functions of the (300) area and for example session management for example, Session Management Function (SMF) or User Plane Function for data transfer. can perform network functions. The first core network agent (311), first core network It is an artificial intelligence agent located within unit (310). The second core network unit (320) is 35 It is another unit that performs a part of the functions of the core network area (300). Second core network agent (321), artificial located within the second core network unit (320) 8 It is an intelligence agent. The third core network unit (330) functions of the core network area (300). It is another unit that performs a part of it. The third core network agent (331), third It is an artificial intelligence agent located within the core network unit (330). Fourth core The network unit (340) performs a part of the functions of the core network area (300). It is another unit. Fourth core network agent (341), fourth core network unit (340) 5 It is an artificial intelligence agent within the cloud environment (400), server, computing unit, Internet through hardware and software components that include storage units and network solutions. hosting applications and providing cloud resources and services as needed on demand. It ensures that it is properly made accessible. The first cloud agent (401) is in the cloud environment (400) is the AI ​​agent positioned. The second cloud agent (402) is located in the cloud environment (400) 10 It is the second AI agent deployed. The third cloud agent (403) is located in the cloud environment (400) It is the third AI agent deployed. The fourth cloud agent (404) is located in the cloud environment. (400) is the fourth AI agent positioned. Inter-agent exploration unit (500), user in the terminal (100), in the access network remote end area (200), in the core network area (300) or AI agents in the cloud environment (400) discover each other and work together 15 Inter-agent reconnaissance in situations requiring their cooperation or mutual service provision. It operates the mechanism. Agent registry server (501), inter-agent reconnaissance unit (500) The abilities of agents that will interact with other agents through the means they support basic protocols, identification methods, and endpoints to which task requests will be sent It records the information and shares it with the relevant units in authorized inquiries. Agent discovery module (502), incoming 20 To discover agents suitable for a collaboration or service request, on the agent registry server (501) Agent information, payload and mission data and network capability statuses found It performs matching between agents by evaluating them together. Network interaction module (503), The communication, sensing, and computing network offered to agents by the next-generation mobile network. Their capabilities are assessed by interacting with relevant mobile network components and key performance 25 It monitors and evaluates the suitability of indicators and service level values, and provides the following: The subject transmits the situations to the agent discovery module (502). The digital twin prediction module (504), network historical and real-time network data received from the network by the interaction module (503) and various Network capabilities to be presented to agents periodically using simulation methods It predicts how it will change, the traffic load on the network or user demands, coverage 30 It estimates quality and resource utilization over different short-term and long-term time intervals. It detects anomalies and malfunctions in the network in advance and makes assumptions for different scenarios. It performs analyses. Agent matches payload database (505), establish cooperation or service It records and stores information about agents that are matched with each other in order to obtain and manage the agents' workload. Agent reconnaissance unit 35, primarily the agent reconnaissance module (502) which uses it as a criterion. (500) shares with other components as needed. 9 The invention enables access to or via next-generation mobile networks, described as sixth-generation. artificial intelligence agents that are directly part of the components of these networks interact with each other. It is geared towards situations where they need to work together or utilize each other's services by exploring and discovering new things. Instead of selecting agents solely based on their abilities, the selection process should also consider the agents' affiliations. the quality of communication, computing and sensing services they receive from the network and the agents' presence on the network 5 Network services are available thanks to the consideration of their location. The situation plays an active role in agent reconnaissance. In this context, connection speed and communication problems will occur. Values ​​such as latency, detection capability, and computational capacity are crucial for the artificial intelligence that agents operate. These are considered network capabilities that can affect the performance of intellectual functions. Mobile communication systems or mobile access systems that can be accessed during the agent registration phase. terminal agent (101), first end field agent (211), second end field agent included in its components (221), first core network agent (311), second core network agent (321), third core network agent (331), fourth core network agent (341), first cloud agent (401), second cloud Agent (402), third cloud agent (403) and fourth cloud agent (404) have the capabilities of 15 the protocols they support, the identification methods, and the endpoints to which mission requests will be sent The points are shared with the agent registry server (501). The agent registry server (501) receives It records the information and transmits it to the relevant units in response to authorized inquiries. To cooperate with agents found on the agent registry server (501) or to obtain 20 of these agents an AI agent wishing to receive service, agent reconnaissance unit (500) The capabilities that the agents to be discovered will have, the protocols to be supported and in the module (502) It transmits basic information including identification methods. The same agent discovery request will be carried out jointly. Communication speed, latency, detection function, and calculation related to the task or service to be provided. network capabilities that next-generation mobile network services need to provide, such as capacity, are also 25 includes. During the reconnaissance process, the agent reconnaissance module (502) is located on the agent registry server (501). Information such as capabilities, supported protocols, and identification methods are used for agent matching. Data contained in the load task database (505) and via the network interaction module (503) 30 Access to communication, sensing, or computing network capabilities of acquired agents By jointly evaluating their situations, they perform the appropriate agent match for the agent scouting request. Agent matching is one of the discovery criteria for agent payloads (505) in the payload database. This allows new requests to be evaluated based on current agent loads. 35 As part of monitoring network capabilities, the network interaction module (503) monitors the relevant mobile network. by interacting with its components, key performance indicators and service level values assesses suitability and provides agents with communication offered by the next-generation mobile network. by monitoring the status of sensing and computing network capabilities to the agent discovery module. (502) transmits. The network interaction module (503) also receives historical and instantaneous network data from the network. transmits its data to the digital twin prediction module (504) and receives data from the digital twin prediction module (504) receives network capability estimation data. 5 Within the scope of prediction, the digital twin prediction module (504) will be presented to the agents as a network. Network interaction in order to predict how competencies will change over time module (503) receives historical and instantaneous network data from the network and various simulations It uses methods such as traffic load or user demands, coverage quality and 10. Resource utilization is estimated for different time intervals in the future, both short-term and long-term. Network anomalies and faults are detected in advance, and hypothetical analyses are performed for different scenarios. This is carried out. The duration of cooperation or service procurement between agents in an agent reconnaissance request. When specified, the network interaction module (503) receives from the digital twin prediction module (504). By transmitting network capability estimation data to the agent discovery module (502), the agent discovery result is stated as 15 This ensures that the subject is determined by taking into account the expected network conditions during the specified period. While the agent matching process initiated by the agent discovery request continues, the network in the agent discovery request... Digital twin estimation module of a non-conformity in terms of competency requirements (504) If detected by the network interaction module (503) via network 20 The interaction module (503) provides information to the agent discovery module (502). The agent discovery module (502) informs the agent. It updates its discovery and assigns the newly discovered agent a collaboration task or service procurement in the relevant context. They transfer it together. New match, payload and mission information are transferred to the agent match payload mission database. (505) by recording subsequent agent discovery requests based on current agent payloads. An evaluation is provided. 25 In a structure of the invention, the access network remote end area (200), end area, remote end area and near can be separated into end areas and access network remote end area (200) and core network area Additional regions can be created between (300) which are described as near end zones. These regional without changing the operation and system structure of the inter-agent discovery unit (500) 30 applicable. The steps involved in the process are as follows:  Accessible via mobile communication systems or located in mobile access system components area terminal agent (101), first end area agent (211), second end area agent (221), first 35 core network agent (311), second core network agent (321), third core network agent (331), fourth core network agent (341), first cloud agent (401), 11 second cloud agent (402), third cloud agent (403) and fourth cloud agent (404) capability, supported protocol, identification method and task endpoint information to which requests will be sent to the agent registry server (501) being transferred and recorded (1001),  By an AI agent that wants to collaborate or obtain services, 5 the capabilities that the agents to be discovered will possess, the protocols to be supported, and for tasks or services to be carried out jointly using identification methods in terms of communication speed, latency, sensing function and computational capacity Agent discovery request including required network capabilities Transfer to module (502) (1002), 10  Agent discovery module (502) located in agent registry server (501) Agent information, agent matches, payload task database data (505) and Communication, sensing and agents obtained via network interaction module (503) The agent's access to computing network capabilities is evaluated together with the other parties. Performing agent matching in accordance with the discovery request (1003), 15  Network interaction module (503) by the digital twin prediction module (504) using historical and real-time network data received from the network and various simulation methods. traffic load or user demands, coverage quality and resources are used. Predicting its usage based on different short-term and long-term time intervals, Early detection of anomalies and malfunctions in the network and 20 different scenarios Performing assumption analyses (1004),  Network capabilities while agent matching continues based on agent discovery request a mismatch originating from the network via the digital twin prediction module (504) If detected by the interaction module (503), the agent is sent to the discovery module (502) The agent discovery is updated by providing information, and the newly discovered agent is given a cooperation mission of 25. or the transfer of the service acquisition along with the relevant context and the new match, load and Recording mission information agent matching payload mission database (505) (1005). The invention's working principle generally involves transferring basic data belonging to artificial intelligence agents to agent 30. The log file is kept on the server (501) and an AI agent cooperates or provides services The process begins with sending an agent discovery request to the agent discovery module (502) for recruitment. Agent discovery module (502) agent information in agent registry server (501) agent matching payload task It uses the current load and duty data in the database (505). Network interaction module (503), key performance indicators and service level 35 obtained from mobile network components. the status of communication, perception and computation network capabilities based on values It identifies and transmits the said data to the agent discovery module (502). The agent discovery module (502) transmits the said data to the agent. 12 along with capability, protocol, identification and network competency requirements in the discovery request. It creates the appropriate agent match by evaluating it. Digital twin prediction module (504), Network capabilities are determined by using historical and real-time network data and simulation methods. It predicts future changes and provides forecast data if the duration of the collaboration or service procurement is specified. The information is transmitted from the network interaction module (503) to the agent discovery module (502). (Continuing 5) Agent Discovery Module (502) when a network capability-related incompatibility is detected during matching It updates the discovery, transfers the mission or service context to the new agent, and provides the updated match, payload, and Mission information agent matches are stored in the payload mission database (505).

Claims

13 REQUESTS 1. Accessed via or within components of next-generation mobile networks AI agents positioned by the mobile network, with their agent capabilities Network capabilities offered to agents are evaluated together to enable them to discover each other. 5 It is a system that enables matching for the purpose of cooperation or service procurement, and its feature is;  Artificial intelligence through software running on at least one electronic device Inter-agent reconnaissance unit (500) which operates the reconnaissance mechanism among its agents,  Artificial intelligence that will interact through the inter-agent discovery unit (500) agents' capabilities, protocols they support, identification 10 records the methods and endpoints to which task requests will be sent, and is authorized. Agent log server sharing in queries (501),  Information about matched AI agents for establishing collaboration or receiving services record agent payload and mission information and inter-agent reconnaissance unit (500) Agent matching payload task database (505) sharing with its components, 15  By interacting with relevant mobile network components, key performance indicators and assessing the appropriateness of service level values ​​and providing future information to agents. Communication, sensing and computing network offered by next-generation mobile network. Network interaction module (503) which monitors the status of competencies,  Agent capabilities to be discovered, as specified in the incoming collaboration or service request, 20 Supported protocols and authentication methods, communication speed, and latency. network capability in terms of duration, sensing function and computational capacity requirements, agent information found in the agent registry server (501), Agent matching data in the payload task database (505) and network Access to network capabilities obtained via interaction module (503) 25 Agents that match agents by evaluating their situations together. Discovery Module (502) It includes.

2. The system is compliant with claim 1 and its feature is; the digital twin prediction module (504), network 30 with historical and instantaneous network data received from the network by the interaction module (503) and various By using simulation methods, traffic load or user demands, coverage quality and resource utilization according to different short-term and long-term time intervals predicts, detects anomalies and malfunctions in the network in advance, and prepares for different scenarios. It is a module that performs hypothesis analyses. 35 14 3. The system compliant with Request 2, its feature being; network interaction module (503), agent discovery Digital twin when the duration of collaboration or service procurement between agents is specified in the request. Network capability forecast data received from the forecast module (504) Agent Discovery It is a module that passes to module (502).

4. The system compliant with Request 2, its feature being; agent discovery module (502), ongoing agent Digital twin prediction of a mismatch stemming from network capabilities in the matching process. Detection by network interaction module (503) via module (504) repeating the agent discovery process, assigning a cooperation task or service to the newly discovered agent. The acquisition is reported along with the relevant context, and new match, payload, and mission information is provided to Agent 10. a unit that ensures the matches are kept up-to-date in the load-duty database (505) It is the fact that.

5. The system complies with Claim 1 and its characteristics are:  Terminal agent (101) located on user terminal (100), 15  access network remote end area (200) first base station (210) and second base First end-field agent (211) and second end-field accessible via station (220) agent (221),  First core network unit located in the core network area (300) (310), second core network unit (320), third core network unit (330) and 20 fourth core network unit (340) and respectively located within these units first core network agent (311), second core network agent (321), third core network agent (331) and fourth core network agent (341),  The first cloud agent (401), located in the cloud environment (400), the second cloud agent (402), third cloud agent (403) and fourth cloud agent (404) 25 It includes.

6. The system complies with Claim 3, and its feature is; agent discovery module (502), network interaction Network capability estimation data received from module (503) of agent discovery result It is the existence of a unit that evaluates in determining this. 30 7. The system complies with Claim 1, and its feature is that the agent discovery module (502) detects agents on the network. a unit that considers their location as an additional criterion in agent matching It is the fact that. 35 8. The system is compliant with claim 5, and its feature is; the access network remote end area (200), end area, access network remote endpoint, which can be separated into remote endpoint and near endpoint. Additional regions with near end areas between (200) and the core network area (300) It is a structure that can accommodate [others].

9. Accessed via or within components of next-generation mobile networks AI agents positioned by the mobile network, with their agent capabilities, 5 Network capabilities offered to agents are evaluated and explored together, fostering collaboration. or at least one electronic device for matching purposes to receive services It is an agent-to-agent reconnaissance method carried out through software being run, and its characteristic feature is;  Accessible via mobile communication systems or in mobile access system components terminal agent (101), first end field agent (211), second end field agent 10 (221), first core network agent (311), second core network agent (321), third core network agent (331), fourth core network agent (341), first cloud agent (401), second cloud agent (402), third cloud agent (403) and Capability by fourth cloud agent (404), supported protocol, Identification method and endpoint information to which task requests will be sent 15 Transfer to the agent log server (501) and recorded (1001),  By an artificial intelligence agent that wants to establish a collaboration or obtain services, the capabilities that the agents to be discovered will possess, the protocols to be supported, and for tasks or services to be carried out jointly using identification methods communication speed, latency, sensing function and computing capacity 20 Agent reconnaissance, which includes network capabilities that must be provided within this scope. Transfer of the request to the agent discovery module (502) (1002),  Agent discovery module (502) on agent registry server (501) Agent information found, agent matches found in the payload task database (505) 25 agents obtained via data and network interaction module (503) Access to communication, sensing and computing network capabilities together agent matching is determined based on the agent scouting request. implementation (1003), It includes the steps of the process.

10. The method is compliant with claim 9 and its feature is; by the digital twin estimation module (504). with the historical and instantaneous network data received from the network by the network interaction module (503) Traffic load or user demands, coverage can be determined using various simulation methods. quality and resource utilization according to different short-term and long-term time intervals Predicting, detecting anomalies and malfunctions in the network in advance, and 35 different Performing assumption analyses for scenarios (1004) process step It includes. 16 11. This method complies with Claim 10 and its characteristic is: agent matching created with an agent discovery request. while continuing, a nonconformity stemming from network capabilities is predicted by digital twin. Detection by network interaction module (503) via module (504) In this case, the agent discovery module (502) is informed and the agent discovery is updated, new 5 the discovered agent's cooperation task or service procurement along with the relevant context. Transfer and new matching, payload and mission information, agent matching, payload / mission data. It is recorded in the base (505) and includes the processing step (1005).

Images