Communication method, apparatus and system

Abstract

Provided in the present application are a communication method, apparatus and system. The method comprises: a first agent receiving first information used for describing an intent of a terminal device, and determining, on the basis of the first information, a first task used for implementing the intent of the terminal device; then, sending to a second agent a first request message for requesting the orchestration of the first task; and finally, receiving an orchestration result of the first task from the second agent, wherein the orchestration result is used for scheduling a network service to execute the first task. By means of the method, collaboration among a plurality of agents and collaborative operation between the agents and mobile network services can be realized, thereby ultimately implementing an intent of a terminal device.

Description

A communication method, apparatus and system Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a communication method, apparatus, and system. Background Technology

[0002] With the continuous development of artificial intelligence (AI) technology, AI models are being applied to mobile communication networks. For example, agents in mobile communication networks can use AI models to process tasks in order to realize the intentions of terminal devices.

[0003] When the intent of a terminal device is complex, multiple agents (e.g., agent 1 and agent 2) can collaborate on task processing. As an example, agent 1 sends a task request to agent 2 (the agent), which then forwards the request to agent 2 via a remote procedure call (RPC) channel. After processing the task, agent 2 sends the result back to agent 2 (the agent) via the RPC channel, which then forwards the result back to agent 1. As another example, agents 1 and 2 can register as clients on a server, with communication between them relayed through the server. However, the above multi-agent collaboration process is cumbersome and inefficient. Summary of the Invention

[0004] This application provides a communication method, apparatus, and system that enables efficient collaboration among multiple intelligent agents to achieve the intent of a terminal device.

[0005] Firstly, a communication method is provided. This method can be executed by a first intelligent agent. Unless otherwise specified, the "first intelligent agent" in this application can refer to the first intelligent agent itself, or a component of the first intelligent agent (e.g., a processor, chip, or chip system, such as a circuit or chip in the first intelligent agent responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core or a system-in-package (SIP) chip), or it can be a logic module or software that can implement all or part of the functions of the first intelligent agent.

[0006] In this method, first information is received, which describes the intent of a terminal device; a first task is determined based on the first information, which is used to implement the intent of the terminal device; a first request message is sent to a second intelligent agent, which requests the orchestration of the first task; and the orchestration result of the first task is received from the second intelligent agent, which is used to schedule network services to execute the first task.

[0007] For example, the first intelligent agent can be a system agent (sys-agent) that supports the application of a large model to understand the intent of the terminal device, decomposes the intent into tasks (e.g., a first task) in one or more service domains (e.g., a perception service domain, a computing power service domain, or a session service domain), and then calls the intelligent agent in each of the one or more service domains to orchestrate the tasks in that service domain. Optionally, the first intelligent agent can decompose the intent of the terminal device into network functional requirements of one or more service domains, and then determine the tasks in that service domain based on the network functional requirements of each service domain. Optionally, the first intelligent agent can also schedule network services based on the task orchestration results (e.g., the orchestration results of the first task) to complete the execution of the task. For example, the first intelligent agent can generate orchestration instructions that can be recognized by network elements (e.g., a first network element) in the network based on the task orchestration results, and send the orchestration instructions to the network element to trigger the execution of the task.

[0008] For example, the second intelligent agent can be a service agent from a different service domain, supporting the invocation of local or remote models for task orchestration and the output of task orchestration results. For instance, the second intelligent agent may include a computational intelligent agent, a connectivity intelligent agent, or a perception intelligent agent.

[0009] Using the above method, after obtaining the intent of the terminal device, the first intelligent agent processes the intent to obtain a first task for realizing the intent. Then, it calls a second intelligent agent to orchestrate the first task, obtaining an orchestration result for scheduling network services. This method not only enables efficient collaboration and communication among multiple intelligent agents, but also achieves collaboration between intelligent agents and mobile network services to realize the intent of the terminal device, meet the service needs of the terminal device, and improve the user experience.

[0010] For example, the first request message includes a description of the first task and / or the output format of the orchestration result of the first task. By using the above method, by carrying the description of the first task in the first request message, the second agent can clearly understand the first task and thus complete the orchestration of the first task; and / or, by carrying the output format of the orchestration result of the first task in the first request message, the second agent can clearly understand the output format of the orchestration result and thus output a task orchestration result that meets the format requirements. That is, the first agent can obtain an effective task orchestration result, improving task orchestration performance.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending second information to a first network element in the network to trigger / cause the first network element to schedule network services, wherein the second information is determined based on the orchestration result of the first task.

[0012] For example, after obtaining the orchestration result of the first task, the first intelligent agent can generate an orchestration instruction that can be recognized by the first network element, and then send the orchestration instruction to the first network element to drive the first network element to schedule network services to execute the first task and ultimately realize the intention of the terminal device.

[0013] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending the arrangement result of the first task to a second network element in the network, the second network element being used to execute the first task.

[0014] For example, after obtaining the orchestration result of the first task, the first intelligent agent can generate an orchestration instruction that the second network element can recognize, and then directly send the orchestration instruction to the second network element to trigger the second network element to execute the first task, ultimately realizing the intention of the terminal device.

[0015] That is, after obtaining the orchestration result of the first task, the first intelligent agent can generate an orchestration instruction (e.g., second information) that can be recognized or parsed by the first network element or the second network element. Then, the first intelligent agent can directly or indirectly call the network service to execute the first task and realize the intention of the terminal device.

[0016] In conjunction with the first aspect, in some implementations of the first aspect, sending a first request message to the second intelligent agent includes: sending a first request message to the second intelligent agent via a message bus.

[0017] In conjunction with the first aspect, in certain implementations of the first aspect, sending a first request message to a second intelligent agent via a message bus includes: sending the first request message and information about the second intelligent agent to the message bus. This information includes the identifier and / or address of the second intelligent agent, and is used by the message bus to send the first request message to the second intelligent agent. For example, the message bus can send the first request message to the second intelligent agent based on the information of the second intelligent agent, requesting the second intelligent agent to orchestrate a first task.

[0018] Optionally, the information of the first request message and the second agent can be carried in a message or signaling. The message bus can process the received first request message, such as format processing, and then forward it to the second agent. No specific details are provided.

[0019] Using the above method, the first and second agents can forward messages via a message bus. The message bus can be decoupled from the agents (e.g., the first or second agent), or in other words, it can act as an independent routing node. In this implementation, after receiving a second request message from the first agent, the message bus can perform fast message routing and forwarding based on the second agent's identifier and / or address, improving the efficiency of collaboration and communication among multiple agents and ultimately realizing the terminal device's intent. In schemes that forward messages between multiple agents through a server, the server needs to manage the multiple agents, such as managing their information (e.g., agent identifiers), which is complex. In contrast, the message bus does not require management of multiple agents and can perform fast message routing and forwarding based on the second agent's information, thus improving the efficiency of collaboration among multiple agents.

[0020] Understandably, the first and second agents can also directly interact via messages. That is, the first agent can directly send a first request message to the second agent, thereby enabling rapid point-to-point communication among multiple agents and ensuring efficient collaboration. This implementation can be understood as the message bus being coupled to the agents (e.g., the first or second agent), or in other words, the message bus can be embedded within the agents as an internal functional module, without limitation.

[0021] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending a second request message to a third network element, the second request message being used to request the discovery of an agent that supports orchestration of the first task; receiving information from a second agent of the third network element, the second agent supporting orchestration of the first task; and sending a first request message to the second agent, including: sending the first request message to the second agent based on the information of the second agent.

[0022] For example, the third network element can be an agent repository or an agent storage node. In this application, the third network element supports functions such as agent registration, agent discovery, and agent information storage.

[0023] Using the above method, the first intelligent agent can discover information about the second intelligent agent that supports the orchestration of the first task from the third network element, such as the identifier and / or address information of the second intelligent agent. Then, based on the information of the second intelligent agent, the first intelligent agent requests the second intelligent agent to perform the orchestration of the first task. This not only realizes the mutual cooperation between multiple intelligent agents, but also realizes the intent of the terminal device through the orchestration of the first task, thus meeting the business needs of the terminal device.

[0024] In conjunction with the first aspect, in some implementations of the first aspect, receiving information from the second intelligent agent from the third network element includes: receiving a third response message from the third network element, the third response message including a first identifier and information of the second intelligent agent, the first identifier being used to identify the third response message as a response to the second request message.

[0025] In conjunction with the first aspect, in some implementations of the first aspect, receiving information from the second intelligent agent of the third network element includes: receiving information from multiple intelligent agents of the third network element, wherein the multiple intelligent agents include the second intelligent agent; the method further includes: determining the second intelligent agent from the multiple intelligent agents based on the information from the multiple intelligent agents.

[0026] In conjunction with the first aspect, in some implementations of the first aspect, the information of the multiple intelligent agents includes at least one of the following: the deployment location of the multiple intelligent agents, the protocols supported by the multiple intelligent agents, the open interfaces supported by the multiple intelligent agents, the models supported by the multiple intelligent agents, or the capabilities of the multiple intelligent agents.

[0027] Optionally, the information of the multiple agents may also include at least one of the following: the service domain of the multiple agents, or the roles of the multiple agents.

[0028] Using the above method, the first agent can obtain information from one or more agents from the third network element, and then select the second agent from among the multiple agents based on the information of the multiple agents, which is beneficial to improving the accuracy of subsequent task orchestration.

[0029] In conjunction with the first aspect, in some implementations of the first aspect, the second request message includes information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, role, or capability.

[0030] Optionally, the second request message may also include information describing at least one of the following characteristics of the requesting agent: supported open interfaces or supported models.

[0031] By using the above method, by carrying information describing the characteristics of the requesting agent in the second request message, the third network element can discover and provide information about the second agent that meets the request. In other words, the second agent meets the characteristics described in the second request message, which facilitates the first agent to request the second agent to orchestrate the first task, thereby improving the accuracy and effectiveness of task orchestration.

[0032] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: sending a first registration request message to a third network element, the first registration request message being used to request registration with the third network element. In other words, the first registration request message is used to request the third network element to store the information of the first intelligent agent.

[0033] Optionally, the third network element sends a first registration response message to the first intelligent agent, which indicates that the first intelligent agent has successfully registered. In other words, the third network element stores the information of the first intelligent agent. For example, the intelligent agent registration management module of the third network element stores the information of the first intelligent agent in the intelligent agent information database of the third network element for subsequent intelligent agent discovery.

[0034] For example, the first registration request message includes at least one of the following: the identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent.

[0035] Using the above method, the first intelligent agent registers its own information with the third network element, so that other intelligent agents can obtain the information of the first intelligent agent through the third network element, which helps to realize the discovery between intelligent agents.

[0036] Secondly, a communication method is provided. This method can be executed by a second intelligent agent. Unless otherwise specified, the "second intelligent agent" in this application can refer to the second intelligent agent itself, or a component of the second intelligent agent (e.g., a processor, chip, or chip system, such as a circuit or chip in the second intelligent agent responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core), or a logic module or software that can implement all or part of the functions of the second intelligent agent).

[0037] In this method, a first request message is received from a first intelligent agent, the first request message being used to request the orchestration of a first task, the first task being used to implement the intent of a terminal device; the orchestration result of the first task is sent to the first intelligent agent, the orchestration result being used to schedule network services to execute the first task.

[0038] Using the above method, after the second agent receives the request message for orchestrating the first task from the first agent, it feeds back the orchestration result of the first task to the first agent. The efficient collaboration between the first agent and the second agent completes the orchestration of the first task, which helps to schedule network services to execute the first task, thereby realizing the intent of the terminal device.

[0039] In conjunction with the second aspect, in some implementations of the second aspect, the method further includes: determining the orchestration result of the first task based on the first request message.

[0040] In conjunction with the second aspect, in some implementations of the second aspect, the first request message includes a description of the first task and / or the output format of the orchestration result of the first task.

[0041] In conjunction with the second aspect, in some implementations of the second aspect, receiving a first request message from the first intelligent agent includes: receiving a first request message from the first intelligent agent via a message bus.

[0042] In conjunction with the second aspect, in some implementations of the second aspect, the method further includes: sending a second registration request message to a third network element, the second registration request message being used to request registration with the third network element. In other words, the second registration request message is used to request the third network element to store the information of the second intelligent agent.

[0043] Optionally, the third network element sends a second registration response message to the second agent, which indicates that the second agent has successfully registered. In other words, the third network element stores the information of the second agent. For example, the agent registration management module of the third network element stores the information of the second agent in the agent information database of the third network element for subsequent agent discovery.

[0044] In conjunction with the second aspect, in some implementations of the second aspect, the second registration request message includes at least one of the following: the identification information of the second intelligent agent, the service domain information of the second intelligent agent, the deployment location information of the second intelligent agent, the role information of the second intelligent agent, the capability information of the second intelligent agent, the protocol information supported by the second intelligent agent, the open interface information supported by the second intelligent agent, or the model information supported by the second intelligent agent.

[0045] The beneficial effects of the second aspect and some implementations thereof can be referred to the relevant descriptions in the first aspect, and will not be repeated here.

[0046] Thirdly, a communication method is provided. This method can be executed by a first network element in the network (e.g., a task-centric function (TCF)). Unless otherwise specified, the "first network element" in this application can refer to the first network element itself, or a component in the first network element (e.g., a processor, chip, or chip system, such as a circuit or chip in the first network element responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core), or a logic module or software that can implement all or part of the functions of the first network element.

[0047] In this method, second information is received from a first intelligent agent. The second information is determined based on the orchestration result of a first task, which is used to realize the intent of the terminal device. Based on the second information, a second network element in the network is scheduled to execute the first task.

[0048] Using the above method, the first network element receives the second information and schedules the second network element in the network to execute the first task in order to realize the intention of the terminal device. This enables the intelligent agent and mobile network services to work together, supports the construction of the mobile network AI native architecture, and provides network services for the terminal device.

[0049] In conjunction with the third aspect, in some implementations of the third aspect, before receiving the second information from the first intelligent agent, the method further includes: receiving the intention of the terminal device; and sending first information to the first intelligent agent, the first information being used to describe the intention of the terminal device.

[0050] In conjunction with the third aspect, in some implementations of the third aspect, scheduling a second network element in the network to perform the first task includes: sending a third request message to the second network element, the third request message being used to request the execution of the first task.

[0051] The beneficial effects of the third aspect and some implementations thereof can be referred to the relevant descriptions in the first aspect, and will not be repeated here.

[0052] Fourthly, a communication method is provided. This method can be executed by a third network element. Unless otherwise specified, the term "third network element" in this application can refer to the third network element itself, or a component within the third network element (e.g., a processor, chip, or chip system, such as a circuit or chip in the third network element responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core), or a logic module or software capable of implementing all or part of the functions of the third network element).

[0053] In this method, a second request message is received from a first intelligent agent, the second request message being used to request the discovery of an intelligent agent that supports orchestration of a first task, the first task being used to implement the intent of a terminal device; information about a second intelligent agent is sent to the first intelligent agent, the second intelligent agent supporting orchestration of the first task.

[0054] Using the above method, the third network element, based on the request of the first intelligent agent, feeds back information about the second intelligent agent that supports the orchestration of the first task, thereby realizing the discovery of the second intelligent agent. This allows the first intelligent agent to call the second intelligent agent to complete the orchestration of the first task, which helps to realize the intent of the terminal device.

[0055] In conjunction with the fourth aspect, in some implementations of the fourth aspect, sending information about the second intelligent agent to the first intelligent agent includes: sending information about multiple intelligent agents to the first intelligent agent, wherein the multiple intelligent agents include the second intelligent agent.

[0056] In conjunction with the fourth aspect, in some implementations of the fourth aspect, the information of the multiple agents includes at least one of the following: the deployment location of the multiple agents, the protocols supported by the multiple agents, the open interfaces supported by the multiple agents, the models supported by the multiple agents, or the capabilities of the multiple agents.

[0057] In conjunction with the fourth aspect, in some implementations of the fourth aspect, the second request message includes information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, role, or capability.

[0058] In conjunction with the fourth aspect, in some implementations of the fourth aspect, the method further includes: receiving a first registration request message from a first intelligent agent, the first registration request message being used to request registration with a third network element; and / or receiving a second registration request message from a second intelligent agent, the second registration request message being used to request registration with a third network element.

[0059] For example, the first registration request message includes at least one of the following: the identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent.

[0060] For example, the second registration request message includes at least one of the following: the identification information of the second intelligent agent, the service domain information of the second intelligent agent, the deployment location information of the second intelligent agent, the role information of the second intelligent agent, the capability information of the second intelligent agent, the protocol information supported by the second intelligent agent, the open interface information supported by the second intelligent agent, or the model information supported by the second intelligent agent.

[0061] In conjunction with the fourth aspect, in some implementations of the fourth aspect, the method further includes: receiving a first deregistration request message from a first intelligent agent, the first deregistration request message being used to request the deletion of information of the first intelligent agent; and / or receiving a second deregistration request message from a second intelligent agent, the second deregistration request message being used to request the deletion of information of the second intelligent agent.

[0062] In conjunction with the fourth aspect, in some implementations of the fourth aspect, the method further includes: receiving a first update request message from a first intelligent agent, the first update request message being used to request an update of the information of the first intelligent agent; and / or receiving a second update request message from a second intelligent agent, the second update request message being used to request an update of the information of the second intelligent agent.

[0063] The beneficial effects of the fourth aspect and some of its implementations can be referred to the relevant descriptions in the first aspect, and will not be repeated here.

[0064] Fifthly, a communication method is provided. In this method, a first intelligent agent receives first information, which describes the intent of a terminal device; the first intelligent agent determines a first task based on the first information, which is used to implement the intent of the terminal device; the first intelligent agent sends a first request message to a second intelligent agent, which requests the orchestration of the first task; the second intelligent agent sends the orchestration result of the first task to the first intelligent agent, which is used to schedule network services to execute the first task.

[0065] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the method further includes: the first intelligent agent sending second information to the first network element in the network, the second information being determined based on the orchestration result of the first task; the first network element scheduling the second network element in the network to execute the first task based on the second information.

[0066] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the first network element schedules the second network element in the network to perform the first task based on the second information, including: the first network element sends a third request message to the second network element, the third request message being used to request the execution of the first task.

[0067] In conjunction with the fifth aspect, in some implementations of the fifth aspect, before the first intelligent agent sends the second information to the first network element in the network, the method further includes: the first network element receiving the intention of the terminal device; the first intelligent agent receiving the first information, including: the first intelligent agent receiving the first information from the first network element.

[0068] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the method further includes: the first intelligent agent sending the arrangement result of the first task to the second network element in the network, the second network element executing the first task; and the second network element executing the first task according to the arrangement result of the first task.

[0069] In conjunction with the fifth aspect, in some implementations of the fifth aspect, before the first intelligent agent sends the first request message to the second intelligent agent, the method further includes: the first intelligent agent sending a second request message to the third network element, the second request message being used to request the discovery of an intelligent agent that supports orchestration of the first task; the third network element sending information about the second intelligent agent to the first intelligent agent, the second intelligent agent supporting orchestration of the first task.

[0070] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the third network element sends information about the second intelligent agent to the first intelligent agent, including: the third network element sends information about multiple intelligent agents to the first intelligent agent, the multiple intelligent agents including the second intelligent agent; the method also includes: the first intelligent agent determines the second intelligent agent from the multiple intelligent agents based on the information of the multiple intelligent agents.

[0071] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the method further includes: a first intelligent agent sending a first registration request message to a third network element, the first registration request message being used to request registration with the third network element; and / or, a second intelligent agent sending a second registration request message to the third network element, the second registration request message being used to request registration with the third network element.

[0072] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the method further includes: the first intelligent agent sending a first deregistration request message to the third network element, the first deregistration request message being used to request the deletion of the information of the first intelligent agent; and / or, the second intelligent agent sending a second deregistration request message to the third network element, the second deregistration request message being used to request the deletion of the information of the second intelligent agent.

[0073] In conjunction with the fifth aspect, in some implementations of the fifth aspect, the method further includes: a first intelligent agent sending a first update request message to a third network element, the first update request message being used to request an update of the information of the first intelligent agent; and / or, a second intelligent agent sending a second update request message to the third network element, the second update request message being used to request an update of the information of the second intelligent agent.

[0074] The beneficial effects of the fifth aspect and some implementations thereof can be referred to the descriptions of the first to fourth aspects, which will not be repeated here.

[0075] Sixthly, a communication method is provided. This method can be executed by a third intelligent agent. Unless otherwise specified, the "third intelligent agent" in this application can refer to the third intelligent agent itself, or a component of the third intelligent agent (e.g., a processor, chip, or chip system, such as a circuit or chip in the third intelligent agent responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core or a system-in-package (SIP) chip), or it can be a logic module or software that can implement all or part of the functions of the third intelligent agent.

[0076] In this method, a third intelligent agent receives first information, which describes the intent of the terminal device; based on the first information, a first task is orchestrated to obtain an orchestration result of the first task, which is used to schedule network services to execute the first task, and the first task is used to realize the intent of the terminal device.

[0077] In conjunction with the sixth aspect, in some implementations of the sixth aspect, the method also includes: sending the orchestration result of the first task to the first network element in the network.

[0078] In conjunction with the sixth aspect, in some implementations of the sixth aspect, the method further includes: sending a request message to a second network element in the network according to the arrangement result of the first task, the request message being used to request the second network element to execute the first task.

[0079] The beneficial effects of the sixth aspect and some implementations thereof can be referred to the relevant descriptions of the first or second aspect, and will not be repeated here.

[0080] Seventhly, a communication method is provided. This method can be executed by a first intelligent agent. Unless otherwise specified, the "first intelligent agent" in this application can refer to the first intelligent agent itself, or a component in the first intelligent agent (e.g., a processor, chip, or chip system, such as a circuit or chip in the first intelligent agent responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core or a system-in-package (SIP) chip), or it can be a logic module or software that can implement all or part of the functions of the first intelligent agent.

[0081] In this method, a first intelligent agent receives first information, which describes the intent of a terminal device; determines a first task based on the first information, which is used to implement the intent of the terminal device; and sends a task request message to a fourth intelligent agent, which is used to request the execution of the first task.

[0082] In conjunction with the seventh aspect, in some implementations of the seventh aspect, the method further includes: sending a discovery request message to a third network element, the discovery request message being used to request the discovery of an agent that supports orchestration of the first task; receiving information from a fourth agent of the third network element, the fourth agent supporting orchestration of the first task; and sending a task request message to the fourth agent, including: sending a task request message to the fourth agent based on the information of the fourth agent.

[0083] For a detailed interpretation of the discovery request message, please refer to the relevant description of the second request message in this application, which will not be repeated here.

[0084] In conjunction with the seventh aspect, in some implementations of the seventh aspect, receiving information from the fourth intelligent agent from the third network element includes: receiving information from multiple intelligent agents of the third network element, wherein the multiple intelligent agents include the fourth intelligent agent; the method further includes: determining the fourth intelligent agent from the multiple intelligent agents based on the information from the multiple intelligent agents.

[0085] In conjunction with the seventh aspect, in some implementations of the seventh aspect, the information of the multiple agents includes at least one of the following: the deployment location of the multiple agents, the protocols supported by the multiple agents, the open interfaces supported by the multiple agents, the models supported by the multiple agents, or the capabilities of the multiple agents.

[0086] In conjunction with aspect seven, in some implementations of aspect seven, the method further includes: sending a first registration request message to the third network element, the first registration request message being used to request registration with the third network element.

[0087] In conjunction with the seventh aspect, in some implementations of the seventh aspect, the first registration request message includes at least one of the following: the identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent.

[0088] The beneficial effects of the seventh aspect and some implementations thereof can be referred to the relevant descriptions in the first aspect, and will not be repeated here.

[0089] Eighthly, a communication method is provided. This method can be executed by a fourth intelligent agent. Unless otherwise specified, the "fourth intelligent agent" in this application can refer to the fourth intelligent agent itself, or a component of the fourth intelligent agent (e.g., a processor, chip, or chip system, such as a circuit or chip in the fourth intelligent agent responsible for communication functions (e.g., a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core), or a logic module or software that can implement all or part of the functions of the fourth intelligent agent.

[0090] In this method, the fourth intelligent agent receives a task request message from the first intelligent agent, the task request message being used to request the execution of a first task, the first task being used to realize the intent of the terminal device; according to the task request message, the fourth intelligent agent sends a request message to the second network element in the network, the request message being used to request the execution of the first task, the second network element being used to execute the first task.

[0091] In conjunction with the eighth aspect, in some implementations of the eighth aspect, the method also includes: performing task orchestration on the first task according to the task request message, and obtaining the orchestration result of the first task.

[0092] In conjunction with the eighth aspect, in some implementations of the eighth aspect, receiving a task request message from the first intelligent agent includes: receiving a task request message from the first intelligent agent via a message bus.

[0093] In conjunction with aspect eight, in some implementations of aspect eight, the method further includes: sending a second registration request message to the third network element, the second registration request message being used to request registration with the third network element.

[0094] In conjunction with aspect eight, in some implementations of aspect eight, the second registration request message includes at least one of the following: the identification information of the fourth intelligent agent, the service domain information of the fourth intelligent agent, the deployment location information of the fourth intelligent agent, the role information of the fourth intelligent agent, the capability information of the fourth intelligent agent, the protocol information supported by the fourth intelligent agent, the open interface information supported by the fourth intelligent agent, or the model information supported by the fourth intelligent agent.

[0095] The beneficial effects of the eighth aspect and some implementations of the eighth aspect can be referred to the relevant description in the second aspect, and will not be repeated here.

[0096] Ninthly, a communication device is provided. This communication device is used to execute the methods described in the first or seventh aspect and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations involved in the first or seventh aspect. These modules, units, or means can be implemented in software, hardware, or a combination of software and hardware.

[0097] In one possible design, the communication device includes: a communication unit for receiving first information, the first information describing the intent of a terminal device; a processing unit for determining a first task based on the first information, the first task being used to implement the intent of the terminal device; the communication unit is further configured to send a first request message to a second intelligent agent, the first request message being used to request the orchestration of the first task; and the communication unit is further configured to receive an orchestration result of the first task from the second intelligent agent, the orchestration result being used to schedule network services to execute the first task.

[0098] In one possible design, the communication device includes: a communication unit for receiving first information, the first information describing the intent of a terminal device; a processing unit for determining a first task based on the first information, the first task being used to implement the intent of the terminal device; and the communication unit is further configured to send a task request message to a fourth intelligent agent, the task request message being used to request the execution of the first task.

[0099] The communication unit can perform the receiving and transmitting processes in the first or seventh aspect described above, and the processing unit can perform other processes in the first or seventh aspect described above besides receiving and transmitting.

[0100] The aforementioned communication device may be a first intelligent agent, or a communication module in the first intelligent agent, or a chip in the first intelligent agent responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module or software that can realize all or part of the functions of the first intelligent agent.

[0101] In a tenth aspect, a communication device is provided. This communication device is used to execute the methods of the second aspect described above and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations involved in the second aspect described above. These modules, units, or means can be implemented in software, hardware, or a combination of software and hardware.

[0102] In one possible design, the communication device includes: a communication unit for receiving a first request message from a first intelligent agent, the first request message being for requesting the orchestration of a first task, the first task being for fulfilling the intent of a terminal device; the communication unit is also configured to send the orchestration result of the first task to the first intelligent agent, the orchestration result being for scheduling network services to execute the first task.

[0103] The communication unit can perform the receiving and transmitting processes described in the second aspect above, and the processing unit can perform other processes described in the second aspect above besides receiving and transmitting.

[0104] The aforementioned communication device may be a second intelligent agent, or a communication module in the second intelligent agent, or a chip in the second intelligent agent responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module or software that can realize all or part of the functions of the second intelligent agent.

[0105] Eleventhly, a communication device is provided. This communication device is used to execute the methods of the third aspect and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations involved in the third aspect. These modules, units, or means can be implemented in software, hardware, or a combination of software and hardware.

[0106] In one possible design, the communication device includes: a communication unit for receiving second information from a first intelligent agent, the second information being determined based on the orchestration result of a first task, the first task being used to realize the intent of a terminal device; and a processing unit for scheduling a second network element in the network to execute the first task based on the second information.

[0107] The communication unit can perform the receiving and transmitting processes described in the third aspect above, and the processing unit can perform other processes described in the third aspect above besides receiving and transmitting.

[0108] The aforementioned communication device may be a first network element, or a communication module in the first network element, or a chip in the first network element responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module, or software that can realize all or part of the functions of the first network element.

[0109] In a twelfth aspect, a communication device is provided. This communication device is used to execute the methods of the fourth aspect and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations involved in the fourth aspect. These modules, units, or means can be implemented in software, hardware, or a combination of software and hardware.

[0110] In one possible design, the communication device includes: a communication unit for receiving a second request message from a first agent, the second request message being for requesting the discovery of an agent that supports orchestration of a first task, the first task being for fulfilling the intent of a terminal device; and the communication unit further for sending information about a second agent to the first agent, the second agent supporting orchestration of the first task.

[0111] The communication unit can perform the receiving and transmitting processes in the fourth aspect mentioned above, and the processing unit can perform other processes in the fourth aspect mentioned above besides receiving and transmitting.

[0112] The aforementioned communication device may be a third network element, or a communication module in a third network element, or a chip in a third network element that is responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module or software that can realize all or part of the functions of a third network element.

[0113] In a thirteenth aspect, a communication device is provided. This communication device is used to execute the methods described in the sixth aspect and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations involved in the fourth aspect, which can be implemented in software, hardware, or a combination of software and hardware.

[0114] In one possible design, the communication device includes: a communication unit for receiving first information, the first information describing the intent of a terminal device; a processing unit for determining a first task based on the first information, the first task being used to realize the intent of the terminal device; and the processing unit is further configured to perform task orchestration on the first task to obtain an orchestration result for the first task, the orchestration result being used to schedule network services to execute the first task.

[0115] The communication unit can perform the receiving and transmitting processes described in the sixth aspect above, and the processing unit can perform other processes described in the sixth aspect above besides receiving and transmitting.

[0116] The aforementioned communication device may be a third intelligent agent, or a communication module in a third intelligent agent, or a chip in a third intelligent agent that is responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module or software that can realize all or part of the functions of a third intelligent agent.

[0117] In a fourteenth aspect, a communication device is provided. This communication device is used to execute the methods described in the eighth aspect and any implementation thereof. For example, the communication device includes modules, units, or means corresponding to the operations described in the ninth aspect. These modules, units, or means can be implemented in software, hardware, or a combination of software and hardware.

[0118] In one possible design, the communication device includes: a communication unit for receiving a task request message from a first intelligent agent, the task request message being used to request the execution of a first task, the first task being used to realize the intent of a terminal device; the communication unit is also used to send a request message to a second network element in the network according to the task request message, the request message being used to request the execution of the first task, the second network element being used to execute the first task.

[0119] The communication unit can perform the receiving and transmitting processes described in the eighth aspect above, and the processing unit can perform other processes described in the eighth aspect above besides receiving and transmitting.

[0120] The aforementioned communication device may be a fourth intelligent agent, or a communication module in a fourth intelligent agent, or a chip in a fourth intelligent agent that is responsible for communication functions, such as a modem chip (also known as a baseband chip) or a SoC or SIP chip containing a modem module, or a logical node, logical module or software that can realize all or part of the functions of a fourth intelligent agent.

[0121] In a fifteenth aspect, a communication device is provided. The communication device includes at least one processor. The at least one processor is capable of executing a computer program or instructions, which, when executed, cause the communication device to implement the methods of any one of the first to ninth aspects and any implementation thereof.

[0122] In one possible design, the communication device may further include at least one interface circuit. This interface circuit is used to implement communication functions within the communication device and / or communication functions between the communication device and other devices or components.

[0123] In one possible design, the communication device may further include at least one interface circuit and / or at least one memory. The at least one processor is coupled to the at least one memory. The at least one memory is used to store part or all of the necessary computer programs or instructions for implementing the functions involved in any of the first to ninth aspects and any implementation thereof. The interface circuit is used to implement the communication functions within the communication device and / or the communication functions between the communication device and other devices or components.

[0124] In one possible design, the at least one processor is used to communicate with other devices or components via at least one interface circuit.

[0125] The aforementioned communication device may be a first intelligent agent, or a communication module in the first intelligent agent, or a chip in the first intelligent agent responsible for communication functions, or a logic node, logic module, or software that can realize all or part of the functions of the first intelligent agent.

[0126] The aforementioned communication device may be a second intelligent agent, or a communication module in the second intelligent agent, or a chip in the second intelligent agent responsible for communication functions, or a logic node, logic module, or software that can realize all or part of the functions of the second intelligent agent.

[0127] The aforementioned communication device may be a third intelligent agent, or a communication module in a third intelligent agent, or a chip in a third intelligent agent responsible for communication functions, or a logic node, logic module, or software that can realize all or part of the functions of a third intelligent agent.

[0128] The aforementioned communication device may be a fourth intelligent agent, or a communication module in a fourth intelligent agent, or a chip in a fourth intelligent agent responsible for communication functions, or a logic node, logic module, or software that can realize all or part of the functions of a fourth intelligent agent.

[0129] The aforementioned communication device may be a first network element, or a communication module in the first network element, or a chip in the first network element responsible for communication functions, or a logic node, logic module, or software that can realize all or part of the functions of the first network element.

[0130] The aforementioned communication device may be a third network element, or a communication module within a third network element, or a chip within a third network element responsible for communication functions, or a logic node, logic module, or software capable of realizing all or part of the functions of a third network element.

[0131] In a sixteenth aspect, a communication system is provided. This communication system includes at least one of the communication devices described in aspects nine through fourteen.

[0132] In a seventeenth aspect, a chip or chip system is provided. The chip or chip system includes at least one processing circuitry for executing a computer program or instructions, causing the chip or chip system to perform the methods described in the first to eighth aspects and any possible implementation thereof.

[0133] The chip or chip system may include output circuits or interfaces for transmitting information or data, and input circuits or interfaces for receiving information or data.

[0134] Eighteenthly, a computer-readable storage medium is provided. This computer-readable storage medium stores computer program code or instructions, which, when executed by a processor, implement the methods described in the first to eighth aspects and any possible implementation thereof.

[0135] In a nineteenth aspect, a computer program product is provided. The computer program product includes computer program code or instructions, wherein when a processor executes the computer program code or instructions, the method in any of the possible implementations of the first to eighth aspects is implemented.

[0136] In a twentieth aspect, a computer program is provided. When the computer program is run, it causes the methods in the first to eighth aspects and any possible implementation thereof to be implemented.

[0137] It should be understood that the beneficial effects of aspects nine through twentieth above can be referenced from aspects one through eight above and any possible implementation thereof, and will not be elaborated here. Attached Figure Description

[0138] Figures 1 and 2 are schematic diagrams of a communication system applicable to embodiments of this application;

[0139] Figure 3 illustrates a scenario of mobile networks evolving into intelligent networks;

[0140] Figure 4 illustrates a schematic diagram of distributed deployment and communication of multiple agents;

[0141] Figure 5 is a schematic diagram of a communication system applicable to an embodiment of this application;

[0142] Figures 6 to 9 are schematic flowcharts of the communication method provided in the embodiments of this application;

[0143] Figure 10 is a schematic diagram of a communication system applicable to an embodiment of this application;

[0144] Figures 11 to 13 are schematic flowcharts of the communication method provided in the embodiments of this application;

[0145] Figure 14 is a schematic diagram of a communication system applicable to an embodiment of this application;

[0146] Figures 15 to 17 are schematic flowcharts of the communication method provided in the embodiments of this application;

[0147] Figure 18 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0148] Figure 19 is a schematic diagram of another communication device provided in an embodiment of this application;

[0149] Figure 20 is a schematic diagram of the chip system provided in an embodiment of this application. Detailed Implementation

[0150] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0151] Before introducing the scheme of this application, the following points should be noted.

[0152] First, in this application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0153] Second, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can mean: a, or, b, or, c, or, a and b, or, a and c, or, b and c, or, a, b, and c. Here, a, b, and c can each be single or multiple.

[0154] Third, in this application, the terms "first," "second," "#1," and "#2," as well as various numerical designations, are used for ease of description and are not intended to limit the scope of the embodiments of this application. For example, they are used to distinguish different messages, rather than to describe a specific order or sequence. It should be understood that such descriptions can be interchanged where appropriate to describe solutions other than those in the embodiments of this application.

[0155] Fourth, in this application, "instruction" or "for instruction" can include both direct and indirect instruction. When describing instruction information as being used to instruct A, it can include whether the instruction information directly or indirectly instructs A, but does not necessarily mean that the instruction information carries A.

[0156] The indication methods involved in the embodiments of this application should be understood to cover various methods that enable the party to be indicated to obtain the information to be indicated. The information to be indicated can be sent as a whole or divided into multiple sub-information and sent separately. Moreover, the sending period and / or sending time of these sub-information can be the same or different. This application does not limit the sending method, for example.

[0157] The "instruction information" in the embodiments of this application can be an explicit instruction, that is, a direct instruction through signaling, or an instruction obtained by combining other rules or parameters with the parameters indicated by the signaling, or by deduction. It can also be an implicit instruction, that is, an instruction obtained based on rules or relationships, or based on other parameters, or by deduction. This application does not specifically limit it in this regard.

[0158] Fifth, in this application, "protocol" can refer to a standard protocol in the field of communications, such as fifth-generation (5G) protocols. th This application does not limit the scope of network protocols such as generation (5G), New Radio (NR) protocols, and related protocols applied in future communication systems. "Predefined" may include predefined terms, such as protocol definitions. "Preconfiguration" can be achieved by pre-storing corresponding codes, tables, or other means of indicating relevant information in the device; this application does not limit the implementation method.

[0159] Sixth, in this application, terms such as "message," "information," "signal," or "information element (IE, which may be abbreviated as information element)" can be used interchangeably. There is no limitation on the name of the message or information, as long as it can achieve the corresponding function.

[0160] "Sending information to XX (device)" can be understood as the destination of the information being that device. This can include sending information to that device directly or indirectly. "Receiving information from XX (device), or receiving information from XX (device)" can be understood as the source of the information being that device. This can include receiving information from that device directly or indirectly. Information may undergo necessary processing between the source and destination, such as format changes, but the destination can understand the valid information from the source. Similar expressions in this application can be interpreted similarly, and will not be elaborated further here.

[0161] "Communication" can also be described as data transmission, information transmission, data processing, etc. "Transmission" includes sending and / or receiving. "Transmission" can be described as output. "Sending" can also be understood as the output of a chip interface, and "receiving" can be understood as the input of a chip interface. In other words, "sending" or "receiving" can occur between devices, for example, device 1 and device 2 sending or receiving data via an air interface. "Sending" or "receiving" can also occur within a device, for example, sending or receiving data between components, modules, chips, software modules, or hardware modules within the device via a bus, wiring, or interface.

[0162] For example, "sending information" can be understood as one device sending information to another device, or it can also be understood as one logic module within a device sending information to another logic module. For instance, "device 1 sending information" can be understood as device 1 sending information to device 2, or it can be understood as logic module 1 within device 1 sending information to logic module 2 within device 1. Similarly, "receiving information" can be understood as device 1 receiving information from device 2, or it can also be understood as logic module 1 within device 1 receiving information from logic module 2. For instance, "device 1 receiving information" can be understood as device 1 receiving information from device 2, or it can be understood as logic module 1 within device 1 receiving information from logic module 2 within device 1.

[0163] Seventh, in this application, the words "exemplarily," "for example," etc., are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as an "example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the word "example" is intended to present concepts in a concrete manner. In the embodiments of this application, "of," "corresponding, relevant," "corresponding," and "associate" may sometimes be used interchangeably, and it should be noted that their intended meanings are consistent unless their distinctions are emphasized.

[0164] The following describes the communication system to which this application applies.

[0165] The technical solutions of this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, 5G systems, New Radio (NR) systems, and future communication systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems.

[0166] In a communication system, a device can send signals to or receive signals from another device. These signals may include reference signals, information, signaling, or data. In this application, "device" can be replaced by an entity, network entity, communication equipment, communication module, node, or communication node.

[0167] Figure 1 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 1, the communication system 100 includes at least one terminal (120a-120j in Figure 1, collectively referred to as 120), a radio access network (RAN), and a core network (CN) 200. The RAN includes at least one RAN node (110a and 110b in Figure 1, collectively referred to as 110). The RAN may also include other RAN nodes, such as wireless relay devices and / or wireless backhaul devices (not shown in Figure 1). The terminal 120 is wirelessly connected to the RAN node 110. The RAN node 110 is wirelessly or wiredly connected to the core network 200. The core network devices in the core network 200 and the RAN node 110 in the RAN can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0168] RAN can be used for the third-generation partner program (3 rd Cellular systems related to the Generation Partnership Project (3GPP), such as fourth-generation (4G) cellular systems. thRAN can be a generation (4G) mobile communication system, a 5G mobile communication system, or a future-oriented evolution system. RAN can also be an open radio access network (O-RAN or ORAN), a cloud radio access network (CRAN), or a wireless fidelity (WiFi) system. RAN can also be a communication system that integrates two or more of the above systems.

[0169] RAN node 110, sometimes also referred to as network equipment, access network equipment, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in communication system 100 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative. For example, network element 120i in Figure 1 can be a helicopter or drone, which can be configured as a mobile base station. For terminals 120j accessing the RAN through network element 120i, network element 120i is a base station; but for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes both referred to as communication devices. For example, network elements 110a and 110b in Figure 1 can be understood as communication devices with base station functions, and network elements 120a-120j can be understood as communication devices with terminal functions.

[0170] In one possible scenario, a RAN node can be a base station (BS), an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a base station in a future mobile communication system, or an access node in a WiFi system. A RAN node can be a macro base station (as shown in Figure 1, 110a), a micro base station or indoor station (as shown in Figure 1, 110b), a relay node or donor node, or a radio controller in a CRAN scenario. Optionally, a RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU).

[0171] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with different RAN nodes each implementing a portion of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-control planes (CU-CPs), CU-user planes (CU-UPs), radio units (RUs), or CU-radio units (CU-RUs), etc. CUs and DUs can be configured separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio equipment or radio units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0172] In different systems, CU (including open CU-CP (O-CU-CP) and open CU-UP (O-CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called an open central unit (O-CU), DU can also be called an open distributed unit (O-DU), CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0173] Terminal 120 can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. A terminal can also be referred to as user equipment (UE), user device, access terminal, user unit, user station, mobile station, mobile station (MS), remote station, remote terminal, mobile device, user terminal, terminal unit, terminal station, terminal device, wireless communication equipment, user agent, or user equipment. A terminal typically contains a communication module, circuit, or chip that performs the corresponding communication functions. The terminal may also be configured with program instructions for performing these communication functions.

[0174] For example, the terminal in this application embodiment can be a mobile phone, a personal digital assistant (PDA) computer, a laptop computer, a tablet computer, a drone, a computer with wireless transceiver capabilities, a machine type communication (MTC) terminal, a virtual reality (VR) terminal, an augmented reality (AR) terminal, an Internet of Things (IoT) terminal, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical care, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home (e.g., game consoles, smart TVs, smart speakers, smart refrigerators, and fitness equipment), a transport vehicle with wireless communication capabilities, a communication module, or a roadside unit (RSU) with terminal capabilities.

[0175] The RAN and terminal 120 can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the scenario in which the RAN and terminal 120 are located.

[0176] Communication between access network devices and terminal devices follows a specific protocol layer structure. This protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: radio resource control (RRC) layer, packet data convergence protocol (PDCP) layer, radio link control (RLC) layer, medium access control (MAC) layer, or physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: service data adaptation protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or physical layer, etc.

[0177] The correspondence between network elements and their achievable protocol layer functions in the ORAN system can be found in Table 1 below.

[0178] Table 1

[0179] CN 200 can be a 5G core network or an evolved 5G core network. Taking a 5G core network as an example, CN 200 includes access and mobility management (AMF) network elements responsible for mobility management and access management services; session management (SMF) network elements responsible for session management; user plane (UPF) network elements responsible for user plane packet routing and forwarding and quality of service (QoS) control; and policy control (PCF) network elements. These core network elements can operate independently or be combined to implement certain control functions; for example, AMF, SMF, and PCF can be combined into a single core network device.

[0180] The communication system 100 provided in this application may further include AI network elements for implementing some or all AI-related operations. AI network elements may also be referred to as AI nodes, AI devices, AI entities, AI modules, AI models, or AI units, etc. The AI ​​network elements may be built into the network elements of the communication system. For example, an AI network element may be an AI module built into: access network equipment, core network equipment, cloud server, or operation, administration, and maintenance (OAM) to implement AI-related functions. The OAM may act as the network management system for the core network equipment and / or the access network equipment. Alternatively, the AI ​​network element may also be an independently configured network element in the communication system. Optionally, the terminal or its built-in chip may also include an AI entity for implementing AI-related functions.

[0181] Figure 2 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 2, network elements in the communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes (RAN nodes), terminals, or one or more devices in OAM, are equipped with one or more AI modules (only one is shown in Figure 2 for clarity). The access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. The CU and / or DU can also be equipped with one or more AI modules. Optionally, the CU can also be split into CU-CP and CU-UP. One or more AI models are provided in CU-CP and / or CU-UP.

[0182] The AI ​​module is used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can implement different functions. The AI ​​module model can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or bias in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The bias in the activation function can also be referred to as the neural network bias.

[0183] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0184] It should be understood that the above naming is defined solely for the purpose of distinguishing different functions and should not be construed as limiting this application. This application does not preclude the possibility of using other naming conventions in 5G networks and other future networks. For example, in future networks, some or all of the above-mentioned network elements may use the terminology from 5G, or they may use other names, etc.

[0185] It is understood that Figure 1 or Figure 2 is merely an example and does not constitute a limitation on the scope of protection of this application. The communication method provided in the embodiments of this application may also involve network elements not shown in Figure 1 or Figure 2, and of course, the communication method provided in the embodiments of this application may also include only some of the network elements shown in Figure 1 or Figure 2.

[0186] To facilitate understanding of the embodiments of this application, the basic concepts involved in this application will be explained first.

[0187] 1. Artificial Intelligence (AI): AI enables machines to possess human-like intelligence, such as allowing machines to use computer hardware and software to simulate certain intelligent human behaviors. To achieve AI, machine learning methods can be employed. In machine learning, machines learn (or train) models using training data. This model represents the mapping between inputs and outputs. The learned model can be used for reasoning (or prediction), that is, it can be used to predict the output corresponding to a given input. This output can also be called the reasoning result (or prediction result).

[0188] 2. Intelligent Agent: This is a concept in the field of artificial intelligence. Any entity capable of independent thought and interaction with its environment can be abstracted as an intelligent agent. In essence, an intelligent agent is a type of AI module. The basic characteristics of an intelligent agent are: it can react to changes in its environment and automatically adjust its behavior and state; different intelligent agents can also interact with other intelligent agents according to their own intentions.

[0189] For example, an intelligent agent can use a large language model (LLM) as its core, which includes a memory module, a tool module, a planning module, and an action module. The memory module implements long-term and / or short-term memory functions; the tool module contains multiple callable external tools; the planning module contains various planning algorithms; for example, the agent can plan external input tasks based on its memory; and the action module supports the agent in performing actions based on the planning results, such as calling up tools.

[0190] In this application, an intelligent agent can generally be considered as a proxy that can autonomously complete a set goal through action. "Intelligent agent" and "intelligence" are inseparable; it possesses some human-like intelligent abilities and behaviors, such as learning, reasoning, decision-making, and execution capabilities. Optionally, the intelligent agent can be replaced with other terms, such as artificial general intelligence (AGI), artificial intelligence, AI controller, intelligent unit, intelligent entity, or intelligent agent, etc. This application does not limit the name of the intelligent agent, as long as it can achieve the corresponding function.

[0191] In this application, the intelligent agent can be integrated into the existing hardware / software of the communication device, or the intelligent agent can be independent of the existing hardware / software of the communication device. The existing hardware / software may include at least one of the following: a chip, a baseband chip, a modem chip, a SoC chip containing a modem core, a SIP chip, a communication module, a chip system, a processor, a logic module, or software, etc.

[0192] The above description of the terminology is for ease of understanding only and does not limit the scope of protection of the embodiments of this application.

[0193] Currently, due to the continuous advancement of AI technology, the reasoning ability of AI models has been significantly improved, promoting the rapid development of intelligent agents based on large models. When the intent of a terminal device is complex, multiple intelligent agents can collaborate to process tasks.

[0194] Understandably, using multiple intelligent agents to solve the problem of complex intents in terminal devices offers several advantages, including but not limited to the following:

[0195] (1) Each intelligent agent can possess knowledge and skills in a certain service domain and can independently complete the business functions of the corresponding service domain, thereby improving the reasoning accuracy of the intelligent agent.

[0196] (2) When multiple intelligent agents cooperate with each other, complex tasks can be decomposed into multiple simple sub-tasks, reducing the processing complexity of a single sub-task.

[0197] (3) Each agent can use different AI models, and each AI model can handle its own strengths, thereby improving the accuracy and efficiency of the entire system's reasoning.

[0198] To facilitate understanding, the following examples, in conjunction with Figures 3 and 4, illustrate how multiple intelligent agents can be used to address the complex intent of terminal devices in the scenario of mobile networks evolving into intelligent networks.

[0199] Figure 3 illustrates a scenario of mobile networks evolving into intelligent networks. As shown in Figure 3, mobile networks can integrate one or more service areas from access, sensing, computing, immersive communication, digital twins, or distributed subnets with AI technology. Based on user intents sent by devices, people, or embodied robots, the AI ​​capabilities of different service areas can translate these intents into specific service flows, thereby invoking or indirectly driving the capabilities of various objects (e.g., third-party services, third-party content, or third-party AI-powered entities) to realize the user intent. This requires mobile networks to have flexible deployment and plug-and-play capabilities, with network deployment and service complexity and flexibility increasing exponentially compared to current mobile networks. Current deterministic task and process orchestration implementation methods may not be able to meet the complexity and flexibility requirements of future networks.

[0200] Figure 4 illustrates a schematic diagram of the distributed deployment and communication of multiple agents.

[0201] As shown in Figure 4(a), Remote Procedure Call (RPC) is used for distributed communication and networking of multiple agents. This architecture includes Node 1 and Node 2, which can be considered as two independent nodes. Node 1 acts as a client, and Node 2 acts as a server. An RPC channel is established between Node 1 and Node 2 to enable remote function or method calls. For example, Agent 1, Agent 2 (proxy), and PRC client are deployed on Node 1. Agent server, Agent 2, and PRC server are deployed on Node 2. Agent 1 can be a complete AI agent with independent AI inference capabilities. Agent 2 (proxy) acts as a local proxy for Agent 2 on Node 1, and all actual actions of Agent 2 (proxy) are performed by Agent 2 on Node 2. The agent server entity handles requests sent by Node 1, such as creating agents and calling agent interfaces / methods.

[0202] During inter-agent collaboration, communication and cooperation between Agent 1 and Agent 2 are achieved through Agent 2 (proxy). For example, Agent 2 (proxy) sends a request from Agent 1 on Node 1 to Agent 2 on Node 2 via an RPC channel. After Agent 2 processes the request, it returns the execution result to Agent 2 (proxy) via the RPC channel, and then Agent 2 (proxy) sends the execution result back to Agent 1. It should be noted that Agent 1 is unaware of Agent 2 on Node 2; it always considers Agent 2 (proxy) to be the agent that actually processes the request.

[0203] In this scheme, the deployment and discovery of multiple agents are implemented using a static deployment method, that is, a proxy agent (e.g., agent 2 (proxy)) is deployed on node 1 to enable collaboration and communication between agent 1 and agent 2, which is inefficient.

[0204] As shown in Figure 4(b), a client / server model is used to implement the distributed deployment of multiple agents. This architecture includes a client and a server. The client includes agent 1 and agent 2. The server is used for the management of agents (e.g., agent 1 and agent 2) and message distribution. Agents register with the server as clients. When information is exchanged between agent 1 and agent 2, messages are forwarded through the server.

[0205] In this scheme, the server acts as the central hub for information management and message routing of the agents. All communication between agents must be completed through the server, and it does not have point-to-point communication capabilities. In other words, this client / server communication mode may increase the communication pressure on the server, especially when facing large-scale communication and collaboration between agents, resulting in a performance bottleneck.

[0206] In summary, while introducing multiple agents to collaborate in mobile networks to address scenarios with complex intents from terminal devices is common, the collaboration and communication between these agents typically involve forwarding through agents (proxy) or servers, resulting in a cumbersome and inefficient process. Therefore, achieving efficient collaboration among multiple agents is a problem that needs to be solved.

[0207] In view of this, this application provides a communication method and apparatus that enables efficient collaboration among multiple intelligent agents to achieve the intent of a terminal device.

[0208] Below, we will first illustrate the network architecture provided in the embodiments of this application with reference to Figure 5.

[0209] Figure 5 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 5(a), an AI mobile network architecture for multi-agent collaboration is proposed. This architecture may include agents, third network elements (e.g., agent repository) and first network elements (e.g., task-centric function (TCF)). The agents in this application include first agents (e.g., system agent sys-agent) and second agents (e.g., service agent).

[0210] Optionally, the architecture may also include at least one of the following: a message bus (e.g., an agent message bus), a second network element in the network (e.g., the second network element supports network services), or a terminal device.

[0211] The first intelligent agent supports understanding the intent of the terminal device using a large application model, decomposing the intent into tasks (e.g., a first task) for one or more service domains (e.g., a perception service domain, a computing power service domain, or a session service domain), and then invoking the intelligent agent in each of the one or more service domains to orchestrate the tasks for that service domain. Optionally, the first intelligent agent can decompose the intent of the terminal device into network functional requirements for one or more service domains, and then determine the tasks for that service domain based on the network functional requirements of each service domain. Optionally, the first intelligent agent can also schedule network services based on the task orchestration results (e.g., the orchestration results of the first task) to complete the execution of the task. For example, the first intelligent agent can generate orchestration instructions that can be recognized by network elements (e.g., the first network element) based on the task orchestration results, and then issue the orchestration instructions to the first network element to trigger the execution of the task. As an example, the functional modules included in the first intelligent agent are shown in Figure 5(b). The network task management module organizes sys-agents and service agents into task intelligent orchestration workgroups, enabling intelligent orchestration collaboration among these agents to obtain instructions that drive the first network element to schedule and execute network services. The agent topology management module manages the topology of local agents, providing functions such as agent selection, discovery, registration, and deregistration. The task orchestration codec module encodes the output of the network task management module, generating instructions that the first network element can understand and recognize, and calls the first network element interface to issue instructions to the first network element to schedule network services for task execution. The agent connection management module manages the underlying transmission connections and maintains the state of information interaction and communication among multiple agents. The underlying transmission protocol can be Transmission Control Protocol (TCP), Hypertext Transfer Protocol (HTTP), or other protocols, without limitation.

[0212] The second intelligent agent can include agents from different service domains, such as connection agents, computing agents, or perception agents. The second intelligent agent receives task orchestration requests from the first intelligent agent, supports calling large models locally or remotely for task orchestration, and outputs the task orchestration results. As an example, the functional modules included in the second intelligent agent are shown in Figure 5(c). The task orchestration module receives task orchestration requests from the first intelligent agent and can use the AI ​​model and knowledge base of its service domain for reasoning and task orchestration to obtain the task orchestration results. The intelligent agent topology management module manages the topology of the local intelligent agent, providing functions such as intelligent agent selection, intelligent agent discovery, intelligent agent registration, and deregistration. The intelligent agent connection management module implements the underlying transmission connection management and state maintenance for information interaction and communication between multiple intelligent agents. Optionally, the second intelligent agent can be viewed as an integrated intelligent agent from multiple service domains, or in other words, the second intelligent agent possesses the service capabilities of multiple intelligent agents; that is, the second intelligent agent supports task orchestration for tasks from multiple service domains and outputs the task orchestration results. For example, the second intelligent agent may include multiple intelligent agents, such as any two or three intelligent agents among connection intelligent agents, computation intelligent agents, or perception intelligent agents, without limitation. Optionally, the second intelligent agent may also schedule network services based on the task orchestration result (e.g., the orchestration result of the first task) to complete the task execution. For example, the second intelligent agent may generate orchestration instructions that can be recognized by network elements (e.g., the first network element) based on the task orchestration result, and then issue the orchestration instructions to the first network element to trigger the execution of the first task.

[0213] The third network element supports automatic online registration, capability registration, and inter-agent discovery for intelligent agents within the network. As an example, the functional modules included in the third network element are shown in Figure 5(d). Specifically, the intelligent agent registration management provides intelligent agent registration and / or deregistration interfaces. When an intelligent agent goes online, it can call the registration interface to report its information to the third network element, such as at least one of the following: intelligent agent identifier, capabilities, roles, open interfaces, supported protocols, or location information. When an intelligent agent goes offline, it can call the deregistration interface to request the deletion of its information from the third network element. The intelligent agent discovery management provides a discovery interface. Intelligent agents (e.g., sys-agents) can call the discovery interface to discover service agents in specific service domains. The input information for the discovery interface can be at least one of the intelligent agent's capabilities, roles, or location information. The intelligent agent information database stores intelligent agent information and provides interfaces for adding, deleting, and modifying information for intelligent agent registration and discovery.

[0214] A message bus can be viewed as a distributed message bus for collaboration and dialogue among multiple intelligent agents. In one implementation, the message bus can be coupled to an intelligent agent (e.g., a first or second intelligent agent), or embedded within the agent as an internal functional module. In this implementation, the first intelligent agent can directly send a first request message to the second intelligent agent, thus achieving rapid point-to-point communication among multiple agents. In another implementation, the message bus can be decoupled from the intelligent agent (e.g., a first or second intelligent agent), or acting as an independent routing node. In this implementation, the first intelligent agent can forward a first request message to the second intelligent agent through the agent's message bus. For example, after receiving a second request message from the first intelligent agent, the message bus can perform rapid message routing and forwarding based on the second intelligent agent's identifier and / or address, thereby improving efficient collaboration among multiple agents and ultimately realizing the intent of the terminal device. In schemes that forward messages between multiple agents through a server, the server needs to manage the multiple agents, such as managing their information (e.g., agent identifiers), making the process complex. In contrast, the message bus does not require management of multiple agents and can quickly route and forward messages based on information from the second agent, which is beneficial to improving the collaboration efficiency among multiple agents. As an example, the functional modules included in the message bus are shown in Figure 5(e). The agent routing module stores the relationship between agent identifiers and agent addresses, and routes and forwards messages based on the agent identifiers. The agent reachability management module stores the reachability status of agent addresses. If the agent address is reachable, the message can be routed; otherwise, message routing fails and an error code is returned. Agent reachability can be checked periodically. The agent connection management module implements the underlying transmission connection management and state maintenance for information interaction and communication among multiple agents.

[0215] Optionally, the agent connection management module in the first and / or second agent can be connected to the agent message bus to realize functions such as routing management, status reporting, and message reception and forwarding.

[0216] The first network element supports receiving task instructions from the first intelligent agent, scheduling network services to execute tasks, and realizing network service functions. The network services involved in this application can be network atomic services or other services, as long as they provide a service interface for the first network element to schedule. For example, network services may include at least one of the following: access services, session services, computing power services, sensing services, or other services.

[0217] A terminal device acts as a mobile terminal requesting network services. In one implementation, the terminal device can send its intent, such as requesting computing resources, to a first intelligent agent or a first network element.

[0218] As one implementation, the first intelligent agent, the second intelligent agent, the first network element, and the third network element in the above network architecture can be deployed independently as independent network elements. The message bus can also be an independent network element, such as an intelligent agent message router deployed independently, or it can be deployed as a functional module within the system intelligent agent and / or service intelligent agent. There are no limitations on this.

[0219] It should be understood that the naming of network elements in the above network architecture is defined only for the convenience of distinguishing different functions and does not constitute a limitation on this application. This application does not exclude the possibility of using other names for various network elements in mobile network architecture and other future networks. For example, the above-mentioned agent storage repository can also be called agent storage node or agent information storage network element, etc., and there is no limitation in this regard.

[0220] It should also be understood that the network architecture shown in Figure 5 is an example given to understand the technical solution and does not constitute a limitation on the scope of protection of this application.

[0221] The following examples, with reference to Figures 6 to 8, illustrate the communication method provided in the embodiments of this application, which can be applied to the aforementioned network architecture. The embodiments of this application use a first network element (e.g., TCF), a first intelligent agent (e.g., a system intelligent agent sys-agent), a second intelligent agent (e.g., a service intelligent agent), and a third network element (e.g., an agent repository) as the executing entities. These executing entities, such as the first network element, the first intelligent agent, the second intelligent agent, and the third network element, can be themselves, or components within them (e.g., processors, chips, or chip systems, such as circuits or chips in the first intelligent agent responsible for communication functions (e.g., modem chips, also known as baseband chips, or SoC chips or SIP chips containing modem cores), or logic modules or software capable of implementing all or part of the functions).

[0222] Figure 6 is a schematic flowchart of a communication method 600 provided in an embodiment of this application. As shown in Figure 6, the method includes the following steps. For details not covered herein, please refer to the relevant descriptions in existing solutions. For the sake of brevity, these details will not be elaborated here.

[0223] S610, the first intelligent agent receives the first information.

[0224] The first information is used to describe the intent of the terminal device, or in other words, to request the implementation of the intent of the terminal device described by the first information.

[0225] For example, the intent could be that the terminal device requests computing resources, and the first information could be that the terminal device requests computing resources.

[0226] Understandably, "intent" refers to the desired outcome for a target object. An intent can include at least one of the following: intent expectation, intent object, intent goal, and intent context. Specifically, the intent expectation refers to the business scenario targeted by the intent, such as entrusting a digital human to perform tasks on behalf of the user, intelligently driving terminal devices, or expanding the computing power of terminal devices; the intent object refers to the target object targeted by the intent, such as terminal devices, base stations, virtual private networks, call applications, computing services, or third-party applications; the intent goal refers to the target value of the performance indicators to be achieved, such as throughput, latency, or handover success rate; and the intent context refers to the constraints on the intent expressed through comparison, inclusion, and association relationships, such as the duration of the intent's effect and the intent's priority.

[0227] In this application, the intent of the terminal device can be replaced by the user intent, or the intent of the user using the terminal device. The first information can be replaced by an intent description, a user intent description, etc. This application does not limit the specific names of the terminal device intent and the first information.

[0228] In this application, the intent description of the terminal device can be a natural language description, such as natural language text, voice, or images, or a formatted intent expression model, such as the intent expression model defined by the intent driven management service (IDMS) of the 3GPP management standards working group. It can also be other descriptive methods such as pictures, text, voice, video, binary code files, or application (APP) installation packages. This application does not limit the form, content, or format of the intent description of the terminal device.

[0229] In one implementation, the first intelligent agent receives first information, including: the first intelligent agent receiving first information from a first network element. For example, a terminal device sends its intention to the first network element. After obtaining the terminal device's intention, the first network element determines that it cannot recognize the terminal device's intention or does not support the decomposition and arrangement of the terminal device's intention, and then sends the first information to the first intelligent agent. That is, the first network element is the network element that sends the terminal device's intention to the first intelligent agent.

[0230] In one example, the first information is carried in a non-access stratum (NAS) message. The NAS message adds an intent information element of the terminal device. That is, the terminal device indicates its intent by sending a NAS message carrying the intent information element to the first network element.

[0231] This application does not limit the specific implementation method of the first network element being unable to recognize the intent of the terminal device or not supporting the decomposition and arrangement of the intent of the terminal device.

[0232] In one example, when the first information is carried in a NAS message, the first network element receives and parses the NAS message to obtain the intent information element. However, the first network element cannot understand the content indicated by the intent information element, such as the information described in the natural language in the intent information element. Therefore, the first network element cannot recognize the intent of the terminal device, or does not support the decomposition and arrangement of the intent of the terminal device, and thus forwards the first information to the first intelligent agent.

[0233] Optionally, this application does not limit the network element that sends the intention of the terminal device to the first intelligent agent; it can be the first network element, other network elements, or third-party applications, etc.

[0234] S620, the first intelligent agent determines the first task based on the first information.

[0235] The first task is used to realize the intent of the terminal device.

[0236] In one implementation, after acquiring initial information, the first intelligent agent can use an AI model, such as a Large Language Model (LLM), to understand the intent of the terminal device, obtain one or more network function requirements, and then plan one or more tasks based on these requirements. One network function requirement can correspond to one or more tasks; this is not limited.

[0237] For example, assuming the terminal device's intent is to request computing resources, the first agent, after understanding the terminal device's intent using LLM, can infer that service agents from both the computing power service domain and the connectivity service domain need to collaborate to orchestrate the first task. For instance, the agent in the computing power service domain (e.g., a computing agent) orchestrates the computing resource request task, and the agent in the connectivity service domain (e.g., a session agent) orchestrates the computing session creation task. Understandably, the first task in this case includes both the computing resource request task and the computing session creation task.

[0238] It should be noted that this application does not limit the number of first tasks determined by the first intelligent agent; there may be one or more.

[0239] Optionally, after determining the first task, the first agent may request a third network element (e.g., agent repository) to discover one or more service agents in one or more service domains to orchestrate the first task. That is, the method 600 further includes the following steps S630 and S640.

[0240] S630, the first intelligent agent sends a second request message to the third network element;

[0241] Correspondingly, the third network element receives the second request message from the first intelligent agent.

[0242] The second request message is used to request the discovery of an agent that supports the orchestration of the first task.

[0243] In one implementation, after determining the first task, the first intelligent agent can invoke the discovery interface of the third network element to discover service intelligent agents (i.e., the second intelligent agent) in a specific service domain. For example, a computing intelligent agent (also called a computing power intelligent agent) in the computing power service domain and a connection intelligent agent (also called a connection intelligent agent) in the connection service domain. The input information of the discovery interface can be one or more pieces of information carried in the second request message.

[0244] For example, the second request message includes information describing at least one of the following characteristics of the requesting agent: role, supported open interfaces, or capabilities. The supported open interfaces can be replaced with the agent's address information, such as an internet protocol address (IP), a fully qualified domain name (FQDN), or a network domain name like a uniform resource locator (URL).

[0245] Optionally, the second request message may also include information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, or supported models.

[0246] Example 1: If the second request message includes a deployment location, such as location A, it means that the first agent requests to discover a service agent deployed at location A that supports orchestration of the first task.

[0247] Example 2: If the second request message includes a service domain, for example, a computing service domain, it means that the first agent requests to find a service agent that supports orchestration of tasks (e.g., computing resource application tasks) in the computing service domain; or, for example, a connectivity service domain, it means that the first agent requests to find a service agent that supports orchestration of tasks (e.g., computing session creation tasks) in the connectivity service domain.

[0248] Example 3: If the second request message includes supported protocols (or session / interaction protocols), such as Transmission Control Protocol TCP, it means that the first agent requests to discover service agents that support TCP and can orchestrate the first task.

[0249] Example 4: If the second request message includes a role, such as a connecting agent, a perceptual agent, or a computational agent, it indicates that the first agent is requesting to discover a connecting agent, a perceptual agent, or a computational agent. It is understood that a connecting agent refers to an agent that supports the orchestration of tasks in the connectivity service domain, a perceptual agent refers to an agent that supports the orchestration of tasks in the perception service domain, and a computational agent refers to an agent that supports the orchestration of tasks in the computation service domain. Optionally, this application does not specifically limit the names of agents in different service domains.

[0250] Example 5: If the second request message includes capabilities, such as computing resource management capability, it means that the first agent requests to find an agent with computing resource management capability; or, for example, computing power session management capability, it means that the first agent requests to find an agent with computing power session management capability.

[0251] Example 6: If the second request message includes supported open interfaces, such as a Uniform Resource Locator URL or an Application Programming Interface (API), it indicates that the first agent is requesting the discovery of an agent-supported URL or API.

[0252] Example 7: If the second request message includes supported models, such as Large Language Model (LLM), CodeLLM, or Multimodal Large Model, it means that the first agent requests that the discovered agent supports LLM, CodeLLM, or Multimodal Large Model.

[0253] It should be understood that the above examples are merely illustrative for the purpose of understanding the technical solutions, and other solutions are not excluded. The above examples can be implemented independently or in combination, and this application does not limit them in this regard.

[0254] In other words, the first agent requests the third network element to discover an agent that supports orchestrating the first task by sending a second request message. Optionally, the agent may also support at least one of the following: deployed at location A, belonging to the computing service domain, supporting TCP, possessing computing resource management capabilities, supporting APIs, or supporting the LLM model, etc.

[0255] S640, the third network element sends the information of the second intelligent agent to the first intelligent agent;

[0256] Correspondingly, the first intelligent agent receives information from the second intelligent agent from the third network element.

[0257] In other words, after receiving the second request message, the third network element discovers the second intelligent agent, that is, the second intelligent agent supports the orchestration of the first task, and then the third network element feeds back the information of the second intelligent agent to the first intelligent agent.

[0258] It should be noted that this application does not limit the number of second intelligent agents; there can be one or more. One intelligent agent can correspond to one service domain, or multiple intelligent agents can correspond to one service domain, or one intelligent agent can correspond to multiple service domains; there is no limitation in this regard. For example, the second intelligent agent includes intelligent agent 1, which belongs to the computing service domain; or, the second intelligent agent includes intelligent agent 1 and intelligent agent 2, where intelligent agent 1 belongs to the computing service domain and intelligent agent 2 belongs to the connectivity service domain; or, the second intelligent agent includes intelligent agent 1, intelligent agent 2, intelligent agent 3, and intelligent agent 4, where intelligent agent 1 and intelligent agent 2 both belong to the computing service domain, and intelligent agent 3 and intelligent agent 4 both belong to the connectivity service domain; there is no limitation in this regard.

[0259] It should be noted that if there are multiple second intelligent agents, for example, if the second intelligent agents include a computing agent and a connection agent, then the first intelligent agent can repeatedly execute the above steps S630 and S640, that is, the first intelligent agent sends two request messages to the third network element, one for discovering the computing agent and the other for discovering the connection agent. Alternatively, the first intelligent agent can also execute the above steps S630 and S640 once, that is, by sending one request message, to discover the computing agent and the connection agent; there is no limitation on this.

[0260] In one implementation, the third network element sends information about the second intelligent agent to the first intelligent agent, including: the third network element sending information about multiple intelligent agents, including the second intelligent agent, to the first intelligent agent; further, the first intelligent agent determines the second intelligent agent from among the multiple intelligent agents based on the information from the multiple intelligent agents. That is, if multiple service intelligent agents meet the discovery requirements, the third network element can feed back the information about the multiple intelligent agents to the first intelligent agent, and then the first intelligent agent can select one or more second intelligent agents from among the multiple intelligent agents to perform the orchestration of the first task.

[0261] For example, the information of the multiple agents includes at least one of the following: the identities of the multiple agents (e.g., the identifiers (IDs) of the agents) or the open interfaces (e.g., URLs or APIs) supported by the multiple agents.

[0262] For example, assuming multiple agents include agent 1, agent 2, and agent 3, the information of the multiple agents may include at least one of the following: the identifiers of the multiple agents (e.g., agent 1 is identified as ID1, agent 2 as ID2, and agent 3 as ID3), and the open interfaces supported by the multiple agents (e.g., agent 1 supports API, agent 2 supports URL and API, and agent 3 supports URL).

[0263] Optionally, the information of the multiple agents may also include at least one of the following: the deployment location of the multiple agents, the service domain of the multiple agents, the protocols supported by the multiple agents, the capabilities of the multiple agents, the roles of the multiple agents, the models supported by the multiple agents, or the service domain of the multiple agents.

[0264] For example, assuming multiple agents include Agent 1 and Agent 2, the information of the multiple agents may include at least one of the following: the deployment location of the multiple agents (e.g., Agent 1 and Agent 2 are both deployed at location A, and Agent 3 is deployed at location B), the service domain of the multiple agents (e.g., Agent 1 belongs to the computing service domain, and Agent 2 and Agent 3 belong to the connectivity service domain), the protocols supported by the multiple agents (e.g., Agent 1 supports TCP, Agent 2 supports HTTP and TCP, and Agent 3 supports HTTP), the capabilities supported by the multiple agents (e.g., Agent 1 supports computing resource management capabilities, and Agent 2 and Agent 3 support computing session management capabilities), the roles of the multiple agents (e.g., Agent 1 is a computing agent, and Agent 2 and Agent 3 are both connectivity agents), the models supported by the multiple agents (e.g., Agent 1, Agent 2, and Agent 3 all support LLM), and the service domain of the multiple agents (e.g., Agent 1 belongs to the computing service domain, and Agent 2 and Agent 3 belong to the connectivity service domain).

[0265] Furthermore, the first intelligent agent can select Intelligent Agent 1 and Intelligent Agent 2 as second intelligent agents to perform the orchestration of the first task. Specifically, Intelligent Agent 1 can orchestrate the task of requesting computing resources, and Intelligent Agent 2 can orchestrate the task of creating computing sessions.

[0266] This application does not limit the implementation method of the first intelligent agent determining the second intelligent agent from multiple intelligent agents. For example, regarding the determination of the aforementioned connecting intelligent agents (e.g., intelligent agent 2 and intelligent agent 3), the first intelligent agent can determine the second intelligent agent based on the open interfaces and / or protocols supported by the intelligent agent. For example, since intelligent agent 2 supports URL and API as open interfaces and HTTP and TCP as protocols, and intelligent agent 2 supports URL as an open interface and HTTP as a protocol, the first intelligent agent can select intelligent agent 2, which has greater flexibility and adaptability (or supports more open interfaces and more protocols), as the second intelligent agent.

[0267] In another implementation, the third network element sends information about the second intelligent agent to the first intelligent agent, including: the third network element sending a third response message to the first intelligent agent, the third response message including a first identifier and information about the second intelligent agent, the first identifier being used to identify the third response message as a response to the second request message.

[0268] For example, the information of the second intelligent agent may include at least one of the following: the identifier of the second intelligent agent, or the open interfaces supported by the second intelligent agent. Optionally, the information of the second intelligent agent may also include at least one of the following: the deployment location of the second intelligent agent, the service domain of the second intelligent agent, the protocols supported by the second intelligent agent, the capabilities of the second intelligent agent, the role of the second intelligent agent, the models supported by the second intelligent agent, or the service domain of the second intelligent agent, as specifically defined above, and will not be described again here.

[0269] Optionally, the second request message in step S630 above may also carry the first identifier. In this application, the first identifier may be replaced with: session identifier, or discovery request session identifier, or agent discovery request session identifier, and its specific name is not limited.

[0270] Based on the above steps, after determining the first task and the second agent, the first agent can request the second agent to orchestrate the first task. That is, the method also includes the following steps S650-S670.

[0271] S650, the first intelligent agent sends a first request message to the second intelligent agent;

[0272] Correspondingly, the second agent receives the first request message from the first agent.

[0273] The first request message is used to request the orchestration of the first task.

[0274] For example, suppose the first intelligent agent is a system intelligent agent (referred to as sys-intelligent agent) and the second intelligent agent is a service intelligent agent, such as including a computing agent and a connection agent. This means that the computing agent and the connection agent jointly complete the first task. The first intelligent agent requests the computing agent and the connection agent to perform task orchestration respectively. For example, it requests the computing agent to orchestrate the computing resource application task and requests the connection agent to perform task orchestration for the computing session creation task.

[0275] For example, the first request message includes a description of the first task and / or the output format of the orchestration result of the first task, as specifically interpreted below.

[0276] (1) Description of the first task;

[0277] For example, assuming the first task includes a computing resource request task and a computing session creation task, and the second agent includes a computing agent and a connection agent, the description information of the first task may include: requesting the computing agent to orchestrate the computing resource request task (e.g., the size or bandwidth of the requested computing resources, storage resources, or network resources), and requesting the connection agent to orchestrate the computing session creation task (e.g., creating a session, the session type being a computing session, etc.).

[0278] In this application, the computing resource application task and the computing session creation task can be regarded as two sub-tasks of the first task. That is, the computing agent completes the computing resource application task, and the connecting agent completes the computing session creation task. Only when both sub-tasks are completed can it be said that the first task is completed.

[0279] Optionally, the computing power resource application task and the computing power session creation task can be described in different ways, and there is no limitation on this.

[0280] It is understood that the computational agent and the connection agent in the embodiments of this application can form an agent collaboration group to jointly execute the orchestration of the first task.

[0281] (2) Output format of the arrangement results of the first task;

[0282] For example, the output format of the orchestration results of the first task can be JSON or YAML, etc., without any restrictions.

[0283] For example, when the output format is described using JSON, the orchestration result of the first task can be as follows.

[0284] It should be noted that the first network element can correctly identify and parse the description information of the first task and / or the output format of the orchestration result of the first task carried in the first request message. Therefore, after receiving the first request message, the first network element can call the local or remote AI model to perform inference and task orchestration.

[0285] In one implementation, the first intelligent agent sends a first request message to the second intelligent agent via a message bus;

[0286] Correspondingly, the second agent receives the first request message from the first agent via the message bus.

[0287] For example, a first agent sends a first request message and information about a second agent to a message bus. This information includes the identifier and / or address of the second agent, and is used by the message bus for message routing. Further, based on the received identifier and / or address of the second agent, the message bus can send the first request message to the second agent.

[0288] In other words, the first and second agents can forward messages through a message bus. The message bus can be decoupled from the agents (e.g., the first or second agent), or in other words, it can act as an independent routing node. In this implementation, after receiving a second request message from the first agent, the message bus can perform fast message routing and forwarding based on the second agent's identifier and / or address, thereby improving efficient collaboration among multiple agents and providing effective network services to terminal devices. Furthermore, compared to the server shown in Figure 4(b) which manages agents and distributes messages, the message bus does not need to manage agents and can simply support fast information forwarding between agents. It can also provide more detailed message distribution information, such as agent capabilities, supported interfaces, and supported protocols, which not only reduces power consumption and signaling overhead but also ensures efficient communication and collaboration among multiple agents.

[0289] Optionally, the message bus can store the correspondence between the identifier of the second intelligent agent and the address of the second intelligent agent. After receiving the identifier of the second intelligent agent from the first intelligent agent, the message bus can determine the address of the second intelligent agent according to the correspondence, and then send a first request message to the second intelligent agent according to the address of the second intelligent agent.

[0290] In another implementation, the first agent sends a first request message to the second agent based on the information from the second agent.

[0291] For example, based on the above steps S630 and S640, the first intelligent agent can obtain the information of the second intelligent agent, the identifier of the second intelligent agent and / or the address of the second intelligent agent from the third network element. Then, based on the identifier of the second intelligent agent and / or the address of the second intelligent agent, the first intelligent agent can send a first request message to the second intelligent agent.

[0292] In other words, the first and second agents can directly interact via messages; that is, the first agent can directly send a first request message to the second agent, thereby achieving rapid point-to-point communication among multiple agents and ensuring efficient collaboration among them. Optionally, this implementation can be understood as follows: the message bus is coupled to the agents (e.g., the first or second agent), or in other words, the message bus can be embedded inside the agents as an internal functional module, without limitation.

[0293] Optionally, the first intelligent agent can store the correspondence between the identifier of the second intelligent agent and the address of the second intelligent agent. After obtaining the identifier of the second intelligent agent from the second intelligent agent, the first intelligent agent can determine the address of the second intelligent agent according to the correspondence, and then send a first request message to the second intelligent agent according to the address of the second intelligent agent.

[0294] For example, assuming the first intelligent agent is a system intelligent agent (referred to as the sys-agent) and the second intelligent agent is a service intelligent agent, such as including a computing agent and a connection agent, then the computing agent and the connection agent jointly complete the first task, or in other words, the computing agent and the connection agent respectively perform task orchestration to complete the first task. Exemplarily, the first intelligent agent can call the API corresponding to the computing agent to send the computing resource application task description information and / or the output format of the task orchestration result to the computing agent, and the first intelligent agent can call the API corresponding to the connection agent to send the computing session creation task description information and / or the output format of the task orchestration result to the connection agent. Optionally, the first intelligent agent can use the message bus for routing and message forwarding to request the computing agent to orchestrate the computing resource application task and to request the connection agent to orchestrate the computing session creation task; this is not limited.

[0295] S660, Optionally, the second intelligent agent determines the arrangement result of the first task based on the first request message.

[0296] For example, after receiving a task orchestration request, the second intelligent agent can invoke a local or remote AI model to perform inference and task orchestration, obtaining the orchestration result of the first task. The orchestration result of the first task includes: interface information of the second network element in the network being invoked. Optionally, it may also include the type of the second network element in the network being invoked, and / or, information such as input parameters. It is understood that the second network element supports the execution of the first task, or in other words, the second network element is the network element that executes the first task; for example, the second network element could be a computing power network element, a session network element, or a sensing network element, etc.

[0297] Optionally, assuming the second intelligent agent includes a computing agent, the orchestration result of the first task may also include at least one of the following: the size of computing resources, the size of storage resources, or the size of network resources, etc.

[0298] S670, the second agent sends the arrangement result of the first task to the first agent;

[0299] Correspondingly, the first agent receives the arrangement result of the first task from the second agent.

[0300] Optionally, after receiving the orchestration result of the first task, the first agent reviews and determines that the orchestration result of the first task meets the requirements. If the orchestration result does not meet the requirements, the first agent can interact with the second agent multiple times until it obtains an orchestration result that meets the requirements.

[0301] The specific implementation method for the first intelligent agent to review whether the arrangement result of the first task meets the requirements is as follows.

[0302] For example, the first agent checks whether the output format of the orchestration result of the first task meets the requirements, such as whether the output format is JSON or YAML. If the format meets the requirements, the orchestration result of the first task meets the requirements; otherwise, if the format does not meet the requirements, the orchestration result of the first task does not meet the requirements.

[0303] For example, the first agent reviews whether the content in the orchestration result of the first task meets the requirements. For instance, it summarizes the orchestration results of the second agent (computation agent and connection agent). If it finds that the information provided by the second agent cannot complete the execution of the first task, such as the computing session establishment task requiring the input of the API gateway address, but the orchestration result does not provide the action of orchestrating to obtain the API gateway address, then the end-to-end business may not be completed. The first agent can consider that the orchestration result of the first task does not meet the requirements, and then can request the second agent to re-orchestrate.

[0304] Optionally, if the orchestration result of the first intelligent agent's review of the first task does not meet the requirements, when the first intelligent agent requests the second intelligent agent to re-orchestrate, it can send the reason for the non-compliance to the second intelligent agent. For example, it can include the content that the second intelligent agent needs to execute in the task description information so that the second intelligent agent can effectively orchestrate the task and obtain the orchestration result that meets the symbol requirements.

[0305] Based on the above implementation, the first intelligent agent can obtain the orchestration result of the first task from the second intelligent agent, and generate orchestration instructions (e.g., second information) that can be recognized or parsed by the first or second network element. Furthermore, the first intelligent agent can directly or indirectly invoke network services to execute the first task, as specifically described below. It should be noted that the following two cases (e.g., steps S680 and S690) and case two (e.g., step S601, not shown in the figure) can be executed selectively, without limitation.

[0306] Scenario 1: The first intelligent agent calls the task orchestration interface of the first network element and sends the orchestration instructions to the first network element, triggering the first network element to call the network service. That is, the first intelligent agent indirectly calls the network service to execute the first task.

[0307] S680, the first intelligent agent sends the second information to the first network element, triggering / causing the first network element to schedule network services to execute the first task;

[0308] Correspondingly, the first network element receives the second information from the first intelligent agent.

[0309] The second information may carry the orchestration result of the first task, including interface information for calling the first network element, such as calling the computing resource allocation interface and / or calling the session creation interface; or, the orchestration result of the first task may also include the type of the first network element in the network being called, and / or the input parameters and other information.

[0310] In this application, the second information is determined based on the orchestration result of the first task. For example, if the orchestration result of the first task meets the requirements, the first agent generates second information that meets the requirements of the first network element based on the orchestration result. For instance, the format of the second information can be JSON or YAML, which the first network element can recognize or parse; this is not limited.

[0311] S690: The first network element, based on the second information, schedules the second network element in the network to perform the first task.

[0312] Understandably, the second network element is used to execute the first task; in other words, the second network element supports the execution of the first task, or the second network element supports the execution of the first task. For example, if the network service can be a computing power service, then the second network element can be a computing power network element; or if the network service can be a connectivity service, then the second network element can be a session network element; or if the network service can be a sensing service, then the second network element can be a sensing network element, etc.

[0313] In this application, the second network element can be one or more, that is, the first network element can schedule one or more second network elements to perform the first task, without limitation.

[0314] In one implementation, the first network element sends a third request message to the second network element, which requests the execution of the first task; correspondingly, after receiving the third request message, the second network element executes the first task and obtains the execution result of the first task.

[0315] Optionally, the second network element may send the execution result of the first task to the first network element.

[0316] For example, suppose the first task includes a computing resource request task and a computing session creation task; the network services include computing power services and session services; and the second network element includes a computing power network element and a session network element. Then, the first network element can call the computing resource allocation interface to request the computing power network element to execute the computing resource request task, and pass request parameters (e.g., the size of computing resources, the size of storage resources, or the size of network resources) to the computing power network element. After completing the allocation of computing resources, the computing power network element can return the resource allocation result to the first network element, including at least one of the following information: API gateway address, software repository IP address, or remote computing power ID. Additionally, the first network element can also call the session creation interface to request the session network element to execute the computing power session creation task, and pass request parameters (e.g., the session type is a computing power session) to the session network element. After creating the computing power session, the session network element can return the session creation result to the first network element, including at least one of the following information: API gateway address, software repository IP address, or computing power session ID.

[0317] Optionally, the resource allocation result can be carried in the computing power resource allocation response message, and the session creation result can be carried in the session creation success response message.

[0318] Optionally, the session network element can interact with the UE to complete the creation of a computing power session. For specific implementation methods, please refer to the relevant descriptions of existing session creation schemes. For the sake of brevity, they will not be described here.

[0319] Scenario 2: The first intelligent agent directly calls the network service to perform the first task.

[0320] S601, the first intelligent agent sends a fourth request message to the second network element in the network. The fourth request message is used to request the execution of the first task. Correspondingly, after receiving the fourth request message, the second network element executes the first task and obtains the execution result of the first task.

[0321] The specific implementation of the second network element in the scheduling network to perform the first task, and the specific implementation of the second network element performing the first task, can be referred to the relevant descriptions of steps S680 and S690 above. For the sake of brevity, they will not be described here.

[0322] In summary, once the first task is orchestrated and executed, the intention of the terminal device can be realized, namely, the terminal device can obtain computing resources. For example, the terminal device can call network computing resources through a created computing session.

[0323] Optionally, in this embodiment, the third network element supports functions such as capability registration and agent discovery for intelligent agents within the network. That is, after an intelligent agent (e.g., the first and / or the second intelligent agent) goes online, it can request that its information be stored in the third network element for subsequent agent discovery. After an intelligent agent (e.g., the first and / or the second intelligent agent) goes offline, it can request that its information be deleted from the third network element. Additionally, an intelligent agent (e.g., the first and / or the second intelligent agent) can also request the third network element to update its information.

[0324] The following examples illustrate the registration process for the first and / or second intelligent agents, using steps S602-S605 (not shown in the figure), the deregistration process for the first and / or second intelligent agents, using steps S606-S609, and the update registration process for the first and / or second intelligent agents, using steps S610-S613.

[0325] S602, the first intelligent agent sends the first registration request message to the third network element;

[0326] Correspondingly, the third network element receives the first registration request message from the first intelligent agent.

[0327] Specifically, the first registration request message is used to request registration with a third network element. In other words, the first registration request message is used to request the third network element to store the information of the first intelligent agent.

[0328] For example, the first registration request message includes at least one of the following: the identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent. For specific interpretations, please refer to the relevant descriptions above, which will not be explained here.

[0329] S603, Optionally, the third network element sends a first registration response message to the first intelligent agent;

[0330] Correspondingly, the first intelligent agent receives the first registration response message from the third network element.

[0331] The first registration response message indicates that the first agent has successfully registered. In other words, the third network element stores the information of the first agent. For example, the agent registration management module in the third network element stores the information of the first agent in the agent information database of the third network element for subsequent agent discovery.

[0332] S604, the second intelligent agent sends a second registration request message to the third network element;

[0333] Correspondingly, the third network element receives the second registration request message from the second intelligent agent.

[0334] Specifically, the second registration request message is used to request registration with the third network element. In other words, the second registration request message is used to request the third network element to store the information of the second intelligent agent.

[0335] For example, the second registration request message includes at least one of the following: the identification information of the second intelligent agent, the service domain information of the second intelligent agent, the deployment location information of the second intelligent agent, the role information of the second intelligent agent, the capability information of the second intelligent agent, the protocol information supported by the second intelligent agent, the open interface information supported by the second intelligent agent, or the model information supported by the second intelligent agent. For specific interpretations, please refer to the relevant descriptions above, which will not be explained here.

[0336] S605, optionally, the third network element sends a second registration response message to the second intelligent agent;

[0337] Correspondingly, the second intelligent agent receives a second registration response message from the third network element.

[0338] The second registration response message indicates that the second agent has successfully registered. In other words, the third network element stores the information of the second agent. For example, the agent registration management module in the third network element stores the information of the second agent in the agent information database of the third network element for subsequent agent discovery.

[0339] It should be noted that this application does not limit the execution order of the above steps S602-S603 and S604-S605. In other words, this application does not limit the execution order of the registration process of the first intelligent agent and the second intelligent agent.

[0340] S606, the first intelligent agent sends the first registration request message to the third network element;

[0341] Correspondingly, the third network element receives the first deregistration request message from the first intelligent agent.

[0342] The first deregistration request message is used to request the deletion of information about the first agent.

[0343] For example, the first registration request message may carry the identifier of the first intelligent agent.

[0344] S607, optionally, the third network element sends a first registration response message to the first intelligent agent;

[0345] Correspondingly, the first intelligent agent receives the first deregistration response message from the third network element.

[0346] The first deregistration response message indicates that the first intelligent agent has successfully registered. In other words, the third network element deletes the information of the first intelligent agent; for example, the intelligent agent registration management module in the third network element deletes the information of the first intelligent agent from the intelligent agent information database in the third network element.

[0347] S608, the second intelligent agent sends a second registration request message to the third network element;

[0348] Correspondingly, the third network element receives the second deregistration request message from the second intelligent agent.

[0349] The second deregistration request message is used to request the deletion of information about the second agent.

[0350] For example, the second registration request message may carry the identifier of the second smart agent.

[0351] S609, optionally, the third network element sends a second deregistration response message to the second intelligent agent;

[0352] Correspondingly, the second intelligent agent receives the second deregistration response message from the third network element.

[0353] The second deregistration response message indicates that the second agent has successfully registered. In other words, the third network element deletes the second agent's information; for example, the agent registration management module in the third network element deletes the second agent's information from the agent information database in the third network element.

[0354] It should be noted that this application does not limit the execution order of the above steps S606-S607 and S608-S609. In other words, this application does not limit the execution order of the deregistration process of the first intelligent agent and the second intelligent agent.

[0355] S610, the first intelligent agent sends the first update request message to the third network element;

[0356] Correspondingly, the third network element receives the first update request message from the first intelligent agent.

[0357] The first update request message is used to request an update to the information of the first agent.

[0358] For example, the first update request message includes at least one of the following: the identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent.

[0359] It should be noted that the information carried in the first update request message is different from the information carried in the first registration request message.

[0360] S611, Optionally, the third network element sends a first update response message to the first intelligent agent;

[0361] Correspondingly, the first intelligent agent receives the first update response message from the third network element.

[0362] The first update response message indicates that the information update of the first intelligent agent was successful or completed. In other words, the third network element updates the information of the first intelligent agent, or the third network element re-stores the information of the first intelligent agent.

[0363] For example, the second update request message includes at least one of the following: the identification information of the second intelligent agent, the service domain information of the second intelligent agent, the deployment location information of the second intelligent agent, the role information of the second intelligent agent, the capability information of the second intelligent agent, the protocol information supported by the second intelligent agent, the open interface information supported by the second intelligent agent, or the model information supported by the second intelligent agent.

[0364] It should be noted that the information carried in the second update request message is different from the information carried in the second registration request message.

[0365] S612, the second intelligent agent sends a second update request message to the third network element;

[0366] Correspondingly, the third network element receives the second update request message from the second intelligent agent.

[0367] The second update request message is used to request an update to the information of the second agent.

[0368] S613, Optionally, the third network element sends a second update response message to the second intelligent agent;

[0369] Correspondingly, the second intelligent agent receives a second update response message from the third network element.

[0370] The second update response message indicates that the information update of the second intelligent agent was successful or completed. In other words, the third network element updates the information of the second intelligent agent, or the third network element re-stores the information of the second intelligent agent.

[0371] It should be noted that this application does not limit the execution order of the above steps S610-S611 and S612-S613. In other words, this application does not limit the execution order of the registration process of the first intelligent agent and the second intelligent agent.

[0372] Based on the above scheme, the first intelligent agent receives first information describing the intent of the terminal device. By processing the intent of the terminal device, a first task for realizing the intent of the terminal device can be obtained. Then, the second intelligent agent is invoked to orchestrate the first task, resulting in an orchestration result for scheduling network services. This method not only enables efficient collaboration and communication among multiple intelligent agents but also achieves collaborative work between intelligent agents and mobile network services to realize the intent of the terminal device, meet the business needs of the terminal device, and improve the user experience.

[0373] For ease of understanding, the specific process applicable to the embodiments of this application is described below with reference to Figures 7 to 9. In the examples below, the first network element is TCF, the first intelligent agent is sys-intelligent agent, the second intelligent agent includes a computing intelligent agent and a connection intelligent agent, the third network element is an intelligent agent storage library, the message bus is an intelligent agent message bus, the network services include computing power services and session services, and the terminal device is UE. It is understood that the process described below is only an example, and the embodiments of this application are not limited thereto. The contents not described in detail below can be referred to the description in the method shown in Figure 6, and will not be repeated below.

[0374] Figure 7 is a schematic flowchart of a communication method 700 provided in this application. As shown in Figure 7, the method includes the following steps.

[0375] S701, the UE sends a network computing resource request message to the TCF;

[0376] Correspondingly, the TCF receives network computing resource request messages from the UE.

[0377] For example, a UE can carry its intention to request computing resources through a NAS message.

[0378] The network computing resource request message describes the UE's intent. For example, it requests computing resources or requests the fulfillment of the UE's intent. Optionally, the network computing resource request message can be replaced by the UE's intent. For a description of the UE's intent, please refer to the description of step S610 in method 600 above. For brevity, it will not be described here again.

[0379] S702, TCF sends an agent orchestration task request message (i.e., first information) to the sys agent;

[0380] Correspondingly, the sys agent receives an agent orchestration task request message from the TCF.

[0381] The intelligent agent orchestration task request message includes the UE's intent. The intelligent agent orchestration task request message is used to describe the UE's intent, or in other words, to request the implementation of the UE's intent described in the intelligent agent orchestration task request message.

[0382] For example, before executing step S702, the TCF determines that it cannot recognize the UE's intent or does not support the decomposition and task orchestration of the UE's intent. For instance, the TCF recognizes that it cannot use existing orchestration strategies to orchestrate the UE's intent and needs to use intelligent orchestration capabilities, thus triggering the execution of step S702.

[0383] The criteria for TCF to determine that it cannot recognize the UE's intent, or cannot decompose and orchestrate the UE's intent, can be found in the relevant description of step S610 of method 600 above. For the sake of brevity, it will not be described here again.

[0384] Furthermore, after receiving the agent orchestration task request message, the sys-agent uses LLM to understand the UE's intent and infers that the computing power service domain and the connectivity service domain need to cooperate to complete the computing power resource application task and the computing power session creation task (i.e., the first task). For the specific implementation method, please refer to the relevant description of step S620 of the above method 600. For the sake of brevity, it will not be described here.

[0385] Optionally, after determining the first task, the sys-agent may request the agent repository to discover one or more service agents in one or more service domains to orchestrate the first task. That is, the method 700 further includes the following steps S703 and S704.

[0386] S703, sys - Agents and agent repositories interact to compute agent discovery request / response messages.

[0387] For example, the sys agent sends a computation agent discovery request message (i.e., a second request message) to the agent repository;

[0388] Correspondingly, the agent repository receives computational agent discovery request messages from sys-agents.

[0389] The computational agent discovery request message is used to request the discovery of computational agents that support the orchestration of computational resource request tasks.

[0390] For example, a computational agent discovery request message includes information describing at least one of the following characteristics of the requesting computational agent: role (e.g., computational agent), supported open interfaces (e.g., API), or capabilities (e.g., computational resource management capabilities).

[0391] Optionally, the computational agent discovery request message may also include information describing at least one of the following characteristics of the requesting computational agent: deployment location (e.g., location A), service domain (e.g., computational service domain), supported protocol (e.g., TCP), or supported model (e.g., LLM).

[0392] For a detailed interpretation of the computational agent discovery request message, please refer to the relevant description of step S630 of method 600 above. For the sake of brevity, it will not be explained here.

[0393] For example, the agent repository sends a computation agent discovery response message (i.e., a third response message) to the sys-agent.

[0394] Correspondingly, the sys agent receives computation agent discovery response messages from the agent repository;

[0395] As an example, the computational agent discovery response message includes information about the computational agent (e.g., a second agent).

[0396] As another example, the computational agent discovery response message includes information from multiple computational agents (i.e., multiple agents, including a second agent).

[0397] For example, the information of the computational agent includes at least one of the following: the identifier of the computational agent (e.g., computational agent ID), or the open interface supported by the computational agent (e.g., URL or API).

[0398] Optionally, the information of the computational agent may also include at least one of the following: the deployment location of the computational agent, the service domain of the computational agent, the protocols supported by the computational agent, the capabilities of the computational agent, the role of the computational agent, the models supported by the computational agent, or the service domain of the computational agent.

[0399] For a detailed interpretation of the information about the computational agent, please refer to the relevant description of step S640 of method 600 above. For the sake of brevity, it will not be explained here.

[0400] Optionally, the computational agent discovery request / response message in step S703 above may also carry a session identifier (also known as a discovery request session identifier, or agent discovery request session identifier, i.e., the first identifier) ​​to identify that the computational agent discovery response message is a response to the computational agent discovery request message.

[0401] It should be noted that the specific implementation of how the agent repository sends the computation agent's information to the sys agent can be found in the relevant description of step S640 of method 600 above, and will not be described here for the sake of brevity.

[0402] In other words, the agent repository can feed back information from one or more computational agents to the sys-agent, meaning that this application does not limit the number of computational agents.

[0403] S704, sys - agent and agent repository interaction connection agent discovery request / response messages.

[0404] For example, the sys agent sends a Connect Agent Discovery Request message (i.e., the second request message) to the agent repository;

[0405] Correspondingly, the agent repository receives connected agent discovery request messages from the sys-agent.

[0406] Among them, the Connected Agent Discovery Request message is used to request the discovery of connected agents that support the orchestration of computing session creation tasks.

[0407] For example, the connected agent discovery request message includes information describing at least one of the following characteristics of the connected agent making the request: role (e.g., connected agent), supported open interfaces (e.g., API and URL), or capabilities (e.g., computing power session management capabilities).

[0408] Optionally, the connected agent discovery request message may also include information describing at least one of the following characteristics of the connected agent making the request: deployment location (e.g., location B), service domain (e.g., connection service domain), supported protocol (e.g., HTTP), or supported model (e.g., LLM).

[0409] For a detailed interpretation of the connection agent discovery request message, please refer to the relevant description of step S630 of method 600 above. For the sake of brevity, it will not be explained here.

[0410] For example, the agent repository sends a connected agent discovery response message (i.e., a third response message) to the sys-agent.

[0411] Correspondingly, the sys agent receives a connected agent discovery response message from the agent repository.

[0412] As an example, the connected agent discovery response message includes information about the connected agent (e.g., a second agent).

[0413] As another example, the connected agent discovery response message includes information from multiple connected agents (i.e., multiple agents, including a second agent).

[0414] For example, the information of the connecting agent includes at least one of the following: the identifier of the connecting agent (e.g., computation agent ID), or the open interface supported by the connecting agent (e.g., URL or API).

[0415] Optionally, the information of the connected agent may also include at least one of the following: the deployment location of the connected agent, the service domain of the connected agent, the protocols supported by the connected agent, the capabilities of the connected agent, the role of the connected agent, the models supported by the connected agent, or the service domain of the connected agent.

[0416] For a detailed explanation of the information regarding the connected agents, please refer to the relevant description of step S640 of method 600 above. For the sake of brevity, it will not be explained here.

[0417] Optionally, the connection agent discovery request / response message in step S704 above may also carry a session identifier (also known as the discovery request session identifier, or agent discovery request session identifier, i.e., the first identifier) ​​to identify that the connection agent discovery response message is a response to the connection agent discovery request message.

[0418] It should be noted that the specific implementation of the agent repository sending the connection agent information to the sys agent can be found in the relevant description of step S640 of method 600 above, and will not be described here for the sake of brevity.

[0419] In other words, the agent repository can feed back information about one or more connected agents to the sys-agent, meaning that this application does not limit the number of connected agents.

[0420] As an example, the computational and connectivity agents discovered above can form a collaborative group to jointly orchestrate the first task.

[0421] It should be noted that this application does not specify the execution order of the above steps S703 and S704, that is, this application does not limit the order in which the sys-agent requests the discovery of computational agents and the connection of agents from the agent repository.

[0422] Based on the above steps, after determining the computing resource application task and the computing session creation task, as well as the corresponding computing agent and connecting agent, the sys agent can request the computing agent to orchestrate the computing resource application task and request the connecting agent to orchestrate the computing session creation task. That is, the method also includes the following steps S705-S708. For details, please refer to the relevant descriptions of steps S650-S670 of the above method 600. For the sake of brevity, they will not be described here.

[0423] S705, sys - Agents interact with each other via the agent message bus, sending computing resource allocation task request / response messages.

[0424] S706, the intelligent agent message bus interacts with computational intelligent agents to send computing resource allocation task request / response messages;

[0425] Among them, the computing resource allocation task request message (i.e., the first request message) is used to request the computing agent to arrange the computing resource application task.

[0426] For example, the computing resource allocation task request message includes at least one of the following: description information of the computing resource allocation task (i.e., description information of the first task), or the output format of the arrangement result of the computing resource allocation task (i.e., the output format of the arrangement result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0427] Correspondingly, after the computational agent performs the task orchestration for requesting computing resources, it feeds back the task orchestration results to the sys agent. For example, the computational agent can forward the task orchestration results to the sys agent via a message bus. The content and specific interpretation of the task orchestration results can be found in the relevant description of step S660 of method 600 above; for brevity, it will not be elaborated here.

[0428] Understandably, in steps S705 and S706 above, the sys-agent and the computational agent send computing resource allocation task request / response messages through the agent message bus. Optionally, the sys-agent and the computational agent can achieve fast point-to-point communication, i.e., without the need for forwarding through the agent message bus, enabling efficient collaboration among multiple agents. In this case, the agent message bus can be understood as being coupled with the sys-agent and the computational agent, and this is not limited.

[0429] S707, sys - Agent-Agent message bus interaction computing power session creation task request / response messages;

[0430] S708, the intelligent agent message bus, interacts with connected intelligent agents to create task request / response messages for computing power sessions.

[0431] Among them, the computing power session creation task request message (i.e., the first request message) is used to request the connecting agent to orchestrate the computing power session creation task.

[0432] For example, the computing power session creation task request message includes at least one of the following: description information of the computing power session creation task (i.e., description information of the first task), or the output format of the orchestration result of the computing power session creation task (i.e., the output format of the orchestration result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0433] Correspondingly, after executing the task orchestration for creating the computing session, the connected agent feeds back the task orchestration results to the sys agent. For example, the connected agent can forward the task orchestration results to the sys agent via the message bus. The content and specific interpretation of the task orchestration results can be found in the relevant description of step S660 of method 600 above, and will not be elaborated here for the sake of brevity.

[0434] Understandably, in steps S707 and S708 above, the sys agent and the connected agent send computing session creation task request / response messages through the agent message bus. Optionally, the sys agent and the connected agent can achieve fast point-to-point communication, i.e., without the need for forwarding through the agent message bus, enabling efficient collaboration among multiple agents. In this case, the agent message bus can be understood as being coupled with the sys agent and the connected agent, and this is not limited.

[0435] It should be noted that this application does not specify the execution order of steps S705-S706 and steps S707-S708, that is, this application does not limit the order in which the sys-agent requests task orchestration from the computational agent and the connection agent.

[0436] Based on this, the sys-agent can receive task orchestration results from the computational agent and the connection agent respectively, and then generate orchestration instructions (e.g., second information) that can be recognized or parsed by the TCF or network function. Furthermore, the sys-agent can directly or indirectly invoke network services to execute the first task.

[0437] S709, the sys-agent sends orchestration instructions (i.e., the second information) to the TCF;

[0438] Correspondingly, the TCF receives orchestration instructions from the sys-agent.

[0439] The orchestration task is used by TCF to schedule network services, such as computing power services and session services, to perform computing power resource allocation tasks and computing power session creation tasks.

[0440] For example, the sys-agent calls the TCF's task orchestration interface to send orchestration instructions to the TCF, carrying interface information for calling computing power network elements and session network elements, such as calling the computing power resource allocation interface and calling the session creation interface.

[0441] For the specific implementation method, please refer to the relevant description of step S680 of method 600 above. For the sake of brevity, it will not be described here again.

[0442] S710, TCF calls the computing power resource allocation interface of the computing power network element to execute the computing power resource allocation task.

[0443] Correspondingly, after executing the computing resource allocation task, the computing network element reports the execution result of the computing resource allocation task to the TCF.

[0444] For example, the TCF parses the orchestration instructions and calls the computing power service's computing power resource allocation interface according to the orchestration instructions, passing resource request parameters such as computing, storage, and network to the computing power network element. After completing the allocation of computing power resources, the computing power network element returns the resource allocation result, which may include at least one of the following: API gateway address, software repository IP address, or remote computing power ID, etc.

[0445] S711, TCF calls the session creation interface of the session network element to execute the computing power session creation task.

[0446] Correspondingly, after executing the computing power session creation task, the session network element reports the execution result of the computing power session creation task to the TCF.

[0447] For example, the TCF calls the session creation interface of the session service according to the orchestration instructions. The parameters passed to the session network element include the session type (e.g., computing power session). After the session network element completes the creation of the computing power session, it returns the session creation result, which may include at least one of the following: API gateway address, software repository IP address, or session ID, etc.

[0448] For details on the implementation of steps S710-S711 above, please refer to the relevant description of step S690 in method 600 above. For the sake of brevity, it will not be described here again.

[0449] It should be noted that this application does not specify the execution order of the above steps S710 and S711, that is, this application does not limit the order of TCF scheduling computing power service and session service.

[0450] Understandably, steps S709-S711 above describe how the sys agent requests the TCF to call network services to perform computing resource allocation and computing session creation tasks, i.e., the sys agent indirectly calls network services to perform computing resource allocation and computing session creation tasks. Optionally, the sys agent can also directly call network services to perform computing resource allocation and computing session creation tasks. For specific implementation methods, please refer to the relevant description of step S601 in method 600 above. For the sake of brevity, it will not be described here.

[0451] S712, the session network element and the UE create a computing power session.

[0452] For specific implementation details, please refer to the relevant descriptions of existing session creation schemes. For the sake of brevity, they will not be described here.

[0453] S713, the session element sends a session creation success response message to the TCF;

[0454] Correspondingly, the session network element sends a session creation success response message to the TCF to indicate that the computing power session creation is complete.

[0455] That is, once the task orchestration and execution are completed, the UE can call upon network computing resources through the created computing session to realize the UE's intent.

[0456] Based on the above scheme, after the sys-agent obtains the UE's intent, it processes the UE's intent to determine the computing resource allocation task and computing session creation task used to realize the UE's intent. Then, it calls the computing agent to orchestrate the computing resource allocation task and the connection agent to orchestrate the computing session creation task, obtaining the task orchestration result. Further, the sys-agent issues an orchestration command to the TCF to trigger the TCF to schedule network services to execute the computing resource allocation task and computing session creation task, ultimately realizing the UE's intent. This method not only enables efficient collaboration and communication among multiple agents but also enables collaborative work between agents and mobile network services to realize the UE's intent and improve user experience. In addition, by introducing an agent repository, automatic agent registration and discovery functions can be realized, and by introducing an agent message bus, efficient collaboration among agents can be achieved.

[0457] Figure 8 is a schematic flowchart of a communication method 800 provided in this application. As shown in Figure 8, the method includes the following steps. Among them, steps S801-S802 describe the registration process of the intelligent agent, steps S803-S804 describe the deregistration process of the intelligent agent, and steps S805-S806 provide examples of the update registration process of the intelligent agent.

[0458] S801, the intelligent agent (e.g., the first intelligent agent and / or the second intelligent agent) sends an intelligent agent registration request message (e.g., the first registration request message or the second registration request message) to the intelligent agent repository (e.g., the third network element);

[0459] Correspondingly, the agent repository receives registration request messages from agents.

[0460] Specifically, the registration request message is used to request registration with a third network element. In other words, the registration request message is used to request the agent's information to be stored in the agent's repository.

[0461] For example, the registration request message includes at least one of the following: the agent's identification information, the agent's service domain information, the agent's deployment location information, the agent's role information, the agent's capability information, the protocol information supported by the agent, the open interface information supported by the agent, or the model information supported by the agent. For specific interpretations, please refer to the relevant description of the above method 600, which will not be explained here.

[0462] S802, Optionally, the agent repository sends a registration response message (e.g., a first registration response message or a second registration response message) to the agent;

[0463] Correspondingly, the agent receives a registration response message from the agent repository.

[0464] The registration response message indicates that the agent has successfully registered. In other words, the agent repository stores agent information. For example, as shown in Figure 5(d), the agent registration management module in the agent repository stores agent information in the agent information database for subsequent agent discovery.

[0465] S803, the agent sends a deregistration request message (e.g., a first deregistration request message or a second deregistration request message) to the agent repository;

[0466] Correspondingly, the agent repository receives deregistration request messages from agents.

[0467] The deregistration request message is used to request the deletion of information about the agent.

[0468] For example, the agent's identifier can be carried in the registration request message.

[0469] S804, Optionally, the agent repository sends a deregistration response message to the agent (e.g., a first deregistration response message or a second deregistration response message);

[0470] Correspondingly, the agent receives a deregistration response message from the agent repository.

[0471] The deregistration response message indicates that the agent has successfully registered. In other words, the agent repository deletes the agent's information; for example, as shown in Figure 5(d), the agent registration management module in the agent repository deletes the agent's information from the agent information database in the agent repository.

[0472] S805, the agent sends an update request message (e.g., a first update request message or a second update request message) to the agent repository;

[0473] Correspondingly, the agent repository receives update request messages from agents.

[0474] The update request message is used to request an update to the agent's information.

[0475] For example, the update request message includes at least one of the following: agent identification information, agent service domain information, agent deployment location information, agent role information, agent capability information, agent supported protocol information, agent supported open interface information, or agent supported model information.

[0476] It should be noted that the information carried in the update request message in step S805 is different from the information carried in the registration request message in step S801.

[0477] S806, Optionally, the agent repository sends an update response message (e.g., a first update response message or a second update response message) to the agent;

[0478] Correspondingly, the agent receives update response messages from the agent's repository.

[0479] The update response message indicates whether the agent's information update was successful or complete. In other words, the agent repository updates the agent's information, or the agent repository re-stores the agent's information.

[0480] As an example, the identifier of the intelligent agent in this application embodiment can be globally encoded using encoding methods such as fully qualified domain name (FQDN), and there is no limitation on this.

[0481] Based on the above scheme, after automatically going online, an agent can request registration from the agent repository; after going offline, it can request to re-register from the agent repository; and it can request to update its registration from the agent repository. In other words, it supports functions such as automatic online registration, capability registration, de-registration, and registration update for agents within the network, as well as functions for subsequent agent discovery. This solves the problem of network AI agent capability registration and automatic networking within mobile networks, and provides a basic multi-agent self-organizing network technology for AI-native mobile networks.

[0482] Figure 9 is a schematic flowchart of a communication method 900 provided in this application. As shown in Figure 9, the method includes the following steps.

[0483] S901, the sys agent sends an agent discovery request message (i.e., the second request message) to the agent repository;

[0484] Correspondingly, the agent repository receives agent discovery request messages from the sys-agent.

[0485] The agent discovery request message is used to request the discovery of an agent that supports the orchestration of the first task, such as a second agent.

[0486] For example, an agent discovery request message includes information describing at least one of the following characteristics of the requesting agent: role, supported open interfaces, or capabilities.

[0487] Optionally, the agent discovery request message may also include information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, or supported models.

[0488] For details regarding the specific meaning of the feature information carried in the agent discovery request message, and the specific implementation of the above steps, please refer to the relevant description of step S630 of method 600 above. For the sake of brevity, it will not be explained here.

[0489] S902, the agent repository sends an agent discovery response message (i.e., a third response message) to the sys-agent;

[0490] Correspondingly, the sys agent receives agent discovery response messages from the agent repository.

[0491] The agent discovery response message includes information about the second agent, the feature information carried in the information about the second agent and its specific meaning, and the specific implementation method of the above steps. For the relevant description of step S640 of the above method 600, it will not be explained here for the sake of brevity.

[0492] S903, the sys agent sends an agent interface call request message (i.e., the first request message) to the service agent (e.g., the second agent);

[0493] Correspondingly, the service agent receives agent interface call request messages from the sys-agent.

[0494] Among them, the agent interface call request message is used to request the orchestration of the first task.

[0495] For example, the agent interface call request message includes at least one of the following: an identifier of the first task, a description of the first task, or the output format of the orchestration result of the first task. For specific interpretations and implementation methods of the above steps, please refer to the relevant description of step S650 of method 600 above. For instance, the sys agent calls an open interface supported by the service agent. The interface functions include task orchestration, pausing tasks, or stopping tasks, and constructs the transmitted content according to the interaction protocol supported by the service agent, such as a task identifier and a task requirement description.

[0496] S904, the service agent sends the agent interface call result (i.e., the orchestration result of the first task) to the sys-agent;

[0497] Correspondingly, the sys agent receives the results of agent interface calls from the service agent.

[0498] For example, after receiving a task orchestration request, the service agent can invoke a local or remote AI model to perform inference and task orchestration, and obtain the agent interface call result. The agent interface call result includes interface information of the invoked network service (e.g., TCF), such as the interface of the invoked network service, the parameters passed, or the order in which the network services were invoked.

[0499] For details on the implementation of the above steps, please refer to the relevant descriptions of steps S660-S670 of method 600 above. For the sake of brevity, they will not be explained here.

[0500] Based on the above solution, agents can discover information about other agents that support task orchestration from the agent repository, solving the problem of mutual discovery between network AI agents within a mobile network. In other words, the agent repository supports mutual discovery functionality between agents within the network, facilitating efficient collaboration and communication among multiple agents and improving the user experience.

[0501] Understandably, the system architecture targeted by the above scheme includes multiple intelligent agents (e.g., a first intelligent agent and a second intelligent agent), which collaborate to understand the intent of the terminal device and, after service orchestration, invoke a first network element (e.g., a TCF) to execute orchestration instructions.

[0502] In the first scenario, in actual network deployment, the first and second agents can be integrated together, or a single agent can perform the functions of multiple agents (e.g., the first and second agents). For example, the first and second agents in the network architecture shown in Figure 5 can be replaced by a third agent (also called a service agent). This third agent supports understanding the intent of the terminal device, generating task orchestration results, and directly or indirectly scheduling network services to execute the task based on the task orchestration results.

[0503] In the second scenario, in actual network deployment, the second agent in the network architecture shown in Figure 5 can be replaced by a fourth agent. This fourth agent can perform the functions of the second agent and can also directly schedule network services to execute the task based on the task orchestration results.

[0504] The following sections will explain the specific implementation of the network architecture and communication method for Case 1, with reference to Figures 10-13. Additionally, the sections will explain the specific implementation of the network architecture and communication method for Case 2, with reference to Figures 14-17. For details not covered, please refer to the relevant descriptions in Figures 5-9 above; these will not be repeated here.

[0505] Figure 10 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 10, the architecture may include a third intelligent agent. This third intelligent agent supports understanding the intentions of terminal devices, generating orchestration instructions, and directly or indirectly invoking network services. The third intelligent agent can support various network services such as computing power services, session services, and perception services. For example, the third intelligent agent can be considered as an integrated / coupled intelligent agent obtained by combining / coupling the first intelligent agent (e.g., the system intelligent agent sys-agent) and the second intelligent agent (e.g., the service intelligent agent) described above; or, in other words, the third intelligent agent can possess the service capabilities of the first and second intelligent agents; or, in other words, the third intelligent agent can be considered as the sum of intelligent agents in different service domains. Specific interpretations of the first and second intelligent agents can be found in the relevant descriptions in Figure 5 above, and will not be repeated here.

[0506] Optionally, the architecture may also include a first network element (e.g., a task-centric function (TCF)), the specific meaning of which can be referred to the relevant description in Figure 5 above, and will not be explained here.

[0507] As an example, a third-party intelligent agent can directly invoke network services, such as access services, session services, computing power services, sensing services, or other services. For instance, after obtaining the intent of a terminal device, the third-party intelligent agent understands the intent, generates task orchestration results, and directly invokes different network elements (e.g., computing power network elements, session network elements, sensing network elements, etc.) to execute different network services.

[0508] As another example, a third intelligent agent can invoke network services through a first network element. For instance, after obtaining the intent of a terminal device, the third intelligent agent understands the intent, generates task orchestration results, and sends the task orchestration results to the first network element, which then invokes different network elements to execute different network services.

[0509] Optionally, the first network element in the above network architecture can be deployed independently as an independent network element.

[0510] It should be understood that the naming of network elements in the above network architecture is defined only for the convenience of distinguishing different functions and does not constitute a limitation on this application. This application does not exclude the possibility of using other names for various network elements in mobile network architecture and other future networks. For example, the above-mentioned agent storage repository can also be called agent storage node or agent information storage network element, etc., and there is no limitation in this regard.

[0511] It should also be understood that the network architecture shown in Figure 10 is an example given to understand the technical solution and does not constitute a limitation on the scope of protection of this application.

[0512] The following description, in conjunction with Figures 11 to 13, illustrates the communication method provided in the embodiments of this application, which can be applied to the network architecture shown in Figure 10. The embodiments of this application use a first network element (e.g., TCF) and a third intelligent agent (e.g., a service agent) as the execution subjects. The execution subjects, such as the first network element and the third intelligent agent, can be the first network element or the third intelligent agent itself, or components within the first network element or the third intelligent agent (e.g., processors, chips, or chip systems), or logical modules or software capable of implementing all or part of the functions.

[0513] Figure 11 is a schematic flowchart of a communication method 1100 provided in an embodiment of this application. As shown in Figure 11, the method includes the following steps. For details not covered herein, please refer to the relevant description in Figure 6 above. For the sake of brevity, these details will not be repeated here.

[0514] S1110, the third intelligent agent receives the first information.

[0515] The first information is used to describe the intent of the terminal device, or in other words, to request the implementation of the intent of the terminal device described by the first information.

[0516] For example, the intent could be that the terminal device requests computing resources, and the first information could be that the terminal device requests computing resources.

[0517] In one implementation, the first intelligent agent receives first information, including: the first intelligent agent receiving first information from a terminal device. The first information is carried in a Non-Access Stratum (NAS) message, which includes an intent element from the terminal device. That is, the terminal device indicates its intent by sending a NAS message carrying the intent element to the third intelligent agent.

[0518] The specific implementation of step S1110 can be referred to step S610 above, and will not be repeated here.

[0519] Optionally, the method 1100 may further include the following steps S1120-S1140. Steps S1120-S1130 and step S1140 can be executed selectively; that is, in one implementation, method 1100 includes steps S1120-S1130, whereby the third intelligent agent indirectly invokes the network service through the first network element; in another implementation, method 1100 includes step S1140, whereby the third intelligent agent directly invokes the network service.

[0520] S1120, the third intelligent agent sends the orchestration result of the first task to the first network element in the network based on the first information. Correspondingly, the first network element receives the orchestration result of the first task from the third intelligent agent.

[0521] The first task is used to realize the intent of the terminal device, and the orchestration result of the first task is used to schedule network services to execute the first task.

[0522] For example, the third intelligent agent determines the first task based on the first information, then performs task orchestration on the first task to obtain the orchestration result of the first task, and then sends the orchestration result of the first task to the first network element in the network. For example, the third intelligent agent can call a local or remote AI model to perform inference and task orchestration to obtain the orchestration result of the first task. For specific implementation methods, please refer to the relevant description of method 600 above, which will not be described here.

[0523] S1130, the first network element schedules the second network element in the network to perform the first task.

[0524] Understandably, the second network element is used to execute the first task; in other words, the second network element supports the execution of the first task, or the second network element supports the execution of the first task.

[0525] In one implementation, the first network element sends a request message to the second network element, which requests the execution of a first task. Correspondingly, after receiving the request message, the second network element executes the first task and obtains the execution result of the first task. Optionally, the second network element may send the execution result of the first task to the first network element.

[0526] The specific implementation of steps S1120-S1130 can be found in steps S680 and S690, and will not be elaborated further.

[0527] S1140, the third intelligent agent sends a request message to the second network element in the network based on the first information;

[0528] Correspondingly, the second network element in the network receives the request message from the third intelligent agent.

[0529] The request message is used to request the execution of a first task, which is used to fulfill the intent of the terminal device.

[0530] In one implementation, the third intelligent agent sends a request message to a second network element in the network, requesting the execution of a first task; correspondingly, the second network element receives the request message, executes the first task, and obtains the execution result of the first task. Optionally, the second network element may send the execution result of the first task to the third intelligent agent.

[0531] For example, the third intelligent agent determines the orchestration result of the first task based on the first information, and sends a request message to the second network element according to the orchestration result of the first task.

[0532] The specific implementation of step S1140 can be found in steps S1120 and S601, and will not be elaborated further.

[0533] The specific implementation of the second network element in the scheduling network to execute the first task, and the specific implementation of the second network element executing the first task, can be referred to the relevant description of method 600 above. For the sake of brevity, it will not be described here.

[0534] For ease of understanding, the specific process applicable to the embodiments of this application is described below with reference to Figures 12 and 13. In the examples below, the first network element is TCF, the third intelligent agent is the serving intelligent agent, the network service includes computing power service and session service, and the terminal device is UE. It should be understood that the process below is only an example, and the embodiments of this application are not limited thereto. The contents not described in detail below can be referred to the description in the method shown in Figure 11 or Figure 7, and will not be repeated below.

[0535] Figure 12 is a schematic flowchart of a communication method 1200 provided in this application. As shown in Figure 12, the method includes the following steps.

[0536] S1201, the UE sends a network computing resource request message to the serving agent;

[0537] Correspondingly, the service agent receives network computing resource request messages from the UE.

[0538] For example, the UE can carry its intention to request computing resources through a NAS message. The network computing resource request message describes the UE's intention; for example, it requests computing resources, or in other words, it requests the fulfillment of the UE's intention. Optionally, the network computing resource request message can be replaced with the UE's intention. A description of the UE's intention can be found in the description of method 600 above, and for brevity, it will not be elaborated further here.

[0539] S1202, Service agent generation and orchestration instructions.

[0540] For example, based on the acquired UE intent (e.g., requesting computing resources), the service agent uses LLM to understand the UE's intent and infers that the computing service domain and the connectivity service domain need to cooperate to complete the computing resource request task and the computing session creation task (i.e., the first task). For the specific implementation method, please refer to the relevant description of step S620 of the method 600 above. For the sake of brevity, it will not be described again.

[0541] S1203, the service agent sends an orchestration instruction to the TCF;

[0542] Correspondingly, the TCF receives orchestration instructions from the service agent.

[0543] The orchestration instructions are used by TCF to schedule network services, such as computing power services and session services, to perform computing power resource allocation tasks and computing power session creation tasks.

[0544] For example, the service agent calls the TCF's task orchestration interface to send orchestration instructions to the TCF, carrying interface information for calling computing power network elements and session network elements, such as calling the computing power resource allocation interface and the session creation interface. For specific implementation details, please refer to the relevant description of step S680 of method 600 above; for brevity, it will not be described further here.

[0545] S1204, TCF calls the computing power resource allocation interface of the computing power network element.

[0546] For example, the TCF parses orchestration instructions and calls the computing power service's computing power resource allocation interface according to the orchestration instructions to execute computing power resource allocation tasks. For example, the TCF transmits resource request parameters such as computing, storage, and network to the computing power network element. After completing the allocation of computing power resources, the computing power network element returns the resource allocation result, which may include at least one of the following: API gateway address, software repository IP address, or remote computing power ID, etc.

[0547] S1205, TCF calls the session creation interface of the session network element.

[0548] For example, the TCF parses the orchestration instructions and calls the session creation interface of the session service according to the orchestration instructions to perform the computing power session creation task. For example, the parameters passed by the TCF to the session network element include the session type (e.g., computing power session). After the session network element completes the computing power session creation, it returns the session creation result, which may include at least one of the following: API gateway address, software repository IP address, or session ID, etc.

[0549] S1206, the session network element and the UE create a computing power session.

[0550] For specific implementation details, please refer to the relevant descriptions of existing session creation schemes. For the sake of brevity, they will not be described here.

[0551] S1207, the session element indicates to the TCF that the session was successfully created.

[0552] For example, the session network element sends a session creation success response message to the TCF; correspondingly, the session network element sends a session creation success response message to the TCF. This session creation success response message indicates that the computing power session creation is complete. That is, task orchestration and execution are complete, and the UE can use the created computing power session to call upon network computing power resources to achieve the UE's intent.

[0553] Figure 13 is a schematic flowchart of a communication method 1300 provided in this application. Compared to the TCF calling the network service in Figure 12 above, the service agent in Figure 13 can directly call the network service. As shown in Figure 13, the method includes the following steps.

[0554] S1301, the UE sends a network computing resource request message to the serving agent;

[0555] Correspondingly, the service agent receives network computing resource request messages from the UE.

[0556] S1302, Service agent generation and orchestration instructions.

[0557] For the specific implementation of steps S1301-S1302 above, please refer to the relevant description of steps S1201-S1202 of method 1200 above, which will not be described here.

[0558] S1303, the service agent calls the computing power resource allocation interface of the computing power network element.

[0559] For example, the service agent parses orchestration instructions and calls the computing power service's computing power resource allocation interface according to the orchestration instructions to execute computing power resource allocation tasks. For example, the service agent passes resource request parameters such as computing, storage, and network to the computing power network element. After completing the allocation of computing power resources, the computing power network element returns the resource allocation result, which may include at least one of the following: API gateway address, software repository IP address, or remote computing power ID, etc.

[0560] S1304, the service agent calls the session creation interface of the session network element.

[0561] For example, the service agent parses the orchestration instructions and calls the session creation interface of the session service according to the orchestration instructions to perform the computing power session creation task. For example, the parameters passed by the service agent to the session network element include the session type (e.g., computing power session). After the session network element completes the computing power session creation, it returns the session creation result, which may include at least one of the following: API gateway address, software repository IP address, or session ID, etc.

[0562] S1305, the session network element and the UE create a computing power session.

[0563] S1306, the session element indicates to the TCF that the session was successfully created.

[0564] For the specific implementation of steps S1305-S1306 above, please refer to the relevant description of steps S1206-S1207 of method 1200 above. For the sake of brevity, it will not be described here again.

[0565] Based on the above scheme, after obtaining the UE's intent, the service agent can process the UE's intent to determine the computing resource allocation task and computing session creation task used to realize the UE's intent, and generate orchestration instructions. Then, it can directly or indirectly (e.g., through TCF) invoke network services to execute the computing resource allocation task and computing session creation task, ultimately realizing the UE's intent. In this method, the service agent can understand the UE's intent and generate orchestration instructions, and then directly or indirectly invoke network services. Compared to realizing the UE's intent through signaling interaction between multiple agents, this reduces signaling overhead and improves the user experience.

[0566] Figure 14 is a schematic diagram of a communication system applicable to an embodiment of this application. As shown in Figure 14, the architecture may include intelligent agents and third network elements (e.g., an agent repository). The intelligent agents in this architecture include a first intelligent agent (e.g., a system agent, sys-agent) and a fourth intelligent agent (e.g., a service agent). The first intelligent agent supports understanding the intentions of terminal devices, and the fourth intelligent agent can generate orchestration instructions and directly invoke network services. The fourth intelligent agent can support various network services such as computing power services, session services, and perception services. For example, the fourth intelligent agent can be viewed as an integrated agent of multiple service domains, or in other words, the fourth intelligent agent possesses the service capabilities of multiple intelligent agents; that is, the fourth intelligent agent supports task orchestration for tasks in multiple service domains and outputs task orchestration results. Furthermore, the fourth intelligent agent may include multiple intelligent agents, such as any two or three intelligent agents selected from connection agents, computing agents, or perception agents. For specific interpretations of the third network element, the first intelligent agent, and the fourth intelligent agent, please refer to the relevant description in Figure 5 above, which will not be elaborated here.

[0567] In this application, the fourth intelligent agent supports direct invocation of network services. For example, the fourth intelligent agent has an interface with a network element that provides network services (e.g., computing power network element, session network element, etc.), and the fourth intelligent agent can directly invoke the network element through this interface.

[0568] Optionally, the architecture may also include at least one of the following: a message bus (e.g., an agent message bus), or a terminal device, etc. For specific interpretations, please refer to the relevant description in Figure 5 above, which will not be explained here.

[0569] Optionally, the first network element in the above network architecture can be deployed independently as an independent network element, and the message bus can be an independent network element, such as an intelligent agent message router deployed independently, or it can be deployed as a functional module within the system intelligent agent and / or service intelligent agent, without limitation.

[0570] It should be understood that the naming of network elements in the above network architecture is defined only for the convenience of distinguishing different functions and does not constitute a limitation on this application. This application does not exclude the possibility of using other names for various network elements in mobile network architecture and other future networks. For example, the above-mentioned agent storage repository can also be called agent storage node or agent information storage network element, etc., and there is no limitation in this regard.

[0571] It should also be understood that the network architecture shown in Figure 14 is an example given to understand the technical solution and does not constitute a limitation on the scope of protection of this application.

[0572] The following description, in conjunction with Figures 15 to 17, illustrates the communication method provided in the embodiments of this application, which can be applied to the network architecture shown in Figure 14. The embodiments of this application use a first intelligent agent (e.g., a system agent) and a fourth intelligent agent (e.g., a service agent) as the execution subjects for illustrative purposes. The execution subjects, such as the first and fourth intelligent agents, can be the first and fourth intelligent agents themselves, or components within the first and fourth intelligent agents (e.g., processors, chips, or chip systems), or logical modules or software capable of implementing all or part of the functions.

[0573] Figure 15 is a schematic flowchart of a communication method 1500 provided in an embodiment of this application. As shown in Figure 15, the method includes the following steps. For details not covered herein, please refer to the relevant descriptions in existing solutions. For the sake of brevity, these details will not be elaborated here.

[0574] S1510, the first intelligent agent receives the first information.

[0575] The specific implementation method of the first intelligent agent receiving the first information, as well as the specific interpretation of the first information, can be referred to the relevant description of step S610 of the above method 600. For the sake of brevity, it will not be explained.

[0576] S1520, the first intelligent agent determines the first task based on the first information.

[0577] The first task is to realize the intent of the terminal device.

[0578] For the specific implementation method, please refer to the relevant description of step S620 of method 600 above. For the sake of brevity, it will not be explained further.

[0579] S1530, the first intelligent agent sends a task request message to the fourth intelligent agent;

[0580] Correspondingly, the fourth agent receives the task request message from the first agent;

[0581] The task request message is used to request the execution of the first task.

[0582] For example, the task request message includes a description of the first task and / or the output format of the orchestration result of the first task. For a detailed explanation, please refer to the relevant description of method 600 above, which will not be explained here.

[0583] S1540, the fourth intelligent agent sends a request message to the second network element in the network according to the task request message;

[0584] Correspondingly, the second network element in the network receives the request message from the fourth intelligent agent.

[0585] The request message is used to request the second network element to perform the first task.

[0586] As an example, the fourth agent orchestrates the first task according to the task request message, obtains the orchestration result of the first task, and then sends a request message to the second network element in the network according to the orchestration result of the first task.

[0587] The orchestration result is used to schedule network services to perform the first task, and the second network element is used to perform the first task.

[0588] For example, after receiving a task request message, the fourth intelligent agent can invoke a local or remote AI model to perform inference and task orchestration, obtaining the orchestration result of the first task. The orchestration result of the first task includes: interface information of the second network element in the network being invoked. Optionally, it may also include the type of the second network element in the network being invoked, and / or, information such as input parameters. It is understood that the second network element supports the execution of the first task, or in other words, the second network element is the network element that executes the first task; for example, the second network element could be a computing power network element, a session network element, or a sensing network element, etc.

[0589] Optionally, the second network element can send the execution result of the first task to the fourth intelligent agent.

[0590] The specific implementation of the second network element in the scheduling network to execute the first task, and the specific implementation of the second network element executing the first task, can be referred to the relevant description of method 600 above. For the sake of brevity, it will not be described here.

[0591] In this embodiment, the third network element supports functions such as capability registration and discovery of intelligent agents (e.g., the first or fourth intelligent agent) within the network. That is, after an intelligent agent (e.g., the first or fourth intelligent agent) goes online, it can request that its information be stored in the third network element for subsequent intelligent agent discovery. After an intelligent agent (e.g., the first or fourth intelligent agent) goes offline, it can request that its information be deleted from the third network element. Additionally, an intelligent agent (e.g., the first or fourth intelligent agent) can also request the third network element to update its information. Specific implementation details can be found in Figures 8 and 9 above; for brevity, they will not be described further here.

[0592] For ease of understanding, the specific process applicable to the embodiments of this application is described below with reference to Figures 16 and 17. In the examples below, the first intelligent agent is the system intelligent agent, the fourth intelligent agent is the service intelligent agent, the network service includes computing power service and session service, and the terminal device is the UE. It should be understood that the process below is only an example, and the embodiments of this application are not limited thereto. The contents not described in detail below can be referred to the description in the method shown in Figure 15, and will not be repeated hereafter.

[0593] Figure 16 is a schematic flowchart of a communication method 1600 provided in an embodiment of this application. As shown in Figure 16, the method includes the following steps. For details not covered herein, please refer to the relevant descriptions in existing solutions. For the sake of brevity, these details will not be elaborated here.

[0594] S1601, the UE sends a network computing resource request message to the system agent;

[0595] Correspondingly, the system's intelligent agent receives network computing resource request messages from the UE.

[0596] For a detailed explanation of the network computing resource request message, please refer to the relevant description of step S1201 of method 1200 above, which will not be explained here.

[0597] S1602, the system agent understands the UE's intent.

[0598] For example, based on the acquired UE intent (e.g., requesting computing resources), the system agent uses LLM to understand the UE intent and infers that the computing service domain and the connectivity service domain need to cooperate to complete the computing resource request task and the computing session creation task (i.e., the first task). For the specific implementation method, please refer to the relevant description of step S620 of the above method 600. For the sake of brevity, it will not be described again.

[0599] S1603, the system agent sends a task request message to the service agent;

[0600] Correspondingly, the service agent receives task request messages from the system agent.

[0601] Among them, the task request message is used to request the execution of computing resource application tasks and computing session creation tasks.

[0602] For example, the task request message includes at least one of the following: description information of the computing resource allocation task (i.e., description information of the first task), or the output format of the arrangement result of the computing resource allocation task (i.e., the output format of the arrangement result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0603] For example, the task request message includes at least one of the following: description information of the computing power session creation task (i.e., description information of the first task), or the output format of the orchestration result of the computing power session creation task (i.e., the output format of the orchestration result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0604] For specific implementation details, please refer to the relevant descriptions of steps S705-S708 of method 700 above, which will not be repeated here.

[0605] S1604, Service Agent Generation and Orchestration Instructions.

[0606] It is understandable that the above steps S1602-S1604 can be regarded as the specific implementation of step S1302 in the above method 1300.

[0607] S1605, the service agent calls the computing resource allocation interface of the computing network element.

[0608] S1606, the service agent calls the session creation interface of the session network element.

[0609] S1607, the session network element and the UE create a computing power session.

[0610] S1608, the session element indicates to the service agent that the session has been successfully created.

[0611] For the specific implementation of steps S605-S1608 above, please refer to the relevant descriptions of steps S1303-S1306 of method 1300 above, which will not be explained here.

[0612] Figure 17 is a schematic flowchart of a communication method 1700 provided in this application. As shown in Figure 17, taking a fourth intelligent agent including a connection intelligent agent and a computational intelligent agent as an example, the method includes the following steps.

[0613] S1701, the UE sends a network computing resource request message to the system agent;

[0614] Correspondingly, the system's intelligent agent receives network computing resource request messages from the UE.

[0615] For the specific implementation method, please refer to the relevant description of step S1601 of method 1600 above, which will not be explained here.

[0616] S1702, the system agent understands the intent of the terminal device and discovers the computational agent and the connection agent.

[0617] For specific implementation details, please refer to the relevant descriptions of steps S703-S704 of method 700 above, which will not be repeated here.

[0618] S1703, the system agent sends a request for computing resource allocation to the computing agent;

[0619] Correspondingly, the computational agent receives computing resource allocation task requests from the system agent.

[0620] Among them, the computing resource allocation task request is used to request the computing agent to arrange the computing resource request task.

[0621] For example, the computing resource allocation task request includes at least one of the following: description information of the computing resource allocation task (i.e., description information of the first task), or the output format of the arrangement result of the computing resource allocation task (i.e., the output format of the arrangement result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0622] S1704, the computing agent calls the computing resource allocation interface of the computing network element.

[0623] For example, the computing agent parses orchestration instructions and calls the computing power service's computing power resource allocation interface according to the orchestration instructions to execute computing power resource allocation tasks. For example, the service agent passes resource request parameters such as computing, storage, and network to the computing power network element. After completing the allocation of computing power resources, the computing power network element returns the resource allocation result, which may include at least one of the following: API gateway address, software repository IP address, or remote computing power ID, etc.

[0624] S1705, The computational agent sends a computational resource allocation task response to the system agent;

[0625] Correspondingly, the system agent receives the computing resource allocation task response from the computational agent.

[0626] S1706, The system agent sends a computing power session creation task request to the connected agent;

[0627] Correspondingly, the connecting agent receives a computing session creation task request from the system agent.

[0628] Among them, the computing power session creation task request message is used to request the connecting agent to orchestrate the computing power session creation task.

[0629] For example, the computing power session creation task request message includes at least one of the following: description information of the computing power session creation task (i.e., description information of the first task), or the output format of the orchestration result of the computing power session creation task (i.e., the output format of the orchestration result of the first task). For specific interpretation, please refer to the relevant description of step S650 of the method 600 above. For the sake of brevity, it will not be explained here.

[0630] S1707 connects the intelligent agent to the computing resource allocation interface of the session network element.

[0631] For example, the connection agent parses the orchestration instructions and calls the session creation interface of the session service according to the orchestration instructions to perform the computing power session creation task. For example, the parameters passed by the service agent to the session network element include the session type (e.g., computing power session). After the session network element completes the computing power session creation, it returns the session creation result, which may include at least one of the following: API gateway address, software repository IP address, or session ID, etc.

[0632] S1708, the session network element and the UE create a computing power session.

[0633] S1709, the session element indicates to the connected agent that the session has been successfully created.

[0634] S1710, The connected agent sends a computing power session creation task response to the system agent;

[0635] Correspondingly, the system agent receives the computing power session creation task response from the connected agent.

[0636] Based on the above scheme, after obtaining the UE's intent, the system agent can process the UE's intent to determine the computing resource allocation task and computing session creation task to realize the UE's intent, and request the service agent (e.g., including the computing agent and the connection agent) to execute the task. Then, the service agent generates orchestration instructions and directly calls network services to execute the computing resource allocation task and computing session creation task, ultimately realizing the UE's intent and improving the user experience.

[0637] The communication method embodiment of this application has been described in detail above with reference to Figures 1 to 17. The communication device embodiment of this application will now be described in detail with reference to Figures 18 to 20. It should be understood that the description of the device embodiment corresponds to the description of the method embodiment. Therefore, any parts not described in detail can be referred to the preceding method embodiment, and for the sake of brevity, will not be repeated here.

[0638] Figure 18 is a schematic diagram of a communication device 1000 provided in an embodiment of this application. As shown in Figure 18, the communication device 1000 includes a transceiver unit 1010 and a processing unit 1020. The transceiver unit 1010 can be used to implement corresponding communication functions. The transceiver unit 1010 can also be referred to as a communication interface or a communication unit. The processing unit 1020 can be used to perform processing.

[0639] Optionally, the device 1000 may further include a storage unit 1030, which can be used to store instructions and / or data, and the processing unit 1020 can read the instructions and / or data in the storage unit 1030 so that the device can implement the aforementioned method embodiments.

[0640] In a first possible design, the device 1000 can be the first intelligent agent in the aforementioned embodiments. The device 1000 can implement the steps or processes corresponding to those executed by the first intelligent agent in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations (such as sending and / or receiving data or messages) of the first intelligent agent in the above method embodiments, and the processing unit 1020 can be used to perform processing-related operations of the first intelligent agent in the above method embodiments, or operations other than transceiver operations (such as operations other than sending and / or receiving data or messages).

[0641] In one possible implementation, the transceiver unit 1010 is configured to receive first information, which describes the intent of the terminal device; the processing unit 1020 is configured to determine a first task based on the first information, which is used to implement the intent of the terminal device; the transceiver unit 1010 is further configured to send a first request message to a second intelligent agent, which requests the orchestration of the first task; the transceiver unit 1010 is further configured to receive the orchestration result of the first task from the second intelligent agent, which is used to schedule network services to execute the first task.

[0642] In one possible implementation, the transceiver unit 1010 is used to receive first information, which describes the intent of the terminal device; the processing unit 1020 is used to determine a first task based on the first information, which is used to implement the intent of the terminal device; the transceiver unit 1010 is also used to send a task request message to a fourth intelligent agent, which is used to request the execution of the first task.

[0643] In a second possible design, the device 1000 can be the second intelligent agent in the foregoing embodiments. The device 1000 can implement the steps or processes corresponding to those executed by the second intelligent agent in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations (such as sending and / or receiving data or messages) of the second intelligent agent in the above method embodiments, and the processing unit 1020 can be used to perform processing-related operations of the second intelligent agent in the above method embodiments, or operations other than transceiver operations (such as operations other than sending and / or receiving data or messages).

[0644] In one possible implementation, the transceiver unit 1010 is configured to receive a first request message from a first intelligent agent, the first request message being used to request the orchestration of a first task, the first task being used to implement the intent of a terminal device; the transceiver unit 1010 is also configured to send the orchestration result of the first task to the first intelligent agent, the orchestration result being used to schedule network services to execute the first task.

[0645] In a third possible design, the device 1000 can be the first network element in the aforementioned embodiments. The device 1000 can implement the steps or processes corresponding to the first network element in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations of the first network element in the above method embodiments (such as sending and / or receiving data or messages), and the processing unit 1020 can be used to perform processing-related operations of the first network element in the above method embodiments, or operations other than transceiver (such as operations other than sending and / or receiving data or messages).

[0646] One possible implementation is that the transceiver unit 1010 is used to receive second information from the first intelligent agent, the second information being determined based on the arrangement result of the first task, the first task being used to realize the intent of the terminal device; the processing unit 1020 is used to schedule the second network element in the network to execute the first task based on the second information.

[0647] In a fourth possible design, the device 1000 can be the third network element in the aforementioned embodiments. This device 1000 can implement the steps or processes corresponding to those performed by the third network element in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations of the third network element in the above method embodiments (such as sending and / or receiving data or messages), and the processing unit 1020 can be used to perform processing-related operations of the third network element in the above method embodiments, or operations other than transceiver (such as operations other than sending and / or receiving data or messages).

[0648] In one possible implementation, the transceiver unit 1010 is configured to receive a second request message from a first intelligent agent, the second request message being used to request the discovery of an intelligent agent that supports orchestration of a first task, the first task being used to implement the intent of a terminal device; the transceiver unit 1010 is also configured to send information about a second intelligent agent to the first intelligent agent, the second intelligent agent being used to orchestrate the first task.

[0649] In a fifth possible design, the device 1000 can be the third intelligent agent in the foregoing embodiments, which can implement the steps or processes corresponding to those executed by the third intelligent agent in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations (such as sending and / or receiving data or messages) of the third intelligent agent in the above method embodiments, and the processing unit 1020 can be used to perform processing-related operations of the third intelligent agent in the above method embodiments, or operations other than transceiver operations (such as operations other than sending and / or receiving data or messages).

[0650] In one possible implementation, the transceiver unit 1010 is used to receive first information, which describes the intent of the terminal device; the processing unit 1020 is used to determine a first task based on the first information, which is used to realize the intent of the terminal device; the processing unit 1020 is also used to perform task orchestration on the first task based on the first information to obtain the orchestration result of the first task, which is used to schedule network services to execute the first task, which is used to realize the intent of the terminal device.

[0651] In a sixth possible design, the device 1000 can be the fourth intelligent agent in the foregoing embodiments. The device 1000 can implement the steps or processes corresponding to those executed by the fourth intelligent agent in the above method embodiments. Specifically, the transceiver unit 1010 can be used to perform transceiver-related operations (such as sending and / or receiving data or messages) of the fourth intelligent agent in the above method embodiments, and the processing unit 1020 can be used to perform processing-related operations of the fourth intelligent agent in the above method embodiments, or operations other than transceiver operations (such as operations other than sending and / or receiving data or messages).

[0652] In one possible implementation, the transceiver unit 1010 is used to receive a task request message from a first intelligent agent, the task request message being used to request the execution of a first task, the first task being used to realize the intent of the terminal device; the communication unit 1010 is also used to send a request message to a second network element in the network according to the task request message, the request message being used to request the execution of the first task, the second network element being used to execute the first task.

[0653] It should be understood that the specific process of each unit performing the above-mentioned corresponding steps has been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0654] It should also be understood that the device 1000 here is embodied in the form of a functional unit. The term "unit" here can refer to an application-specific integrated circuit (ASIC), electronic circuitry, a processor (e.g., a shared processor, a proprietary processor, or a group processor, etc.) and memory for executing one or more software or firmware programs, integrated logic circuitry, and / or other suitable components supporting the described functions. In an alternative example, those skilled in the art will understand that the device 1000 can be specifically the communication device in the above embodiments, and can be used to execute the various processes and / or steps corresponding to the communication device in the above method embodiments; to avoid repetition, these will not be described again here.

[0655] The apparatus 1000 of each of the above-described schemes has the function of implementing the corresponding steps performed by the communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the above-described methods. The functions can be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions; for example, a transceiver unit can be replaced by a transceiver (e.g., the transmitting unit in the transceiver unit can be replaced by a transmitter, and the receiving unit in the transceiver unit can be replaced by a receiver), and other units, such as processing units, can be replaced by processors, each executing the transceiver operations and related processing operations in the respective method embodiments.

[0656] In addition, the transceiver unit 1010 may also be a transceiver circuit (for example, it may include a receiving circuit and a transmitting circuit), and the processing unit may be a processing circuit.

[0657] It should be noted that the device in Figure 18 can be the communication device in the aforementioned embodiments (such as the first intelligent agent, the second intelligent agent, the first network element, or the third network element), or it can be a chip or a chip system, such as a system on a chip (SoC). The transceiver unit can be an input / output circuit or a communication interface; the processing unit is a processor, microprocessor, or integrated circuit integrated on the chip, and is not limited thereto.

[0658] Figure 19 is a schematic diagram of another communication device 1100 provided in an embodiment of this application. As shown in Figure 19, the device 1100 includes a processor 1110 and a transceiver 1130. The processor 1110 is coupled to a memory 1120, which is used to store computer programs or instructions and / or data. The processor 1110 is used to execute the computer programs or instructions stored in the memory 1120, or to read the data stored in the memory 1120, in order to execute the methods in the above method embodiments.

[0659] Optionally, there may be one or more processors 1110.

[0660] Optionally, the memory 1120 may be one or more.

[0661] Alternatively, the memory 1120 can be integrated with the processor 1110, or it can be set separately.

[0662] Optionally, transceiver 1130 is used for receiving and / or transmitting signals. For example, processor 1110 is used to control transceiver 1130 to receive and / or transmit signals.

[0663] As an example, processor 1110 may have the functions of processing unit 1010 shown in FIG18, memory 1120 may have the functions of storage unit 1030, and transceiver 1130 may have the functions of transceiver unit 1010 shown in FIG18.

[0664] As one approach, the device 1100 is used to implement the operations performed by the communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the various method embodiments described above.

[0665] For example, processor 1110 is used to execute computer programs or instructions stored in memory 1120 to implement the relevant operations of the communication device in the various method embodiments described above.

[0666] It should be understood that the processor mentioned in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0667] It should also be understood that the memory mentioned in the embodiments of this application can be volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM). For example, RAM can be used as an external cache. By way of example and not limitation, RAM includes the following forms: static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0668] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) can be integrated into the processor.

[0669] It should also be noted that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0670] Figure 20 is a schematic diagram of a chip system 1200 provided in an embodiment of this application. As shown in Figure 20, the chip system 1200 (or processing system) includes logic circuitry 1210 and an input / output interface 1220.

[0671] The logic circuit 1210 can be a processing circuit in the chip system 1200. The logic circuit 1210 can be coupled to a memory unit, calling instructions from the memory unit, enabling the chip system 1200 to implement the methods and functions of the embodiments of this application. The input / output interface 1220 can be an input / output circuit in the chip system 1200, outputting processed information from the chip system 1200, or inputting data or signaling information to be processed into the chip system 1200 for processing.

[0672] As one approach, the chip system 1200 is used to implement operations performed by communication devices (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the various method embodiments described above.

[0673] For example, logic circuit 1210 is used to implement processing-related operations performed by communication devices (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the above method embodiments; input / output interface 1220 is used to implement sending and / or receiving-related operations performed by communication devices (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the above method embodiments.

[0674] This application also provides a computer-readable storage medium storing a computer program or instructions for implementing the methods executed by a communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) in the above-described method embodiments. For example, when the computer program or instructions are run on the communication device, they cause the communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) to execute the above-described methods.

[0675] This application also provides a computer program product comprising instructions that, when executed by a computer, implement the methods described above, which are performed by a communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element). For example, when the computer program or instructions are run on the communication device, the communication device (such as a first intelligent agent, a second intelligent agent, a first network element, or a third network element) performs the methods described above.

[0676] This application also provides a communication system, which includes at least one of the first intelligent agent, second intelligent agent, first network element, or third network element in the above embodiments.

[0677] The explanations and beneficial effects of the relevant contents in any of the devices provided above can be found in the corresponding method embodiments provided above, and will not be repeated here.

[0678] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed between each other may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0679] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. For example, the computer can be a personal computer, a server, or a network device, etc. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media, or semiconductor media (e.g., solid-state disks, SSDs). For example, the aforementioned available media include, but are not limited to, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, and other media capable of storing program code.

[0680] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

A communication method, characterized in that, Applied to the first intelligent agent, including: Receive first information, which describes the intent of the terminal device; Based on the first information, a first task is determined, and the first task is used to realize the intent of the terminal device; Send a first request message to the second intelligent agent, the first request message being used to request the orchestration of the first task; The first task is received from the second intelligent agent, and the orchestration result is used to schedule network services to perform the first task. The method according to claim 1, characterized in that, The method further includes: Send a second message to a first network element in the network, the second message being determined based on the orchestration result of the first task. The method according to claim 1 or 2, characterized in that, The method further includes: The orchestration result of the first task is sent to a second network element in the network, and the second network element is used to execute the first task. The method according to any one of claims 1 to 3 is characterized in that, Sending the first request message to the second intelligent agent includes: The first request message is sent to the second intelligent agent via the message bus. The method according to claim 4, characterized in that, Sending the first request message to the second intelligent agent via the message bus includes: The message bus sends the first request message and information about the second agent. The information about the second agent includes the identifier and / or the address of the second agent. The information about the second agent is used by the message bus to send the first request message to the second agent. The method according to any one of claims 1 to 5, characterized in that, The method further includes: Send a second request message to the third network element, the second request message being used to request the discovery of an intelligent agent that supports the orchestration of the first task; The second intelligent agent receives information from the third network element, and the second intelligent agent supports the orchestration of the first task; Sending the first request message to the second intelligent agent includes: Based on the information from the second intelligent agent, the first request message is sent to the second intelligent agent. The method according to claim 6, characterized in that, The receiving of information from the second intelligent agent from the third network element includes: Receive information from multiple intelligent agents of the third network element, the multiple intelligent agents including the second intelligent agent; The method further includes: determining the second intelligent agent from the plurality of intelligent agents based on the information of the plurality of intelligent agents. The method according to claim 7, characterized in that, The information of the multiple intelligent agents includes at least one of the following: the deployment location of the multiple intelligent agents, the protocols supported by the multiple intelligent agents, the open interfaces supported by the multiple intelligent agents, the models supported by the multiple intelligent agents, or the capabilities of the multiple intelligent agents. The method according to any one of claims 6 to 8, characterized in that, The second request message includes information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, role, or capability. The method according to any one of claims 1 to 9, characterized in that, The method further includes: A first registration request message is sent to the third network element, the first registration request message being used to request registration with the third network element. The method according to claim 10, characterized in that, The first registration request message includes at least one of the following: The identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent. The method according to any one of claims 1 to 11, characterized in that, The first request message includes a description of the first task and / or the output format of the orchestration result of the first task. A communication method, characterized in that, Applied to second intelligent agents, including: Receive a first request message from a first intelligent agent, the first request message being used to request the orchestration of a first task, the first task being used to implement the intent of a terminal device; The orchestration result of the first task is sent to the first intelligent agent, and the orchestration result is used to schedule network services to execute the first task. The method according to claim 13, characterized in that, The method further includes: Based on the first request message, determine the orchestration result of the first task. The method according to claim 13 or 14 is characterized in that, Receiving the first request message from the first intelligent agent includes: The first request message is received from the first intelligent agent via the message bus. The method according to any one of claims 13 to 15, characterized in that, The method further includes: A second registration request message is sent to the third network element, the second registration request message being used to request registration with the third network element. The method according to claim 16, characterized in that, The second registration request message includes at least one of the following: The identification information of the second intelligent agent, the service domain information of the second intelligent agent, the deployment location information of the second intelligent agent, the role information of the second intelligent agent, the capability information of the second intelligent agent, the protocol information supported by the second intelligent agent, the open interface information supported by the second intelligent agent, or the model information supported by the second intelligent agent. A communication method, characterized in that, The first network element applied in the network includes: Receive second information from the first intelligent agent, the second information being determined based on the orchestration result of the first task, the first task being used to realize the intent of the terminal device; Based on the second information, the second network element in the network is scheduled to execute the first task. The method according to claim 18, characterized in that, Before receiving the second information from the first intelligent agent, the method further includes: The intention of receiving the terminal device; Send first information to the first intelligent agent, the first information being used to describe the intent of the terminal device. The method according to claim 18 or 19, characterized in that, The scheduling of the second network element in the network to execute the first task includes: A third request message is sent to the second network element, the third request message being used to request the execution of the first task. A communication method, characterized in that, Applied to third-party network elements, including: Receive a second request message from a first intelligent agent, the second request message being used to request the discovery of an intelligent agent that supports orchestration of a first task, the first task being used to implement the intent of a terminal device; The first agent sends information about the second agent, which supports orchestrating the first task. The method according to claim 21, characterized in that, Sending the information of the second agent to the first agent includes: Send information about multiple agents to the first agent, the multiple agents including the second agent. The method according to claim 21 or 22 is characterized in that, The second request message includes information describing at least one of the following characteristics of the requesting agent: deployment location, service domain, supported protocols, role, or capability. The method according to any one of claims 21 to 23 is characterized in that, The method further includes: Receive a first registration request message from the first intelligent agent, the first registration request message being used to request registration with the third network element; and / or, A second registration request message is received from the second intelligent agent, the second registration request message being used to request registration with the third network element. The method according to any one of claims 21 to 24 is characterized in that, The method further includes: Receive a first deregistration request message from the first agent, the first deregistration request message being used to request the deletion of the first agent's information; and / or, Receive a second deregistration request message from the second agent, the second deregistration request message being used to request the deletion of information of the second agent. The method according to any one of claims 21 to 25 is characterized in that, The method further includes: Receive a first update request message from the first agent, the first update request message being used to request an update to the information of the first agent; and / or, Receive a second update request message from the second agent, the second update request message being used to request an update to the information of the second agent. A communication method, characterized in that, Applied to third-party intelligent agents, including: Receive first information, which describes the intent of the terminal device; Based on the first information, the first task is orchestrated to obtain the orchestration result of the first task. The orchestration result is used to schedule network services to execute the first task, and the first task is used to realize the intent of the terminal device. The method according to claim 27, characterized in that, The method further includes: The orchestration result of the first task is sent to the first network element in the network. The method according to claim 27, characterized in that, The method further includes: Based on the scheduling result of the first task, a request message is sent to the second network element in the network, the request message being used to request the second network element to execute the first task. A communication method, characterized in that, Applied to the first intelligent agent, including: Receive first information, which describes the intent of the terminal device; Based on the first information, a first task is determined, and the first task is used to realize the intent of the terminal device; A task request message is sent to the fourth intelligent agent, the task request message being used to request the execution of the first task. The method according to claim 30, characterized in that, The method further includes: Send a discovery request message to a third network element, the discovery request message being used to request the discovery of an intelligent agent that supports orchestration of the first task; The fourth intelligent agent receives information from the third network element, and the fourth intelligent agent supports the orchestration of the first task; Sending the task request message to the fourth intelligent agent includes: Based on the information from the fourth intelligent agent, the task request message is sent to the fourth intelligent agent. The method according to claim 31, characterized in that, The receiving of information from the third network element by the fourth intelligent agent includes: Receive information from multiple intelligent agents from the third network element, the multiple intelligent agents including the fourth intelligent agent; The method further includes: determining the fourth intelligent agent from among the plurality of intelligent agents based on the information of the plurality of intelligent agents. The method according to claim 32, characterized in that, The information of the multiple intelligent agents includes at least one of the following: the deployment location of the multiple intelligent agents, the protocols supported by the multiple intelligent agents, the open interfaces supported by the multiple intelligent agents, the models supported by the multiple intelligent agents, or the capabilities of the multiple intelligent agents. The method according to any one of claims 30 to 33, characterized in that, The method further includes: A first registration request message is sent to the third network element, the first registration request message being used to request registration with the third network element. The method according to claim 34, characterized in that, The first registration request message includes at least one of the following: The identification information of the first intelligent agent, the service domain information of the first intelligent agent, the deployment location information of the first intelligent agent, the role information of the first intelligent agent, the capability information of the first intelligent agent, the protocol information supported by the first intelligent agent, the open interface information supported by the first intelligent agent, or the model information supported by the first intelligent agent. A communication method, characterized in that, Applied to the fourth intelligent agent, including: Receive a task request message from a first intelligent agent, the task request message being used to request the execution of a first task, the first task being used to realize the intent of the terminal device; According to the task request message, a request message is sent to a second network element in the network. The request message is used to request the execution of the first task, and the second network element is used to execute the first task. The method according to claim 36, characterized in that, Receiving the task request message from the first intelligent agent includes: The task request message is received from the first intelligent agent via the message bus. The method according to claim 36 or 37, characterized in that, The method further includes: A second registration request message is sent to the third network element, the second registration request message being used to request registration with the third network element. The method according to claim 38, characterized in that, The second registration request message includes at least one of the following: The identification information of the fourth intelligent agent, the service domain information of the fourth intelligent agent, the deployment location information of the fourth intelligent agent, the role information of the fourth intelligent agent, the capability information of the fourth intelligent agent, the protocol information supported by the fourth intelligent agent, the open interface information supported by the fourth intelligent agent, or the model information supported by the fourth intelligent agent. A communication device, characterized in that, It includes modules or units for performing the method as described in any one of claims 1 to 12, or modules or units for performing the method as described in any one of claims 13 to 17, or modules or units for performing the method as described in any one of claims 18 to 20, or modules or units for performing the method as described in any one of claims 21 to 26, or modules or units for performing the method as described in any one of claims 27 to 29, or modules or units for performing the method as described in any one of claims 30 to 35, or modules or units for performing the method as described in any one of claims 36 to 39. A communication device, characterized in that, The method includes at least one processor for executing a computer program or instructions to cause the method as described in any one of claims 1 to 26 to be executed, or to cause the method as described in any one of claims 27 to 29 to be executed, or to cause the method as described in any one of claims 30 to 39 to be executed. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program or instructions that, when executed by a processor, implement the method as described in any one of claims 1 to 26, or the method as described in any one of claims 27 to 29, or the method as described in any one of claims 30 to 39. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the method as described in any one of claims 1 to 26, or implement the method as described in any one of claims 27 to 29, or implement the method as described in any one of claims 30 to 39.

Images