Authentication and Authorization Methods and Systems for Intelligent Agents

By performing dual authentication of agents and users and operator policy verification on the network device side, the security and reliability issues in agent applications are resolved, enabling secure access for agents and differentiated traffic protection, thereby improving the security and interactive experience of agent applications.

CN123122872APending Publication Date: 2026-11-17INSPUR TIANYUAN COMM INFORMATION SYST CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611103361.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-23
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, mobile communication networks only provide services through terminal authentication mechanisms, which makes it difficult to guarantee the security, reliability, and interactive experience of intelligent agent applications. There are risks of intelligent agent impersonation attacks and unauthorized operations, and there is a lack of differentiated protection for intelligent agent interaction traffic.

Method used

The network device receives service access requests from agent identifiers and user identifiers, performs dual identity verification, including authentication of the target agent and associated user, verifies processing permissions according to operator policies, and establishes differentiated data transmission channels to support secure access and network service invocation for agents.

Benefits of technology

It effectively reduces the risk of agents being impersonated, improves the security, reliability and interactive experience of agent applications, and achieves differentiated traffic protection for agents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122872A_ABST
    Figure CN123122872A_ABST
Patent Text Reader

Abstract

This invention provides an authentication and authorization method and system for intelligent agents. The method includes: receiving a service access request transmitted by a terminal device; authenticating a target intelligent agent based on an intelligent agent identifier in the service access request; if the target intelligent agent passes authentication, authenticating an associated user based on a user identifier in the service access request; if the associated user passes authentication, and if the target intelligent agent is confirmed to have processing authority for the target service according to the target operator's policy, then establishing a target data transmission channel according to the target operator's policy, and providing the target service to the target intelligent agent based on the target data transmission channel. This invention achieves dual identity verification for the target intelligent agent and its associated users, as well as access control based on operator policies, improving the security, reliability, and interactive experience of intelligent agent applications.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of communication security technology, and in particular to an authentication and authorization method and system for intelligent agents. Background Technology

[0002] With the deep integration of artificial intelligence technology and communication networks, intelligent agents are widely used in scenarios such as intelligent service scheduling because of their ability to interpret user intentions and autonomously execute operations across devices.

[0003] Existing mobile communication networks primarily provide services to users through terminal authentication mechanisms. Under this mechanism, only authentication and authorization of user devices to access the home network and localized services are supported. This is to verify the identity and legitimacy of the terminal device to grant network access permissions. This can easily lead to other users using intelligent agents to carry out impersonation attacks or unauthorized operations caused by the intelligent agent's misinterpretation of the user's intent, threatening network and user security and making it difficult to guarantee the security, reliability, and interactive experience of intelligent agent applications. Summary of the Invention

[0004] This invention provides an authentication and authorization method and system for intelligent agents, which solves the shortcomings of existing technologies that only support authentication and authorization for user devices to access home networks and localized services, making it difficult to guarantee the security, reliability, and interactive experience of intelligent agent applications, thereby effectively improving the security, reliability, and interactive experience of intelligent agent applications.

[0005] This invention provides an authentication and authorization method for intelligent agents, applied to network devices, the method comprising: Receive a service access request transmitted by a terminal device; the service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent, the target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent; The target intelligent agent is authenticated based on the intelligent agent identifier; If the target intelligent agent is determined to be authenticated, the associated user is authenticated based on the user identifier; If the associated user is authenticated, the system verifies whether the target intelligent agent has the authority to process the target service requested by the service access request, based on the target operator policy corresponding to the target intelligent agent. If it is determined that the target agent has the authority to process the target service, a target data transmission channel is established according to the target operator's policy, and the target service is provided to the target agent based on the target data transmission channel.

[0006] According to the present invention, an authentication and authorization method for an intelligent agent includes authenticating the target intelligent agent based on the intelligent agent identifier, comprising: Based on the agent identifier, the target authentication information corresponding to the target agent is searched in the pre-stored set of authorized agent information; the target authentication information includes registration status information, authentication credential information, and device binding information. If the target authentication information is found, the system verifies whether the target intelligent agent has completed registration based on the registration status information, verifies whether the authentication credential of the target intelligent agent is valid based on the authentication credential information, and verifies whether the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information. If the target intelligent agent has completed registration, the authentication credentials of the target intelligent agent are valid, and the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information, then the target intelligent agent is determined to have passed authentication.

[0007] According to the present invention, an authentication and authorization method for an intelligent agent includes authenticating the associated user based on the user identifier, comprising: Obtain the real-time identity credential information of the associated user and the real-time device feature information of the terminal device; Based on the user identifier and the agent identifier, the target user information corresponding to the target agent is searched in the pre-stored authorized agent information set; the target user information includes legitimate identity credential information and legitimate device feature information; If the target user information is found, the consistency between the real-time identity credential information and the legitimate identity credential information is verified to obtain the first verification result. The consistency between the real-time device feature information and the legitimate device feature information is verified to obtain a second verification result; If both the first and second verification results are successful, the associated user is determined to have passed authentication.

[0008] According to the present invention, an authentication and authorization method for an intelligent agent includes verifying whether the target intelligent agent has the authority to process the target service requested by the service access request, based on the target operator policy corresponding to the target intelligent agent. Based on the agent identifier, the target operator policy corresponding to the target agent is searched in the pre-stored set of authorized agent information; If the target operator's policy is found, the resource access scope and network service call permissions of the target intelligent agent are obtained from the target operator's policy. Based on the resource access scope, verify whether the target agent has resource access permissions to the target service; Based on the network service invocation permissions, verify whether the target intelligent agent has the invocation permissions for the network service corresponding to the target service; If the target intelligent agent has the resource access permission and the calling permission, it is determined that the target intelligent agent has the processing permission for the target service.

[0009] According to the present invention, an authentication and authorization method for an intelligent agent, wherein establishing a target data transmission channel based on the target operator's policy includes: Obtain the traffic differentiation strategy from the target operator's strategy; If the initiator of the service access request is determined to be the target intelligent agent according to the traffic differentiation strategy, the traffic configuration characteristics corresponding to the target intelligent agent are determined according to the traffic differentiation strategy. The target data transmission channel is established based on the traffic configuration characteristics corresponding to the target intelligent agent.

[0010] According to the present invention, an authentication and authorization method for an intelligent agent, wherein establishing the target data transmission channel based on the traffic configuration characteristics corresponding to the target intelligent agent includes: Assign quality of service parameters to the service access request that are adapted to the traffic configuration features corresponding to the target agent; The target data transmission channel is established based on the service quality parameters.

[0011] According to the present invention, an authentication and authorization method for an intelligent agent is provided, wherein the target data transmission channel is used to transmit multimodal data according to the quality of service parameters; The multimodal data includes multiple data sources such as image data, audio data, and real-time interactive data required for visual rendering. The service quality parameters include a latency time less than a preset latency time threshold, a bandwidth higher than a preset bandwidth threshold, and a reliability value higher than a preset reliability threshold.

[0012] According to the present invention, an authentication and authorization method for an intelligent agent further includes: If the target agent fails to authenticate, the associated user fails to authenticate, or the target agent does not have the authority to process the target service, the service access request is rejected, and the abnormal behavior of the target agent is recorded. If, based on the abnormal behavior records, it is determined that the number of abnormal behaviors generated by the target intelligent agent within a preset time period reaches a preset threshold, an abnormal handling operation is performed. The exception handling operation includes at least one of the following operations: Trigger a security alert for the target intelligent agent; Reduce the service quality corresponding to the target intelligent agent; A re-authentication request is sent to the associated user, which instructs the associated user to verify the abnormal behavior of the target intelligent agent.

[0013] The present invention also provides an authentication and authorization system for intelligent agents, comprising: A request receiving module is used to receive a service access request transmitted by a terminal device; the service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent, the target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent; The agent authentication module is used to authenticate the target agent based on the agent identifier; The user authentication module is used to authenticate the associated user based on the user identifier when it is determined that the target intelligent agent has passed the authentication. The authorization verification module is used to verify, in the case that the associated user has passed the authentication, whether the target intelligent agent has the processing authority for the target service requested by the service access request, according to the target operator policy corresponding to the target intelligent agent. The service response module is used to establish a target data transmission channel according to the target operator's policy, and provide the target service to the target intelligent agent based on the target data transmission channel, when it is determined that the target intelligent agent has the processing authority for the target service.

[0014] According to the present invention, an authentication and authorization system for an intelligent agent further includes a pre-configuration module and a security protection module. The pre-configuration module is used to pre-store a set of authorized intelligent agent information; the set of authorized intelligent agent information includes the authentication information, user information and operator policies of each authorized intelligent agent; The security protection module is configured to reject the service access request and record the abnormal behavior of the target intelligent agent when the target intelligent agent fails authentication, the associated user fails authentication, or the target intelligent agent does not have the authority to process the target service; and to perform an anomaly handling operation when the number of abnormal behaviors generated by the target intelligent agent within a preset time period reaches a preset threshold based on the abnormal behavior records; the anomaly handling operation includes at least one of the following operations: triggering a security alarm for the target intelligent agent; reducing the service quality corresponding to the target intelligent agent; and sending a re-authentication request to the associated user, wherein the re-authentication request is used to request the associated user to verify the abnormal behavior of the target intelligent agent.

[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the authentication and authorization method for the intelligent agent as described above.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the authentication and authorization method for an intelligent agent as described above.

[0017] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the authentication and authorization method for an intelligent agent as described above.

[0018] The authentication and authorization method and system for intelligent agents provided by this invention receive service access requests carrying intelligent agent identifiers and user identifiers on the network device side, and sequentially authenticate the target intelligent agent, authenticate the associated user, and verify processing permissions according to the target operator's policies. Finally, after successful verification, a target data transmission channel is established and the target service is provided. This achieves dual identity verification for the target intelligent agent and its associated user, as well as permission control based on operator policies. It can distinguish whether the initiator of the service access request is the terminal device itself or the target intelligent agent running on the terminal device, and it can also support secure access and network service calls for third-party intelligent agents. This effectively reduces the risk of impersonation attacks caused by stolen credentials and device intrusion. Furthermore, it can carry differentiated services for target intelligent agents that have passed authentication and authorization verification, so that the data transmission guarantee obtained by the target intelligent agent can be adapted to the traffic characteristics of the target intelligent agent, thereby improving the security, reliability, and interactive experience of intelligent agent applications. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the authentication and authorization method for intelligent agents provided by the present invention.

[0021] Figure 2 This is a schematic diagram of the intelligent agent authentication and authorization system provided by the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] All actions involving the acquisition of signal information or data in this application are carried out in accordance with the relevant data protection laws and policies of the country where the application is located, and with the authorization granted by the owner of the relevant device.

[0025] With the deep integration of artificial intelligence technology and communication networks, intelligent agents in the 6th Generation Mobile Communication Technology (6G) network are widely used in many scenarios due to their ability to interpret user intentions and autonomously perform operations across devices.

[0026] However, in the current mobile communication system, the authentication system represented by the 5th Generation Mobile Communication Technology (5G) mainly focuses on the authentication and authorization between the terminal device (User Equipment, UE) and the home network. That is, it only supports the authentication and authorization of the user equipment to access the home network and localized services, in order to verify the legitimacy of the terminal device's identity and grant network access permissions.

[0027] When dealing with the growing demand for artificial intelligence (AI) applications, this mechanism only verifies the identity and legitimacy of the terminal device, which makes it impossible for the network side to effectively distinguish whether the service access request comes from the terminal device's native system or from an intelligent agent deployed on the device, and it lacks a dedicated access control mechanism for intelligent agents.

[0028] However, while intelligent agents possess the ability to interpret user intent and autonomously execute operations across devices, their operational logic presents a dual risk in actual deployment: malicious users may exploit intelligent agents to carry out impersonation attacks; and misinterpretation of user intent by intelligent agents may lead to unauthorized and unauthorized operations.

[0029] When intelligent agents are deployed on user devices or third-party devices, existing technologies make it difficult to verify the legitimacy of the intelligent agent's identity and the real control of the associated user. As a result, there are security risks such as abuse of authority and impersonation attacks, which cannot meet the security needs of large-scale and diversified intelligent agent applications in future 6G networks.

[0030] Furthermore, existing resource allocation and traffic scheduling logic typically treats all uplink traffic as homogeneous data, lacking the ability to differentiate and guarantee traffic specific to intelligent agents.

[0031] In summary, in complex application scenarios, existing technologies have technical shortcomings that cannot simultaneously ensure the security, reliability, and real-time interactive experience of intelligent agent applications.

[0032] Therefore, there is an urgent need for an authentication and authorization scheme that can achieve dual authentication between the agent and the associated user, support secure access for the agent, and ensure differentiated protection of the agent's unique interactive traffic.

[0033] In response, this application provides an authentication and authorization method for intelligent agents. This method achieves dual identity verification and permission control for intelligent agents and associated communication network users through a multi-dimensional authentication and authorization process involving intelligent agent authentication, communication network user authentication, and service processing permission review. It also supports secure access and network service invocation for third-party intelligent agents and ensures the differentiation of interaction traffic unique to intelligent agents. This ensures that only legally authenticated intelligent agents and communication network users access network resources and subscribe to services, significantly reducing the risk of impersonation attacks and improving the security, reliability, and interactive experience of intelligent agent applications in future 6G networks.

[0034] Figure 1 This is a flowchart illustrating the authentication and authorization method for intelligent agents provided by the present invention.

[0035] It should be noted that the method provided in this application is implemented by a network device. In actual communication network architectures, this network device is typically a network element device or authentication control server deployed on the core network side, possessing global policy management and security verification capabilities. This method is mainly applied in highly intelligent communication scenarios such as future 6G networks to achieve fine-grained, multi-dimensional security control over various intelligent agents accessing the network and requesting services.

[0036] like Figure 1 As shown, the method includes: Step 110: Receive a service access request transmitted by the terminal device; the service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent, the target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent.

[0037] The terminal device here refers to physical hardware with network access and computing capabilities, such as virtual reality (VR) glasses, smartwatches, smartphones, or in-vehicle terminals.

[0038] A target intelligent agent refers to a software program or artificial intelligence application entity that actually generates a network service request and actively initiates a call. It should be noted that the target intelligent agent can be a native intelligent agent that comes with the terminal device at the factory, or a third-party intelligent agent that is subsequently installed and deployed on the device.

[0039] Associated users refer to users who currently physically possess and legally control the terminal device and the target intelligent agent, that is, the objects whose intentions are interpreted and whose operations are performed on their behalf by the target intelligent agent.

[0040] It should be noted that this service access request is a request sent by the target intelligent agent to the network device through the underlying communication module of the terminal device during the request initiation phase, when the associated user wears or operates the terminal device and triggers the start of a specific target service. It is used to request access to network resources or call network services.

[0041] The service access request carries at least the agent identifier of the target intelligent agent and the user identifier of its corresponding associated user. The agent identifier is a unique identifier used to identify the identity of the target intelligent agent, such as a digital certificate, serial number, or hash string; the user identifier is an identifier used to uniquely identify the identity of the associated user, such as an account code or identity credential digest. This embodiment does not specifically limit the specific identifiers.

[0042] Step 120: Authenticate the target intelligent agent based on the intelligent agent identifier.

[0043] Step 130: If the target intelligent agent is determined to be authenticated, the associated user is authenticated based on the user identifier.

[0044] Optionally, upon receiving a service access request, the network device first performs dual authentication operations for both the agent and the associated user.

[0045] Specifically, network devices can first authenticate the target intelligent agent based on the agent's identifier to perform identity verification at the application entity level. The purpose of this authentication is to confirm that the target intelligent agent initiating the service access request is indeed a legitimate application entity authorized by the network side, and not a forged or unregistered illegal program.

[0046] It should be noted that during step 120, the network device can parse the agent identifier of the target agent from the received service access request and compare it with the relevant information of authorized agents pre-stored on the network device side to verify whether the target agent is a legitimate entity and not a maliciously forged or unauthorized illegal program. This allows for the timely interception of malicious probing requests initiated due to the agent itself being tampered with or forged. If the comparison result shows that the target agent is an authorized and legitimate agent, the target agent is determined to have passed authentication; otherwise, the target agent is determined to have failed authentication.

[0047] It should be noted that network device authentication of the target agent is a prerequisite for associated user authentication. That is, the network device will only proceed to step 130, authenticating the associated user, if the target agent is confirmed to be authenticated. If the target agent fails authentication, the network device does not need to authenticate the associated user and can directly terminate the authentication and authorization process for the current service access request. By implementing agent authentication upfront, service access requests initiated by unauthorized agents can be filtered out at the application entity level, avoiding the continued consumption of subsequent user verification and authorization review resources on unauthorized agents.

[0048] Furthermore, because intelligent agents possess the characteristic of performing operations autonomously on behalf of users, malicious users may launch impersonation attacks using legitimate intelligent agents if the terminal device is lost or compromised. Therefore, after confirming that the target intelligent agent has passed authentication, the network device can also parse the user identifier of the associated user from the service access request and authenticate the associated user controlling the target intelligent agent based on this user identifier. For example, the network device can use the user identifier carried in the service access request to compare or verify with reference data in the pre-stored authorized intelligent agent information set in the network device to verify whether the associated user currently driving the target intelligent agent to initiate the request is the legitimate owner or authorized user of the intelligent agent. If the authentication determines that the associated user is the legitimate owner or authorized user, the associated user is determined to have passed authentication; otherwise, the associated user is determined to have failed authentication.

[0049] Therefore, by using a two-step authentication mechanism that first authenticates the target intelligent agent and then authenticates the associated user, network devices can perform dual identity verification on service access requests at both the application entity level and the user level. This not only confirms the legitimacy of the target intelligent agent initiating the request but also confirms the real control of the associated user who controls the target intelligent agent, thereby effectively reducing the risk of impersonation attacks caused by the impersonation of intelligent agents or the theft of user credentials.

[0050] Step 140: If the associated user is authenticated, verify whether the target intelligent agent has the processing authority for the target service requested by the service access request, according to the target operator policy corresponding to the target intelligent agent.

[0051] The target service here refers to the network resources or network services that the target intelligent agent requests to access or invoke through a service access request, such as data subscription services, content rendering services, or computing power invocation services provided by network devices.

[0052] Target operator policy refers to the set of policies pre-configured on the network device side for the target intelligent agent, used to constrain the boundaries of the target intelligent agent's executable behavior in the network. It includes at least the range of resources the target intelligent agent is allowed to access, the network services it is allowed to invoke, and rules for differentiating and protecting traffic initiated by the target intelligent agent. Processing permissions refer to the operational permissions that the target intelligent agent is allowed to access or invoke target services.

[0053] Optionally, after both identity verifications are passed, that is, after the target intelligent agent and its corresponding associated users are both authenticated, the authorization review stage can be entered.

[0054] During the authorization and verification phase, the network device obtains the target operator policy corresponding to the target agent and verifies, based on the target operator policy, whether the target agent is allowed to process the target service requested by the current service access request. If the verification result shows that the target agent is allowed to process the target service, it is determined that the target agent has the authority to process the target service; otherwise, it is determined that the target agent does not have the authority to process the target service.

[0055] It should be noted that the authorization verification here is contingent upon the associated user being authenticated. That is, only after both the target agent and the associated user have been authenticated will the network device continue to verify permissions according to the target operator's policy; if the associated user has not been authenticated, the network device will not execute the authorization verification process.

[0056] Therefore, by conducting authorization review based on the target operator's policies after dual authentication, network devices can further control the behavioral permissions of target intelligent agents on the basis of confirming the legitimacy of the identity. This prevents target intelligent agents that have passed identity verification from accessing target services they are not authorized to access due to defects in their own algorithm logic or misinterpretation of user instructions, thereby preventing security risks caused by abuse of permissions.

[0057] Step 150: If it is determined that the target intelligent agent has the processing authority for the target service, a target data transmission channel is established according to the target operator's policy, and the target service is provided to the target intelligent agent based on the target data transmission channel.

[0058] The target data transmission channel here refers to the communication link established by the network device to provide target services to the target intelligent agent, used to transmit target service-related data between the network device and the target intelligent agent.

[0059] Optionally, if it is determined that the target intelligent agent has the authority to process the target service, the network device establishes a corresponding data transmission channel for the target intelligent agent according to the target operator's policy, that is, the target data transmission channel, and provides the target service to the target intelligent agent based on the target data transmission channel.

[0060] In one possible implementation, the network device can configure corresponding quality of service parameters for the target data transmission channel according to the traffic guarantee rules constrained in the target operator's policy, so that the established target data transmission channel can adapt to the target agent's transmission needs for the target service, thereby providing the target service to the target agent stably and reliably based on the target data transmission channel.

[0061] Thus, by establishing target data transmission channels and providing target services based on the target operator's policies, network devices achieve differentiated service delivery for target intelligent agents that have passed authentication and authorization audits. This ensures that the data transmission guarantee obtained by the target intelligent agent is compatible with the traffic characteristics of the target intelligent agent, thereby ensuring service security while also taking into account service reliability.

[0062] The method provided in this invention receives service access requests carrying agent and user identifiers on the network device side, and sequentially authenticates the target agent, authenticates the associated user, and verifies processing permissions according to the target operator's policies. Finally, after successful verification, a target data transmission channel is established and the target service is provided. This achieves dual identity verification for the target agent and its associated user, as well as permission control based on operator policies. It can distinguish whether the initiator of the service access request is the terminal device itself or the target agent running on the terminal device, and supports secure access and network service calls for third-party agents. It effectively reduces the risk of impersonation attacks caused by stolen credentials and device intrusion. Furthermore, it can carry differentiated services for target agents that have passed authentication and authorization verification, so that the data transmission guarantee obtained by the target agent can be adapted to the traffic characteristics of the target agent, thereby improving the security, reliability, and interactive experience of agent applications.

[0063] Based on the above embodiments, in one possible implementation, authenticating the target intelligent agent according to the intelligent agent identifier includes: Based on the agent identifier, the target authentication information corresponding to the target agent is searched in the pre-stored set of authorized agent information; the target authentication information includes registration status information, authentication credential information, and device binding information. If the target authentication information is found, the system verifies whether the target intelligent agent has completed registration based on the registration status information, verifies whether the authentication credential of the target intelligent agent is valid based on the authentication credential information, and verifies whether the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information. If the target intelligent agent has completed registration, the authentication credentials of the target intelligent agent are valid, and the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information, then the target intelligent agent is determined to have passed authentication.

[0064] The authorized agent information set here refers to the data set pre-stored by network devices during the pre-configuration phase to record relevant information of each authorized agent. This includes authentication information, user information, and operator policies for each authorized agent. Authentication information includes, but is not limited to, registration status information, authentication credential information, and device binding information; user information includes, but is not limited to, legitimate identity credential information and legitimate device characteristic information; operator policies include, but are not limited to, resource access scope, network service call permissions, and traffic differentiation policies. This authorized agent information set can be formed through batch configuration and dynamic updates based on configuration information uploaded by the security management party.

[0065] Target authentication information refers to the information in the authorized intelligent agent information set corresponding to the target intelligent agent, used to verify the legitimacy of the target intelligent agent's identity. It includes at least registration status information, authentication credential information, and device binding information. Specifically, registration status information indicates whether the target intelligent agent has completed registration on the network device side; authentication credential information indicates the legitimate authentication credentials issued to the target intelligent agent by the network device during the pre-configuration phase, such as a digital certificate or key generated and issued to the authorized intelligent agent by the network device during the registration phase; and device binding information indicates the pre-defined binding relationship between the target intelligent agent, the terminal device, and the network device.

[0066] Optionally, during the authentication process of the target agent, the network device uses the agent identifier parsed from the service access request as a lookup index to search for the target authentication information corresponding to the agent identifier in a pre-stored set of authorized agent information.

[0067] If the target authentication information is found, the network device verifies whether the target intelligent agent has completed registration based on the registration status information in the target authentication information; verifies whether the authentication credentials currently held by the target intelligent agent are valid based on the authentication credential information in the target authentication information, such as verifying whether the authentication credentials currently held by the target intelligent agent are consistent with the pre-stored authentication credentials and whether they are within the validity period; and verifies whether the actual binding relationship between the target intelligent agent and the terminal device and network device matches the preset binding relationship recorded in the device binding information.

[0068] If the above verification results confirm that the target intelligent agent has completed registration, the target intelligent agent's authentication credentials are valid, and the binding relationship between the target intelligent agent and the terminal device and network device matches the preset binding relationship, then the target intelligent agent is determined to have passed authentication. If any of the above three conditions are not met, or if the target authentication information is not found in the authorized intelligent agent information set, then the target intelligent agent is determined to have failed authentication.

[0069] The method provided in this embodiment comprehensively verifies the target intelligent agent from three dimensions: registration status, validity of authentication credentials, and device binding relationship. This not only confirms that the target intelligent agent has been legally registered on the network side and holds valid credentials, but also prevents impersonation attacks launched after legitimate credentials are implanted into unauthorized devices by verifying the binding relationship between the target intelligent agent, terminal device, and network device. This further improves the reliability of intelligent agent authentication at the application entity level.

[0070] Based on the above embodiments, in one possible implementation, authenticating the associated user based on the user identifier includes: Obtain the real-time identity credential information of the associated user and the real-time device feature information of the terminal device; Based on the user identifier and the agent identifier, the target user information corresponding to the target agent is searched in the pre-stored authorized agent information set; the target user information includes legitimate identity credential information and legitimate device feature information; If the target user information is found, the consistency between the real-time identity credential information and the legitimate identity credential information is verified to obtain the first verification result. The consistency between the real-time device feature information and the legitimate device feature information is verified to obtain a second verification result; If both the first and second verification results are successful, the associated user is determined to have passed authentication.

[0071] The real-time identity credential information here refers to the credential information collected by the network device at the current authentication time to represent the current identity of the associated user, such as the account password, biometric information, or identity credential summary entered by the associated user; the real-time device feature information refers to the information collected by the network device at the current authentication time to represent the current features of the terminal device, such as the hardware identifier or security chip information of the terminal device.

[0072] Target user information refers to the information in the authorized intelligent agent information set that corresponds to the target intelligent agent and is used to verify the legitimacy of the associated user. It includes at least legitimate identity credential information and legitimate device feature information. Among them, the legitimate identity credential information is the pre-stored legitimate identity information of the associated user; the legitimate device feature information is the pre-stored feature information of the legitimate terminal device bound to the target intelligent agent.

[0073] Optionally, network devices may employ a multi-factor authentication mechanism to authenticate associated users.

[0074] Specifically, the network device first obtains the real-time identity credential information of the associated user and the real-time device characteristic information of the terminal device; then, using the user identifier and agent identifier parsed from the service access request as the search index, it searches for the target user information corresponding to the target agent in the pre-stored set of authorized agent information.

[0075] When the target user information is found, the network device verifies the consistency between the real-time identity credential information and the legitimate identity credential information in the target user information to obtain a first verification result that characterizes whether the associated user's identity credential is legitimate. In addition, it can also verify the consistency between the real-time device feature information and the legitimate device feature information in the target user information to obtain a second verification result that characterizes whether the terminal device is legitimate.

[0076] If both the first and second verification results are successful, the network device determines that the associated user has passed authentication; if either the first or second verification result is unsuccessful, or if the target user information is not found in the authorized agent information set, the network device determines that the associated user has failed authentication.

[0077] The method provided in this embodiment combines the verification of the identity credentials of the associated user with the verification of the device characteristics of the terminal device to construct a multi-factor verification mechanism for associated users. Even if the identity credentials of the associated user are stolen, the attacker will not be able to pass the double consistency verification because it is difficult for the attacker to have legitimate terminal device characteristics at the same time. This further reduces the risk of impersonation attacks caused by the theft of user credentials or the intrusion of terminal devices at the user level.

[0078] Based on the above embodiments, in one possible implementation, verifying whether the target intelligent agent has the processing authority for the target service requested by the service access request, according to the target operator policy corresponding to the target intelligent agent, includes: Based on the agent identifier, the target operator policy corresponding to the target agent is searched in the pre-stored set of authorized agent information; If the target operator's policy is found, the resource access scope and network service call permissions of the target intelligent agent are obtained from the target operator's policy. Based on the resource access scope, verify whether the target agent has resource access permissions to the target service; Based on the network service invocation permissions, verify whether the target intelligent agent has the invocation permissions for the network service corresponding to the target service; If the target intelligent agent has the resource access permission and the calling permission, it is determined that the target intelligent agent has the processing permission for the target service.

[0079] The resource access scope here refers to the range of network resources that the target agent is allowed to access, as constrained by the target operator's policies; network service invocation permission refers to the network service permissions that the target agent is allowed to invoke, as constrained by the target operator's policies. Resource access permission refers to the target agent's permission to access the resources corresponding to the target service; invocation permission refers to the target agent's permission to invoke the network service corresponding to the target service.

[0080] Optionally, during the authorization review phase, the network device uses the agent identifier parsed from the service access request as a lookup index to search for the target operator policy corresponding to the target agent in a pre-stored set of authorized agent information.

[0081] Upon finding the target operator's policy, the network device obtains the target agent's resource access scope and network service invocation permissions from the target operator's policy. Based on the resource access scope, it verifies whether the target agent has the resource access permissions for the target service, that is, whether the resources corresponding to the target service fall within the resource access scope that the target agent is allowed to access. In addition, based on the network service invocation permissions, it can also verify whether the target agent has the invocation permissions for the network service corresponding to the target service, that is, whether the network service corresponding to the target service is a network service that the target agent is allowed to invoke.

[0082] If the target intelligent agent possesses both resource access permission and invocation permission, the network device determines that the target intelligent agent has the authority to process the target service; if the target intelligent agent does not possess either resource access permission or invocation permission, the network device determines that the target intelligent agent does not have the authority to process the target service.

[0083] The method provided in this embodiment verifies the processing permissions of the target intelligent agent in a fine-grained manner from two dimensions: the scope of resource access and the permission to call network services. This can strictly constrain the behavior of the target intelligent agent within the permission range specified by the target operator's policy, preventing the target intelligent agent from accessing resources it does not have the right to access and from illegally calling network services it does not have the right to call. This achieves fine-grained control over the behavior permissions of the target intelligent agent.

[0084] Based on the above embodiments, in one possible implementation, establishing the target data transmission channel according to the target operator's strategy includes: Obtain the traffic differentiation strategy from the target operator's strategy; If the initiator of the service access request is determined to be the target intelligent agent according to the traffic differentiation strategy, the traffic configuration characteristics corresponding to the target intelligent agent are determined according to the traffic differentiation strategy. The target data transmission channel is established based on the traffic configuration characteristics corresponding to the target intelligent agent.

[0085] The traffic differentiation strategy here refers to the rules included in the target operator's strategy for distinguishing whether the initiator of a service access request is the terminal device itself or a target intelligent agent running on the terminal device, as well as the rules that can assign different traffic configuration characteristics to different initiators. Traffic configuration characteristics refer to traffic configuration parameters adapted to the traffic initiated by the target intelligent agent, such as the transmission rate, transmission priority, and other configuration characteristics corresponding to the traffic initiated by the target intelligent agent.

[0086] Optionally, during the establishment of the target data transmission channel, the network device first obtains the traffic differentiation policy from the target operator's policy; then, the network device identifies the initiator of the current service access request based on the traffic differentiation policy.

[0087] When the network device determines that the initiator of the service access request is the target intelligent agent rather than the terminal device itself based on the traffic differentiation strategy, the network device determines the traffic configuration characteristics corresponding to the target intelligent agent according to the traffic differentiation strategy, and then establishes a target data transmission channel based on the traffic configuration characteristics corresponding to the target intelligent agent, so that the established target data transmission channel is adapted to the transmission requirements of the traffic initiated by the target intelligent agent.

[0088] The method provided in this embodiment introduces a traffic differentiation strategy into the operator's policy, enabling network devices to accurately distinguish whether the initiator of the service access request is the terminal device itself or the target intelligent agent running on the terminal device. It then determines the corresponding traffic configuration characteristics for the traffic initiated by the target intelligent agent and establishes a target data transmission channel. This achieves differentiated carrying for different traffic needs of the terminal device and the target intelligent agent, meets the unique traffic transmission needs of the target intelligent agent, and effectively improves the interactive experience.

[0089] Based on the above embodiments, in one possible implementation, establishing the target data transmission channel according to the traffic configuration characteristics corresponding to the target intelligent agent includes: Assign quality of service parameters to the service access request that are adapted to the traffic configuration features corresponding to the target agent; The target data transmission channel is established based on the service quality parameters.

[0090] The Quality of Service (QoS) parameter here refers to parameters used to characterize the level of transmission service assurance of a data transmission channel, such as latency, bandwidth, and reliability values.

[0091] Optionally, after obtaining the traffic configuration characteristics corresponding to the target agent, the network device can allocate service quality parameters that are compatible with the traffic configuration characteristics to the current service access request. That is, based on the traffic configuration characteristics of the traffic initiated by the target agent, the network device determines the service quality parameters such as latency, bandwidth, and reliability values ​​that can meet the transmission requirements of the traffic. Subsequently, the network device establishes a target data transmission channel according to the allocated service quality parameters, so that the established target data transmission channel can transmit data according to the transmission service guarantee level constrained by the service quality parameters.

[0092] The method provided in this embodiment allocates service quality parameters that are compatible with the traffic configuration characteristics of the target intelligent agent to the service access request, and establishes a target data transmission channel accordingly. This achieves differentiated service quality assurance for the traffic characteristics of the target intelligent agent, so that the transmission service assurance level of the target data transmission channel can be accurately matched with the traffic transmission needs of the target intelligent agent, effectively improving the interactive experience.

[0093] Based on the above embodiments, in one possible implementation, the target data transmission channel is used to transmit multimodal data according to the quality of service parameters; The multimodal data includes multiple data sources such as image data, audio data, and real-time interactive data required for visual rendering. The service quality parameters include a latency time less than a preset latency time threshold, a bandwidth higher than a preset bandwidth threshold, and a reliability value higher than a preset reliability threshold.

[0094] Multimodal data here refers to data containing multiple modalities, including at least several types such as image data, audio data, and real-time interactive data required for visual rendering. The latency threshold is a pre-set upper limit for measuring whether the latency of the target data transmission channel meets the transmission requirements; the bandwidth threshold is a pre-set lower limit for measuring whether the bandwidth of the target data transmission channel meets the transmission requirements; and the reliability threshold is a pre-set lower limit for measuring whether the reliability of the target data transmission channel meets the transmission requirements.

[0095] Optionally, in order to adapt to the high-specification requirements of the target intelligent agent for multimodal data transmission in big data scenarios such as virtual reality and content rendering, the target data transmission channel is configured to transmit multimodal data according to the quality of service parameters, and the quality of service parameters corresponding to the target data transmission channel meet the following conditions: the latency is less than a preset latency threshold, that is, the target data transmission channel has ultra-low latency characteristics; the bandwidth is higher than a preset bandwidth threshold, that is, the target data transmission channel has high bandwidth characteristics; and the reliability value is higher than a preset reliability threshold, that is, the target data transmission channel has high reliability characteristics.

[0096] Therefore, by configuring the target data transmission channel with quality of service parameters that have ultra-low latency, high bandwidth, and high reliability, the target data transmission channel can stably carry the transmission of multimodal data such as image data, audio data, and real-time interactive data required for visual rendering.

[0097] The method provided in this embodiment, by configuring the target data transmission channel to have ultra-low latency, high bandwidth, and high reliability characteristics and using it to transmit multimodal data, can meet the high-specification requirements of the target intelligent agent for multimodal data transmission in large data volume scenarios, thereby ensuring the real-time rendering and interactive experience of the target intelligent agent application.

[0098] Based on the above embodiments, in one possible implementation, the method further includes: If the target agent fails to authenticate, the associated user fails to authenticate, or the target agent does not have the authority to process the target service, the service access request is rejected, and the abnormal behavior of the target agent is recorded. If, based on the abnormal behavior records, it is determined that the number of abnormal behaviors generated by the target intelligent agent within a preset time period reaches a preset threshold, an abnormal handling operation is performed. The exception handling operation includes at least one of the following operations: Trigger a security alert for the target intelligent agent; Reduce the service quality corresponding to the target intelligent agent; A re-authentication request is sent to the associated user, which instructs the associated user to verify the abnormal behavior of the target intelligent agent.

[0099] Abnormal behavior here refers to actions that do not meet security requirements during the authentication and authorization process of the target intelligent agent. Examples include authentication failures such as the target intelligent agent failing authentication, authentication failures of associated users failing authentication, or unauthorized access requests by the target intelligent agent to a target service it is not authorized to handle. The frequency threshold is a pre-set upper limit on the number of abnormal behaviors generated by the target intelligent agent within a preset time period that warrant further action.

[0100] Optionally, if any of the following three scenarios occur: the target intelligent agent fails to be authenticated, the associated user fails to be authenticated, or the target intelligent agent does not have the authority to process the target service, the network device rejects the current service access request and returns a rejection response to the terminal device that initiated the service access request. At the same time, it records the abnormal behavior corresponding to the target intelligent agent to form an abnormal behavior record.

[0101] Furthermore, the network device can also count the number of abnormal behaviors generated by the target agent within a preset time period based on abnormal behavior records. If the number of abnormal behaviors generated by the target agent within the preset time period reaches a preset threshold, the network device performs at least one of the following anomaly handling operations: Trigger a security alert for the target intelligent agent to prompt the administrator to handle the abnormal behavior of the target intelligent agent in a timely manner; Lower the service quality parameters corresponding to the target intelligent agent to limit the level of service assurance that the target intelligent agent can obtain; A re-authentication request is sent to the associated user. This re-authentication request instructs the associated user to verify the abnormal behavior of the target intelligent agent to confirm whether the abnormal behavior was caused by the associated user's own operation.

[0102] The method provided in this embodiment records the abnormal behavior generated by the target intelligent agent during the authentication and authorization process, and performs abnormal handling operations such as security alarms, reducing service quality parameters, or sending re-authentication requests when the number of abnormal behaviors reaches a threshold. This constructs a security protection mechanism for the abnormal behavior of the target intelligent agent, which can promptly detect and handle the continuous abnormal behavior of the target intelligent agent, thereby further improving the security of network services in intelligent agent application scenarios.

[0103] The authentication and authorization system for intelligent agents provided by this invention is described below. The authentication and authorization system for intelligent agents described below can be referred to in correspondence with the authentication and authorization method for intelligent agents described above.

[0104] Figure 2 This is a schematic diagram of the intelligent agent authentication and authorization system provided by the present invention; as shown below. Figure 2As shown, the system includes at least a network device for performing an authentication and authorization method for an intelligent agent; the system includes: The request receiving module 210 is used to receive a service access request transmitted by the terminal device; the service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent, the target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent; The agent authentication module 220 is used to authenticate the target agent based on the agent identifier; The user authentication module 230 is used to authenticate the associated user based on the user identifier when it is determined that the target intelligent agent has passed the authentication. The authorization verification module 240 is used to verify, in the case that the associated user has passed the authentication, whether the target intelligent agent has the processing authority for the target service requested by the service access request, according to the target operator policy corresponding to the target intelligent agent. The service response module 250 is used to establish a target data transmission channel according to the target operator's policy, and provide the target service to the target intelligent agent based on the target data transmission channel, when it is determined that the target intelligent agent has the processing authority for the target service.

[0105] The system provided in this invention receives service access requests carrying agent and user identifiers on the network device side, and sequentially authenticates the target agent, authenticates the associated user, and verifies processing permissions according to the target operator's policies. Finally, after successful verification, it establishes a target data transmission channel and provides the target service. This achieves dual identity verification for the target agent and its associated user, as well as permission control based on operator policies. It can distinguish whether the initiator of the service access request is the terminal device itself or the target agent running on the terminal device. It also supports secure access and network service calls for third-party agents, effectively reducing the risk of impersonation attacks caused by stolen credentials and device intrusion. Furthermore, it can carry differentiated services for target agents that have passed authentication and authorization verification, ensuring that the data transmission guarantee obtained by the target agent is adapted to the traffic characteristics of the target agent, thereby improving the security, reliability, and interactive experience of agent applications.

[0106] like Figure 2 As shown, in some embodiments, the system further includes a pre-configuration module and a security protection module; The pre-configuration module 200 is used to pre-store an authorized agent information set; the authorized agent information set includes the authentication information, user information and operator policies of each authorized agent. The security protection module 260 is used to reject the service access request and record the abnormal behavior of the target intelligent agent when the target intelligent agent fails authentication, the associated user fails authentication, or the target intelligent agent does not have the processing authority for the target service; and to perform an anomaly handling operation when it is determined from the anomaly behavior record that the number of anomalies generated by the target intelligent agent within a preset time period reaches a preset threshold; the anomaly handling operation includes at least one of the following operations: triggering a security alarm for the target intelligent agent; reducing the service quality corresponding to the target intelligent agent; and sending a re-authentication request to the associated user, the re-authentication request being used to request the associated user to verify the abnormal behavior of the target intelligent agent.

[0107] The system provided in this embodiment records the abnormal behaviors generated by the target intelligent agent during the authentication and authorization process, and performs abnormal handling operations such as security alarms, reducing service quality parameters, or sending re-authentication requests when the number of abnormal behaviors reaches a threshold. This system constructs a security protection mechanism for abnormal behaviors of the target intelligent agent, which can promptly detect and handle the continuous abnormal behaviors of the target intelligent agent, thereby further improving the security of network services in intelligent agent application scenarios.

[0108] The system provided by this invention is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0109] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communications bus 340. The processor 310 can invoke logical instructions in the memory 330 to execute an authentication and authorization method for an intelligent agent. This method includes: receiving a service access request transmitted by a terminal device; the service access request carrying an intelligent agent identifier of a target intelligent agent and a user identifier of an associated user corresponding to the target intelligent agent, wherein the target intelligent agent is an intelligent agent running on the terminal device that initiates the service access request, and the associated user is a user controlling the target intelligent agent; authenticating the target intelligent agent based on the intelligent agent identifier; authenticating the associated user based on the user identifier if the target intelligent agent passes authentication; verifying whether the target intelligent agent has the processing authority for the target service requested by the service access request according to the target operator policy corresponding to the target intelligent agent if the target intelligent agent has the processing authority for the target service; and establishing a target data transmission channel according to the target operator policy and providing the target service to the target intelligent agent based on the target data transmission channel if the target intelligent agent has the processing authority for the target service.

[0110] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the agent authentication and authorization method provided by the above methods. The method includes: receiving a service access request transmitted by a terminal device; the service access request carrying an agent identifier of a target agent and a user identifier of an associated user corresponding to the target agent, wherein the target agent is an agent running on the terminal device that initiates the service access request, and the associated user is a user controlling the target agent; authenticating the target agent according to the agent identifier; authenticating the associated user according to the user identifier if the target agent passes authentication; verifying whether the target agent has the processing authority for the target service requested by the service access request according to the target operator policy corresponding to the target agent if the associated user passes authentication; and establishing a target data transmission channel according to the target operator policy and providing the target service to the target agent based on the target data transmission channel if the target agent has the processing authority for the target service.

[0112] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements an authentication and authorization method for an intelligent agent provided by the methods described above. The method includes: receiving a service access request transmitted by a terminal device; the service access request carrying an intelligent agent identifier of a target intelligent agent and a user identifier of an associated user corresponding to the target intelligent agent, wherein the target intelligent agent is an intelligent agent running on the terminal device that initiates the service access request, and the associated user is a user controlling the target intelligent agent; authenticating the target intelligent agent based on the intelligent agent identifier; authenticating the associated user based on the user identifier if the target intelligent agent passes authentication; verifying whether the target intelligent agent has the processing authority for the target service requested by the service access request according to a target operator policy corresponding to the target intelligent agent if the associated user passes authentication; and establishing a target data transmission channel according to the target operator policy and providing the target service to the target intelligent agent based on the target data transmission channel if the target intelligent agent has the processing authority for the target service.

[0113] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0114] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An authentication authorization method of an intelligent body, characterized by, Applied to network devices, the method includes: Receive a service access request transmitted by a terminal device; the service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent, the target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent; The target intelligent agent is authenticated based on the intelligent agent identifier; If the target intelligent agent is determined to be authenticated, the associated user is authenticated based on the user identifier; If the associated user is authenticated, the system verifies whether the target intelligent agent has the authority to process the target service requested by the service access request, based on the target operator policy corresponding to the target intelligent agent. If it is determined that the target agent has the authority to process the target service, a target data transmission channel is established according to the target operator's policy, and the target service is provided to the target agent based on the target data transmission channel.

2. The method of claim 1, wherein, The step of authenticating the target intelligent agent based on the intelligent agent identifier includes: Based on the agent identifier, the target authentication information corresponding to the target agent is searched in the pre-stored set of authorized agent information; the target authentication information includes registration status information, authentication credential information, and device binding information. If the target authentication information is found, the system verifies whether the target intelligent agent has completed registration based on the registration status information, verifies whether the authentication credential of the target intelligent agent is valid based on the authentication credential information, and verifies whether the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information. If the target intelligent agent has completed registration, the authentication credentials of the target intelligent agent are valid, and the binding relationship between the target intelligent agent and the terminal device and the network device matches the preset binding relationship in the device binding information, then the target intelligent agent is determined to have passed authentication.

3. The method of claim 1, wherein, The step of authenticating the associated user based on the user identifier includes: Obtain the real-time identity credential information of the associated user and the real-time device feature information of the terminal device; Based on the user identifier and the agent identifier, the target user information corresponding to the target agent is searched in the pre-stored authorized agent information set; the target user information includes legitimate identity credential information and legitimate device feature information; If the target user information is found, the consistency between the real-time identity credential information and the legitimate identity credential information is verified to obtain the first verification result. The consistency between the real-time device feature information and the legitimate device feature information is verified to obtain a second verification result; If both the first and second verification results are successful, the associated user is determined to have passed authentication.

4. The method of claim 1-3, wherein, The step of verifying whether the target intelligent agent has the authority to process the target service requested by the service access request, based on the target operator policy corresponding to the target intelligent agent, includes: Based on the agent identifier, the target operator policy corresponding to the target agent is searched in the pre-stored set of authorized agent information; If the target operator's policy is found, the resource access scope and network service call permissions of the target intelligent agent are obtained from the target operator's policy. Based on the resource access scope, verify whether the target agent has resource access permissions to the target service; Based on the network service invocation permissions, verify whether the target intelligent agent has the invocation permissions for the network service corresponding to the target service; If the target intelligent agent has the resource access permission and the calling permission, it is determined that the target intelligent agent has the processing permission for the target service.

5. The method of claim 1-3, wherein, The step of establishing a target data transmission channel according to the target operator's strategy includes: From the target operator's policy, obtain the traffic differentiation policy; If the initiator of the service access request is determined to be the target intelligent agent according to the traffic differentiation strategy, the traffic configuration characteristics corresponding to the target intelligent agent are determined according to the traffic differentiation strategy. The target data transmission channel is established based on the traffic configuration characteristics corresponding to the target intelligent agent.

6. The method of claim 5, wherein, The step of establishing the target data transmission channel based on the traffic configuration characteristics corresponding to the target intelligent agent includes: Assign quality of service parameters to the service access request that are adapted to the traffic configuration features corresponding to the target agent; The target data transmission channel is established based on the service quality parameters.

7. The method of claim 6, wherein, The target data transmission channel is used to transmit multimodal data according to the quality of service parameters; The multimodal data includes multiple data sources such as image data, audio data, and real-time interactive data required for visual rendering. The service quality parameters include a latency time less than a preset latency time threshold, a bandwidth higher than a preset bandwidth threshold, and a reliability value higher than a preset reliability threshold.

8. The method of claim 1-3, wherein, The method further includes: If the target agent fails to authenticate, the associated user fails to authenticate, or the target agent does not have the authority to process the target service, the service access request is rejected, and the abnormal behavior of the target agent is recorded. If, based on the abnormal behavior records, it is determined that the number of abnormal behaviors generated by the target intelligent agent within a preset time period reaches a preset threshold, an abnormal handling operation is performed. The exception handling operation includes at least one of the following operations: Trigger a security alert for the target intelligent agent; Reduce the service quality corresponding to the target intelligent agent; A re-authentication request is sent to the associated user, which instructs the associated user to verify the abnormal behavior of the target intelligent agent.

9. An authentication authorization system for an agent, characterized by include: The request receiving module is used to receive service access requests transmitted by terminal devices. The service access request carries the agent identifier of the target agent and the user identifier of the associated user corresponding to the target agent. The target agent is the agent running on the terminal device that initiates the service access request, and the associated user is the user who controls the target agent. The agent authentication module is used to authenticate the target agent based on the agent identifier; The user authentication module is used to authenticate the associated user based on the user identifier when it is determined that the target intelligent agent has passed the authentication. The authorization verification module is used to verify, in the case that the associated user has passed the authentication, whether the target intelligent agent has the processing authority for the target service requested by the service access request, according to the target operator policy corresponding to the target intelligent agent. The service response module is used to establish a target data transmission channel according to the target operator's policy, and provide the target service to the target intelligent agent based on the target data transmission channel, when it is determined that the target intelligent agent has the processing authority for the target service.

10. The authentication authorization system of an agent according to claim 9, wherein, The system also includes a pre-configuration module and a security protection module; The pre-configuration module is used to pre-store a set of authorized intelligent agent information; the set of authorized intelligent agent information includes the authentication information, user information and operator policies of each authorized intelligent agent; The security protection module is used to reject the service access request and record the abnormal behavior of the target intelligent agent when the target intelligent agent fails to be authenticated, the associated user fails to be authenticated, or the target intelligent agent does not have the processing authority to the target service. If, based on the abnormal behavior records, it is determined that the number of abnormal behaviors generated by the target intelligent agent within a preset time period reaches a preset threshold, an abnormal handling operation is performed. The abnormal handling operation includes at least one of the following operations: triggering a security alarm for the target intelligent agent; reducing the service quality corresponding to the target intelligent agent; and sending a re-authentication request to the associated user, wherein the re-authentication request is used to request the associated user to verify the abnormal behavior of the target intelligent agent.