Model authorization method and apparatus, and readable storage medium

Abstract

The present application relates to the technical field of communications, and in particular, to a model authorization method and apparatus, and a readable storage medium. The method comprises: an MTLF1 receives a first model acquisition request from an AnLF, comprising an analytics ID; the MTLF1 sends a token acquisition request to an NRF, comprising the analytics ID, an AnLF ID, and an MTLF2 ID, an MTLF2 being used to provide a model corresponding to the analytics ID; upon verifying that the AnLF has permission to acquire the model from the MTLF2, the NRF returns a first token to the MTLF1; the MTLF1 sends a second model acquisition request to the MTLF2, comprising the analytics ID, the AnLF ID, and the first token; and once the first token is successfully verified, the MTLF2 sends information of the model to the AnLF. By using the present application, model authorization and model acquisition in a scenario involving entrusted acquisition of a model are facilitated, and the security of the model is improved.

Description

Model authorization method, device and readable storage medium

[0001] This application claims priority to the Chinese patent application with application number 202311428355.6 filed with the State Intellectual Property Office of China on October 30, 2023, and priority to the Chinese patent application with the invention name “Model Authorization Method, Device and Readable Storage Medium”, all contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a model authorization method, device, and readable storage medium. Background Art

[0003] The network data analytics function (NWDAF) network element has functions such as data collection, model training, data analysis, and model reasoning. The NWDAF network element containing the analytics logical function (AnLF) can be used to infer and export analytical information and expose analytical services. The NWDAF network element containing the model training logical function (MTLF) can be used to train machine learning (ML) models or artificial intelligence (AI) models and expose new training services; for example, providing trained AI models or ML models. The analytics data repository functional (ADRF) network element can provide storage, deletion, and retrieval services for AI / ML models.

[0004] AI / ML models can be stored in ADRF network elements in the following way: the consumer (for example, NWDAF network element containing MTLF) sends Nadrf_MLModelManagement_Storage Request (Nadrf machine learning model management storage request) to the ADRF network element, which contains the model or the address of the model to be stored; the ADRF network element returns a response and provides a result indication. AI / ML models can be retrieved from ADRF network elements in the following way: the consumer (e.g., NWDAF network element with AnLF) sends a Nadrf_MLModelManagement_Retrieval Request (Nadrf Machine Learning Model Management Retrieval Request) to the ADRF network element, which includes the Analytics Identifier (ID) and token. The ADRF network element authenticates the consumer (e.g., NWDAF network element with MTLF) and verifies whether the token and the consumer's (e.g., NWDAF network element with MTLF) network function (NF) instance ID (NF ID) are included in the allowed NF consumer (NFc) list of the ML model. If both verifications are successful, the ADRF network element returns a response indicating the result.

[0005] Currently, authorization is required before obtaining or retrieving a model, but model authorization in some scenarios still has security issues.

[0006] Summary of the Invention

[0007] The embodiments of the present application provide a model authorization method, device and readable storage medium, which help to implement model authorization and model acquisition in model delegation acquisition scenarios, and improve the security of models in model delegation acquisition scenarios.

[0008] The present application is introduced below from different aspects. It should be understood that the implementation methods and beneficial effects of the following different aspects can be referenced to each other.

[0009] In a first aspect, the present application provides a model authorization method, which includes: a first model training logic function network element receives a first model acquisition request from an analysis logic function network element, the first model acquisition request including an analysis identifier, and the first model acquisition request can be used to request information about a model corresponding to the analysis identifier; the first model training logic function network element sends a token acquisition request to a network repository function (NRF) network element, the token acquisition request including the analysis identifier, an identifier of the analysis logic function network element, and an identifier of a second model training logic function network element, and the second model training logic function network element can be used to provide information about a model corresponding to the analysis identifier; the first model training logic function network element receives a first token from the NRF network element, the first token including an identifier of the analysis logic function network element and an identifier of the second model training logic function network element; the first model training logic function network element sends a second model acquisition request to the second model training logic function network element, the second model acquisition request including the analysis identifier, the identifier of the analysis logic function network element, and the first token, and the second model acquisition request is used to obtain information about the model corresponding to the analysis identifier for the analysis logic function network element.

[0010] In this application, the identifiers of various network elements may be network function instance identifiers (NF instance IDs), which will not be described in detail below. For example, the identifier of the analysis logic function network element refers to the NF instance ID of the analysis logic function network element, and the identifier of the second model training logic function network element refers to the NF instance ID of the second model training logic function network element. Among them, the NF instance ID can be used to uniquely identify a network function instance (NF Instance ID: Unique identity of the NF Instance).

[0011] Exemplarily, the first token in the second model acquisition request can be used to verify the second model acquisition request. Alternatively, the first token in the second model acquisition request can be used to verify the various identifiers in the second model acquisition request. Alternatively, the first token in the second model acquisition request can be used to verify the authority of the analysis logic function network element to obtain the model (the model can be the model corresponding to the above analysis ID) from the second model training logic function network element. Alternatively, the first token in the second model acquisition request can be used to verify the authority of the first model training logic function network element to obtain the model (the model can be the model corresponding to the above analysis ID) from the second model training logic function network element for (or on behalf of) the analysis logic function network element.

[0012] Exemplarily, the first model training logic function network element may be a first network element having a model training logic function, such as a first NWDAF network element containing MTLF (NWDAF1 containing MTLF), which may be referred to as MTLF1. The second model training logic function network element may be a second network element having a model training logic function, such as a second NWDAF network element containing MTLF (NWDAF2 containing MTLF), which may be referred to as MTLF2. The analysis logic function network element may be a network element having an analysis logic function, such as a NWDAF network element containing AnLF (NWDAF containing AnLF), which may be referred to as AnLF.

[0013] Exemplarily, the above-mentioned token acquisition request can be used to request a first token. In other words, the token acquisition request can be used to request (NRF network element) to authorize the analysis logic function network element to obtain the model corresponding to the analysis ID from the second model training logic function network element. In other words, the token acquisition request can be used for the first model training logic function network element to request the first token for (or on behalf of) the analysis logic function network element to obtain the model corresponding to the analysis ID. The first token can be used to indicate that the analysis logic function network element is authorized to obtain information about the model corresponding to the analysis ID from the second model training logic function network element. Alternatively, the first token can be used to indicate that the first model training logic function network element is authorized to obtain information about the model corresponding to the analysis ID from the second model training logic function network element for (or on behalf of) the analysis logic function network element.

[0014] The existing technology only considers the situation where NF service consumers directly obtain / retrieve models from model providers (such as NWDAF containing MTLF). However, in the scenario of model delegation, the model provider or model producer cannot verify whether the model consumer has the authority to obtain the model it requests, which poses a security issue.

[0015] After receiving the model acquisition request from the analysis logic function network element, the first model training logic function network element of the present application requests an authorization token (i.e., the above-mentioned first token) from the NRF for (or on behalf of) the analysis logic function network element. After obtaining the authorization token, the first model training logic function network element requests the second model training logic function network element for (or on behalf of) the analysis logic function network element for information about a model that meets the requirements. The request includes the identification and authorization token of the analysis logic function network element; this helps to realize model authorization and model acquisition in a delegated acquisition scenario, and authorizes the actual model consumer (i.e., the analysis logic function network element) in the model delegated acquisition scenario, thereby improving the security of the model.

[0016] In combination with the first aspect, in a possible implementation, the above-mentioned token acquisition request also includes one or more of the following: an identifier of the first model training logic function network element, a supplier identifier of the analysis logic function network element, or a first indication information. Among them, the supplier identifier of the analysis logic function network element can be used by the NRF network element to verify whether it can be authorized to obtain the model corresponding to the analysis ID from the second model training logic function network element. The first indication information can be used to indicate that the first model training logic function network element requests the first token for (or on behalf of behalf) the analysis logic function network element. In other words, the first indication information can be used to indicate that the NRF network element is requested to authorize the analysis logic function network to obtain the model in the second model training logic function network element. Alternatively, the first indication information can be used to indicate that the token acquisition request is an access token acquisition request (Nnrf_AccessToken_Get request) for the model delegation acquisition scenario.

[0017] Including the vendor identifier of the analysis logic function network element in the token acquisition request in this application is helpful for the NRF network element to verify whether the analysis logic function network element has the authority to obtain the model from the second model training logic function network element. Including the first indication information in the token acquisition request in this application can make the meaning of the token acquisition request clearer and enable the NRF network element to clarify its own behavior.

[0018] In combination with the first aspect, in one possible implementation, after the NRF network element receives the token acquisition request from the first model training logic function network element, it can verify whether the supplier identifier of the analysis logic function network element is included in the interoperability indicator corresponding to the above-mentioned analysis ID of the second model training logic function network element. In other words, after the NRF network element receives the above-mentioned token acquisition request, it can verify whether the analysis logic function network element has the authority to obtain the model from the second model training logic function network element. For the specific verification method, please refer to the description of the method embodiment below, which is not described in detail here. If the supplier identifier of the analysis logic function network element is in the interoperability indicator corresponding to the above-mentioned analysis ID of the second model training logic function network element, it means that the analysis logic function network element has the authority to obtain the model from the second model training logic function network element, and the NRF network element sends a first token to the first model training logic function network element.

[0019] In combination with the first aspect, in a possible implementation, the above-mentioned first token also includes one or more of the following: the above-mentioned analysis identifier, the identifier of the first model training logic function network element, the supplier identifier of the analysis logic function network element, or the second indication information. The second indication information can be used to indicate that the analysis logic function network element is authorized to obtain the information of the model corresponding to the above-mentioned analysis ID from the second model training logic function network element. Alternatively, the second indication information can be used to indicate that the first model training logic function network element is authorized to be (or represent behalf) the analysis logic function network element to obtain the information of the model corresponding to the analysis ID from the second model training logic function network element. Alternatively, the second indication information can be used to indicate that the first token is an authorization token for the model delegation acquisition scenario.

[0020] The present application includes the second indication information in the first token, which can make the meaning of the first token clearer and help the first model training logic function network element to clarify the role of the first token.

[0021] In combination with the first aspect, in a possible implementation, the above-mentioned second model acquisition request also includes one or more of the following: an identifier of the first model training logic function network element, an address of the analysis logic function network element, a supplier identifier of the analysis logic function network element, or a third indication information. Among them, the address of the analysis logic function network element can be carried by the subscription endpoint address. The subscription endpoint address can be used to represent the address for receiving model notification messages. The third indication information can be used to indicate that the first model training logic function network element obtains information about the model corresponding to the above-mentioned analysis ID for (or on behalf of) the analysis logic function network element. Alternatively, the third indication information can be used to indicate that the second model acquisition request is an ML model provision request (Nnwdaf_MLModelProvision request) for a model delegation acquisition scenario.

[0022] The present application includes the address of the analysis logic function network element in the second model acquisition request, which is beneficial for the second model training logic function network element to determine the recipient of the model notification message. The present application includes the supplier identifier of the analysis logic function network element in the second model acquisition request, which is beneficial for the second model training logic function network element to verify whether the analysis logic function network element has the authority to obtain the model from itself. The present application includes third indication information in the second model acquisition request, which can make the meaning of the second model acquisition request clearer, so that the second model training logic function network element can clarify its own behavior.

[0023] In combination with the first aspect, in a possible implementation method, after the second model training logic function network element receives the second model acquisition request, it can verify the first token in the second model acquisition request. For the specific verification method, please refer to the description of the method embodiment below, which is not described in detail here. If the first token verification is successful, the second model training logic function network element can send a model notification message to the analysis logic function network element. The model notification message may include information about the model corresponding to the analysis ID, such as: model identifier, the address of the model in the second model training logic function network element, or model file, etc. Exemplarily, the model notification message may also include the identifier of the ADRF network element that stores the model.

[0024] In combination with the first aspect, in a possible implementation, before the first model training logic function network element sends a token acquisition request to the NRF network element, the method also includes: the first model training logic function network element sends a network element discovery request to the NRF network element, and the network element discovery request includes the analysis identifier, and the identifier of the analysis logic function network element and / or the supplier identifier of the analysis logic function network element; the first model training logic function network element receives a network element discovery response from the NRF network element, and the network element discovery response includes a candidate network element list, and the network element list includes the second model training logic function network element, and the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element includes the supplier identifier of the analysis logic function network element.

[0025] It is understood that during the network element registration process, the first model training logic function network element and the second model training logic function network element can register their respective analysis IDs and interoperability indicators with the NRF network element. The analysis logic function network element can register its own analysis ID and vendor ID with the NRF network element. The interoperability indicator can be used to indicate a list of NWDAF providers (or vendors) that are allowed to retrieve ML models from the NWDAF containing MTLF.

[0026] In combination with the first aspect, in a possible implementation, after the NRF network element receives the network element discovery request, it can obtain the vendor identifier of the analysis logic function network element based on the network element discovery request; then, it can determine the candidate network element list based on the NF configuration files of each network element stored locally; and then send a network element discovery response to the first model training logic function network element, and the network element discovery response includes the candidate network element list. The manner in which the NRF network element determines the candidate network element list can be found in the description of the method embodiment below and is not described in detail here. The interoperability indicator corresponding to the above-mentioned analysis ID of each model training logic function network element in the candidate network element list includes the vendor identifier of the analysis logic function network element. The second model training logic function network element can be any network element in the candidate network element list.

[0027] In the second aspect, the present application provides a model authorization method, which includes: an NRF network element receives a token acquisition request from a first model training logic function network element, the token acquisition request includes an analysis identifier, an identifier of the analysis logic function network element, and an identifier of a second model training logic function network element, and the second model training logic function network element can be used to provide information about a model corresponding to the analysis identifier; the NRF network element verifies whether the supplier identifier of the analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element; if the supplier identifier is included in the interoperability indicator, it means that the analysis logic function network element has the authority to obtain the model from the second model training logic function network element, then the NRF network element sends a first token to the first model training logic function network element, and the first token includes the identifier of the analysis logic function network element and the identifier of the second model training logic function network element.

[0028] Exemplarily, the above-mentioned token acquisition request can be used to request a first token. In other words, the token acquisition request can be used to request (NRF network element) to authorize the analysis logic function network element to obtain the model corresponding to the analysis ID from the second model training logic function network element. In other words, the token acquisition request can be used for the first model training logic function network element to request the first token for (or on behalf of) the analysis logic function network element to obtain the model corresponding to the analysis ID. The first token can be used to indicate that the analysis logic function network element is authorized to obtain information about the model corresponding to the analysis ID from the second model training logic function network element. Alternatively, the first token can be used to indicate that the first model training logic function network element is authorized to obtain information about the model corresponding to the analysis ID from the second model training logic function network element for (or on behalf of) the analysis logic function network element.

[0029] The first model training logic function network element of the present application requests an authorization token (i.e., the above-mentioned first token) from the NRF for (or on behalf of) the analysis logic function network element. The request includes at least the identifier of the analysis logic function network element and the identifier of the second model training logic function network element. After the NRF verifies that the analysis logic function network element has the authority to obtain the model from the second model training logic function network element, it returns the authorization token (i.e., the above-mentioned first token) to the first model training logic function network element; this is beneficial for the first model training logic function network element to obtain the model from the second model training logic function network element for (or on behalf of behalf) the analysis logic function network element, which can improve the security of the model.

[0030] In conjunction with the second aspect, in one possible implementation, before the NRF network element verifies whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element, the method further includes: the NRF network element obtains the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element from the NF configuration file corresponding to the identifier of the second model training logic function network element. The interoperability indicator can be used to indicate a list of NWDAF providers (or vendors) that are allowed to retrieve ML models from the NWDAF containing MTLF.

[0031] In combination with the second aspect, in a possible implementation method, before the NRF network element verifies whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator of the second model training logic function network element, the method also includes: the NRF network element obtains the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

[0032] In combination with the second aspect, in a possible implementation, the above-mentioned token acquisition request also includes one or more of the following: an identifier of the first model training logic function network element, a supplier identifier of the analysis logic function network element, or a first indication information. Among them, the analysis ID can be the identifier of the analysis service corresponding to the model for which authorization is requested. The supplier identifier of the analysis logic function network element can be used by the NRF network element to verify whether it can be authorized to obtain the model corresponding to the analysis ID from the second model training logic function network element. The first indication information can be used to indicate that the first model training logic function network element requests the first token for (or on behalf of behalf) the analysis logic function network element. In other words, the first indication information can be used to indicate that the NRF network element is requested to authorize the analysis logic function network to obtain the model in the second model training logic function network element. Alternatively, the first indication information can be used to indicate that the token acquisition request is an access token acquisition request (Nnrf_AccessToken_Get request) for the model delegation acquisition scenario.

[0033] In combination with the second aspect, in a possible implementation, the above-mentioned first token also includes one or more of the following: the above-mentioned analysis identifier, the identifier of the first model training logic function network element, the supplier identifier of the analysis logic function network element, or the second indication information. The second indication information can be used to indicate that the analysis logic function network element is authorized to obtain the information of the model corresponding to the above-mentioned analysis ID from the second model training logic function network element. Alternatively, the second indication information can be used to indicate that the first model training logic function network element is authorized to obtain the information of the model corresponding to the analysis ID from the second model training logic function network element as (or on behalf of behalf) the analysis logic function network element. Alternatively, the second indication information can be used to indicate that the first token is an authorization token for the model delegation acquisition scenario.

[0034] In conjunction with the second aspect, in a possible implementation, before the NRF network element receives the token acquisition request from the first model training logic function network element, the method further includes: the NRF network element receives a network element discovery request from the first model training logic function network element, the network element discovery request including the analysis identifier, and the identifier of the analysis logic function network element and / or the supplier identifier of the analysis logic function network element; the NRF network element obtains the supplier identifier of the analysis logic function network element based on the network element discovery request; the NRF network element determines a candidate network element list based on the stored NF configuration files of each network element; the NRF network element sends a network element discovery response to the first model training logic function network element, the network element discovery response including the candidate network element list. The interoperability indicator corresponding to the above-mentioned analysis ID of each model training logic function network element in the candidate network element list includes the supplier identifier of the analysis logic function network element. The second model training logic function network element can be any network element in the candidate network element list, so the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element includes the supplier identifier of the analysis logic function network element.

[0035] Exemplarily, the network element discovery request includes an identifier of the analysis logic function network element. The NRF network element obtains the vendor identifier of the analysis logic function network element based on the network element discovery request, including: the NRF network element obtains the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

[0036] On the third aspect, the present application provides a model authorization method, which includes: the second model training logic function network element receives a second model acquisition request from the first model training logic function network element, the second model acquisition request includes an analysis identifier, an identifier of the analysis logic function network element, and a first token, and the second model acquisition request is used to obtain information about the model corresponding to the analysis identifier for the analysis logic function network element; the second model training logic function network element verifies the first token; if the first token verification is successful, the second model training logic function network element sends a model notification message to the analysis logic function network element, and the model notification message includes information about the model corresponding to the analysis identifier. Wherein, the specific verification method of the first token is described in the method embodiment below and is not described in detail here.

[0037] After receiving a request from the first model training logic function network element to obtain a model for or on behalf of the analysis logic function network element, the second model training logic function network element in this application verifies the validity of the token. After the token is verified, the second model training logic function network element sends information about the model that meets the requirements to the analysis logic function network element. This improves model acquisition in the model delegation acquisition scenario and verifies the actual model consumer (i.e., the analysis logic function network element) in the model delegation acquisition scenario, thereby improving the security of the model.

[0038] In combination with the third aspect, in a possible implementation, the above-mentioned second model acquisition request also includes one or more of the following: the identifier of the first model training logic function network element, the address of the analysis logic function network element, the supplier identifier of the analysis logic function network element, or a third indication information. Among them, the address of the analysis logic function network element can be carried by the subscription endpoint address. The subscription endpoint address can be used to represent the address for receiving model notification messages. The third indication information can be used to indicate that the first model training logic function network element obtains information about the model corresponding to the above-mentioned analysis ID for (or on behalf of behalf) the analysis logic function network element. Alternatively, the third indication information can be used to indicate that the second model acquisition request is an ML model provision request (Nnwdaf_MLModelProvision request) for the model delegation acquisition scenario.

[0039] In combination with the third aspect, in a possible implementation, the above-mentioned model notification message also includes an identifier of the ADRF network element that stores the model.

[0040] In conjunction with the third aspect, in one possible implementation, if the first token verification passes, the second model training logic function network element may add the identifier of the analysis logic function network element to an allowed NF consumer list (allowed NFc list). The allowed NFc list is associated with the model corresponding to the analysis ID. In other words, the allowed NFc list is a list of network function instance identifiers that are allowed to obtain / retrieve the model corresponding to the analysis ID.

[0041] After the first token verification is passed, this application adds the identification of the analysis logic function network element to the allowed NF consumer list, which is conducive to the subsequent analysis logic function network element to obtain the model again.

[0042] In a fourth aspect, the present application provides a model authorization method, which includes: a source analysis logic function network element receives an analysis context transfer request from a target analysis logic function network element, the analysis context transfer request includes an analysis context identifier, and the analysis context transfer request is used to request the transfer of information of a first model in the analysis context; the source analysis logic function network element sends a token acquisition request to an NRF network element, the token acquisition request includes an analysis identifier corresponding to the first model, an identifier of the target analysis logic function network element, and an identifier of a model training logic function network element, and the model training logic function network element is used to provide information of the first model; the source analysis logic function network element receives a first token from the NRF network element, the first token includes an identifier of the target analysis logic function network element, and an identifier of the model training logic function network element; the source analysis logic function network element sends a model acquisition request to the model training logic function network element, the model acquisition request includes the analysis identifier, the identifier of the target analysis logic function network element, and the first token, and the model acquisition request is used to obtain information of the first model for the target analysis logic function network element.

[0043] It can be understood that the identifier of the above analysis context can be a subscription correlation identifier (Subscription Correlation ID). The subscription correlation identifier can be used to identify the analysis subscription requesting the relevant analysis context. The analysis context can include model-related information, such as: the identifier of the model producer / provider / trainer (such as NWDAF containing MTLF), the identifier of the model, the file address of the model, the analysis identifier of the model, etc. It can be understood that the analysis context can include the identifiers of multiple model producers / providers / trainers (such as NWDAF containing MTLF). For the sake of simplicity, this application takes a model producer / provider / trainer that provides a model as an example for explanation. The above analysis context transfer request can be used to request the transfer of information of one or more models in the analysis context. For the sake of simplicity, the embodiment of the present application takes the example of the analysis context transfer request being used to request the transfer of information of the first model in the above analysis context for explanation.

[0044] Exemplarily, the first token in the above-mentioned model acquisition request can be used to verify the model acquisition request. Alternatively, the first token in the model acquisition request can be used to verify the various identifiers in the model acquisition request. Alternatively, the first token in the model acquisition request can be used to verify the target analysis logic function network element's permission to obtain the above-mentioned first model from the model training logic function network element. Alternatively, the first token in the model acquisition request can be used to verify that the source analysis logic function network element is (or acts on behalf of) the target analysis logic function network element's permission to obtain the above-mentioned first model from the model training logic function network element.

[0045] Exemplarily, the source analysis logic function network element may be a network element having an analysis logic function network element, such as a source NWDAF network element containing AnLF (SourceNWDAF containing AnLF), which may be referred to as source AnLF (Source AnLF). The target analysis logic function network element may be another network element having an analysis logic function network element, such as a target NWDAF network element containing AnLF (TargetNWDAF containing AnLF), which may be referred to as target AnLF (Target AnLF). The model training logic function network element may be a network element having a model training logic function, such as a NWDAF network element containing MTLF (NWDAF containing MTLF), which may be referred to as MTLF.

[0046] Exemplarily, the token acquisition request can be used to request a first token. In other words, the token acquisition request can be used to request (NRF network element) to authorize the target analysis logic function network element to obtain the above-mentioned first model from the model training logic function network element. In other words, the token acquisition request can be used by the source analysis logic function network element to request the first token for (or on behalf of) the target analysis logic function network element to obtain the above-mentioned first model. The first token can be used to indicate that the target analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element. Alternatively, the first token can be used to indicate that the source analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element for (or on behalf of) the target analysis logic function network element.

[0047] After receiving the analysis context transfer request from the target analysis logic function network element, the source analysis logic function network element of the present application requests an authorization token (i.e., the above-mentioned first token) from the NRF for (or on behalf of) the target analysis logic function network element. The request includes the identifier of the target analysis logic function network element and the identifier of the model training logic function network element. After obtaining the authorization token, the source analysis logic function network element requests the model training logic function network element for (or on behalf of) the target analysis logic function network element. The request includes the identifier and authorization token of the target analysis logic function network element. This helps to realize model authorization and model acquisition in the model delegation acquisition scenario, and authorizes the actual model consumer (i.e., the target analysis logic function network element) in the model delegation acquisition scenario, thereby improving the security of the model.

[0048] In conjunction with the fourth aspect, in one possible implementation, after receiving the analysis context transfer request, the source analysis logic function network element may return analysis context information to the target analysis logic function network element. The analysis context information may include other information in addition to the model-related information in the analysis context transfer request, such as an active data source identifier (ID), a subscription association identifier, etc.

[0049] Exemplarily, model-related information includes, but is not limited to: the identifier of the model producer / provider / trainer (such as NWDAF containing MTLF), the identifier of the model, the file address of the model, or the analysis identifier corresponding to the model.

[0050] In conjunction with the fourth aspect, in one possible implementation, the analysis context transfer request further includes one or more of the following: a second operation indicator of the target analysis logic function network element, or a vendor identifier of the target analysis logic function network element. The second operation indicator may be used to indicate models that the target analysis logic function network element can use and the providers / producers of these models.

[0051] For example, if the operation indicator corresponds to AnLF, that is, one AnLF corresponds to one operation indicator. The operation indicator may include a list of vendor identifiers (or vendor list), or be described as a list of NWDAF providers (or suppliers). AnLF allows retrieval or use of models provided by vendors in the vendor list. The operation indicator also indicates that AnLF supports the use of vendor-provided models for network elements (e.g., NWDAF) of vendors in the vendor list. That is, the operation indicator applies to each analysis identifier.

[0052] Exemplarily, if the operation indicator corresponds to the analysis identifier, that is, an AnLF can correspond to one or more operation indicators. Different operation indicators can correspond to different analysis identifiers. Among them, the operation indicator corresponding to an analysis identifier includes a manufacturer identifier list (or manufacturer list), or is described as a NWDAF provider (or supplier) list. AnLF allows the retrieval or use of models provided by the manufacturers in the manufacturer list. The operation indicator also indicates that AnLF supports the use of models provided by the manufacturer for the network elements (such as NWDAF) of the manufacturers in the manufacturer list, and these models correspond to a certain analysis identifier.

[0053] This application provides a new operation indicator for indicating which models of which manufacturers are supported by the analysis logic function network element, which is helpful for subsequent verification of whether the actual consumers of the model have the ability or authority to use the specified model.

[0054] In conjunction with the fourth aspect, in one possible implementation, before the source analysis logic function network element sends a token acquisition request to the NRF network element, the method further includes: the source analysis logic function network element determining that the provider of the first model is located in the second operation indicator of the target analysis logic function network element. In other words, the source analysis logic function network element can confirm that the target analysis logic function network element has the ability or permission to use the model in the analysis context.

[0055] In conjunction with the fourth aspect, in one possible implementation, before the source analysis logic function network element sends a token acquisition request to the NRF network element, the method further includes: the source analysis logic function network element determining that a vendor identifier of the target analysis logic function network element is included in an interoperability indicator of the information provider of the first model (or the producer of the first model, such as the MTLF). In other words, the source analysis logic function network element can determine whether the target analysis logic function network element can retrieve / acquire the first model.

[0056] In conjunction with the fourth aspect, in one possible implementation, before the source analysis logic function network element receives the analysis context transfer request from the target analysis logic function network element, the method further includes: the source analysis logic function network element sends a network element registration request to the NRF network element, where the network element registration request includes a first operation indicator of the source analysis logic function network element. The first operation indicator can be used to indicate models that the source analysis logic function network element can use and the providers / producers of these models.

[0057] The source analysis logic function network element of the present application registers its own operation indicator with the NRF network element in the network element registration process, which is beneficial to the subsequent verification of the subsequent NRF network element.

[0058] In combination with the fourth aspect, in a possible implementation, the above-mentioned token acquisition request also includes one or more of the following: an identifier of the source analysis logic function network element, a supplier identifier of the target analysis logic function network element, an identifier of the above-mentioned first model, or a first indication information. The supplier identifier of the target analysis logic function network element can be used by the NRF network element to verify whether it can be authorized to obtain the above-mentioned first model from the model training logic function network element. The first indication information can be used to indicate that the source analysis logic function network element requests the first token for (or on behalf of behalf) the target analysis logic function network element. In other words, the first indication information can be used to indicate that the NRF network element is requested to authorize the target analysis logic function network to obtain the first model in the model training logic function network element. Alternatively, the first indication information can be used to indicate that the token acquisition request is an access token acquisition request (Nnrf_AccessToken_Get request) for a model delegation acquisition scenario.

[0059] Including the vendor identifier of the target analysis logic function network element in the token acquisition request in this application is helpful for the NRF network element to verify whether the target analysis logic function network element has the authority to obtain the first model from the model training logic function network element. Including the first indication information in the token acquisition request in this application can make the meaning of the token acquisition request clearer and enable the NRF network element to clarify its own behavior.

[0060] In combination with the fourth aspect, in a possible implementation method, after the NRF network element receives the token acquisition request from the source analysis logic function network element, it can verify whether the supplier identifier of the target analysis logic function network element is included in the interoperability indicator corresponding to the above analysis identifier of the model training logic function network element. In other words, after the NRF network element receives the above token acquisition request, it can verify whether the target analysis logic function network element has the authority to obtain the above first model from the model training logic function network element. If the supplier identifier is included in the interoperability indicator, it means that the target analysis logic function network element has the authority to obtain the first model from the model training logic function network element, and the NRF network element sends a first token to the source analysis logic function network element. The first token includes the identifier of the target analysis logic function network element and the identifier of the model training logic function network element.

[0061] In combination with the fourth aspect, in a possible implementation, the above-mentioned first token also includes one or more of the following: the above-mentioned analysis identifier, the identifier of the source analysis logic function network element, the supplier identifier of the target analysis logic function network element, the identifier of the above-mentioned first model, or the second indication information. The second indication information can be used to indicate that the target analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element. Alternatively, the second indication information can be used to indicate that the source analysis logic function network element is authorized to obtain the information of the first model from the model training logic function network element for (or on behalf of) the target analysis logic function network element. Alternatively, the second indication information can be used to indicate that the first token is an authorization token for the model delegation acquisition scenario.

[0062] The present application includes the second indication information in the first token, which can make the meaning of the first token clearer and help the model training logic function network element to clarify the role of the first token.

[0063] In combination with the fourth aspect, in a possible implementation, the above-mentioned model acquisition request also includes one or more of the following: the identifier of the source analysis logic function network element, the address of the target analysis logic function network element, the identifier of the above-mentioned first model, the supplier identifier of the target analysis logic function network element, or a third indication information. Among them, the address of the target analysis logic function network element can be carried by the subscription endpoint address. The subscription endpoint address can be used to represent the address for receiving model notification messages. The third indication information can be used to indicate that the source analysis logic function network element obtains the information of the above-mentioned first model for (or on behalf of behalf) the target analysis logic function network element. Alternatively, the third indication information can be used to indicate that the model acquisition request is an ML model provision request (Nnwdaf_MLModelProvision request) for the model delegation acquisition scenario.

[0064] The present application includes the address of the target analysis logic function network element in the model acquisition request, which is beneficial for the model training logic function network element to determine the recipient of the model notification message. The present application includes the supplier identifier of the target analysis logic function network element in the model acquisition request, which is beneficial for the model training logic function network element to verify whether the target analysis logic function network element has the authority to obtain the model from itself. The present application includes the third indication information in the model acquisition request, which can make the meaning of the model acquisition request clearer, so that the model training logic function network element can clarify its own behavior.

[0065] In conjunction with the fourth aspect, in a possible implementation method, after the model training logic function network element receives the model acquisition request, it can verify the first token in the model acquisition request. For the specific verification method, please refer to the description of the method embodiment below, which is not described in detail here. If the first token verification is successful, the model training logic function network element can send a model notification message to the target analysis logic function network element. The model notification message may include information about the above-mentioned first model, such as: the identifier of the first model, the address of the first model in the model training logic function network element, or the model file of the first model, etc. Exemplarily, the model notification message may also include the identifier of the ADRF network element that stores the first model.

[0066] In a fifth aspect, the present application provides a model authorization method, which includes: an NRF network element receives a token acquisition request from a source analysis logic function network element, the token acquisition request includes an analysis identifier, an identifier of a target analysis logic function network element, and an identifier of a model training logic function network element, and the model training logic function network element is used to provide information of a first model corresponding to the analysis identifier; the NRF network element verifies whether the supplier identifier of the target analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the model training logic function network element; if the supplier identifier is included in the interoperability indicator, the NRF network element sends a first token to the source analysis logic function network element, and the first token includes the identifier of the target analysis logic function network element and the identifier of the model training logic function network element.

[0067] Exemplarily, the token acquisition request can be used to request a first token. In other words, the token acquisition request can be used to request (NRF network element) to authorize the target analysis logic function network element to obtain the above-mentioned first model from the model training logic function network element. In other words, the token acquisition request can be used by the source analysis logic function network element to request the first token for (or on behalf of) the target analysis logic function network element to obtain the above-mentioned first model. The first token can be used to indicate that the target analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element. Alternatively, the first token can be used to indicate that the source analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element for (or on behalf of) the target analysis logic function network element.

[0068] The NRF network element of the present application receives a token acquisition request from the source analysis logic function network element, which includes the identifier of the target analysis logic function network element and the identifier of the model training logic function network element. After the NRF verifies that the target analysis logic function network element has the authority to obtain the first model from the model training logic function network element, it returns the authorization token (i.e. the above-mentioned first token) to the source analysis logic function network element; this is beneficial for the source analysis logic function network element to obtain the model from the model training logic function network element for (or on behalf of) the target analysis logic function network element, which can improve the security of the model.

[0069] In combination with the fifth aspect, in a possible implementation method, before the NRF network element verifies whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator of the model training logic function network element, the method also includes: the NRF network element obtains the interoperability indicator corresponding to the analysis identifier of the model training logic function network element from the NF configuration file corresponding to the identifier of the model training logic function network element.

[0070] In combination with the fifth aspect, in a possible implementation method, before the NRF network element verifies whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator of the model training logic function network element, the method also includes: the NRF network element obtains the vendor identifier of the target analysis logic function network element from the NF configuration file corresponding to the identifier of the target analysis logic function network element.

[0071] In combination with the fifth aspect, in a possible implementation, the above-mentioned token acquisition request also includes one or more of the following: an identifier of the source analysis logic function network element, a supplier identifier of the target analysis logic function network element, an identifier of the above-mentioned first model, or a first indication information. The supplier identifier of the target analysis logic function network element can be used by the NRF network element to verify whether it can be authorized to obtain the above-mentioned first model from the model training logic function network element. The first indication information can be used to indicate that the source analysis logic function network element requests the first token for (or on behalf of behalf) the target analysis logic function network element. In other words, the first indication information can be used to indicate that the NRF network element is requested to authorize the target analysis logic function network element to obtain the first model in the model training logic function network element. Alternatively, the first indication information can be used to indicate that the token acquisition request is an access token acquisition request (Nnrf_AccessToken_Get request) for a model delegation acquisition scenario.

[0072] In combination with the fifth aspect, in a possible implementation, the above-mentioned first token also includes one or more of the following: the above-mentioned analysis identifier, the identifier of the source analysis logic function network element, the supplier identifier of the target analysis logic function network element, the identifier of the above-mentioned first model, or the second indication information. The second indication information can be used to indicate that the target analysis logic function network element is authorized to obtain the information of the above-mentioned first model from the model training logic function network element. Alternatively, the second indication information can be used to indicate that the source analysis logic function network element is authorized to obtain the information of the first model from the model training logic function network element for (or on behalf of) the target analysis logic function network element. Alternatively, the second indication information can be used to indicate that the first token is an authorization token for the model delegation acquisition scenario.

[0073] In conjunction with the fifth aspect, in one possible implementation, before the NRF network element receives a token acquisition request from the source analysis logic function network element, the method further includes: the NRF network element receives a first network element registration request from the source analysis logic function network element, where the first network element registration request includes a first operation indicator of the source analysis logic function network element. The first operation indicator is used to indicate models that the source analysis logic function network element can use and providers / producers of these models.

[0074] In conjunction with the fifth aspect, in one possible implementation, before the NRF network element receives a token acquisition request from the source analysis logic function network element, the method further includes: the NRF network element receives a second network element registration request from the target analysis logic function network element, where the second network element registration request includes a second operation indicator of the target analysis logic function network element. The second operation indicator is used to indicate the providers of models that can be used by the target analysis logic function network element and the providers / producers of these models.

[0075] In the sixth aspect, the present application provides a model authorization method, which includes: a model training logic function network element receives a model acquisition request from a source analysis logic function network element, the model acquisition request includes an analysis identifier, an identifier of a target analysis logic function network element, and a first token, and the model acquisition request is used to obtain information about a first model corresponding to the analysis identifier for the target analysis logic function network element; the model training logic function network element verifies the first token; if the first token verification is successful, the model training logic function network element sends a model notification message to the target analysis logic function network element, and the model notification message includes information about the above-mentioned first model. Wherein, the verification method of the first token refers to the description of the method embodiment below and is not described in detail here.

[0076] After the model training logic function network element of this application receives a request from the source analysis logic function network element to obtain the first model for or on behalf of the target analysis logic function network element, it verifies the validity of the token. After the token is verified, the information of the first model is sent to the target analysis logic function network element. This improves the model acquisition in the model delegation acquisition scenario and verifies the actual model consumer (i.e., the target analysis logic function network element) in the model delegation acquisition scenario, thereby improving the security of the model.

[0077] In combination with the sixth aspect, in a possible implementation, the above-mentioned model acquisition request also includes one or more of the following: the identifier of the source analysis logic function network element, the address of the target analysis logic function network element, the identifier of the above-mentioned first model, the supplier identifier of the target analysis logic function network element, or a third indication information. Among them, the address of the target analysis logic function network element can be carried by the subscription endpoint address. The subscription endpoint address can be used to represent the address for receiving model notification messages. The third indication information can be used to indicate that the source analysis logic function network element obtains the information of the above-mentioned first model for (or on behalf of behalf) the target analysis logic function network element. Alternatively, the third indication information can be used to indicate that the model acquisition request is an ML model provision request (Nnwdaf_MLModelProvision request) for the model delegation acquisition scenario.

[0078] In combination with the sixth aspect, in a possible implementation, the above-mentioned model notification message also includes an identifier of the ADRF network element that stores the first model.

[0079] In conjunction with the sixth aspect, in one possible implementation, if the first token verification passes, the model training logic function network element may add the identifier of the target analysis logic function network element to an allowed NF consumer list (allowed NFc list). The allowed NFc list is associated with the first model. In other words, the allowed NFc list is a list of network function instance identifiers that are allowed to obtain / retrieve the first model.

[0080] After the first token verification is passed, this application adds the identifier of the target analysis logic function network element to the allowed NF consumer list, which is conducive to the subsequent target analysis logic function network element to obtain the first model again.

[0081] In the seventh aspect, the present application provides a communication device, which may be a first model training logic function network element, or a second model training logic function network element, or an NRF network element, or a chip therein. The communication device includes a unit and / or module for executing the method provided by any one of the first to third aspects, or any possible implementation of any one of the aspects, such as a transceiver unit and / or a processing unit. The transceiver unit is used to send and receive various information or signaling, and thus can also achieve the beneficial effects (or advantages) of the method provided by any one of the first to third aspects.

[0082] In an eighth aspect, the present application provides a communication device, which may be a source analysis logic function network element, a target analysis logic function network element, an NRF network element, or a chip therein. The communication device includes a unit and / or module for executing the method provided by any one of the fourth to sixth aspects, or any possible implementation of any one of the aspects, such as a transceiver unit and / or a processing unit. The transceiver unit is used to send and receive various information or signaling, and thus can also achieve the beneficial effects (or advantages) of the method provided by any one of the fourth to sixth aspects.

[0083] In a ninth aspect, the present application provides a communication device, comprising a processor configured to execute the method described in any one of the first to third aspects, or any possible implementation thereof. Alternatively, the processor is configured to execute a program stored in a memory, and when the program is executed, the method described in any one of the first to third aspects, or any possible implementation thereof, is executed.

[0084] In combination with the ninth aspect, in a possible implementation, the memory is located outside the above-mentioned communication device.

[0085] In combination with the ninth aspect, in a possible implementation, the memory is located within the above-mentioned communication device.

[0086] In the present application, the processor and the memory may also be integrated into one device, that is, the processor and the memory may also be integrated together.

[0087] In combination with the ninth aspect, in one possible implementation, the communication device also includes a transceiver, which is used to send or receive various information, such as: receiving a first model acquisition request, sending and receiving a token acquisition request, sending and receiving a second model acquisition request, sending and receiving a first token, and so on.

[0088] In a tenth aspect, the present application provides a communication device, comprising a processor configured to execute the method described in any one of the fourth to sixth aspects, or any possible implementation thereof. Alternatively, the processor is configured to execute a program stored in a memory, and when the program is executed, the method described in any one of the first to sixth aspects, or any possible implementation thereof, is executed.

[0089] In combination with the tenth aspect, in a possible implementation, the memory is located outside the above-mentioned communication device.

[0090] In combination with the tenth aspect, in a possible implementation, the memory is located within the above-mentioned communication device.

[0091] In the present application, the processor and the memory may also be integrated into one device, that is, the processor and the memory may also be integrated together.

[0092] In combination with the tenth aspect, in one possible implementation, the communication device also includes a transceiver, which is used to send or receive various information, such as: receiving a context transfer request, sending and receiving a token acquisition request, sending and receiving a model acquisition request, sending and receiving a first token, and so on.

[0093] In the eleventh aspect, the present application provides a communication device, which may include a processor and an interface circuit, and the processor is connected to the interface circuit. Wherein, the interface circuit is used to interact (or receive and send or input and output) information or data, and the processor is used to run program instructions so that the communication device performs the method described in any one of the first to sixth aspects above, or any possible implementation of any one of them. Wherein, the interface circuit may be a communication interface, or a transceiver. The transceiver may be a radio frequency module in a communication device, or a combination of a radio frequency module and an antenna, or an input and output interface of a chip or circuit.

[0094] In the twelfth aspect, the present application provides a readable storage medium having program instructions stored thereon, which, when executed on a communication device, enables the communication device to execute the model authorization method described in any one of the above-mentioned first to sixth aspects, or any possible implementation method of any one of the aspects.

[0095] In a thirteenth aspect, the present application provides a program product comprising instructions, which, when executed, enables the model authorization method described in any possible implementation of any of the first to sixth aspects to be executed.

[0096] In a fourteenth aspect, the present application provides a communication device, which can be implemented in the form of a chip or in the form of a device, and the device includes a processor. The processor is used to read and execute a program stored in a memory to execute one or more of any aspects of the first to sixth aspects, or a model authorization method provided by one or more of any possible implementation methods of any aspect. Optionally, the device also includes a memory, which is connected to the processor via a circuit. Further optionally, the device also includes a communication interface, and the processor is connected to the communication interface. The communication interface is used to receive information and / or signaling to be processed, and the processor obtains the information and / or signaling from the communication interface, processes the information and / or signaling, and outputs the processing results through the communication interface. The communication interface can be an input and output interface.

[0097] Optionally, the processor and memory may be physically independent units, or the memory may be integrated with the processor.

[0098] In a fifteenth aspect, the present application provides a communication system, which includes the above-mentioned first model training logic function network element and NRF network element, and optionally includes a second model training logic function network element and / or an analysis logic function network element.

[0099] In the sixteenth aspect, the present application provides a communication system, which includes the above-mentioned source analysis logic function network element and NRF network element, and optionally includes a model training logic function network element and / or a target analysis logic function network element.

[0100] The technical effects achieved in the above-mentioned aspects can be referred to each other or to the beneficial effects in the method embodiments shown below, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] FIG1 is a schematic diagram of a system architecture provided by an embodiment of the present application;

[0102] FIG2 is a schematic diagram of a secure and authorized AI / ML model sharing process provided by an embodiment of the present application;

[0103] FIG3 is a flow chart of a model authorization method provided in an embodiment of the present application;

[0104] FIG4 is another flow chart of the model authorization method provided in an embodiment of the present application;

[0105] FIG5 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0106] FIG6 is another schematic structural diagram of a communication device provided in an embodiment of the present application;

[0107] FIG7 is another structural diagram of a communication device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0108] In the description of this application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or "one or more 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, or c can mean: a, b, c; a and b; a and c; b and c; or a, b, and c. Among them, a, b, and c can be single or multiple.

[0109] In the description of this application, words such as "first" and "second" are used only to distinguish different objects and do not limit the quantity or execution order. Moreover, words such as "first" and "second" do not necessarily mean different. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to the process, method, product, or device.

[0110] In this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described in this application as "exemplary," "for example," or "for example" should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "for example," or "for example" is intended to present the relevant concepts in a concrete way.

[0111] It should be understood that in this application, "when", "if" and "if" all mean that the device will perform corresponding processing under certain objective circumstances, and do not limit the time. It does not require that the device must perform a judgment action when it is implemented, nor does it mean that there are other limitations.

[0112] Elements used in the singular herein are intended to mean "one or more" rather than "one and only one" unless specifically stated otherwise.

[0113] Additionally, the terms "system" and "network" are often used interchangeably herein.

[0114] It is understood that in the various embodiments of the present application, expressions such as "A corresponds to B," "A corresponds to / is associated with B," and the like all indicate that there is a corresponding relationship between A and B, and that B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information.

[0115] The following is a brief description of the network elements and system architecture involved in this application.

[0116] The network data analytics function (NWDAF) network element supports data collection from other network function and application function (AF) network elements, supports data collection from operation administration and maintenance (OAM) network elements, and supports the provision of analysis information to other network function and AF network elements.

[0117] The NWDAF network element has functions such as data collection, model training, data analysis, and model reasoning. The NWDAF network element can be used to collect relevant data from network function network elements, third-party service servers, terminal devices, or network management systems, perform data analysis based on the relevant data to obtain analysis results, and provide analysis results to the network function network elements, third-party service servers, terminal devices, or network management systems. The analysis results can assist the network in selecting service quality parameters for the service, or assist the network in executing traffic routing, or assist the network in selecting background data transmission strategies, etc. In addition, the NWDAF network element can also be used to collect relevant data from network function network elements, third-party service servers, terminal devices, or network management systems, and perform model training based on the relevant data to obtain an artificial intelligence (AI) model or a machine learning (ML) model, and provide the AI ​​model / ML model to other NWDAF network elements. The AI ​​model / ML model can be used to assist the NWDAF network element in generating data analysis results based on the relevant data.

[0118] NWDAF network elements containing AnLF can be used to perform inference and derive analytical information (i.e., derive statistics and / or predictions based on analytics consumer requests) and expose analytical services. NWDAF network elements containing MTLF can be used to train machine learning (ML) models and expose new training services; for example, providing trained ML models.

[0119] For the convenience of description in this application, "NWDAF network element containing AnLF (NWDAF containing AnLF)" can be simply expressed as "AnLF". In some cases, the two can be used interchangeably; "NWDAF network element containing MTLF (NWDAF containing MTLF)" can be simply expressed as "MTLF". In some cases, the two can be used interchangeably.

[0120] The analytics data repository functional (ADRF) network element can provide storage, deletion and retrieval services for AI / ML models. The ADRF network element can also provide storage and retrieval services for data, as well as storage and retrieval services for analysis. For example, the consumer sends a data management storage request (Nadrf Data Management Storage Request) to the ADRF network element, which contains the data or analysis that needs to be stored; or, the consumer sends a data management storage subscription request (Nadrf Data Management Storage Subscription Request) to the ADRF network element, requesting the ADRF network element to subscribe to the data or analysis for storage, and then the ADRF network element subscribes to the NWDAF network element or the data collection control function (DCCF) network element to obtain the data or analysis. These analyses or data can be provided as notifications using the DCCF data management (Ndccf_DataManagement) service, the NWDAF data management (Nnwdaf_DataManagement) service, or the messaging framework adaptor function (MFAF) data management (Nmfaf_3caDataManagement) service.

[0121] The network repository function (NRF) network element can provide registration and discovery functions, enabling network functions (NFs) to discover each other and communicate through application programming interfaces (APIs).

[0122] The technical solution provided in this application can be applied to various communication systems deployed with virtual network functions. For example: fifth generation (5G) communication systems or new radio (NR), equipped with network functions virtualization infrastructure (NFVI) or other long term evolution (LTE) networks with virtual network functions, MulteFire networks (creating new wireless networks by independently running LTE technology on unlicensed spectrum (such as the global 5GHz unlicensed spectrum)), or home base station networks, mobile networks with wireless fidelity (Wi-Fi) access, wideband code division multiple access (WCDMA) networks, fixed-mobile converged networks (fixed access networks access mobile networks), and other future communication systems, such as sixth generation mobile communication systems.

[0123] For example, the technical solution provided in this application can be applied to the 5G system architecture defined in the 3rd Generation Partnership Project Technical Specifications (3GPP TS) 23.288.

[0124] Referring to Figure 1 , which is a schematic diagram of the system architecture provided by an embodiment of the present application, the system architecture 100 includes, but is not limited to, an access network, a core network, a data network (DN) 140, and an application function (AF) 141. The access network, which can be used to implement functions related to wireless access, may include a radio access network (RAN) 120 and user equipment (UE) 110. The core network may include but is not limited to the following logical functions: user plane function (UPF) 130, network exposure function (NEF) 131, network repository function (NRF) 132, policy control function (PCF) 133, unified data management (UDM) function 134, unified data repository (UDR) function 135, network data analytics function (NWDAF) 136, authentication server function (AUSF) 137, access and mobility management function (AMF) 138, or session management function (SMF) 139, etc. It can be understood that "Nnef", "Nnrf", "Npcf", "Nudm", "Nudr", "Nnwdaf", "Naf", "Nausf", "Namf", and "Nsmf" in Figure 1 represent the names of service-oriented interfaces. For details, please refer to the relevant description in the 3GPP standard protocol, which is not explained in detail here.

[0125] Optionally, the UE may access the data network by establishing a session from the UE to the RAN, then to the UPF, and then to the data network (DN), ie, a protocol data unit (PDU) session (PDU session).

[0126] A UE can be a terminal device, such as a mobile phone, IoT terminal device, smart terminal, vehicle-mounted terminal, vehicle-mounted device, wearable device, multimedia device, streaming media device, etc. The RAN provides wireless access for terminal devices and includes, but is not limited to, 5G base stations (Next-Generation Node B, gNB), wireless base stations (evolved Node B, eNodeB or eNB) in LTE networks, wireless fidelity access points (Wi-Fi APs), worldwide interoperability for microwave access base stations (WiMAX BSs), and relay stations. In the 5G RAN architecture, a gNB can include a centralized unit (CU) and a distributed unit (DU). The gNB can also include a radio unit (RU). The CU and DU can be understood as a logical functional division of the base station. The CU and DU can be physically separated or deployed together. For example, multiple DUs can share a CU, or a single DU can be connected to multiple CUs. The CU and DU can be connected via the F1 interface.

[0127] AMF is mainly responsible for mobility management in mobile networks, such as user location update, user network registration, user switching, etc. SMF is mainly responsible for session management in mobile networks, such as session establishment, modification, and release. Specific functions include allocating Internet Protocol (IP) addresses to users and selecting UPFs that provide message forwarding functions. PCF is responsible for providing policies to AMF and SMF, such as Quality of Service (QoS) policies and slice selection policies. UDM can be used to store user data, such as contract information, authentication or authorization information. UPF is mainly responsible for processing user messages, such as forwarding and billing.

[0128] It is understood that the various network functions shown in Figure 1 can refer to relevant protocols or standards, etc., and this application does not expand on them. It should also be understood that N1, N2, N3, N4, N6, etc. shown in Figure 1 are all interface sequence numbers. For example, the meaning of the above interface sequence numbers can be found in the meaning defined in the 3GPP standard protocol, and this application does not limit the meaning of the above interface sequence numbers.

[0129] To better understand the technical solutions of the embodiments of the present application, the following briefly introduces the detailed procedure for secured and authorized AI / ML model sharing between different vendors.

[0130] See Figure 2, which is a schematic diagram of the secure and authorized AI / ML model sharing process provided by an embodiment of the present application. As shown in Figure 2, the secure and authorized AI / ML model sharing process between different suppliers includes but is not limited to:

[0131] Step 0a, registering NWDAF containing MTLF, that is, NWDAF containing MTLF registers NWDAF containing MTLF with NRF. Specifically, NWDAF containing MTLF sends an analysis ID (Analytics ID) and an interoperability indicator (Interoperability indicator) to NRF. Among them, the interoperability indicator is specifically a model interoperability indicator (such as: ML model interoperability indicator, ML Model interoperability indicator). The NF service producer (NFp), that is, NWDAF containing MTLF, uses the interoperability indicator to register its network function (NF) profile in the NRF network element according to the analysis ID. Among them, the interoperability indicator can be used to indicate a list of NWDAF providers (or vendors) that are allowed to retrieve ML models from NWDAF containing MTLF.

[0132] Step 0b: Register NWDAF containing AnLF, that is, NWDAF containing AnLF registers NWDAF containing AnLF with NRF. Specifically, NWDAF containing AnLF sends the analysis ID and AnLF vendor ID to NRF.

[0133] Step 0c, conditionally encrypt the ML model. Unless the AI / ML model generator (NWDAF containing AnLF) and the storage platform (ADRF) are part of the same system and belong to the same vendor and operator security domain, the model will be stored in an encrypted format.

[0134] Step 1: The NWDAF containing MTLF sends a model storage request (e.g., Nadrf_MLModelManagement_StorageRequest) to the ADRF. The model storage request includes one or more of the following: the NWDAF containing MTLF ID, the model ID, the model address in the NWDAF containing MTLF (e.g., uniform resource locator 1 (URL1)), and the allowed NFc list. The allowed NFc list indicates the list of network function instance identifiers (NF instance IDs) that are allowed to obtain / retrieve / access / query the model.

[0135] Step 2: ADRF sends a response to NWDAF containing MTLF. The response includes the model ID and the model address in ADRF (such as URL2).

[0136] Step 3: Discovery of NWDAF containing MTLF. The NF service consumer (NFc), such as NWDAF containing AnLF, sends a network element discovery request (e.g., Nnrf_NFDiscovery_Request) to the NRF to select an appropriate NF service provider (NFp), such as NWDAF containing MTLF. The network element discovery request includes the analysis ID.

[0137] In step 4a, the NF service consumer (NFc), such as the NWDAF containing AnLF, sends a token acquisition request (e.g., Nnrf_AccessToken_Get request) to the NRF to request an access token from the NRF. The token acquisition request includes, but is not limited to, the vendor ID and / or analysis ID of the NWDAF containing AnLF.

[0138] In step 4b, the NRF checks whether the NWDAF containing AnLF is authorized to access the service requested in the NWDAF containing MTLF, verifies whether the vendor ID of the NF service consumer (NFc) is contained in the NWADF containing the interoperability indicator of the MTLF analysis ID, and grants token 1 (token1) based on the vendor ID provided by the NF service consumer (NFc) during registration.

[0139] In step 5, the NF service consumer (NFc) sends a model provision request (e.g., Nnwdaf_MLModelProvision request) to the NWDAF containing MTLF to retrieve the ML model corresponding to the analysis ID. The model provision request contains one or more of the following: analysis ID, vendor ID, and token 1.

[0140] In step 6a, the NFp (e.g., NWDAF containing MTLF) authorizes the NFc and stores the NFc ID. The NWDAF containing MTLF authenticates the NF service consumer (NFc) and verifies the access token 1, ensuring that the analysis ID is included in the access token 1. If the verification is successful, the NWDAF containing MTLF determines the ML model to be shared for the requested analysis ID and stores the NF instance ID of the NWDAF containing AnLF as part of the allowed NFc list corresponding to the ML model.

[0141] In step 6b, if the ML model to be shared is determined to be stored in the ADRF and the NF service consumer (NFc) is not yet in the allowed NFc list stored in the ADRF, the NWDAF containing MTLF triggers an update at the ADRF by sending a model storage update message (e.g., Nadrf_MLModelManagement_StorageRequest) to the ADRF. The model storage update message contains one or more of the following: the NWDAF containing MTLF ID, the model ID, or the allowed NFc list. The ADRF then stores the allowed NFc list for the ML model corresponding to the model ID.

[0142] In step 6c, ADRF sends a response to NWDAF containing MTLF, which includes the model ID.

[0143] In step 7, the NWDAF containing MTLF sends a model provision response (e.g., Nnwdaf_MLModelProvision Response) to the NF service consumer (NFc). The model provision response contains the model ID and / or the address of the ML model to be shared. This address can be the address where the ML model is stored in the NWDAF containing MTLF (i.e., URL1), or the address where the ML model is stored in the Active Directory Response Framework (ADRF) (i.e., URL2). If the model is stored in the ADRF, this message may also contain the AD RFID.

[0144] In step 8a, the NF service consumer (NFc) requests an access token from the NRF to authorize it to retrieve the model stored in the ADRF.

[0145] In step 8b, the NRF verifies whether the NF service consumer (NFc), such as NWDAF containing AnLF, is authorized to access the model provided by ADRF. If the verification is successful, the NRF will grant token 2 based on the information provided in the NF profile of ADRF.

[0146] In step 9, the NF service consumer (NFc), such as NWDAF containing AnLF, retrieves the target model by sending a model retrieval / acquisition request (e.g., Nadrf_MLModelManagement_Retrieval Request). The model retrieval / acquisition request includes the analysis ID and / or token2.

[0147] In step 10, ADRF authenticates the NF service consumer (NFc) and verifies the access token (token2). ADRF also verifies whether the NF ID of the NF service consumer (NFc) is included in the allowed NFc list for the ML model. If verification is successful, ADRF sends a model retrieval response (e.g., Nadrf_MLModelManagement_Retrieval Response) to the NF service consumer (NFc). This model retrieval response contains the address where the model is stored in ADRF.

[0148] Step 11: Conditionally decrypt the ML model. The NF service consumer (NFc) retrieves the ML model from ADRF and decrypts the model according to the vendor’s implementation.

[0149] It can be understood that the full name of the “interoperability indicator” in this application is ML Model interoperability indicator (ML Model interoperability indicator), which is referred to as interoperability indicator for brevity.

[0150] As shown in Figure 2, the AI / ML model sharing process only considers the scenario where NF service consumers directly obtain / retrieve models from model providers (e.g., NWDAF containing MTLF). However, for scenarios involving delegated model retrieval, it's important to consider how to authorize consumers. Currently, in delegated model retrieval scenarios, model providers or model producers cannot verify whether model consumers have permission to retrieve the requested models.

[0151] In view of this, the embodiments of the present application provide a model authorization method, device and readable storage medium, which help to realize model authorization and model acquisition in the model delegation acquisition scenario, and improve the security of the model in the model delegation acquisition scenario.

[0152] In one possible implementation, the "model delegation acquisition scenario" mentioned in this application can be understood as network element A requesting a model from network element C on behalf of (or on behalf of) network element B. For example, in some scenarios, the first NWDAF network element containing MTLF may not be able to generate a model that meets the requirements of a consumer (e.g., NWDAF containing AnLF), and the consumer cannot directly request the model from the provider of the model. In this case, the first NWDAF network element containing MTLF can request the model from the provider of the model (e.g., the second NWDAF network element containing MTLF) on behalf of the consumer. Alternatively, if the training of the ML model is triggered by a request from an NWDAF containing AnLF, the NWDAF containing MTLF determines that a federated learning (FL) mechanism is required, but it cannot act as a FL server. The NWDAF containing MTLF can discover the FL server NWDAF and request the FL server NWDAF to provide the trained ML model. The NWDAF containing AnLF subscription endpoint address is provided in the request message sent to the FL server NWDAF. The FL server NWDAF can determine to start the FL process before providing the ML model. After the ML model is successfully trained, the FL server NWDAF sends the ML model information to the notification endpoint (e.g., NWDAF containing AnLF). For another example, during analysis subscription transfer or analysis context transfer, the source NWDAF network element containing AnLF can request the transferred model from the model provider (e.g., NWDAF containing MTLF) for the target NWDAF network element containing AnLF.

[0153] The technical solution provided in this application will be described in detail below with reference to more drawings.

[0154] The technical solutions provided in this application are described through a plurality of embodiments, with specific reference to the description of each embodiment below. Among them, the same or similar parts between each embodiment or implementation can refer to each other. In each embodiment in this application, and each implementation method / implementation method / implementation method in each embodiment, if there is no special explanation and logical conflict, the terms and / or descriptions between different embodiments and each implementation method / implementation method / implementation method in each embodiment are consistent and can be referenced to each other, and the technical features in different embodiments and each implementation method / implementation method / implementation method in each embodiment can be combined to form new embodiments, implementation methods, implementation methods, or implementation methods according to their inherent logical relationships. The implementation methods of this application described below do not constitute a limitation on the scope of protection of this application.

[0155] In each embodiment of the present application, "network element A sends information A to network element B" can be understood as the destination end of the information A or the intermediate network element in the transmission path between the destination end and the network element B, which may include directly or indirectly sending information to network element B. "Network element B receives information A from network element A" can be understood as the source end of the information A or the intermediate network element in the transmission path between the source end and the network element A, which may include directly or indirectly receiving information from network element A. The information may be processed as necessary between the source end and the destination end of the information transmission, such as format changes, but the destination end can understand the valid information from the source end. Similar expressions in this application can be understood similarly and will not be elaborated here.

[0156] It should be understood that, in this application, indication includes direct indication (also known as explicit indication) and implicit indication. Direct indication of information A refers to including information A; implicit indication of information A refers to indicating information A through the correspondence between information A and information B and the direct indication of information B. The correspondence between information A and information B can be predefined, pre-stored, pre-burned, or pre-configured.

[0157] It should be understood that, in this application, information D is determined based on information C, which includes information D being determined solely based on information C, as well as information D being determined based on information C and other information. Furthermore, information C being used to determine information D may also include indirect determination, such as information D being determined based on information E, which in turn is determined based on information C.

[0158] It's understood that the NWDAF containing AnLF can request the NWDAF containing MTLF to train a model for a specific function, which can be represented by an "analysis ID." In other words, the NWDAF containing AnLF can send a model training request to the NWDAF containing MTLF, including the analysis ID. The NWDAF containing MTLF then trains the model corresponding to the analysis ID. Therefore, there's a corresponding relationship between the analysis ID and the model, and they can be associated through the model's functionality.

[0159] Each embodiment is described in detail below.

[0160] See Figure 3, which is a flow chart of the model authorization method provided in an embodiment of the present application. In this method, the first model training logic function network element may be the first network element with the model training logic function, such as the first NWDAF network element containing MTLF (NWDAF1containing MTLF), which may be referred to as MTLF1. The second model training logic function network element may be the second network element with the model training logic function, such as the second NWDAF network element containing MTLF (NWDAF2 containing MTLF), which may be referred to as MTLF2. The analysis logic function network element may be a network element with the analysis logic function, such as the NWDAF network element containing AnLF (NWDAF containing AnLF), which may be referred to as AnLF. This method mainly introduces MTLF1 requesting a model from MTLF2 for AnLF.

[0161] As shown in Figure 3, the model authorization method includes but is not limited to the following steps:

[0162] S101, the analysis logic function network element (such as AnLF) sends a first model acquisition request to the first model training logic function network element (such as MTLF1), where the first model acquisition request includes an analysis identifier, and is used to request information of a model corresponding to the analysis identifier.

[0163] Correspondingly, the first model training logic function network element (such as MTLF1) receives the first model acquisition request.

[0164] S102: The first model training logic function network element (e.g., MTLF1) sends a token acquisition request to the NRF network element. The token acquisition request includes one or more of the following: the analysis identifier, the identifier of the analysis logic function network element (e.g., AnLF ID), or the identifier of the second model training logic function network element (e.g., MTLF2 ID). The second model training logic function network element (e.g., MTLF2) can be used to train and / or provide information about the model corresponding to the analysis identifier.

[0165] Correspondingly, the NRF network element receives the token acquisition request.

[0166] In one possible implementation, the above-mentioned first model acquisition request may include but is not limited to: an analysis identifier (analysis ID), an address of an analysis logic function network element (such as AnLF) (such as a URL or a fully qualified domain name (FQDN)), and / or token1. The first model acquisition request can be used to request information about the model corresponding to the analysis ID. Exemplarily, the address of the analysis logic function network element (such as AnLF) can be carried by a subscription endpoint address (Subscription endpoint address). The subscription endpoint address can be used to represent the address for receiving model notification messages. The token1 can be used by the first model training logic function network element (such as MTLF1) to verify the identity of the analysis logic function network element (such as AnLF). If the token1 verification is successful, the first model training logic function network element (such as MTLF1) can determine the model to be shared for the requested analysis ID. Among them, the method for obtaining and using token1 can refer to the relevant steps in the aforementioned Figure 2, which will not be repeated here.

[0167] In one possible implementation, after the first model training logic function network element (such as MTLF1) receives the first model acquisition request, it determines that the second model training logic function network element produces / trains / provides the corresponding model for the analysis logic function network element. Specifically, the first model training logic function network element decides to find other model training logic function network elements (MTLF) to provide models for the analysis logic function network element (such as AnLF) based on local policies, such as when it finds that it cannot generate the model corresponding to the analysis ID. Exemplarily, the first model training logic function network element (such as MTLF1) can obtain a candidate network element list from the NRF network element through the network element discovery process. The candidate network element list may include one or more model training logic function network elements (MTLFs). The first model training logic function network element (such as MTLF1) can select a model training logic function network element (MTLF) from the candidate network element list to provide the model corresponding to the above-mentioned analysis ID to the analysis logic function network element (such as AnLF). For ease of description, the model training logic function network element (MTLF) selected from the candidate network element list is referred to as the second model training logic function network element (e.g., MTLF2). In other words, the second model training logic function network element (e.g., MTLF2) can produce / train / provide information about the model corresponding to the above analysis ID to the analysis logic function network element (e.g., AnLF).

[0168] Furthermore, the first model training logic function network element (such as MTLF1) can send a token acquisition request to the NRF network element. The token acquisition request may include but is not limited to: the above-mentioned analysis ID, the identifier of the analysis logic function network element (such as AnLF ID), and / or the identifier of the second model training logic function network element (such as MTLF2 ID), etc. Among them, the analysis ID can be the identifier of the analysis service corresponding to the model for which authorization is requested. Exemplarily, the token acquisition request can be used to request the first token (token2). In other words, the token acquisition request can be used to request (NRF network element) to authorize the analysis logic function network element (such as AnLF) to obtain the model corresponding to the analysis ID from the second model training logic function network element (such as MTLF2). In other words, the token acquisition request can be used by the first model training logic function network element (such as MTLF1) to obtain the model corresponding to the analysis ID for (or on behalf of behalf) the analysis logic function network element (such as AnLF) to request the first token.

[0169] In one possible implementation, the token acquisition request may also include one or more of the following: an identifier of the first model training logic function network element (such as MTLF1 ID), a vendor identifier of the analysis logic function network element (such as AnLF's Vendor ID), or a first indication information. The vendor identifier of the analysis logic function network element (such as AnLF's Vendor ID) can be used by the NRF network element to verify whether it can be authorized to obtain the model corresponding to the analysis ID from the second model training logic function network element (such as MTLF2). For the specific verification method, please refer to the description below. The first indication information can be used to indicate that the first model training logic function network element (such as MTLF1) requests the first token for (or on behalf of) the analysis logic function network element (such as AnLF). In other words, the first indication information can be used to indicate that the NRF network element is requested to authorize the analysis logic function network element (such as AnLF) to obtain the model in the second model training logic function network element (such as MTLF2). Alternatively, the first indication information can be used to indicate that the token acquisition request is an access token acquisition request (Nnrf_AccessToken_Get request) for the model delegation acquisition scenario.

[0170] It can be understood that the model in this application can be an ML model or an AI model without limitation.

[0171] It is understood that the IDs of the various network elements mentioned in this application may be network function instance identifiers (NF instance IDs), which will not be described in detail below. Among them, the NF instance ID can be used to uniquely identify a network function instance (NF Instance ID: Unique identity of the NF Instance).

[0172] In one possible implementation, before step S101, the model authorization method further includes: the analysis logic function network element (such as AnLF), the first model training logic function network element (such as MTLF1) and the second model training logic function network element (such as MTLF2) can respectively send registration request messages to the NRF network element. Exemplarily, the first model training logic function network element (such as MTLF1) and the second model training logic function network element (such as MTLF2) register their respective analysis IDs and interoperability indicators with the NRF network element. The specific registration process can be seen in step 0a of Figure 2 above, which is not described here. The analysis logic function network element (such as AnLF) registers its own analysis ID and vendor ID with the NRF network element. The specific registration process can be seen in step 0b of Figure 2 above, which is not described here. Among them, the interoperability indicator can be used to represent a list of NWDAF providers (or vendors) that are allowed to retrieve ML models from NWDAF containing MTLF.

[0173] In one possible implementation, the method for the first model training logic function network element (such as MTLF1) to obtain the candidate network element list may include: the first model training logic function network element (such as MTLF1) sends a network element discovery request to the NRF network element, where the network element discovery request includes one or more of the following: the above-mentioned analysis ID, the identifier of the analysis logic function network element (such as AnLF ID) and / or the vendor identifier of the analysis logic function network element (such as AnLF Vendor ID). After receiving the network element discovery request, the NRF network element can obtain the vendor identifier of the analysis logic function network element (such as AnLF Vendor ID) based on the network element discovery request, and then determine the candidate network element list based on the NF configuration files of each network element stored locally. Exemplarily, if the network element discovery request includes the identifier of the analysis logic function network element (such as AnLF ID), the NRF network element can determine the NF configuration file 1 corresponding to the identifier of the analysis logic function network element (such as AnLF ID) from multiple locally stored NF configuration files, and then obtain the vendor identifier of the analysis logic function network element (such as AnLF Vendor ID) from the NF configuration file 1. It can be understood that if the network element discovery request includes the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF), the NRF network element can obtain the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF) from the network element discovery request without having to search the locally stored NF configuration file. After the NRF network element obtains the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF), it can then determine the NF configuration files of multiple model producers (i.e., NWDAF containing MTLF) stored locally. The interoperability indicator corresponding to the above analysis ID includes the NF configuration file of the model producer (i.e., NWDAF containing MTLF) of the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF). These model producers (i.e., NWDAF containing MTLF) are the candidate network element list. In other words, the interoperability indicator corresponding to the above analysis ID of each model training logic function network element (such as MTLF) in the candidate network element list includes the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF). The NRF network element may send a network element discovery response to the first model training logic function network element (such as MTLF1), where the network element discovery response includes a candidate network element list.

[0174] S103, the NRF network element verifies whether the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF) is included in the interoperability indicator corresponding to the above analysis identifier of the second model training logic function network element (such as MTLF2).

[0175] S104, if the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF) is in the interoperability indicator corresponding to the above-mentioned analysis identifier of the second model training logic function network element (such as MTLF2), the NRF network element sends a first token to the first model training logic function network element (such as MTLF1), and the first token includes one or more of the following: the identifier of the analysis logic function network element (such as AnLF ID), or the identifier of the second model training logic function network element (such as MTLF2 ID).

[0176] Correspondingly, the first model training logic function network element (such as MTLF1) receives the first token.

[0177] In one possible implementation, after receiving the token acquisition request, the NRF network element can verify whether the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF) is included in the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element (such as MTLF2). In other words, after receiving the token acquisition request, the NRF network element can verify whether the analysis logic function network element (such as AnLF) has the authority to obtain the model from the second model training logic function network element (such as MTLF2). Exemplarily, the token acquisition request includes one or more of the following: analysis ID, identifier of the analysis logic function network element (such as AnLF ID), or identifier of the second model training logic function network element (such as MTLF2 ID). The NRF network element can determine the NF configuration file 1 corresponding to the identifier of the analysis logic function network element (such as AnLF ID) from multiple NF configuration files stored locally, and can obtain the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF) from the NF configuration file 1. It can be understood that when the token acquisition request includes the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF), the NRF network element can obtain the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF) from the token acquisition request without searching the locally stored NF configuration file. The NRF network element can then determine the NF configuration file 2 corresponding to the identifier of the second model training logic function network element (such as MTLF2 ID) from the multiple locally stored NF configuration files, and can obtain the interoperability indicator corresponding to the above-mentioned analysis ID (here, the analysis ID carried in the token acquisition request) from the NF configuration file 2.

[0178] Then, the NRF network element can verify whether the interoperability indicator corresponding to the analysis ID contains the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF). If the interoperability indicator corresponding to the analysis ID in the NF configuration file 2 contains the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF), it means that the analysis logic function network element (such as AnLF) has the authority to obtain the model from the second model training logic function network element (such as MTLF2), then the NRF network element can generate an authorization token for the model (i.e., a first token, token2), and can send the first token (token2) to the first model training logic function network element (such as MTLF1). The first token (token2) may include part or all of the content in the above-mentioned token acquisition request. Exemplarily, the first token may include but is not limited to one or more of the following: the above-mentioned analysis ID, the identifier of the analysis logic function network element (such as AnLF ID), or the identifier of the second model training logic function network element (such as MTLF2ID). The first token can be used to indicate authorization for the analysis logic function network element (such as AnLF) to obtain information about the model corresponding to the analysis ID from the second model training logic function network element (such as MTLF2). Alternatively, the first token can be used to indicate authorization for the first model training logic function network element (such as MTLF1) to be (or represent behalf) the analysis logic function network element (such as AnLF) to obtain information about the model corresponding to the analysis ID from the second model training logic function network element (such as MTLF2). It can be understood that the authorization information of the NRF network element is for each analysis ID (Per Analytics ID), or the above-mentioned first token uniquely corresponds to the analysis ID in the above-mentioned token acquisition request. For different analysis IDs, the NRF network element can have different authorization information, such as tokens, and the token and analysis ID can correspond one to one.

[0179] In a possible implementation, if the token acquisition request also includes the identifier of the first model training logic function network element (such as MTLF1ID) and the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF). Accordingly, the first token (token2) may also include the identifier of the first model training logic function network element (such as MTLF1 ID) and the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF). Exemplarily, the first token may also include second indication information. The second indication information may be used to indicate authorization for the analysis logic function network element (such as AnLF) to obtain information about the model corresponding to the above analysis ID from the second model training logic function network element (such as MTLF2). Alternatively, the second indication information may be used to indicate authorization for the first model training logic function network element (such as MTLF1) to obtain information about the model corresponding to the analysis ID from the second model training logic function network element (such as MTLF2) for (or on behalf of) the analysis logic function network element (such as AnLF). Alternatively, the second indication information may be used to indicate that the first token is an authorization token for a model delegation acquisition scenario.

[0180] In one possible implementation, if the interoperability indicator corresponding to the analysis ID in the NF configuration file 2 does not include the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF), it means that the analysis logic function network element (such as AnLF) does not have the authority to obtain the model from the second model training logic function network element (such as MTLF2), and the NRF network element can send a response message to the first model training logic function network element (such as MTLF1) to reject the above token acquisition request. The response message can carry the reason for the rejection, such as the analysis logic function network element (such as AnLF) cannot obtain the model from the second model training logic function network element (such as MTLF2), or the vendor identifier of the analysis logic function network element (such as the Vendor ID of AnLF) is not included in the interoperability indicator of the second model training logic function network element (such as MTLF2), etc.

[0181] In one possible implementation, when the token acquisition request includes the first indication information, after receiving the token acquisition request, the NRF network element may determine, based on the first indication information, that the token acquisition request is a request from the first model training logic function network element (e.g., MTLF1) for the NRF network element to authorize the analysis logic function network element (e.g., AnLF) to obtain the model of the second model training logic function network element (e.g., MTLF2). The verification of step S103 is then performed.

[0182] S105: The first model training logic function network element (e.g., MTLF1) sends a second model acquisition request to the second model training logic function network element (e.g., MTLF2). The second model acquisition request includes one or more of the following: the analysis identifier, an identifier of the analysis logic function network element (e.g., AnLF ID), or the first token. The second model acquisition request is used to obtain information about the model corresponding to the analysis identifier for the analysis logic function network element (e.g., AnLF). The first token can be used to verify the second model acquisition request.

[0183] Correspondingly, the second model training logic function network element (such as MTLF2) receives the second model acquisition request.

[0184] S106: The second model training logic function network element (such as MTLF2) verifies the first token.

[0185] S107, when the first token verification is passed, the second model training logic function network element (such as MTLF2) sends a model notification message to the analysis logic function network element (such as AnLF), and the model notification message includes information about the model corresponding to the analysis identifier.

[0186] Correspondingly, the analysis logic function network element (such as AnLF) receives the model notification message.

[0187] In one possible implementation, after the first model training logic function network element (such as MTLF1) receives the above-mentioned first token (token2), it can send a second model acquisition request to the second model training logic function network element (such as MTLF2). The second model acquisition request can be used to obtain information about the model corresponding to the above-mentioned analysis ID for the analysis logic function network element (such as AnLF). The second model acquisition request may include but is not limited to one or more of the following: the above-mentioned analysis ID (the analysis ID carried in the instruction sign acquisition request here is also the analysis ID carried in the first model acquisition request), the identifier of the analysis logic function network element (such as AnLF ID), or the first token (token2). The first token can be used to verify the second model acquisition request, or the first token can be used to verify various identifiers in the second model acquisition request. Alternatively, the first token can be used to verify the authority of the analysis logic function network element (such as AnLF) to obtain the model (the model can be the model corresponding to the above-mentioned analysis ID) from the second model training logic function network element (such as MTLF2). Alternatively, the first token can be used to verify the authority of the first model training logic function network element (such as MTLF1) to be (or represent behalf) the analysis logic function network element (such as AnLF) to obtain a model (the model can be the model corresponding to the above-mentioned analysis ID) from the second model training logic function network element (such as MTLF2).

[0188] In one possible implementation, after receiving the second model acquisition request, the second model training logic function network element (such as MTLF2) can verify the above-mentioned first token (token2). Exemplarily, the second model training logic function network element (such as MTLF2) can verify whether the analysis ID in the second model acquisition request is consistent with (or identical to, matching) the analysis ID in the first token (token2), and whether the identifier of the analysis logic function network element in the second model acquisition request (such as AnLF ID) is consistent with (or identical to, matching) the identifier of the analysis logic function network element in the first token (token2) (such as AnLF ID). If they are all consistent (or identical), it can be said that the first token (token2) has passed the verification. Here, the match can be interpreted as: the analysis ID(s) in the second model acquisition request is located in the analysis ID(s) contained in the first token.

[0189] In one possible implementation, the second model acquisition request further includes one or more of the following: an identifier of the first model training logic function network element (such as MTLF1 ID), an address of the analysis logic function network element (such as AnLF) (such as a URL or FQDN), a vendor identifier of the analysis logic function network element (such as AnLF's Vendor ID), or a third indication information. Exemplarily, the address of the analysis logic function network element (such as AnLF) can be carried by a subscription endpoint address. The subscription endpoint address can be used to indicate the address for receiving model notification messages. The third indication information can be used to instruct the first model training logic function network element (such as MTLF1) to obtain information about the model corresponding to the above-mentioned analysis ID for (or on behalf of) the analysis logic function network element (such as AnLF). Alternatively, the third indication information can be used to indicate that the second model acquisition request is an ML model provision request (Nnwdaf_MLModelProvision request) for a model delegation acquisition scenario.

[0190] Correspondingly, after the second model training logic function network element (such as MTLF2) receives the second model acquisition request, when verifying the first token, in addition to verifying whether the analysis ID and the identifier of the analysis logic function network element (such as AnLF ID) in the second model acquisition request are consistent (or the same) with the analysis ID and the identifier of the analysis logic function network element (such as AnLF ID) in the first token (token2), it can also verify one or more of the following: whether the identifier of the first model training logic function network element in the second model acquisition request (such as MTLF1 ID) is consistent (or the same) with the identifier of the first model training logic function network element in the first token (token2) (such as MTLF1 ID), and whether the vendor identifier of the analysis logic function network element in the second model acquisition request (such as Vendor ID of AnLF) is consistent (or the same) with the vendor identifier of the analysis logic function network element in the first token (token2) (such as Vendor ID of AnLF). The second model training logic function network element (such as MTLF2) can also confirm whether the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF) in the second model acquisition request is included in the interoperability indicator corresponding to the above-mentioned analysis ID of the second model training logic function network element (such as MTLF2). If all are consistent (or the same), and the vendor identifier of the analysis logic function network element (such as Vendor ID of AnLF) is included in the interoperability indicator corresponding to the above-mentioned analysis ID of the second model training logic function network element (such as MTLF2), it can be said that the first token (token2) has been verified.

[0191] In one possible implementation, when the above-mentioned first token verification is passed, the second model training logic function network element (such as MTLF2) can send a model notification message to the analysis logic function network element (such as AnLF). It can be understood that the second model training logic function network element (such as MTLF2) can send the model notification message to the analysis logic function network element (such as AnLF) either directly or through an intermediate network element, and the embodiment of the present application does not limit this. Exemplarily, the second model training logic function network element (such as MTLF2) can send a model notification message to the analysis logic function network element (such as AnLF) through the first model training logic function network element (such as MTLF1). For example: the second model training logic function network element (such as MTLF2) sends a model notification message to the first model training logic function network element (such as MTLF1), and the first model training logic function network element (such as MTLF1) can then forward it to the analysis logic function network element (such as AnLF). The forwarding here can be transparent transmission or forwarding after processing the model notification message, and the embodiment of the present application does not limit this. However, the model notification message received by the analysis logic function network element (such as AnLF) may include information about the model corresponding to the analysis ID. Exemplarily, the model notification message may also include the identifier of the ADRF network element storing the model (such as ADRFID). After receiving the model notification message, the analysis logic function network element (such as AnLF) may obtain (or download) the model file of the model based on the model notification message and save it locally.

[0192] In this application, "model information" may include but is not limited to one or more of the following: model identification (model ID), the address of the model in the second model training logic function network element (such as MTLF2), or model file, etc.

[0193] The model ID mentioned in this application can be used to uniquely identify a model (ML Model identifier: unique ML Model identifier). The address of the model in the second model training logical function network element (such as MTLF2) can be used by the analysis logical function network element (such as AnLF) to obtain (e.g., download) the model file from the second model training logical function network element (such as MTLF2) according to the address and store it locally.

[0194] In one possible implementation, when the first token verification is passed, the second model training logic function network element (such as MTLF2) may further add the identifier of the analysis logic function network element (such as AnLF ID) to the allowed NF consumer list (allowed NFc list). The allowed NFc list is associated with the model corresponding to the analysis ID. In other words, the allowed NFc list is a list of network function instance identifiers that are allowed to obtain / retrieve the model corresponding to the analysis ID. Exemplarily, the second model training logic function network element (such as MTLF2) may store the identifier of the analysis logic function network element (such as AnLF ID) as part of the allowed NFc list of the model, and may send a Nadrf_MLModelManagement_StorageRequest (Nadrf machine learning model management storage request) to the ADRF network element to trigger an update at the ADRF network element. The Nadrf_MLModelManagement_StorageRequest may include the identifier of the second model training logic function network element (such as MTLF2 ID), the model ID, and the allowed NFc list.

[0195] In various embodiments of this application, the "allowed NFc list" can be understood as a list of network function instance identifiers (NF instance IDs) that are allowed to obtain / retrieve / access / query models. The "allowed NFc list" can also be referred to as the "allowed NF instance ID list" or the "allowed network function list"; this application does not impose any restrictions.

[0196] After receiving the model acquisition request from AnLF, MTLF1 of the embodiment of the present application finds that it cannot provide a model that meets the requirements according to its internal strategy. MTLF1 can request a model that meets the requirements from other MTLFs (such as MTLF2) for (or on behalf of) AnLF. For example: MTLF1 first requests an authorization token (i.e., the above-mentioned first token) from NRF for (or on behalf of) AnLF. The request includes AnLF ID and MTLF2 ID. After NRF verifies that AnLF has the authority to obtain the model from MTLF2, it returns the authorization token (i.e., the above-mentioned first token) to MTLF1; after obtaining the authorization token, MTLF1 requests information about a model that meets the requirements from MTLF2 for (or on behalf of) AnLF. The request includes AnLF ID and authorization token; after MTLF2 verifies that the authorization token is passed, it sends information about a model that meets the requirements to AnLF. Through the above process, it is helpful to realize model authorization and model acquisition in the model delegation acquisition scenario, and to authorize the actual model consumer (i.e., AnLF) in the model delegation acquisition scenario, thereby improving the security of the model.

[0197] Refer to Figure 4, which is another flow chart of the model authorization method provided in an embodiment of the present application. In this method, the source analysis logic function network element may be a network element having an analysis logic function network element, such as: a source NWDAF network element containing AnLF (SourceNWDAF containing AnLF), which may be referred to as source AnLF (Source AnLF). The target analysis logic function network element may be another network element having an analysis logic function network element, such as: a target NWDAF network element containing AnLF (TargetNWDAF containing AnLF), which may be referred to as target AnLF (Target AnLF). The model training logic function network element may be a network element having a model training logic function, such as: a NWDAF network element containing MTLF (NWDAF containing MTLF), which may be referred to as MTLF. This method mainly introduces that in the process of analysis subscription transfer or analysis context transfer, the source AnLF requests the MTLF for the model in the analysis context for the target AnLF.

[0198] As shown in Figure 4, the model authorization method includes but is not limited to the following steps:

[0199] S201, the target analysis logic function network element (such as Target AnLF) sends an analysis context transfer request to the source analysis logic function network element (such as Source AnLF), where the analysis context transfer request includes an identification of the analysis context, and is used to request the transfer of information of the first model in the analysis context.

[0200] Correspondingly, the source analysis logic function network element (such as Source AnLF) receives the analysis context transfer request.

[0201] In one possible implementation, before step S201, the model authorization method further includes: the source analysis logic function network element (e.g., Source AnLF) and the target analysis logic function network element (e.g., Target AnLF) may each send a registration request message to the NRF network element. Exemplarily, the source analysis logic function network element (e.g., Source AnLF) may send a registration request message 1 to the NRF network element. The registration request message 1 includes the analysis ID and vendor ID of the source analysis logic function network element (e.g., Source AnLF), and optionally also includes a first operation indicator of the source analysis logic function network element (e.g., Source AnLF). The NRF network element may store the information contained in the registration request message 1 in the NF configuration file of the source analysis logic function network element (e.g., Source AnLF). Furthermore, exemplarily, the target analysis logic function network element (e.g., Target AnLF) may send a registration request message 2 to the NRF network element. The registration request message 2 includes the analysis ID and vendor ID of the target analysis logic function network element (e.g., Target AnLF), and optionally also includes a second operation indicator of the target analysis logic function network element (e.g., Target AnLF). The NRF network element may store the information contained in the registration request message 2 in the NF configuration file of the target analysis logic function network element (such as Target AnLF). The first operation indicator and the second operation indicator may be a new operation indicator provided by an embodiment of the present application. For example, the first operation indicator may be used to indicate a model that the source analysis logic function network element (such as Source AnLF) can use, and the second operation indicator may be used to indicate a model that the target analysis logic function network element (such as Target AnLF) can use.

[0202] The following describes the operation indicators provided in the embodiments of the present application.

[0203] For example, the operation indicator may correspond to an analysis logic function network element (such as AnLF), that is, the operation indicator is at the AnLF granularity. Alternatively, the operation indicator may correspond to an analysis identifier, that is, the operation indicator is at the analysis identifier granularity.

[0204] In one possible implementation, if an operation indicator corresponds to an AnLF, that is, one AnLF corresponds to one operation indicator. The operation indicator may include a list of vendor identifiers (or a vendor list), or may be described as a list of NWDAF providers (or suppliers). AnLF allows retrieval or use of models provided by vendors in the vendor list. The operation indicator also indicates that AnLF supports the use of vendor-provided models for network elements (e.g., NWDAF) of vendors in the vendor list. That is, the operation indicator applies to each analysis identifier.

[0205] For example, the operation indicator includes the identifier of manufacturer 1 and the identifier of manufacturer 2, which means that AnLF can use the model provided by manufacturer 1 and the model provided by manufacturer 2. Alternatively, it can be understood that if the manufacturer of MTLF1 is manufacturer 1 or manufacturer 2, AnLF can use the model trained by MTLF1. For example, assuming that the models provided by manufacturer 1 include a model corresponding to analysis identifier 1 and a model corresponding to analysis identifier 2, AnLF can use both models.

[0206] It can be understood that the model provided by Manufacturer 1 can be understood as a model produced (or trained) by Manufacturer 1, or a model produced (or trained) by MTLF, and the manufacturer of the MTLF is Manufacturer 1.

[0207] In another possible implementation, if the operation indicator corresponds to the analysis identifier, that is, one AnLF can correspond to one or more operation indicators. Different operation indicators can correspond to different analysis identifiers. Among them, the operation indicator corresponding to an analysis identifier includes a manufacturer identifier list (or manufacturer list), or is described as a NWDAF provider (or supplier) list. AnLF allows the retrieval or use of models provided by manufacturers in the manufacturer list. The operation indicator also indicates that AnLF supports the use of models provided by the manufacturer for the network elements of the manufacturers in the manufacturer list (such as: NWDAF), and these models correspond to a certain analysis identifier.

[0208] For example, it is assumed that the operation indicator of AnLF includes an operation indicator a corresponding to the analysis identifier 1, and an operation indicator b corresponding to the analysis identifier 2. Among them, the operation indicator a corresponding to the analysis identifier 1 includes the identifier of manufacturer 1 and the identifier of manufacturer 2, that is, AnLF can use the model provided by manufacturer 1 and the model provided by manufacturer 2, and the model corresponds to the analysis identifier 1. For example, assuming that the model provided by manufacturer 1 includes the model corresponding to the analysis identifier 1 and the model corresponding to the analysis identifier 2, then AnLF can use the model corresponding to the analysis identifier 1, but cannot use the model corresponding to the analysis identifier 2. The operation indicator b corresponding to the analysis identifier 2 includes the identifier of manufacturer 4, that is, AnLF can use the model provided by manufacturer 4, and the model corresponds to the analysis identifier 2. For example, assuming that the model provided by manufacturer 4 includes the model corresponding to the analysis identifier 1 and the model corresponding to the analysis identifier 2, then AnLF can use the model corresponding to the analysis identifier 2, but cannot use the model corresponding to the analysis identifier 1.

[0209] Therefore, the first operation indicator may include a list of vendor identifiers (or a vendor list). The first operation indicator may indicate that the source analysis logic function network element (e.g., Source AnLF) supports the use of models provided by vendors in the vendor list. Alternatively, the first operation indicator may indicate that the source analysis logic function network element (e.g., Source AnLF) supports the use of models corresponding to a certain analysis identifier provided by vendors in the vendor list. In other words, the first operation indicator can be used to indicate which model manufacturers the source analysis logic function network element (such as Source AnLF) supports the use of models provided by. Alternatively, the first operation indicator can be used to indicate which model manufacturers the source analysis logic function network element (such as Source AnLF) supports the use of which models provided by. Similarly, the above-mentioned second operation indicator can also include a manufacturer identifier list (or manufacturer list). The second operation indicator can indicate that the target analysis logic function network element (such as Target AnLF) supports the use of models provided by manufacturers in the manufacturer list. Alternatively, the second operation indicator can indicate that the target analysis logic function network element (such as Target AnLF) supports the use of models provided by manufacturers in the manufacturer list corresponding to a certain analysis identifier. In other words, the second operation indicator can be used to indicate which model manufacturers the target analysis logic function network element (such as Target AnLF) supports the use of models provided by. Alternatively, the second operation indicator can be used to indicate which model manufacturers the target analysis logic function network element (such as Target AnLF) supports the use of which models provided by.

[0210] In one possible implementation, the target analysis logic function network element (such as Target AnLF) can send an analysis context transfer request to the source analysis logic function network element (such as Source AnLF). The analysis context transfer request may include but is not limited to an analysis context identifier, such as a subscription correlation identifier (Subscription Correlation ID). The subscription correlation identifier can be used to identify the analysis subscription for which the related analytics context is requested (Subscription Correlation ID: identifies the analytics subscription for which the related analytics context is requested). The analysis context may include model-related information, such as: the identifier of the model producer / provider / trainer (such as NWDAF containing MTLF), the identifier of the model, the file address of the model, or the analysis identifier corresponding to the model. It can be understood that the analysis context may include the identifiers of multiple model producers / providers / trainers (such as NWDAF containing MTLF). For the sake of simplicity, the embodiment of the present application takes a model producer / provider / trainer that provides a model as an example for explanation. The above-mentioned analysis context transfer request can be used to request the transfer of information of one or more models in the analysis context. For the sake of simplicity, the embodiment of the present application takes the example of the analysis context transfer request being used to request the transfer of information of the first model in the above-mentioned analysis context.

[0211] In one possible implementation, after receiving the analysis context transfer request, the source analysis logic function network element (e.g., Source AnLF) may return analysis context information to the target analysis logic function network element (e.g., Target AnLF). The analysis context information may include other information in the analysis context transfer request in addition to the model-related information, such as the active data source identifier (ID), subscription association identifier, etc.

[0212] In one possible implementation, the analysis context transfer request may further include one or more of the following: a second operation indicator of the target analysis logic function network element (such as Target AnLF), or a vendor identifier of the target analysis logic function network element (such as Vendor ID of Target AnLF). For an explanation of the second operation indicator, please refer to the previous description and will not be repeated here.

[0213] In one possible implementation, if the analysis context transfer request includes a second operation indicator, after the source analysis logic function network element (such as Source AnLF) receives the analysis context transfer request, it can determine whether the second operation indicator of the target analysis logic function network element (such as Target AnLF) belongs to the first operation indicator of the source analysis logic function network element (such as Source AnLF). In other words, the source analysis logic function network element (such as Source AnLF) can confirm whether the target analysis logic function network element (such as Target AnLF) has the ability or authority to use the model in the analysis context. Exemplarily, the source analysis logic function network element (such as Source AnLF) can determine whether the provider / producer of the first model is located in the second operation indicator. If the provider / producer of the first model is located in the second operation indicator, it means that the target analysis logic function network element (such as Target AnLF) can use the first model in the analysis context. The source analysis logic function network element (such as Source AnLF) can then return the analysis context information to the target analysis logic function network element (such as Target AnLF). For the description of the analysis context information, please refer to the previous description and will not be repeated here.

[0214] In another possible implementation, if the above-mentioned analysis context transfer request includes the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF), after the source analysis logic function network element (such as Source AnLF) receives the above-mentioned analysis context transfer request, it can determine whether the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) is included in the interoperability indicator of the information provider of the above-mentioned first model (such as MTLF). If the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) is included in the interoperability indicator of the information provider of the first model, it means that the target analysis logic function network element (such as Target AnLF) can obtain / retrieve / query the first model. It can be understood that the information provider of the first model can be the producer of the first model, or it can be not the producer of the first model but only the provider of the first model (it does not produce the first model). The source analysis logic function network element (such as Source AnLF) can then return the analysis context information to the target analysis logic function network element (such as Target AnLF). For the description of the analysis context information, please refer to the previous description and will not be repeated here. It is understood that the interoperability indicator can be per Analytics ID, or one interoperability indicator corresponds to one Analytics ID, and there can be a one-to-one correspondence between the interoperability indicator and the Analytics ID. Different interoperability indicators can be used for different Analytics IDs. Of course, the interoperability indicator can also be per Model Maker (MTLF), or one interoperability indicator corresponds to one Model Maker (MTLF), and there can be a one-to-one correspondence between the interoperability indicator and the Model Maker (MTLF). Different Model Makers (MTLF) can have different interoperability indicators.

[0215] In one possible implementation, before step S201 (and after the network element registration process), the model authorization method further includes: the NWDAF service consumer (NWDAF service consumer) sends an analysis subscription service request to the source analysis logic function network element (such as Source AnLF), carrying the analysis ID. After receiving the analysis subscription service request, the source analysis logic function network element (such as Source AnLF) can subscribe to the model associated with the analysis ID from the model training logic function network element (such as MTLF). The model training logic function network element (such as MTLF) returns a corresponding response message, carrying the subscription association identifier 1. Optionally, the model training logic function network element (such as MTLF) can send the model ID associated with the analysis ID to the source analysis logic function network element (such as Source AnLF), and the source analysis logic function network element (such as Source AnLF) can obtain the model identified by the model ID through the model ID. The source analysis logic function network element (such as Source AnLF) sends an analysis subscription service response to the NWDAF service consumer, including the subscription association identifier 2 (i.e., the identifier of the above-mentioned analysis context). The NWDAF service consumer decides to initiate the analysis context transfer process and determines the target analysis logic function network element (such as Target AnLF). The NWDAF service consumer sends an analysis subscription request to the target analysis logic function network element (such as Target AnLF), including the subscription association identifier 2 (i.e., the identifier of the above-mentioned analysis context). After receiving the analysis subscription request, the target analysis logic function network element (such as Target AnLF) can execute step S201.

[0216] S202: The source analysis logic function network element (e.g., Source AnLF) sends a token acquisition request to the NRF network element. The token acquisition request includes one or more of the following: an analysis identifier corresponding to the first model, an identifier of the target analysis logic function network element (e.g., Target AnLF ID), or an identifier of the model training logic function network element (e.g., MTLF ID). The model training logic function network element (e.g., MTLF) can be used to train and / or provide information about the first model.

[0217] Correspondingly, the NRF network element receives the token acquisition request.

[0218] In one possible implementation, after the source analysis logic function network element (such as Source AnLF) receives the above-mentioned analysis context transfer request, it determines a model training logic function network element (such as MTLF) to produce / train / provide the first model for the target analysis logic function network element (such as Target AnLF ID). In other words, the producer / trainer / provider of the first model is this model training logic function network element. The source analysis logic function network element (such as Source AnLF) can send a token acquisition request to the NRF network element. The token acquisition request includes one or more of the following: the analysis identifier corresponding to the above-mentioned first model, the identifier of the target analysis logic function network element (such as Target AnLF ID), or the identifier of this model training logic function network element (such as MTLF ID). Among them, the analysis ID can be the identifier of the analysis service corresponding to the first model for which authorization is requested. For the role of the token acquisition request, please refer to the relevant description in the embodiment shown in Figure 3 above.

[0219] It can be understood that the target analysis logic function network element (such as Target AnLF) in the embodiment of the present application can be analogous to the analysis logic function network element (such as AnLF) in the embodiment shown in Figure 3 above, and the source analysis logic function network element (such as Source AnLF) in the embodiment of the present application can be analogous to the first model training logic function network element (such as MTLF1) in the embodiment shown in Figure 3 above. The model training logic function network element (such as MTLF) in the embodiment of the present application can be analogous to the second model training logic function network element (such as MTLF2) in the embodiment shown in Figure 3 above. No further details will be given below.

[0220] In a possible implementation, the token acquisition request may also include one or more of the following: an identifier of the source analysis logic function network element (such as Source AnLF ID), a vendor identifier of the target analysis logic function network element (such as Vendor ID of Target AnLF), an identifier of the above-mentioned first model, or a first indication information. Among them, the vendor identifier of the target analysis logic function network element (such as Vendor ID of Target AnLF) can be used by the NRF network element to verify whether it can be authorized to obtain the information of the above-mentioned first model from the model training logic function network element (such as MTLF). For the specific verification method, please refer to the description below. Regarding the role of the first indication information, please refer to the relevant description in the embodiment shown in Figure 3 above, which will not be repeated here.

[0221] S203, the NRF network element verifies whether the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) is included in the interoperability indicator corresponding to the above analysis identifier of the model training logic function network element (such as MTLF).

[0222] S204, if the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) is included in the interoperability indicator corresponding to the above-mentioned analysis identifier of the model training logic function network element (such as MTLF), the NRF network element sends a first token to the source analysis logic function network element (such as Source AnLF), and the first token includes the identifier of the target analysis logic function network element (such as Target AnLF ID) and / or the identifier of the model training logic function network element (such as MTLF ID).

[0223] Correspondingly, the source analysis logic function network element (such as Source AnLF) receives the first token.

[0224] In one possible implementation, after receiving the above-mentioned token acquisition request, the NRF network element can verify whether the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) is included in the interoperability indicator corresponding to the above-mentioned analysis identifier of the model training logic function network element (such as MTLF). In other words, after receiving the above-mentioned token acquisition request, the NRF network element can verify whether the target analysis logic function network element (such as Target AnLF) has the authority to obtain the above-mentioned first model from the model training logic function network element (such as MTLF). Exemplarily, the NRF network element can determine the NF configuration file 3 corresponding to the target analysis logic function network element (such as Target AnLF) from multiple NF configuration files stored locally, and can obtain the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) from the NF configuration file 3. It can be understood that when the above-mentioned token acquisition request includes the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF), the NRF network element can obtain the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF) from the token acquisition request without searching the locally stored NF configuration file. The NRF network element can then determine the NF configuration file 4 corresponding to the identifier of the model training logic function network element (such as MTLF ID) from the multiple locally stored NF configuration files, and can obtain the interoperability indicator corresponding to the above-mentioned analysis ID (here the analysis ID carried in the token acquisition request) from the NF configuration file 4.

[0225] Then, the NRF network element can verify whether the interoperability indicator corresponding to the analysis ID contains the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF). If the interoperability indicator corresponding to the analysis ID in the NF configuration file 4 contains the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF), it means that the target analysis logic function network element (such as TargetAnLF) has the authority to obtain the model from the model training logic function network element (such as MTLF), then the NRF network element can generate an authorization token for the model (i.e., the first token, token2), and can send the first token (token2) to the source analysis logic function network element (such as Source AnLF). Among them, the first token (token2) may include part or all of the content in the above-mentioned token acquisition request. Exemplarily, the first token may include but is not limited to one or more of the following: the analysis ID corresponding to the above-mentioned first model, the identifier of the target analysis logic function network element (such as Target AnLF ID), or the identifier of the model training logic function network element (such as MTLF ID). For the description of the first token, please refer to the relevant description in the embodiment shown in Figure 3 above, which will not be repeated here.

[0226] In one possible implementation, the first token may further include one or more of the following: the analysis ID, the identifier of the source analysis logic function network element (such as Source AnLF ID), the vendor identifier of the target analysis logic function network element (such as Vendor ID of Target AnLF), the identifier of the first model, or the second indication information. For the role of the second indication information, reference may be made to the relevant description of the embodiment shown in FIG3 , which will not be repeated here.

[0227] In one possible implementation, if the interoperability indicator corresponding to the analysis ID in the NF configuration file 4 does not include the vendor identifier of the target analysis logic function network element (such as the Vendor ID of Target AnLF), it means that the target analysis logic function network element (such as Target AnLF) does not have the authority to obtain the model from the model training logic function network element (such as MTLF). The NRF network element can send a response message to the source analysis logic function network element (such as Source AnLF) to reject the above token acquisition request. For the description of the response message, please refer to the relevant description in the embodiment shown in Figure 3 above, which will not be repeated here.

[0228] In one possible implementation, when the token acquisition request includes the first indication information, after receiving the token acquisition request, the NRF network element may determine, based on the first indication information, that the token acquisition request is a request from the source analysis logic function network element (e.g., Source AnLF) for the NRF network element to authorize the target analysis logic function network element (e.g., Target AnLF) to obtain the model of the model training logic function network element (e.g., MTLF). The verification of step S203 is then performed.

[0229] S205: The source analysis logic function network element (e.g., Source AnLF) sends a model acquisition request to the model training logic function network element (e.g., MTLF). The model acquisition request includes one or more of the following: the analysis identifier, the identifier of the target analysis logic function network element (e.g., Target AnLF ID), or the first token. The model acquisition request is used to obtain information about the first model for the target analysis logic function network element (e.g., Target AnLF). The first token can be used to verify the model acquisition request.

[0230] Correspondingly, the model training logic function network element (such as MTLF) receives the model acquisition request.

[0231] S206: The model training logic function network element (such as MTLF) verifies the first token.

[0232] S207, when the first token verification is passed, the model training logic function network element (such as MTLF) sends a model notification message to the target analysis logic function network element (such as Target AnLF), and the model notification message includes the information of the first model.

[0233] Correspondingly, the target analysis logic function network element (such as Target AnLF) receives the model notification message.

[0234] In one possible implementation, after receiving the first token, the source analysis logic function network element (e.g., Source AnLF) may send a model acquisition request to the model training logic function network element (e.g., MTLF). The model acquisition request is used to obtain information about the first model for the target analysis logic function network element (e.g., Target AnLF). The model acquisition request may include, but is not limited to, one or more of the following: an analysis identifier corresponding to the first model, an identifier of the target analysis logic function network element (e.g., Target AnLF ID), or the first token. The first token may be used to verify the model acquisition request, or the first token may be used to verify various identifiers in the model acquisition request. Alternatively, the first token may be used to verify the target analysis logic function network element (e.g., Target AnLF)'s permission to obtain the first model from the model training logic function network element (e.g., MTLF). Alternatively, the first token may be used to verify the source analysis logic function network element (e.g., Source AnLF)'s permission to obtain the first model from the model training logic function network element (e.g., MTLF) on behalf of (or on behalf of) the target analysis logic function network element (e.g., Target AnLF).

[0235] In one possible implementation, after receiving the model acquisition request, the model training logic function network element (such as MTLF) can verify the first token (token2). The way in which the model training logic function network element (such as MTLF) verifies the first token is the same as the way in which the second model training logic function network element (such as MTLF2) verifies the first token in the embodiment shown in Figure 3 above, and will not be repeated here.

[0236] In a possible implementation, the above-mentioned model acquisition request may also include one or more of the following: the identifier of the source analysis logic function network element (such as Source AnLF ID), the address of the target analysis logic function network element (such as URL or FQDN), the identifier of the above-mentioned first model, the vendor identifier of the target analysis logic function network element (such as Vendor ID of Target AnLF), or a third indication information. Exemplarily, the address of the target analysis logic function network element can be carried by the subscription endpoint address. The subscription endpoint address can be used to represent the address for receiving model notification messages. Among them, the description of the third indication information can refer to the relevant description in the embodiment shown in Figure 3 above, which will not be repeated here. It can be understood that the model acquisition request in the embodiment of the present application can be analogous to the second model acquisition request in the embodiment shown in Figure 3 above.

[0237] In one possible implementation, when the above-mentioned first token verification is passed, the model training logic function network element (such as MTLF) can send a model notification message to the target analysis logic function network element (such as Target AnLF). It can be understood that the model training logic function network element (such as MTLF) can send the model notification message to the target analysis logic function network element (such as Target AnLF) either directly or through an intermediate network element, and the embodiment of the present application does not limit this. Exemplarily, the model training logic function network element (such as MTLF) can send a model notification message to the target analysis logic function network element (such as Target AnLF) through the source analysis logic function network element (such as Source AnLF). For example: the model training logic function network element (such as MTLF) sends a model notification message to the source analysis logic function network element (such as Source AnLF), and the source analysis logic function network element (such as Source AnLF) can then forward it to the target analysis logic function network element (such as Target AnLF). The forwarding here can be transparent transmission, or it can be forwarding after processing the model notification message, and the embodiment of the present application does not limit this. However, the model notification message received by the target analysis logic function network element (such as Target AnLF) may include information of the above-mentioned first model, such as the ID of the first model, the address of the first model in the model training logic function network element (such as MTLF), or the model file, etc. Exemplarily, the model notification message may also include the identifier of the ADRF network element that stores the first model (such as ADRFID). After the analysis logic function network element (such as AnLF) receives the model notification message, it can obtain (or download) the model file of the first model to local storage based on the model notification message.

[0238] In one possible implementation, if the first token verification is successful, the model training logic function network element (e.g., MTLF) may also add the identifier of the target analysis logic function network element (e.g., Target AnLF ID) to an allowed NF consumer list (allowed NFc list). The allowed NFc list is associated with the first model. In other words, the allowed NFc list is a list of network function instance identifiers that are allowed to obtain / retrieve the first model.

[0239] After receiving the analysis context transfer request from Target AnLF, Source AnLF in the embodiment of the present application requests an authorization token (i.e., the above-mentioned first token) from NRF for (or on behalf of behalf) Target AnLF. The request includes Target AnLF ID and MTLF ID. After NRF verifies that Target AnLF has the authority to obtain the first model in the analysis context from MTLF, it returns the authorization token (i.e., the above-mentioned first token) to Source AnLF. After obtaining the authorization token, Source AnLF requests information about the first model from MTLF for (or on behalf of behalf) Target AnLF. The request includes TargetAnLF ID and authorization token. After MTLF verifies that the authorization token is passed, it sends the information about the first model to Target AnLF. The above process helps to realize model authorization and model acquisition in the model delegation acquisition scenario, and authorizes the actual model consumer (i.e., Target AnLF) in the model delegation acquisition scenario, thereby improving the security of the model.

[0240] The above content elaborates on the method of the present application in detail. In order to facilitate better implementation of the above scheme of the embodiment of the present application, the embodiment of the present application also provides corresponding devices or equipment.

[0241] The embodiment of the present application can divide the functional modules of each network element of the present application according to the above-mentioned method example, and also divide the functional modules of the above-mentioned network elements according to the above-mentioned method example. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation. The communication device of the embodiment of the present application will be described in detail below with reference to Figures 5 to 7.

[0242] Referring to Figure 5 , Figure 5 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application. As shown in Figure 5 , the communication device includes a transceiver unit 10 and a processing unit 20. The transceiver unit 10 can implement corresponding communication functions, and the processing unit 20 is used for data processing. For example, the transceiver unit 10 can also be referred to as a communication interface or a communication unit.

[0243] In some embodiments of the present application, the communication device may be the first model training logic function network element shown above. That is, the communication device shown in Figure 5 may be used to execute the steps or functions performed by the first model training logic function network element in the above method embodiment. Exemplarily, the communication device may be the first model training logic function network element or a chip or functional module configured in the first model training logic function network element, etc., which is not limited in the embodiments of the present application. The transceiver unit 10 is used to execute the operations related to the transmission and reception of the first model training logic function network element in the above method embodiment, and the processing unit 20 is used to execute the operations related to the processing of the first model training logic function network element in the above method embodiment.

[0244] Exemplarily, the transceiver unit 10 is used to receive a first model acquisition request from the analysis logic function network element, the first model acquisition request including an analysis identifier, the first model acquisition request being used to request information of the model corresponding to the analysis identifier; the transceiver unit 10 is also used to send a token acquisition request to the network storage function NRF network element, the token acquisition request including the analysis identifier, the identifier of the analysis logic function network element, and the identifier of the second model training logic function network element, the second model training logic function network element being used to provide information of the model; the transceiver unit 10 is also used to receive a first token from the NRF network element, the first token including the identifier of the analysis logic function network element and the identifier of the second model training logic function network element; the transceiver unit 10 is also used to send a second model acquisition request to the second model training logic function network element, the second model acquisition request including the analysis identifier, the identifier of the analysis logic function network element, and the first token, the second model acquisition request being used to obtain information of the model for the analysis logic function network element, and the first token being used to verify the second model acquisition request.

[0245] Exemplarily, the processing unit 20 is used to generate various information sent by the transceiver unit 10, such as a token acquisition request and a second model acquisition request; the processing unit 20 is also used to control the transceiver unit 10 to send or receive various information.

[0246] Exemplarily, the transceiver unit 10 is also used to send a network element discovery request to the NRF network element, and the network element discovery request includes the analysis identifier, and the identifier of the analysis logic function network element and / or the vendor identifier of the analysis logic function network element; the transceiver unit 10 is also used to receive a network element discovery response from the NRF network element, and the network element discovery response includes a candidate network element list, and the network element list includes the second model training logic function network element, and the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element includes the vendor identifier of the analysis logic function network element.

[0247] In the embodiment of the present application, for specific descriptions of the first model acquisition request, token acquisition request, first token, second model acquisition request, network element discovery request, network element discovery response, and each network element, please refer to the method embodiment shown in Figure 3 above, and will not be described in detail here.

[0248] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiment of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG3 above, and no further details will be given here. In addition, the technical effects of the embodiment of the present application refer to the technical effects of the method embodiment shown in FIG3 above, and for the sake of brevity, no further details will be given here.

[0249] Reusing Figure 5, in some other embodiments of the present application, the communication device may be the source analysis logic function network element shown above. That is, the communication device shown in Figure 5 can be used to execute the steps or functions performed by the source analysis logic function network element in the above method embodiment. Exemplarily, the communication device may be a source analysis logic function network element or a chip or functional module configured in the source analysis logic function network element, etc., which is not limited in the embodiments of the present application. The transceiver unit 10 is used to execute the operations related to the transmission and reception of the source analysis logic function network element in the above method embodiment, and the processing unit 20 is used to execute the operations related to the processing of the source analysis logic function network element in the above method embodiment.

[0250] Exemplarily, the transceiver unit 10 is used to receive an analysis context transfer request from a target analysis logic function network element, the analysis context transfer request including an identification of the analysis context, and the analysis context transfer request is used to request the transfer of information of the first model in the analysis context; the transceiver unit 10 is also used to send a token acquisition request to the NRF network element, the token acquisition request including the analysis identification corresponding to the first model, the identification of the target analysis logic function network element, and the identification of the model training logic function network element, the model training logic function network element is used to provide information of the first model; the transceiver unit 10 is also used to receive a first token from the NRF network element, the first token including the identification of the target analysis logic function network element, and the identification of the model training logic function network element; the transceiver unit 10 is also used to send a model acquisition request to the model training logic function network element, the model acquisition request including the analysis identification, the identification of the target analysis logic function network element, and the first token, the model acquisition request is used to obtain information of the first model for the target analysis logic function network element, and the first token is used to verify the model acquisition request.

[0251] Exemplarily, the processing unit 20 is used to generate various information sent by the transceiver unit 10, such as a token acquisition request and a model acquisition request; the processing unit 20 is also used to control the transceiver unit 10 to send or receive various information.

[0252] Exemplarily, the processing unit 20 is further configured to determine whether the provider of the first model is located in a second operation indicator of the target analysis logic function network element. The second operation indicator is used to indicate the provider of the model that can be used by the target analysis logic function network element.

[0253] Exemplarily, the processing unit 20 is further configured to determine whether a supplier identifier of the target analysis logic function network element is included in the interoperability indicator of the information provider of the first model.

[0254] Exemplarily, the transceiver unit 10 is further used to send a network element registration request to the NRF network element, where the network element registration request includes a first operation indicator of the source analysis logic function network element, where the first operation indicator is used to indicate a provider of a model that can be used by the source analysis logic function network element.

[0255] In an embodiment of the present application, for specific descriptions of the context transfer request, token acquisition request, first token, model acquisition request, network element registration request, first operation indicator, second operation indicator, and each network element, please refer to the method embodiment shown in Figure 4 above, and will not be described in detail here.

[0256] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiments of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG4 above, which will not be described in detail here. Furthermore, the technical effects of the embodiments of the present application refer to the technical effects of the method embodiment shown in FIG4 above, which will not be described here for the sake of brevity.

[0257] Reusing Figure 5, in some other embodiments of the present application, the communication device may be the NRF network element shown above. That is, the communication device shown in Figure 5 can be used to execute the steps or functions performed by the NRF network element in the above method embodiment. Exemplarily, the communication device may be an NRF network element or a chip or functional module configured in the NRF network element, etc., which is not limited in the embodiments of the present application. The transceiver unit 10 is used to execute the operations related to NRF network element transceiver in the above method embodiment, and the processing unit 20 is used to execute the operations related to NRF network element processing in the above method embodiment.

[0258] Exemplarily, the transceiver unit 10 is used to receive a token acquisition request from a first model training logic function network element, where the token acquisition request includes an analysis identifier, an identifier of the analysis logic function network element, and an identifier of a second model training logic function network element, where the second model training logic function network element is used to provide information about a model corresponding to the analysis identifier; the processing unit 20 is used to verify whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element; the transceiver unit 10 is also used to send a first token to the first model training logic function network element when the vendor identifier is included in the interoperability indicator, where the first token includes the identifier of the analysis logic function network element and the identifier of the second model training logic function network element.

[0259] Exemplarily, the processing unit 20 is further configured to obtain an interoperability indicator corresponding to the analysis identifier of the second model training logical function network element from an NF configuration file corresponding to the identifier of the second model training logical function network element.

[0260] Exemplarily, the processing unit 20 is further configured to obtain the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

[0261] Exemplarily, the transceiver unit 10 is also used to receive a network element discovery request from the first model training logic function network element, the network element discovery request including the analysis identifier, and the identifier of the analysis logic function network element and / or the vendor identifier of the analysis logic function network element; the processing unit 20 is also used to obtain the vendor identifier of the analysis logic function network element based on the network element discovery request; the processing unit 20 is also used to determine a candidate network element list based on the stored NF configuration files of each network element, the network element list including the second model training logic function network element, the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element including the vendor identifier of the analysis logic function network element; the transceiver unit 10 is also used to send a network element discovery response to the first model training logic function network element, the network element discovery response including the candidate network element list.

[0262] Exemplarily, the processing unit 20 is further specifically configured to obtain the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

[0263] In the embodiment of the present application, for specific descriptions of the token acquisition request, the first token, the network element discovery request, the network element discovery response, and each network element, please refer to the method embodiment shown in Figure 3 above, and will not be described in detail here.

[0264] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiment of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG3 above, and no further details will be given here. In addition, the technical effects of the embodiment of the present application refer to the technical effects of the method embodiment shown in FIG3 above, and for the sake of brevity, no further details will be given here.

[0265] Exemplarily, the transceiver unit 10 is used to receive a token acquisition request from a source analysis logic function network element, where the token acquisition request includes an analysis identifier, an identifier of a target analysis logic function network element, and an identifier of a model training logic function network element, and the model training logic function network element is used to provide information of a first model corresponding to the analysis identifier; the processing unit 20 is used to verify whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the model training logic function network element; the transceiver unit 10 is also used to send a first token to the source analysis logic function network element when the vendor identifier is included in the interoperability indicator, where the first token includes the identifier of the target analysis logic function network element and the identifier of the model training logic function network element.

[0266] Exemplarily, the processing unit 20 is further used to obtain the interoperability indicator corresponding to the analysis identifier of the model training logical function network element from the NF configuration file corresponding to the identifier of the model training logical function network element.

[0267] Exemplarily, the processing unit 20 is further configured to obtain the vendor identifier of the target analysis logic function network element from the NF configuration file corresponding to the identifier of the target analysis logic function network element.

[0268] Exemplarily, the transceiver unit 10 is also used to receive a first network element registration request from the source analysis logic function network element, where the first network element registration request includes a first operation indicator of the source analysis logic function network element, where the first operation indicator is used to indicate a provider of a model that the source analysis logic function network element can use.

[0269] Exemplarily, the transceiver unit 10 is also used to receive a second network element registration request from the target analysis logic function network element, where the second network element registration request includes a second operation indicator of the target analysis logic function network element, where the second operation indicator is used to indicate the provider of the model that the target analysis logic function network element can use.

[0270] In the embodiment of the present application, for specific descriptions of the token acquisition request, the first token, the first network element registration request, the second network element registration request, and each network element, please refer to the method embodiment shown in Figure 4 above, and will not be described in detail here.

[0271] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiments of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG4 above, which will not be described in detail here. Furthermore, the technical effects of the embodiments of the present application refer to the technical effects of the method embodiment shown in FIG4 above, which will not be described here for the sake of brevity.

[0272] Reusing Figure 5, in some further embodiments of the present application, the communication device may be the second model training logic function network element shown above. That is, the communication device shown in Figure 5 can be used to execute the steps or functions performed by the second model training logic function network element in the above method embodiment. Exemplarily, the communication device may be the second model training logic function network element or a chip or functional module configured in the second model training logic function network element, etc., which is not limited in the embodiments of the present application. The transceiver unit 10 is used to execute the operations related to the transmission and reception of the second model training logic function network element in the above method embodiment, and the processing unit 20 is used to execute the operations related to the processing of the second model training logic function network element in the above method embodiment.

[0273] Exemplarily, the transceiver unit 10 is used to receive a second model acquisition request from the first model training logic function network element, the second model acquisition request including an analysis identifier, an identifier of the analysis logic function network element, and a first token, the second model acquisition request being used to obtain information of a model corresponding to the analysis identifier for the analysis logic function network element; the processing unit 20 is used to verify the first token; the transceiver unit 10 is also used to send a model notification message to the analysis logic function network element when the first token verification is passed, the model notification message including information of the model corresponding to the analysis identifier.

[0274] In the embodiment of the present application, for specific descriptions of the second model acquisition request, model notification message, and each network element, etc., please refer to the method embodiment shown in Figure 3 above, and will not be described in detail here.

[0275] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiment of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG3 above, and no further details will be given here. In addition, the technical effects of the embodiment of the present application refer to the technical effects of the method embodiment shown in FIG3 above, and for the sake of brevity, no further details will be given here.

[0276] Reusing Figure 5, in some further embodiments of the present application, the communication device may be the model training logic function network element shown above. That is, the communication device shown in Figure 5 can be used to execute the steps or functions performed by the model training logic function network element in the above method embodiment. Exemplarily, the communication device may be a model training logic function network element or a chip or functional module configured in the model training logic function network element, etc., which is not limited in the embodiments of the present application. The transceiver unit 10 is used to execute operations related to the transmission and reception of the model training logic function network element in the above method embodiment, and the processing unit 20 is used to execute operations related to the processing of the model training logic function network element in the above method embodiment.

[0277] Exemplarily, the transceiver unit 10 is used to receive a model acquisition request from a source analysis logic function network element, where the model acquisition request includes an analysis identifier, an identifier of a target analysis logic function network element, and a first token, where the model acquisition request is used to obtain information of a first model corresponding to the analysis identifier for the target analysis logic function network element; the processing unit 20 is used to verify the first token; the transceiver unit 10 is also used to send a model notification message to the target analysis logic function network element when the first token verification is successful, where the model notification message includes information of the above-mentioned first model.

[0278] In the embodiment of the present application, for specific descriptions of the model acquisition request, model notification message, and each network element, please refer to the method embodiment shown in Figure 4 above, and will not be described in detail here.

[0279] It is understood that the specific descriptions of the transceiver unit and the processing unit shown in the embodiments of the present application are merely examples. For the specific functions or execution steps of the transceiver unit and the processing unit, reference can be made to the method embodiment shown in FIG4 above, which will not be described in detail here. Furthermore, the technical effects of the embodiments of the present application refer to the technical effects of the method embodiment shown in FIG4 above, which will not be described here for the sake of brevity.

[0280] The above describes the communication device according to the embodiment of the present application. The following describes possible product forms of the communication device. It should be understood that any product having the functions of the communication device described in FIG. 5 falls within the scope of protection of the embodiment of the present application. It should also be understood that the following description is merely illustrative and does not limit the product forms of the communication device according to the embodiment of the present application to these examples.

[0281] In one possible implementation, in the communication device shown in FIG5 , the processing unit 20 may be one or more processors, and the transceiver unit 10 may be a transceiver. Alternatively, the transceiver unit 10 may be a transmitting unit and a receiving unit, wherein the transmitting unit may be a transmitter and the receiving unit may be a receiver, and the transmitting unit and receiving unit are integrated into a single device, such as a transceiver. In embodiments of the present application, the processor and transceiver may be coupled, and the connection method between the processor and transceiver is not limited in embodiments of the present application. During the execution of the above-described method, the process of sending information in the above-described method can be understood as the process of the processor outputting the above-described information. When outputting the above-described information, the processor outputs the above-described information to the transceiver for transmission by the transceiver. After being output by the processor, the above-described information may require further processing before reaching the transceiver. Similarly, the process of receiving information in the above-described method can be understood as the process of the processor receiving the above-described information. When the processor receives the input information, the transceiver receives the above-described information and inputs it into the processor. Furthermore, after the transceiver receives the above-described information, the above-described information may require further processing before being input into the processor.

[0282] Refer to Figure 6, which is another structural diagram of the communication device provided in an embodiment of the present application. As shown in Figure 6, the communication device provided in an embodiment of the present application can be used to implement the method described in any of the above method embodiments, and please refer to the description in the above method embodiments. The communication device can be the aforementioned first model training logic function network element, analysis logic function network element, second model training logic function network element, NRF network element, source analysis logic function network element, target analysis logic function network element, model training logic function network element, or a chip or circuit therein. Exemplarily, the communication device includes one or more processors 1001 and a transceiver 1002. The communication device may further include a memory 1003. In one implementation, the communication device also includes an input and output device (not shown in the figure).

[0283] Processor 1001 is primarily used to process communication protocols and communication data, control the entire communication device, execute software programs, and process software program data. Memory 1003 is primarily used to store software programs and data. Transceiver 1002 may include control circuitry and an antenna. The control circuitry is primarily used to convert baseband signals into radio frequency signals and process radio frequency signals. The antenna is primarily used to transmit and receive radio frequency signals in the form of electromagnetic waves. Input / output devices, such as a touch screen, display, and keyboard, are primarily used to receive user input and output data to the user.

[0284] When the communication device is powered on, the processor 1001 can read the software program in the memory 1003, interpret and execute the instructions of the software program, and process the data of the software program. When data needs to be sent wirelessly, the processor 1001 performs baseband processing on the data to be sent and outputs the baseband signal to the radio frequency circuit. The radio frequency circuit performs radio frequency processing on the baseband signal and then transmits the radio frequency signal to the outside in the form of electromagnetic waves through the antenna. When data is sent to the communication device, the radio frequency circuit receives the radio frequency signal through the antenna, converts the radio frequency signal into a baseband signal, and outputs the baseband signal to the processor 1001. The processor 1001 converts the baseband signal into data and processes the data.

[0285] In another implementation, the RF circuit and antenna may be provided independently of the processor performing baseband processing. For example, in a distributed scenario, the RF circuit and antenna may be remotely arranged independent of the communication device.

[0286] The processor 1001 , the transceiver 1002 , and the memory 1003 may be connected via a communication bus.

[0287] Exemplarily, when the communication device is used to execute the steps, methods, or functions of analyzing the execution of logical function network elements in the above-mentioned method embodiment 1 (such as Figure 3), the processor 1001 can be used to generate a first model acquisition request, and / or to execute other processes of the technology described in this document; the transceiver 1002 can be used to execute step S101 in Figure 3, and / or other processes of the technology described in this document.

[0288] Exemplarily, when the communication device is used to execute the steps, methods, or functions performed by the first model training logic function network element in the above-mentioned method embodiment 1 (such as Figure 3), the processor 1001 can be used to generate a token acquisition request and a second model acquisition request, and / or for executing other processes of the technology described in this document; the transceiver 1002 can be used to execute steps S102 and S105 in Figure 3, and / or for other processes of the technology described in this document.

[0289] Exemplarily, when the communication device is used to execute the steps, methods, or functions performed by the NRF network element in the first embodiment of the above method (such as Figure 3), the processor 1001 can be used to execute step S103 in Figure 3, and / or for executing other processes of the technology described herein; the transceiver 1002 can be used to execute step S104 in Figure 3, and / or for other processes of the technology described herein.

[0290] Exemplarily, when the communication device is used to execute the steps, methods, or functions performed by the second model training logic function network element in the first embodiment of the above-mentioned method (such as Figure 3), the processor 1001 can be used to execute step S106 in Figure 3, and / or used to execute other processes of the technology described in this document; the transceiver 1002 can be used to execute step S107 in Figure 3, and / or used for other processes of the technology described in this document.

[0291] Exemplarily, when the communication device is used to execute the steps, methods or functions performed by the target analysis logic function network element in the second embodiment of the above-mentioned method (such as Figure 4), the processor 1001 can be used to generate an analysis context transfer request, and / or to execute other processes of the technology described in this document; the transceiver 1002 can be used to execute step S201 in Figure 4, and / or other processes of the technology described in this document.

[0292] Exemplarily, when the communication device is used to execute the steps or methods or functions performed by the source analysis logic function network element in the second embodiment of the above method (such as Figure 4), the processor 1001 can be used to generate a token acquisition request and a model acquisition request, and / or for executing other processes of the technology described in this document; the transceiver 1002 can be used to execute steps S202 and S205 in Figure 4, and / or for other processes of the technology described in this document.

[0293] Exemplarily, when the communication device is used to execute the steps, methods, or functions performed by the NRF network element in the second embodiment of the above method (such as Figure 4), the processor 1001 can be used to execute step S203 in Figure 4, and / or for executing other processes of the technology described herein; the transceiver 1002 can be used to execute step S204 in Figure 4, and / or for other processes of the technology described herein.

[0294] Exemplarily, when the communication device is used to execute the steps, methods, or functions performed by the model training logic function network element in the second embodiment of the above method (such as Figure 4), the processor 1001 can be used to execute step S206 in Figure 4, and / or used to execute other processes of the technology described in this document; the transceiver 1002 can be used to execute step S207 in Figure 4, and / or used for other processes of the technology described in this document.

[0295] In any of the above implementations, the processor 1001 may include a transceiver for implementing receiving and transmitting functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing the receiving and transmitting functions may be separate or integrated. The transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or the transceiver circuit, interface, or interface circuit may be used for transmitting or delivering signals.

[0296] In any of the above implementations, the processor 1001 may store instructions, which may be computer programs. The computer programs, when executed on the processor 1001, may cause the communication device to perform the methods described in the above method embodiments. The computer programs may be embedded in the processor 1001, in which case the processor 1001 may be implemented by hardware.

[0297] In one implementation, the communication device may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiment. The processor and transceiver described in this application can be implemented in an integrated circuit (IC), an analog IC, a radio frequency integrated circuit (RFIC), a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (nMetal-oxide-semiconductor, NMOS), P-channel metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (bipolar junction transistor, BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0298] It is understood that the communication device shown in the embodiment of the present application may also have more components than those in Figure 6, and the embodiment of the present application is not limited to this. The methods performed by the processor and transceiver shown above are only examples. For the specific steps performed by the processor and transceiver, please refer to the description of the various method embodiments above.

[0299] In another possible implementation, in the communication device shown in Figure 5, the processing unit 20 can be one or more logic circuits, and the transceiver unit 10 can be an input / output interface, or a communication interface, or an interface circuit, or an interface, etc. Or the transceiver unit 10 can also be a sending unit and a receiving unit, the sending unit can be an output interface, and the receiving unit can be an input interface, and the sending unit and the receiving unit are integrated into one unit, such as an input / output interface. Referring to Figure 7, Figure 7 is another structural diagram of the communication device provided in an embodiment of the present application. As shown in Figure 7, the communication device shown in Figure 7 includes a logic circuit 901 and an interface 902. That is, the above-mentioned processing unit 20 can be implemented with a logic circuit 901, and the transceiver unit 10 can be implemented with an interface 902. Among them, the logic circuit 901 can be a chip, a processing circuit, an integrated circuit or a system on chip (SoC) chip, etc., and the interface 902 can be a communication interface, an input / output interface, a pin, etc. Exemplarily, Figure 7 is shown as an example of the above-mentioned communication device being a chip, and the chip includes a logic circuit 901 and an interface 902.

[0300] In the embodiment of the present application, the logic circuit and the interface may also be coupled to each other. The embodiment of the present application does not limit the specific connection method between the logic circuit and the interface.

[0301] Exemplarily, when the communication device is used to execute the method, function or step performed by the first model training logic function network element in the aforementioned method embodiment 1 (such as Figure 3), interface 902 is used to input a first model acquisition request; logic circuit 901 is used to generate a token acquisition request; interface 902 is also used to output the token acquisition request and input the first token; logic circuit 901 is also used to generate a second model acquisition request; interface 902 is also used to output the second model acquisition request.

[0302] Exemplarily, when the communication device is used to execute the method, function or step performed by the NRF network element in the aforementioned method embodiment 1 (such as Figure 3), interface 902 is used to input a token acquisition request; logic circuit 901 is used to verify whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator of the second model training logic function network element; interface 902 is used to output a first token.

[0303] In the embodiment of the present application, for specific descriptions of the first model acquisition request, token acquisition request, first token, second model acquisition request, etc., please refer to the method embodiment 1 shown above (as shown in Figure 3), and will not be described in detail here.

[0304] Exemplarily, when the communication device is used to execute the method, function or step performed by the source analysis logic function network element in the aforementioned method embodiment 2 (such as Figure 4), interface 902 is used to input an analysis context transfer request; logic circuit 901 is used to generate a token acquisition request; interface 902 is also used to output the token acquisition request and input a first token; logic circuit 901 is also used to generate a model acquisition request; interface 902 is also used to output the model acquisition request.

[0305] Exemplarily, when the communication device is used to execute the method, function or step performed by the NRF network element in the second embodiment of the aforementioned method (such as Figure 4), interface 902 is used to input a token acquisition request; logic circuit 901 is used to verify whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator of the model training logic function network element; interface 902 is used to output a first token.

[0306] In the embodiment of the present application, for specific descriptions of the context transfer request, token acquisition request, first token, model acquisition request, etc., please refer to the method embodiment 2 shown above (as shown in Figure 4), and will not be described in detail here.

[0307] An embodiment of the present application also provides a communication system, which includes at least two of a first model training logic function network element, a second model training logic function network element, an analysis logic function network element, and an NRF network element. The first model training logic function network element, the second model training logic function network element, the analysis logic function network element, and at least two of the NRF network element can be used to execute the method in the method embodiment shown in the aforementioned Figure 3.

[0308] An embodiment of the present application also provides a communication system, which includes at least two of a source analysis logic function network element, a target analysis logic function network element, a model training logic function network element, and an NRF network element. The source analysis logic function network element, the target analysis logic function network element, the model training logic function network element, and at least two of the NRF network elements can be used to execute the method in the method embodiment shown in the aforementioned Figure 4.

[0309] In addition, the present application also provides a computer program, which is used to implement the operations and / or processing performed by the first model training logic function network element in the method provided by the present application.

[0310] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the second model training logic function network element in the method provided by the present application.

[0311] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the analysis logic function network element in the method provided by the present application.

[0312] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the NRF network element in the method provided by the present application.

[0313] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the source analysis logic function network element in the method provided by the present application.

[0314] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the target analysis logic function network element in the method provided by the present application.

[0315] The present application also provides a computer program, which is used to implement the operations and / or processing performed by the model training logic function network element in the method provided by the present application.

[0316] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the first model training logic function network element in the method provided by the present application.

[0317] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the second model training logic function network element in the method provided by the present application.

[0318] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the analysis logic function network element in the method provided by the present application.

[0319] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the NRF network element in the method provided by the present application.

[0320] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the source analysis logic function network element in the method provided by the present application.

[0321] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the target analysis logic function network element in the method provided by the present application.

[0322] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, it enables the computer to execute the operations and / or processing performed by the model training logic function network element in the method provided by the present application.

[0323] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the first model training logic function network element in the method provided by the present application are executed.

[0324] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the second model training logic function network element in the method provided by the present application are executed.

[0325] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the analysis logic function network element in the method provided by the present application are executed.

[0326] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program is run on a computer, the operations and / or processing performed by the NRF network element in the method provided by the present application are executed.

[0327] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the source analysis logic function network element in the method provided by the present application are executed.

[0328] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the target analysis logic function network element in the method provided by the present application are executed.

[0329] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program runs on a computer, the operations and / or processing performed by the model training logic function network element in the method provided by the present application are executed.

[0330] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0331] The units described as separate components may or may not be physically separate, and 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 units may be selected according to actual needs to achieve the technical effects of the solutions provided in the embodiments of the present application.

[0332] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0333] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a readable storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.

[0334] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A model authorization method, characterized in that: include: The first model training logic function network element receives a first model acquisition request from the analysis logic function network element, where the first model acquisition request includes an analysis identifier, and the first model acquisition request is used to request information of a model corresponding to the analysis identifier; The first model training logic function network element sends a token acquisition request to the network storage function NRF network element, wherein the token acquisition request includes the analysis identifier, the identifier of the analysis logic function network element, and the identifier of the second model training logic function network element, and the second model training logic function network element is used to provide information about the model; The first model training logic function network element receives a first token from the NRF network element, where the first token includes an identifier of the analysis logic function network element and an identifier of the second model training logic function network element; The first model training logic function network element sends a second model acquisition request to the second model training logic function network element, the second model acquisition request includes the analysis identifier, the identifier of the analysis logic function network element, and the first token, the second model acquisition request is used to obtain information of the model for the analysis logic function network element, and the first token is used to verify the second model acquisition request.

2. The method according to claim 1, characterized in that The token acquisition request further includes one or more of the following: an identifier of the first model training logic function network element, a supplier identifier of the analysis logic function network element, or first indication information; The first indication information is used to instruct the first model training logic function network element to request a first token for the analysis logic function network element.

3. The method according to claim 1 or 2, characterized in that: The first token further includes one or more of the following: the analysis identifier, the identifier of the first model training logic function network element, the supplier identifier of the analysis logic function network element, or second indication information; The second indication information is used to indicate authorization for the analysis logic function network element to obtain information about the model from the second model training logic function network element.

4. The method according to any one of claims 1 to 3, characterized in that The second model acquisition request further includes one or more of the following: an identifier of the first model training logic function network element, an address of the analysis logic function network element, an identifier of a supplier of the analysis logic function network element, or third indication information; The third indication information is used to instruct the first model training logical function network element to obtain information about the model for the analysis logical function network element.

5. The method according to any one of claims 1 to 4, characterized in that Before the first model training logic function network element sends a token acquisition request to the network storage function NRF network element, the method further includes: The first model training logic function network element sends a network element discovery request to the NRF network element, where the network element discovery request includes the analysis identifier, and an identifier of the analysis logic function network element and / or a supplier identifier of the analysis logic function network element; The first model training logic function network element receives a network element discovery response from the NRF network element, the network element discovery response includes a candidate network element list, the network element list includes a second model training logic function network element, and the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element includes the vendor identifier of the analysis logic function network element.

6. A model authorization method, characterized in that: include: The network storage function NRF network element receives a token acquisition request from the first model training logic function network element, wherein the token acquisition request includes an analysis identifier, an identifier of the analysis logic function network element, and an identifier of a second model training logic function network element, wherein the second model training logic function network element is used to provide information of a model corresponding to the analysis identifier; The NRF network element verifies whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element; If the vendor identifier is included in the interoperability indicator, the NRF network element sends a first token to the first model training logic function network element, wherein the first token includes an identifier of the analysis logic function network element and an identifier of the second model training logic function network element.

7. The method according to claim 6, characterized in that Before the NRF network element verifies whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator of the second model training logic function network element, the method further includes: The NRF network element obtains the interoperability indicator corresponding to the analysis identifier of the second model training logical function network element from the network function NF configuration file corresponding to the identifier of the second model training logical function network element.

8. The method according to claim 6 or 7, characterized in that: Before the NRF network element verifies whether the vendor identifier of the analysis logic function network element is included in the interoperability indicator of the second model training logic function network element, the method further includes: The NRF network element obtains the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

9. The method according to any one of claims 6 to 8, characterized in that The token acquisition request further includes one or more of the following: an identifier of the first model training logic function network element, a supplier identifier of the analysis logic function network element, or first indication information; The first indication information is used to instruct the first model training logic function network element to request a first token for the analysis logic function network element.

10. The method according to any one of claims 6 to 9, characterized in that The first token further includes one or more of the following: the analysis identifier, the identifier of the first model training logic function network element, the supplier identifier of the analysis logic function network element, or second indication information; The second indication information is used to indicate authorization for the analysis logic function network element to obtain information about the model from the second model training logic function network element.

11. The method according to any one of claims 6 to 10, characterized in that Before the network storage function NRF network element receives the token acquisition request from the first model training logic function network element, the method further includes: The NRF network element receives a network element discovery request from the first model training logic function network element, where the network element discovery request includes the analysis identifier, and an identifier of the analysis logic function network element and / or a supplier identifier of the analysis logic function network element; The NRF network element obtains the vendor identifier of the analysis logic function network element based on the network element discovery request; The NRF network element determines a candidate network element list based on the stored NF configuration files of each network element, wherein the network element list includes a second model training logic function network element, and the interoperability indicator corresponding to the analysis identifier of the second model training logic function network element includes a vendor identifier of the analysis logic function network element; The NRF network element sends a network element discovery response to the first model training logic function network element, and the network element discovery response includes the candidate network element list.

12. The method according to claim 11, characterized in that The network element discovery request includes the identification of the analysis logic function network element; The NRF network element obtains the vendor identifier of the analysis logic function network element based on the network element discovery request, including: The NRF network element obtains the vendor identifier of the analysis logic function network element from the NF configuration file corresponding to the identifier of the analysis logic function network element.

13. A model authorization method, characterized in that: include: The source analysis logic function network element receives an analysis context transfer request from the target analysis logic function network element, wherein the analysis context transfer request includes an identification of an analysis context, and the analysis context transfer request is used to request to transfer information of a first model in the analysis context; The source analysis logic function network element sends a token acquisition request to the network storage function NRF network element, wherein the token acquisition request includes an analysis identifier corresponding to the first model, an identifier of the target analysis logic function network element, and an identifier of a model training logic function network element, and the model training logic function network element is used to provide information of the first model; The source analysis logic function network element receives a first token from the NRF network element, where the first token includes an identifier of the target analysis logic function network element and an identifier of the model training logic function network element; The source analysis logic function network element sends a model acquisition request to the model training logic function network element, and the model acquisition request includes the analysis identifier, the identifier of the target analysis logic function network element, and the first token. The model acquisition request is used to obtain information of the first model for the target analysis logic function network element, and the first token is used to verify the model acquisition request.

14. The method according to claim 13, characterized in that The analysis context transfer request also includes one or more of the following: a second operation indicator of the target analysis logic function network element, or a supplier identifier of the target analysis logic function network element; the second operation indicator is used to indicate the provider of the model that can be used by the target analysis logic function network element.

15. The method according to claim 14, characterized in that Before the source analysis logic function network element sends a token acquisition request to the network storage function NRF network element, the method further includes: The source analysis logic function network element determines whether a provider of the first model is located in a second operation indicator of the target analysis logic function network element.

16. The method according to claim 14, characterized in that Before the source analysis logic function network element sends a token acquisition request to the network storage function NRF network element, the method further includes: The source analysis logic function determines that a vendor identifier of the target analysis logic function network element is contained in an interoperability indicator of an information provider of the first model.

17. The method according to any one of claims 13 to 16, characterized in that Before the source analysis logic function network element receives the analysis context transfer request from the target analysis logic function network element, the method further includes: The source analysis logic function network element sends a network element registration request to the NRF network element, where the network element registration request includes a first operation indicator of the source analysis logic function network element, where the first operation indicator is used to indicate a provider of a model that can be used by the source analysis logic function network element.

18. The method according to any one of claims 13 to 17, characterized in that The token acquisition request further includes one or more of the following: an identifier of the source analysis logic function network element, a supplier identifier of the target analysis logic function network element, an identifier of the first model, or first indication information; The first indication information is used to instruct the source analysis logic function network element to request a first token for the target analysis logic function network element.

19. The method according to any one of claims 13 to 18, characterized in that The first token further includes one or more of the following: the analysis identifier, the identifier of the source analysis logic function network element, the supplier identifier of the target analysis logic function network element, the identifier of the first model, or the second indication information; The second indication information is used to indicate authorization for the target analysis logic function network element to obtain information of the first model from the model training logic function network element.

20. The method according to any one of claims 13 to 19, characterized in that The model acquisition request further includes one or more of the following: an identifier of the source analysis logic function network element, an address of the target analysis logic function network element, an identifier of the first model, an identifier of a supplier of the target analysis logic function network element, or third indication information; The third indication information is used to instruct the source analysis logic function network element to obtain information of the first model for the target analysis logic function network element.

21. A model authorization method, characterized in that: include: The network storage function NRF network element receives a token acquisition request from a source analysis logic function network element, wherein the token acquisition request includes an analysis identifier, an identifier of a target analysis logic function network element, and an identifier of a model training logic function network element, and the model training logic function network element is used to provide information of a first model corresponding to the analysis identifier; The NRF network element verifies whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator corresponding to the analysis identifier of the model training logic function network element; If the vendor identifier is included in the interoperability indicator, the NRF network element sends a first token to the source analysis logic function network element, wherein the first token includes an identifier of the target analysis logic function network element and an identifier of the model training logic function network element.

22. The method according to claim 21, characterized in that Before the NRF network element verifies whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator of the model training logic function network element, the method further includes: The NRF network element obtains the interoperability indicator corresponding to the analysis identifier of the model training logical function network element from the network function NF configuration file corresponding to the identifier of the model training logical function network element.

23. The method according to claim 21 or 22, characterized in that Before the NRF network element verifies whether the vendor identifier of the target analysis logic function network element is included in the interoperability indicator of the model training logic function network element, the method further includes: The NRF network element obtains the supplier identifier of the target analysis logic function network element from the NF configuration file corresponding to the identifier of the target analysis logic function network element.

24. The method according to any one of claims 21 to 23, characterized in that The token acquisition request further includes one or more of the following: an identifier of the source analysis logic function network element, a supplier identifier of the target analysis logic function network element, an identifier of the first model, or first indication information; The first indication information is used to instruct the source analysis logic function network element to request a first token for the target analysis logic function network element.

25. The method according to any one of claims 21 to 24, characterized in that The first token further includes one or more of the following: the analysis identifier, the identifier of the source analysis logic function network element, the supplier identifier of the target analysis logic function network element, the identifier of the first model, or the second indication information; The second indication information is used to indicate authorization for the target analysis logic function network element to obtain information of the first model from the model training logic function network element.

26. The method according to any one of claims 21 to 25, characterized in that Before the network storage function NRF network element receives the token acquisition request from the source analysis logic function network element, the method further includes: The NRF network element receives a first network element registration request from a source analysis logic function network element, where the first network element registration request includes a first operation indicator of the source analysis logic function network element, where the first operation indicator is used to indicate a provider of a model that the source analysis logic function network element can use.

27. The method according to any one of claims 21 to 26, characterized in that Before the network storage function NRF network element receives the token acquisition request from the source analysis logic function network element, the method further includes: The NRF network element receives a second network element registration request from a target analysis logic function network element, where the second network element registration request includes a second operation indicator of the target analysis logic function network element, where the second operation indicator is used to indicate a provider of a model that the target analysis logic function network element can use.

28. A communication device, characterized in that: Comprising units or modules for executing the method according to any one of claims 1 to 27.

29. A readable storage medium, characterized in that: The readable storage medium stores program instructions, and when the program instructions are executed on a communication device, the communication device executes the method according to any one of claims 1 to 27.

30. A communication system, characterized in that: include: A first model training logic function network element for executing the method described in any one of claims 1 to 5 and an NRF network element for executing the method described in any one of claims 6 to 12; or the communication system includes: a source analysis logic function network element for executing the method described in any one of claims 13 to 20 and an NRF network element for executing the method described in any one of claims 21 to 27.

31. A program product, characterized in that The method comprises instructions which, when executed, cause the method according to any one of claims 1 to 27 to be performed.

32. A communication device, characterized in that: The device comprises a processor, wherein the processor is used to read and execute a program stored in a memory to perform the method according to any one of claims 1 to 27.

Images